HK1066937B - Method and apparatus for processing data in a multiple-input multiple-output (mimo) communication system utilizing channel state information - Google Patents

Description

Method and apparatus for processing data with channel state information in a multiple-input multiple-output (MIMO) communication system

Technical Field

The present invention relates generally to data communications, and more particularly to a novel and improved method and apparatus for processing data with channel state information in a multiple-input multiple-output (MIMO) communication system to provide improved system performance.

Background

Wireless communication systems are widely deployed to provide various types of communication such as voice, data, and so on. These systems may be based on Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Orthogonal Frequency Division Multiplexing (OFDM), or some other multiplexing technique. OFDM systems may provide high performance for certain channel environments.

In terrestrial communication systems (e.g., cellular systems, broadcast systems, multi-channel multipoint distribution systems (MMDS), and others), the RF modulated signal from a transmitter unit may reach a receiver unit via multiple transmission paths. The characteristics of the transmission path typically vary over time due to a number of factors such as fading and multipath.

To provide diversity against the effects of degraded paths and improve performance, multiple transmit and receive antennas may be used for data transmission. If the transmission paths between the transmitting and receiving antennas are linearly independent (i.e. the transmission on one path does not form a linear combination of the transmissions on the other paths), which is generally true to some extent, the likelihood of correctly receiving a data transmission increases with the number of antennas. In general, as the number of transmit and receive antennas increases, diversity also increases and performance improves.

Multiple Input Multiple Output (MIMO) communication systems using multiple roots (N)_T) Transmitting antenna and multiple antennas (N)_R) The receive antennas are used for data transmission. From N_TRoot transmitting antenna and N_RThe MIMO channel formed by the root receiving antenna may be decomposed into N_CA separate channel, wherein N_C≤min{N_T，N_R}。N_CEach of the individual channels is also referred to as a spatial subchannel of the MIMO channel and corresponds to a dimension. MIMO systems may provide improved performance (e.g., increased transmission capacity) if the additional dimensionalities created by the multiple transmit and receive antennas are utilized.

There is therefore a need in the art for techniques to process data at both the transmitter and receiver units to take advantage of the additional dimensionality created by the MIMO system to provide improved system performance.

Disclosure of Invention

Aspects of the present invention provide techniques for processing received signals at a receiving unit in a multiple-input multiple-output (MIMO) system to recover transmitted signals and adjusting data processing at a transmitter unit based on estimated characteristics of a MIMO channel used for data transmission. In one aspect, the received signal is processed using a "successive cancellation" receiver processing technique (described below). On the other hand, channel characteristics are estimated and reported back to the transmitter system and used to adjust (i.e., adapt) the processing (e.g., coding, modulation, etc.) of the data prior to transmission. High performance can be achieved for MIMO systems by using a combination of successive cancellation receiver processing techniques and adaptive transmitter processing techniques.

Certain embodiments of the present invention provide a method of transmitting data from a transmitter unit to a receiver unit in a MIMO communication system. According to the method, at a receiver unit, a number of signals are initially received via a number of receive antennas, each received signal comprising a combination of one or more signals transmitted from a transmitter unit. The received signal is processed in accordance with successive cancellation receiver processing techniques to provide a number of decoded data streams, which are estimates of the data streams transmitted from the transmitter unit. Channel State Information (CSI) indicative of characteristics of the MIMO channel used to transmit the data streams is also determined and transmitted back to the transmitter unit. At the transmitter unit, each data stream is adaptively processed prior to transmission over the MIMO channel in accordance with the received CSI.

Successive cancellation receiver processing schemes typically perform multiple iterations to provide decoded data streams, one iteration for each decoded data stream. For each iteration, a number of input signals for the iteration are processed in accordance with a particular linear or non-linear processing scheme to provide one or more symbol streams. One of the symbol streams is then selected and processed to provide a decoded data stream. A plurality of modified signals are also derived from the input channels, the modified signals having components resulting from substantially removing (i.e., canceling) the decoded data streams. The input signal of the first iteration is the received signal and the input signal of each subsequent iteration is the modified signal from the previous iteration.

Various linear and non-linear processing schemes may be used to process the input signal. For non-dispersive channels (i.e., with smooth fading), it is possible to use a Channel Correlation Matrix Inversion (CCMI) technique, a Minimum Mean Square Error (MMSE) technique, or some other technique. While for time-dispersive channels (i.e., fading with frequency selectivity), an MMSE linear equalizer (MMSE-LE), a Decision Feedback Equalizer (DFE), a Maximum Likelihood Sequence Estimator (MLSE), or some other technique may be used.

The available CSI may include, for example, a signal-to-noise-plus-interference ratio (SNR) for each transmission channel to be used for data transmission. At the transmitter unit, the data for each transmission channel may be encoded according to the CSI associated with that channel, and the encoded data for each transmission channel may be further modulated according to a modulation scheme selected based on the CSI.

The present invention also provides methods, systems, and apparatus that enable various aspects, embodiments, and features of the invention, as described in more detail below.

Drawings

The features, nature, and advantages of the present invention will become more apparent from the detailed description set forth below when taken in conjunction with the drawings in which like reference characters identify correspondingly throughout and wherein:

FIG. 1 is a diagram of a multiple-input multiple-output (MIMO) communication system in which aspects and embodiments of the present invention may be implemented;

fig. 2 is a block diagram of an embodiment of a MIMO transmitter system capable of processing data for transmission in accordance with available CSI;

FIG. 3 is a block diagram of an embodiment of a MIMO transmitter system using Orthogonal Frequency Diversity Modulation (OFDM);

FIG. 4 is a flowchart illustrating a process for processing N_RReceiving signal to recover N_TA flow diagram of a successive cancellation receiver processing technique for each transmitted signal;

FIG. 5 is a block diagram of a receiver system capable of implementing aspects and embodiments of the invention;

FIGS. 6A, 6B, and 6C are block diagrams of three channel MIMO/data processors, which can implement the CCMI, MMSE, and DFE techniques, respectively;

FIG. 7 is a block diagram of an embodiment of a Receive (RX) data processor;

fig. 8 is a block diagram of an interference canceller; and

fig. 9A, 9B, and 9C are graphs illustrating the performance of various receiver and transmitter processing schemes.

Detailed Description

Fig. 1 is a diagram of a multiple-input multiple-output (MIMO) communication system 100 in which various aspects and embodiments of the invention may be implemented. The system 100 includes a first system 110 in communication with a second system 150. System 100 may be used to use a combination of antenna, frequency, and time diversity (described below) to increase spectral efficiency, improve performance, and increase flexibility. In an aspect, system 150 can be configured to determine characteristics of a MIMO channel and report Channel State Information (CSI) indicative of the channel characteristics determined in this manner back to system 110, and system 110 can be configured to adjust the processing (e.g., coding and modulation) of data prior to transmission based on the available CSI. On the other hand, the system 150 may be used to process data transmissions from the system 110 in a manner that provides high performance, as will be described in greater detail below.

At system 110, a data source 112 provides data (i.e., information bits) to a Transmit (TX) data processor 114, which encodes the data according to a particular coding scheme, interleaves (i.e., reorders) the encoded data according to a particular interleaving scheme, and maps the interleaved bits to modulation symbols for one or more transmission channels over which the data is transmitted. The interleaving provides time diversity for the coded bits, allows data to be transmitted according to an average signal-to-noise-and-interference (SNR) ratio of a transmission channel used for data transmission, combat fading, and further removes correlation between the coded bits used to form each modulation symbol. Interleaving may further provide frequency diversity if the coded bits are transmitted on multiple frequency subchannels. In one aspect, as shown in fig. 1, encoding, interleaving, and symbol mapping (or a combination thereof) are performed according to the CSI available to the system 110.

The encoding, interleaving, and symbol mapping at the transmitter system 110 may be implemented according to a number of schemes. One particular approach is described in U.S. patent application serial No. 09/776075 entitled "CODING schedule FOR A WIRELESS COMMUNICATION SYSTEM," filed on 2/1/2001, assigned to the assignee of the present invention and incorporated herein by reference. Another scheme will be further described below.

MIMO system 100 employs multiple antennas at both the transmit and receive ends of the communication link. These transmit and receive antennas may be used to provide various forms of spatial diversity (i.e., antenna diversity), including transmit diversity and receive diversity. Spatial diversity is characterized by the use of multiple transmit antennas and one or more receive antennas. Transmit diversity is characterized by the transmission of data over multiple transmit antennas. In general, additional processing is performed on the data transmitted from the transmit antennas to achieve the desired diversity. For example, data transmitted from different transmit antennas may be delayed or rearranged in time, encoded and interleaved over available transmit antennas, and so on. Receive diversity is characterized by receiving the transmitted signals on multiple receive antennas, and achieves diversity by simply receiving the signals via different signal paths.

System 100 may be used for a number of different communication modes, each using antenna, frequency or time diversity, or a combination thereof. The communication modes may include, for example, "diversity" communication modes and "MIMO" communication modes. The diversity communication mode uses diversity to improve the reliability of the communication link. In a common application of diversity communication mode, also referred to as "pure" diversity communication mode, data is transmitted from all available transmit antennas to a trusted receiver system. Pure diversity communication modes may be used in situations where data rate requirements are low or where SNR is low or both. The MIMO communication mode uses antenna diversity (i.e., multiple transmit antennas and multiple receive antennas) at both ends of the communication link and is generally used to both improve reliability and increase capacity of the communication link. The MIMO communication mode may also use frequency and/or time diversity in combination with antenna diversity.

System 100 may utilize Orthogonal Frequency Division Modulation (OFDM), which effectively partitions the operating frequency band into a number (N)_L) A frequency subchannel (i.e., a frequency segment). At each time slot (i.e., a particular time interval that may depend on the frequency subchannel bandwidth), the modulation symbols may be at N_LA strip frequency subchannel is transmitted on each strip.

System 100 may be used to transmit data over a number of transmission channels. As described above, a MIMO channel may be decomposed into N_CA separate channel, wherein N_C≤min{N_T，N_R}。N_CEach of the individual channels is also referred to as a spatial subchannel of the MIMO channel. For MIMO systems that do not use OFDM, there is typically only one frequency subchannel and each spatial subchannel may be referred to as a "transmission channel". For a MIMO system using OFDM, each spatial subchannel of each frequency subchannel may be referred to as a transmission channel.

MIMO systems may provide improved performance if additional dimensions are used that may be produced by multiple transmit and receive antennas. While this does not necessarily require that the CSI at the transmitter be known, system efficiency and performance may be improved when the transmitter is equipped with CSI, which is indicative of the transmission characteristics from the transmit antennas to the receive antennas. The processing of data prior to transmission at the transmitter depends on whether CSI is available.

The available CSI may include, for example, a signal-to-noise-plus-interference ratio (SNR) for each transmission channel (i.e., an SNR for each spatial subchannel of a MIMO system without OFDM, or an SNR for each spatial subchannel for each frequency subchannel of a MIMO system with OFDM). In this case, it is possible to adaptively process the data at the transmitter of each transmission channel according to the SNR of the channel (e.g., by selecting an appropriate coding and modulation scheme).

For MIMO systems that do not use OFDM, TX MIMO processor 120 receives and demultiplexes the modulation symbols from TX data processor 114 and provides a stream of modulation symbols, one modulation symbol per slot, for each transmit antenna. For a MIMO system using OFDM, TX MIMO processor 120 provides a stream of vectors of modulation symbols for each transmit antenna, each vector comprising N for a given time slot_LN of the frequency sub-channels_LAnd a modulation symbol. Each stream of modulation symbols or vectors of modulation symbols is received and modulated by a respective Modulator (MOD)122 and transmitted via an associated antenna 124.

At the receiver system 150, a number of receive antennas 152 receive the transmitted signals and provide received signals to corresponding demodulators (DEMODs) 154. Each demodulator 154 performs the inverse of the processing performed at the modulator 122. The modulation symbols from all demodulators 154 are provided to a Receive (RX) MIMO/data processor 156 and processed to recover the transmitted data streams. An RX MIMO/data processor 156 performs inverse processing to that performed by TX data processor 114 and TX MIMO processor 120 and provides decoded data to a data sink 160. The processing by the receiver system 150 is described in further detail below.

The spatial subchannels of a MIMO system (or more specifically, the transmission channels within a MIMO system with or without OFDM) typically experience different link conditions (e.g., different fading and multipath effects) and may achieve different SNRs. Thus, the capacity of a transmission channel may vary from channel to channel. The capacity may be quantified in terms of the information bit rate (i.e., the number of information bits PER modulation symbol) that may be transmitted on each transmission channel for a particular level of performance (e.g., a particular Bit Error Rate (BER) or Packet Error Rate (PER)). Furthermore, link conditions typically change over time. Therefore, the information bit rate supported by the transmission channel also changes with time. To more fully utilize the capacity of the transmission channel, CSI describing the link conditions may be determined (typically at the receiver unit) and provided to the transmitter unit so that the processing may be adjusted (or adapted) accordingly. The CSI may comprise any type of information that characterizes the communication link and may be reported through various mechanisms as described below. For simplicity, various aspects and embodiments of the present invention are described below, wherein the CSI comprises SNR. Techniques for determining and using CSI to provide improved system performance are described below.

MIMO transmitter system with CSI processing

Fig. 2 is a block diagram of an embodiment of a MIMO transmitter system 110a that does not use OFDM but can adjust its processing based on CSI available to the transmitter system (i.e., as reported by receiver system 150). Transmitter system 110a is an embodiment of the transmitter portion of system 110 in fig. 1. System 110a includes (1) a TX data processor 114a that receives and processes information bits to provide modulation symbols, and (2) a pair N_TA TX MIMO processor 120a that demultiplexes the modulation symbols for the root transmit antennas.

In the particular embodiment shown in fig. 2, TX data processor 114a includes a demultiplexer 208 coupled to a number of channel data processors 210, one for N_CEach of the transmission channels. Demultiplexer 208 receives and demultiplexes the aggregate information bits into a plurality (up to N)_C) And (4) data flow. One data stream for each transport channel to be used for data transmission, each data stream being provided to a respective channel data processor 210.

In the embodiment shown in fig. 2, each channel data processor 210 includes an encoder 212, a channel interleaver 214, and a symbol mapping element 216. Encoder 212 receives and encodes information bits within a received data stream according to a particular coding scheme to provide coded bits. Channel interleaver 214 interleaves the coded bits according to a particular interleaving scheme to provide diversity. And symbol mapping element 216 maps the interleaved bits into modulation symbols for the transmission channel used to transmit the data stream.

Pilot data (e.g., a known pattern of data) may also be encoded and multiplexed with the processed information bits. The processed pilot data may be transmitted (e.g., in a Time Division Multiplexed (TDM) fashion) over all or a subset of the transmission channels used to transmit the information bits. The pilot data may be used at the receiver to perform channel estimation, as described below.

As shown in fig. 2, it is possible to adjust the data coding, interleaving, and modulation (or a combination thereof) based on the available CSI (e.g., as reported by receiver system 150). In one coding and modulation scheme, adaptive coding is achieved by using a fixed base code (e.g., a Turbo code of rate 1/3) and adjusting puncturing to achieve a desired coding rate, as supported by the SNR of the transmission channel used to transmit the data. For this scheme, puncturing may be performed after channel interleaving. In another coding and modulation scheme, different coding schemes may be used for the reported CSI. For example, each data stream may be encoded with a separate encoding. According to this scheme, it is possible to use a "successive cancellation" receiver processing scheme to detect and decode the data stream to derive a more reliable estimate of the transmitted data stream, as described in detail below.

Symbol mapping element 216 may be designed to group the interleaved sets of bits to form non-binary symbols and map each non-binary symbol to a point within a signal constellation corresponding to a particular modulation scheme (e.g., QPSK, M-PSK, M-QAM, or some other scheme) selected for the transmission channel. Each mapped signal point corresponds to a modulation symbol.

The number of information bits that may be transmitted for each modulation symbol for a particular performance level (e.g., one percent PER) depends on the SNR of the transmission channel. Thus, it is possible to select the coding and modulation scheme for each transmission channel based on the available CSI. It is also possible to adjust the channel interleaving according to the available CSI.

Table 1 lists various combinations of code rates and modulation schemes that may be used for many SNR ranges. The supported bit rate for each transport channel may be obtained using any of a number of possible combinations of coding rates and modulation schemes. For example, one information bit per modulation symbol may be obtained by (1)1/2 code rate and QPSK modulation (2)1/3 code rate and 8-PSK modulation, (3)1/4 code rate and 16-QAM modulation, or some other combination of code rate and modulation scheme. In Table 1, QPSK, 16-QAM and 64-QAM are used for the listed SNR ranges. Other modulation schemes such as 8-PSK, 32-QAM, 128-QAM, etc. may also be used and are within the scope of the present invention.

TABLE 1

SNR Range	Information bits/symbols #	Modulation code element	# of coded bits/symbols	Coding rate
					1.5-4.4	1	QPSK	2	1/2
4.4-6.4	1.5	QPSK	2	3/4
					6.4-8.35	2	16-QAM	4	1/2
8.35-10.4	2.5	16-QAM	4	5/8
					10.4-12.3	3	16-QAM	4	3/4
12.3-14.15	3.5	64-QAM	6	7/12
					14.15-15.55	4	64-QAM	6	2/3
15.55-17.35	4.5	64-QAM	6	3/4
					＞17.35	5	64-QAM	6	5/6

The modulation symbols from TX data processor 114a are provided to a TX MIMO processor 120a, which is an embodiment of TX MIMO processor 120 in fig. 1. Within TX MIMO processor 120a, a demultiplexer 222 is coupled from N_CMultiple channel data processor 210 receives (up to) N_CA stream of modulation symbols and demultiplexing the received modulation symbols into a plurality (N)_T) The stream of modulation symbols, one for each antenna used to transmit the modulation symbols. Each modulation symbol stream is provided to a respective modulator 122. Each modulator 122 converts the modulation symbols into an analog signal and further amplifies, filters, quadrature modulates, and upconverts the signal to generate a modulated signal suitable for transmission over a wireless link.

MIMO transmitter system with OFDM

Fig. 3 is a block diagram of an embodiment of a MIMO transmitter system 110c that uses OFDM and can adjust its processing based on the available CSI. Within TX data processor 114c, the information bits to be transmitted are demultiplexed into a number (up to N)_L) Frequency subchannel data streams, one stream for each frequency subchannel to be used for data transmission. Each frequencyThe sub-channel data streams are provided to respective frequency sub-channel data processors 310.

Each data processor 310 processes data for a respective frequency subchannel of the OFDM system. Each data processor 310 may be implemented similar to the TX data processor shown in fig. 2. For this design, design processor 310 includes a demultiplexer that demultiplexes the frequency subchannel data stream into a number of data substreams, one substream for each spatial subchannel used by the frequency subchannel. Each data stream is then encoded, interleaved, and symbol mapped by a corresponding channel data processor to generate modulation symbols for that particular transmission channel (i.e., the spatial subchannels of the frequency subchannel). The coding and modulation for each transport channel may be adjusted based on the available CSI (e.g., as reported by the receiver system). Thus, each frequency subchannel data processor 310 is (up to) N_CStripe space subchannel provides (up to) N_CAnd a modulation symbol.

For a MIMO system using OFDM, modulation symbols may be transmitted from multiple transmit antennas on multiple frequency subchannels. Within MIMO processor 120c, N from each data processor 310_CThe streams of modulation symbols are provided to respective channel MIMO processors 322, which process the received modulation symbols in accordance with the available CSI.

Each channel MIMO processor 322 assigns N per slot_CDemultiplexing of a modulation symbol into N_TN of root transmitting antenna_TAnd a modulation symbol. Each combiner 324 receives up to N_TThe modulation symbols for the frequency subchannels, combine the symbols for each slot into a vector of modulation symbols V, and provide the vector of modulation symbols to the next processing stage (i.e., the corresponding modulator 122).

Thus, MIMO processor 120c receives and processes the modulation symbols to provide N_TVector of modulation symbols V₁To V_NtThere is one vector of modulation symbols for each transmit antenna. Each modulation symbol vector V covers a single time slot, and each element of the modulation symbol vector V is associated with a specific frequencyThe rate subchannel is correlated, and the particular frequency subchannel has a unique subcarrier on which the modulation symbol is conveyed.

Fig. 3 also shows an embodiment of the modulator 122 for OFDM. Vector of modulation symbols V from MIMO processor₁To V_NtAre provided to modulators 122a through 122t, respectively. In the embodiment shown in fig. 3, each modulator 122 includes an Inverse Fast Fourier Transform (IFFT)320, a cyclic prefix generator 322, and an upconverter 324.

IFFT 320 converts each received vector of modulation symbols into its time-domain representation (referred to as an OFDM symbol) using an IFFT. The IFFT 320 can be designed to perform an IFFT on any number of frequency subchannels (e.g., 8, 16, 32, etc.). In one embodiment, for each vector of modulation symbols that is translated into an OFDM symbol, cyclic prefix generator 322 repeats a portion of the time-domain representation of the OFDM symbol to form a "transmission symbol" for the particular transmit antenna. The cyclic prefix ensures that the transmission symbol retains its orthogonality in the presence of multipath delay spread, thereby improving performance against deteriorating multipath effects. The implementation of the IFFT 320 and cyclic prefix generator 322 is known in the art and will not be described in detail herein.

The time domain representation (i.e., the transmission symbols for each antenna) from each cyclic prefix generator 322 is then processed (e.g., converted to analog, modulated, amplified, and filtered) by an upconverter 324 to generate a modulated signal, which is then transmitted from the corresponding antenna 124.

OFDM modulation is described in detail in the following paper entitled "Multicarrier modulation for Data Transmission: an Idea while Time Has Come, author John A.C. Bingham, published in the IEEE journal of communications at 1990, 5 months, and incorporated herein by reference.

Fig. 2 and 3 show two designs of a MIMO transmitter capable of implementing aspects of the present invention. Other transmitter designs are possible and within the scope of the invention. Some such transmitter designs are detailed in the following applications: U.S. patent application Ser. No. 09/532492 entitled "HIGH EFFICIENCY, HIGH PERFORMANCE COMMUNICATIONS SYSTEM EMPLOYING MULTIPLE-CARRIER MODULATION" filed on 3/22/2000; the above-mentioned U.S. patent application serial No. 09/776075; AND U.S. patent application Ser. No. 09/826481, "METHOD AND APPATUS FOR UTILIZING CHANNEL STATE INFORMATION INA WIRELESS COMMUNICATION SYSTEM", filed 3, 23/3/2001; all of which are assigned to the assignee of the present invention and incorporated herein by reference. These applications further describe MIMO processing and CSI processing.

In general, transmitter system 110 encodes and modulates data for each transmission channel based on information describing the transmission performance of the channel. The information form is typically CSI. The CSI for the transmission channel used for data transmission is typically determined at the receiver system and reported back to the transmitter system, which then uses this information to adjust the coding and modulation accordingly. The techniques described herein may be applied to any number of parallel transmission channels supported by MIMO, OFDM, or any other communication scheme capable of supporting multiple parallel transmission channels (e.g., a CDMA scheme).

MIMO receiver system

Aspects of the present invention provide techniques for (1) processing a received signal at a receiver system within a MIMO system according to a successive cancellation receiver processing scheme to recover transmitted data, and (2) adjusting data processing at the transmitter system according to estimated characteristics of the MIMO channel. In one aspect, the received signal is processed using a successive cancellation receiver processing technique (described below). On the other hand, channel characteristics are estimated at the receiver system and reported back to the transmitter system, which uses this information to adjust (i.e., adapt) the data processing (e.g., coding, modulation, etc.). High performance can be achieved for MIMO systems by a combination of adaptive successive cancellation receiver processing techniques and adaptive transmitter processing techniques.

FIG. 4 is a flow diagram illustrating a successive cancellation receiver processing technique for processing N_RReceiving signal to recover N_THair setting deviceAnd emitting the signal. For simplicity, the following description of FIG. 4 assumes that (1) the number of transmission channels (i.e., spatial subchannels of a MIMO system that does not use OFDM) is equal to the number of transmit antennas (i.e., N)_C＝N_T) And (2) transmitting an independent data stream from each transmit antenna.

First, at step 412, the receiver system pair N_RThe signals are subjected to linear and/or nonlinear spatial processing in an attempt to separate the multiple transmitted signals included in the received signal. If the MIMO channel is "non-dispersive" (i.e., frequency non-selective or smooth fading), linear spatial processing may be performed on the received signal. If the MIMO channel is "time-dispersive" (i.e., frequency-selective fading), additional linear or non-linear time processing (i.e., equalization) of the received signal may be necessary or desirable. Spatial processing may be based on a Channel Correlation Matrix Inversion (CCMI) technique, a minimum mean square error technique (MMSE), or some other technique. The space-time processing may be based on an MMSE linear equalizer (MMSE-LE), a Decision Feedback Equalizer (DFE), a Maximum Likelihood Sequence Estimator (MLSE), or some other technique. Some such spatial and space-time processing techniques are described in further detail below. The amount of signal separation that can be achieved depends on the amount of correlation between the transmitted signals, with larger signal separations being possible if the transmitted signals are less correlated.

The initial spatial or space-time processing step provides N_TA "post-processed" signal, which is N_TAn estimate of the transmitted signal. N is then determined at step 414_TSNR of the post-processed signal. The SNR is detailed further below. In one embodiment, the SNRs are arranged in order of highest to lowest SNR, and the post-processed signal with the highest SNR is selected and further processed (i.e., "detected") in step 416 to obtain a decoded data stream. Detection typically includes demodulation, deinterleaving, and decoding of the selected post-processed signal. The decoded data stream is an estimate of the data stream that was transmitted on the recovered transmit signal in this iteration. The particular post-processed signal to be detected may also be selected according to some other scheme (e.g., the particular signal may be specifically identified by the transmitter system).

In step 418, it is determined whether all transmitted signals have been recovered. If all transmitted signals have been recovered, the receiver processing terminates. Otherwise, the interference generated by the decoded data stream is removed from the received signal, thereby generating a "modified" signal for the next iteration for recovering the next transmitted signal.

In step 420, the decoded data streams are used to form an estimate of the interference given by the transmitted signal corresponding to the decoded data streams on each received signal. The interference may be estimated as follows: the decoded data stream is first encoded, the re-encoded data is interleaved, and the interleaved data is symbol mapped (using the same coding, interleaving, and modulation schemes used for the data stream at the transmitter) to obtain a "re-modulated" symbol stream. The remodulated symbol stream is from N to the front_TTransmitted from the root transmitting antenna and transmitted by N_REstimates of the stream of modulation symbols received by the receive antennas are received. Thus, the remodulated symbol stream and the estimated channel response vectorh _iInner N_REach of the elements is convolved to derive N resulting from the recovered transmit signal_RAn interfering signal. Vector quantityh _iIs (N)_R×N_T) A specific column of a channel coefficient matrix H, the matrix H representing N at a specific time_TRoot transmitting antenna and N_RAn estimate of the MIMO channel response for the root receive antenna, and possibly derived from pilot signals transmitted along with the data. Then in step 422, from N_RSubtracting N from the corresponding received signal_RAn interference signal to derive N_RA modified signal. These modified signals represent the signals at the receive antennas if the components due to the decoded data streams have not been transmitted (i.e., assuming that interference cancellation is effectively performed).

Then to N_RA modifying signal (instead of N)_RReceived signals) repeats the processing performed in steps 412 through 416 to recover another transmitted signal. Step 4 is therefore repeated for each transmitted signal to be recovered12 to 416 and if there is no further transmitted signal to recover steps 420 and 422 are performed.

Thus, the successive cancellation receiver processing technique performs multiple iterations, one for each transmitted signal to be recovered. Each iteration (except the last) is processed in two steps to recover one transmitted signal and to generate a modified signal for the next iteration. In the first step, for N_RThe received signals are spatially or space-time processed to provide N_RThe post-processed signals, and one of the post-processed signals is detected to recover a data stream corresponding to the transmitted signal. In a second step, which need not be performed for the last iteration, the interference due to the decoded data stream is cancelled from the received signal to derive a modified signal with the recovered components removed.

First, the input signal for the first iteration is the received signal, which may be expressed as:

img id="idf0001" file="C0280965500211.GIF" wi="99" he="102" img-content="drawing" img-format="GIF"/formula (1)

WhereinrIs N_RA vector of received signals, andr ¹is N for the first iteration of a successive cancellation receiver processing scheme_RAn input signal. These input signals are linear or non-linearProcessing to provide a post-processed signal, which may be represented as:

img id="idf0002" file="C0280965500212.GIF" wi="76" he="108" img-content="drawing" img-format="GIF"/formula (2)

Whereinx ¹Is N from the first iteration_RA vector of post-processed signals. It is possible to estimate the SNR of the post-processed signal, which can be expressed as:

img id="idf0003" file="C0280965500213.GIF" wi="156" he="26" img-content="drawing" img-format="GIF"/formula (3)

Selecting one of the post-processed signals for further processingThe step(s) (e.g., post-processed signal with highest SNR) are processed to provide a decoded data stream. This decoded data stream is then used to estimate the interference generated by the recovered signali ¹This may be expressed as:

img id="idf0004" file="C0280965500214.GIF" wi="69" he="108" img-content="drawing" img-format="GIF"/formula (4)

Then from the input signal vector of this iterationr ¹Subtracting interference fromTo derive an input signal vector comprising the next iterationr ²The modification signal of (2). The interference cancellation may be expressed as:

img id="idf0006" file="C0280965500216.GIF" wi="181" he="108" img-content="drawing" img-format="GIF"/formula (5)

The same process is then repeated for the next iteration, where the vectors arer ²Including this iteration is the input signal.

According to a successive cancellation receiver processing scheme, one transmitted signal is recovered for each iteration, and the SNR, γ, of the ith transmitted signal recovered in the kth iteration may be provided_i ^kAs the CSI of the transmission channel used to transmit the recovered signal. For example, if the first post-processed signal x is recovered in the first iteration₁ ¹Then the second post-processed signal x is recovered in the second iteration₂ ²And so on, and recover the Nth iteration in the last iteration_TA post-processed signal x_NT ^NTThen, the CSI that may be reported for these recovered signals may be expressed as:img id="idf0007" file="C0280965500221.GIF" wi="140" he="21" img-content="drawing" img-format="GIF"/

successively processing the original N by using successive cancellation receiver processing techniques_ROne received signal to recover one transmitted signal at a time. In addition, each recovered transmitted signal is removed (i.e., canceled) from the received signal prior to processing to recover the next transmitted signal. If the transmitted data stream is decoded error-free (or with minimal error), and if the channel response estimate is reasonably accurate, it is effective to cancel the interference generated by the previously recovered transmitted signal from the received signal. Interference cancellation generally improves the SNR of each transmitted signal that is subsequently recovered. In this way, a higher performance can be obtained for all transmitted signals (possibly except for the first transmitted signal to be recovered).

It is possible to illustrate the improvement in SNR of a recovered transmitted signal using a successive cancellation receiver processing technique. In this example, a pair of cross-polarized antennas is employed at the transmitter and receiver, the MIMO channel is line-of-sight, and four independent data streams are transmitted on the vertical and horizontal components of the pair of cross-polarized transmit antennas. For simplicity, cross-polarization isolation is assumed to be ideal so that the vertical and horizontal components do not interfere with each other at the receiver.

The receiver first receives four signals on the vertical and horizontal components of the cross-polarized pair of receive antennas and processes the four received signals. The received signals on the vertical elements of the cross-polarized antenna are highly correlated, while the received signals on the horizontal elements are also highly correlated.

The ability to null interference is compromised when there is a strong linear correlation between two or more transmit-receive antenna pairs that make up a MIMO channel. In this case, linear spatial processing cannot successfully separate the four independent data streams transmitted on the vertical and horizontal components of the cross-polarized antenna pair. In particular, the vertical component on each cross-polarized transmit antenna interferes with the vertical component on the other cross-polarized transmit antenna. The resulting SNR for each of the four transmitted signals may be poor due to the associated interference from another antenna having the same polarization. Thus, the capacity of a transmitted signal based on linear spatial processing alone may be severely constrained by the associated interference signal.

When examining the eigenmodes (eigenmodes) of this example MIMO channel, only two non-zero eigenmodes (i.e., vertical and horizontal polarizations) can be seen. The "full-CSI" processing scheme then transmits only two independent data streams with these two eigenmodes. The capacity (capacity) thus obtained can be expressed as:

Capacity＝2·log₂(1+λ_i/σ²)

wherein λ_i/σ²Is the ratio of the received signal power to the thermal noise power of the i-th eigenmode. Thus, the capacity of the full-CSI processing scheme for this example MIMO channel is equivalent to the capacity of two parallel white Gaussian noise (AWGN) channels, each having a channel defined by λ_i/σ²Given the SNR.

According to the successive cancellation receiver processing technique, the linear spatial processing performed in step 412 first results in an SNR of 0dB or less for each of the four transmitted signals (due to noise plus interference from other transmitted signals on the same polarization). The total capacity will be poor if no additional receiver processing is performed.

However, by applying successive spatial processing and interference cancellation, the SNR of the subsequently recovered transmitted signal may be improved. For example, the first signaling to be recovered may be the vertical polarization from the first cross-polarized transmit antenna. If interference cancellation is assumed to be performed efficiently (i.e., zero or minimal decision error and accurate channel estimation), the signal no longer (or minimally) interferes with the remaining three (not yet recovered) transmitted signals. Removing this vertical polarization interference improves the SNR of another signal transmitted on the vertical polarization that has not yet been recovered. For the sake of this simple example it is assumed that cross-polarization isolation is ideal and that the two signals transmitted on the horizontal polarization do not interfere with the signals transmitted on the vertical polarization. Thus, it is possible to recover the signal emitted on the vertical polarization of the second cross-polarized transmit antenna at a SNR limited (theoretically) by the thermal noise power, according to an efficient interference cancellation.

In the above example, removing interference from the vertical polarization does not affect the SNR of the two signals transmitted on the horizontal polarization. Thus, successive spatial processing and interference cancellation are also applied for the two signals transmitted in the horizontal polarization. This results in a first recovered signal on the horizontal polarization having a low SNR, while a second recovered signal on the horizontal polarization has an SNR that is also (theoretically) limited by thermal noise.

Since successive spatial processing and interference cancellation are performed, two transmit signals with low SNR rarely contribute to the total capacity, while two transmit signals with high SNR greatly contribute to the total capacity.

Non-dispersive and dispersive channel

Different receive and (possibly) transmit processing schemes may be used depending on the characteristics of the MIMO channel, which may be either non-dispersive or dispersive. Non-dispersive MIMO channels experience smooth fading (i.e., frequency non-selective fading), which is more likely when the system bandwidth is narrow. Dispersive MIMO channels experience frequency-nonselective fading (i.e., different amounts of attenuation across the system bandwidth), which is wide at the system bandwidth and more likely for certain operating conditions and environments. Successive cancellation receiver processing techniques may be advantageously used for both non-dispersive and dispersive MIMO channels.

For non-dispersive MIMO channels, linear spatial processing techniques such as CCMI and MMSE may be used to process the received signal before demodulation and decoding. These linear spatial processing techniques may be employed at the receiver to remove undesired signals or to maximize the received signal-to-interference-plus-noise ratio of each constituent signal in the presence of noise and interference from other signals. Is provided withThe ability to effectively remove undesired signals or optimize the signal-to-interference-plus-noise ratio depends on the channel coefficient matrixHThe matrix describes the channel response between the transmit and receive antennas. Successive cancellation receiver processing techniques (e.g., with CCMI or MMSE) can be effectively used for non-dispersive MIMO channels.

For dispersive MIMO channels, time dispersion within the channel causes inter-symbol interference (ISI). To improve performance, a wideband receiver attempting to recover a particular transmitted data stream needs to improve "crosstalk" from other transmitted signals as well as intersymbol interference from all transmitted signals. The successive cancellation receiver processing technique can be extended to handle dispersive MIMO channels. To handle crosstalk and intersymbol interference, it is possible to replace the spatial processing in the narrowband receiver (which handles crosstalk well but does not handle intersymbol interference efficiently) with space-time processing in the wideband receiver. Within the wideband receiver, successive cancellation receiver processing techniques may be employed in a manner similar to that described in fig. 4. However, the spatial processing performed in step 412 is replaced with space-time processing.

In an embodiment, an MMSE linear equalizer (MMSE-LE) may be used for space-time processing within the wideband receiver. The space-time processing assumes a form similar to the spatial processing of the narrowband channel by using MMSE-LE techniques. However, each "filter tap" within the spatial processor includes more than one tap, as described in more detail below. When estimating the channel (i.e. channel coefficient matrix)H) When accurate, MMSE-LE techniques are most effective in space-time processing.

In another embodiment, a Decision Feedback Equalizer (DFE) may be used for space-time processing at the wideband receiver. The DFE is a non-linear equalizer that is effective for channels with severe amplitude distortion and uses decision feedback to cancel interference from symbols that have been detected. If the data stream can be decoded error-free (or with minimal error), it is possible to effectively cancel the intersymbol interference generated by the modulation symbols corresponding to the decoded data bits.

In yet another embodiment, a Maximum Likelihood Sequence Estimator (MLSE) may be used for space-time processing.

DFE and MLSE techniques may reduce or may eliminate performance degradation when channel estimation is inaccurate. DFE and MLSE techniques are detailed by s.l. ariyavistakul et al in a paper entitled "OptimumSpace-Time Processors with dispersion Interference: the Unified analytical and Required Filter Span ", published in IEEE journal on communications, 7 < 7 > 1999, No. 7, which is incorporated herein by reference.

It is also possible to advantageously employ adaptive transmitter processing and successive cancellation receiver processing based on available CSI for dispersive MIMO channels. The SNR of the recovered transmitted signal from the output of each space-time processing stage may include the CSI of the transmitted signal. This information may be fed back to the transmitter to assist in selecting the appropriate coding and modulation schemes for the data streams associated with the transmitted signals.

Receiver structure

Fig. 5 is a block diagram of a receiver system 150a capable of implementing various aspects and embodiments of the invention. Receiver system 150a implements a successive cancellation receiver processing technique to receive and recover the transmitted signal. From (up to) N_TThe transmission signal of the root transmitting antenna is composed of N_REach of the root antennas 152a through 152r receives and is routed to a respective demodulator (DEMOD)154 (also referred to as a front-end processor). For example, receive antenna 152a may receive multiple transmit signals from multiple transmit antennas, and receive antenna 152r may similarly receive multiple transmit signals. Each demodulator 154 conditions (e.g., filters and amplifies) a respective received signal, downconverts the conditioned signal to an intermediate frequency or baseband, and digitizes the downconverted signal to provide samples. Each demodulator 154 may also demodulate samples with a received pilot to generate a stream of received modulation symbols, which is provided to an RX MIMO/data processor 156.

If OFDM is used for data transmission, each demodulator 154 also performs the same operations as the modulator 122 shown in FIG. 3The processes of (1) are complementary processes. In this case, each demodulator 154 includes an FFT processor (not shown) that generates a sample-transformed representation and provides a stream of vectors of modulation symbols, each vector including N_LN of one frequency subchannel_LOne modulation symbol and one vector is provided for each slot. From all N_RThe vector streams of modulation symbols from the FFT processors of the multiple demodulators are then provided to a demultiplexer (not shown in fig. 5) that "channelizes" the vector stream of modulation symbols from each FFT processor into multiple (up to N)_L) A stream of modulation symbols. For transmit processing schemes in which each frequency subchannel is processed independently (as shown in fig. 3), the demultiplexer also passes (up to) N_LEach of the individual modulation symbol streams is provided to a respective RX MIMO/data processor 156.

For MIMO systems using OFDM, one RX MIMO/data processor 156 may be used for N used for data transmission_LEach of the frequency sub-channels is processed from N_RThe modulated symbol streams for the root receive antennas. For MIMO systems that do not use OFDM, one RX MIMO/data processor 156 may be used to process data from N_RN of root receiving antenna_RA stream of modulation symbols.

In the embodiment illustrated in fig. 5, RX MIMO/data processor 156 includes a number of successive (i.e., cascaded) receiver processing stages 510, one for each transmission channel used for data transmission. In a transmit processing scheme, one data stream is transmitted on each transport channel, and each data stream is processed independently (e.g., with its own coding and modulation scheme) and transmitted from a corresponding transmit antenna. For this transmit processing scheme, the number of data streams is equal to the number of transmission channels, which is equal to the number of transmit antennas used for data transmission (which may be a subset of the available transmit antennas). For simplicity, the RXMIMO/data processor 156 is described for this transmit processing scheme.

Each receiver processing stage 510 (the final stage 510n of processing) includes a channel coupled to an interference canceller 530A MIMO/data processor 520 and a final stage 510n includes only a channel MIMO/data processor 520 n. For the first receiver processing stage 510a, a channel MIMO/data processor 520a receives and processes N from demodulators 154a to 154r_RModulates the symbol stream and provides a decoded data stream for the first transmission channel (or first transmitted signal). And for each of the second through last stages 510b through 510N, the channel MIMO/data processor 520 of that stage receives and processes N from the interference canceller of the previous stage_RA stream of modified symbols to derive a decoded data stream for the transmission channel being processed by that stage. Each channel MIMO/data processor 520 may also provide CSI (e.g., SNR) for the associated transmission channel.

For the first receiver processing stage 510a, the interference canceller 530a operates from all N_RA demodulator 154 receiving N_RA stream of modulation symbols. And for each of the second through last stages, interference canceller 530 receives N from the interference canceller in the previous stage_RA modified symbol stream. Each interference canceller 530 also receives decoded data streams from channel MIMO/data processors 520 at the same stage and performs processing (e.g., encoding, interleaving, modulation, channel response, etc.) to derive N_RA remodulated symbol stream that is an estimate of an interference component of the received modulated symbol stream that was generated as a result of the decoded data stream. The remodulated symbol stream is then subtracted from the received modulated symbol stream to derive N_RModified symbol streams that include all but the subtracted (i.e., cancelled) interference component. Then, N_RThe modified symbol streams are provided to the next stage.

In fig. 5, a controller 540 is shown coupled to RX MIMO/data processor 156 and may be used to direct steps in successive cancellation receiver processing performed by processor 156.

Fig. 5 shows a receiver structure that may be used in a direct manner when each data stream is transmitted over a respective transmit antenna (i.e., one data stream corresponds to). In this case, each receiver processing stage 510 may be used to recover one transmitted signal and provide a decoded data stream corresponding to the recovered transmitted signal.

For some other transmit processing schemes, the data streams may be transmitted over multiple antennas, frequency subchannels, and/or time intervals, providing spatial, frequency, and time diversity, respectively. For these schemes, receiver processing initially derives a stream of received modulation symbols for the transmitted signal on each transmit antenna for each frequency subchannel. Modulation symbols for multiple transmit antennas, frequency subchannels, and/or time intervals may be combined in a manner that is inverse to the demultiplexing performed at the transmitter system. The stream of combined modulation symbols is then processed to provide an associated decoded data stream.

Spatial processing techniques for non-dispersive channels

As described above, many linear spatial processing techniques may be used to process signals received over non-dispersive channels to recover each transmitted signal stream from interference caused by other transmitted signal streams. These techniques include CCMI, MMSE, and possibly others. Within each channel MIMO/data processor 520 for N_RThe input signals are subjected to linear spatial processing. For the first receiver processing stage 510a, the input signal is from N_RN of root receiving antenna_RA received signal. And for each subsequent stage the input signal is N from the interference canceller of the previous stage_RThe modified signal, as described above. For simplicity, the CCMI and MMSE techniques are described for the first stage. However, each subsequent stage of processing proceeds in a similar manner to the appropriate substitution of the input signal. More specifically, in each subsequent stage, it is assumed that the interference detected in the preceding stage has been cancelled, and therefore the dimensionality of the channel coefficient matrix is reduced in each stage described below.

In the presence of N_TRoot transmitting antenna and N_RIn MIMO systems with root receiving antennas, N_RThe received signal at the output of the root receive antenna can be expressed as:

r＝ Hx+ nformula (6)

WhereinrIs a vector of received symbols (i.e., N from the MIMO channel)_RX 1 vector output as derived from the receive antennas).HIs a matrix of the channel coefficients and,xis a vector of transmitted symbols (i.e., N for a MIMO channel)_TX 1 vector input), andnis N representing noise plus interference_RX 1 vector. Receiving a symbol vectorrInvolving the passage of N in a particular time slot_RRoot receiving antenna from N_RN received by a signal_RAnd a modulation symbol. Likewise, a symbol vector is transmittedxInvolving the passage of N in a particular time slot_TN transmitted by root transmitting antenna_TN within one signal_TAnd a modulation symbol.

Channel coefficient matrixHIt can also be written as:

img id="idf0008" file="C0280965500272.GIF" wi="146" he="21" img-content="drawing" img-format="GIF"/formula (6a)

Wherein the vectorh _iComprises the ith hairAnd antenna dependent channel coefficients. In each subsequent step of the successive cancellation process, the column vector in equation (6a) associated with the previously cancelled signal is removed. For simplicity, assuming cancellation of the transmitted signals in the order in which the associated channel coefficient vectors are listed in equation (6a), at the kth stage of the successive cancellation process, the channel coefficient matrix is:

img id="idf0009" file="C0280965500273.GIF" wi="162" he="21" img-content="drawing" img-format="GIF"/formula (6b)

CCMI technique

For CCMI spatial processing techniques, the receiver system first pairs a vector of received symbolsrA channel matched filter operation is performed. The matched filtered output may be expressed as:

H ^H r＝ H ^H Hx+ H ^H nformula (7)

Where the superscript "H" represents transpose and complex conjugate. Square matrixRPossibly for representing a channel coefficient matrixHWith its conjugate transposeH ^HProduct of (i.e. the product of) R＝ H ^H H)。

Channel coefficient matrixHPossibly derived from, for example, pilot symbols transmitted with the data. In order to perform "best" reception and estimate the SNR of a transmission channel, it is often convenient to insert some known symbols into the transmitted data stream and transmit these known symbols over one or more transmission channels. These known symbols are also referred to as pilot symbols or pilot signals. Methods for estimating a single transmission channel from pilot signals and/or data transmissions are found in many papers in the art. One such channel estimation method is described by f.ling in a paper entitled "Optimal Reception, Performance Bound, and current-Rate Analysis of references-associated CDMA Communications with Applications", published in the IEEE proceedings on Communications, 10.1999. This or some other channel estimation method may be extended to a matrix form to derive a channel coefficient matrix, as is known in the artH。

By matching filtered vectorsH ^H rAndRis multiplied to obtain an estimate of the transmitted symbol vectorx', which can be expressed as:

x′＝ R ^-1 H ^H r＝ x+ R ^-1 H ^H n＝ x+ n' formula (8)

As can be seen from the above equation, it is possible to estimate the symbol vector by comparing the received symbol vectorsrMatched filtering (i.e. multiplication by a matrix)H ^H) Then, the filtering result is multiplied by an inverse square matrixR ^-1And recovering the transmitted symbol vectorx。

For CFor CMI techniques, the SNR of the processed received symbol vector (i.e.,xthe ith element of' may be expressed as:

img id="idf0010" file="C0280965500284.GIF" wi="89" he="53" img-content="drawing" img-format="GIF"/formula (9)

If the ith transmitted symbolOn average equal to one (1.0), the SNR of the processed received symbol vector may be expressed as:

may be determined by scaling up the ith element of the received symbol vectorBut the noise variance is normalized.

If the streams of modulation symbols are replicated and transmitted over multiple transmit antennas, the modulation symbols may be summed together to form a combined modulation symbol. For example, if a data stream is transmitted from all antennas, the summation corresponds to all N_TThe modulation symbols for the root transmit antenna, and the combined modulation symbols may be represented as:

formula (10)

Alternatively, the transmitter may transmit one or more data streams over a number of transmission channels using the same coding and modulation schemes on some or all of the transmit antennas. Thus, only one SNR (e.g., average SNR) may be required for a transmission channel to which a common coding and modulation scheme is applied. For example, if the same coding and modulation scheme is applied on all transmit antennas, the SNR of the combined modulation symbol can be derived: SNR_total. Then, the SNR_totalThere will be a maximum combined SNR equal to that from N_TThe sum of the SNRs of the modulation symbols for the root transmit antennas. The combined SNR can be expressed as:

formula (11)

Fig. 6A is a block diagram of an embodiment of a channel MIMO/data processor 520x that can implement the CCMI techniques described above. Channel MIMO/data processor 520x includes a processor 610 (which performs CCMI processing) coupled to an RX data processor 620.

Within processor 610x, a vector of modulation symbols is receivedrThe filtering is performed by a matched filter 614, which pre-filters each vectorrMatrix of channel coefficients transposed with conjugationH ^H(as shown in equation (7) above). Channel coefficient matrixHCan be estimated from the pilot signal in a manner similar to that used by conventional pilot-assisted single and multi-carrier systems, as is known in the art. As shown above, the matrixRAccording to the formulaR＝ H ^H HTo calculate. The filtered vector is further coupled to an inverse square matrix using a multiplier 616R ^-1Multiplying to form a vector of transmit modulation symbolsxIs estimated byx', as shown in the above equation (8).

For some transmit processing schemes, the estimated streams of modulation symbols correspond to multiple transmit antennas used for transmission of the data streams and are provided to a combiner 618, which combines the redundant information in time, space, and frequency. Then, the combined modulation symbolsx"is provided to an RX data processor 620. For some other transmit processing schemes, the estimated modulation symbolsx' may be provided to RX data processor 620 directly (not shown in fig. 6A).

Thus, processor 610x generates a number of independent symbol streams, which correspond to a number of data streams transmitted from the transmitter system. Each symbol stream includes recovered modulation symbols that correspond to and are estimates of the mapped modulation symbols at the transmitter system. The (recovered) symbol stream is then provided to an RX data processor 620.

As described above, each stage 510 within RX MIMO/data processor 156 recovers and decodes one of the transmitted signals included in the input signal for that stage (e.g., the transmitted signal with the best SNR). CSI processor 626 estimates the SNR of the transmitted signal and may obtain these estimates based on equations (9) and (11) above. CSI processor 626 then provides CSI (e.g., SNR) for the transmitted signal that has been selected (e.g., "best") for recovery and decoding and further provides control signals that identify the selected transmitted signal.

Fig. 7 is a block diagram of an embodiment of RX data processor 620. In this embodiment, selector 710 within RX data processor 620 receives a number of symbol streams from preceding linear spatial processors and extracts the symbol stream corresponding to the selected transmit signal, as indicated by the control signal from CSI processor 626. In another embodiment, RX data processor 620 is provided with a stream of symbols corresponding to the selected transmit signal and stream extraction is performed by combiner 618 based on a control signal from CSI processor 626. In any case, the extracted stream of modulation symbols is provided to a demodulation element 712.

For the transmitter embodiment shown in fig. 2, in which the data streams for each transmission channel are independently encoded and modulated based on the SNR of the channel, the recovered modulation symbols for the selected transmission channel are demodulated according to a demodulation scheme (e.g., M-PSK, M-QAM) that is complementary to the modulation scheme used for the transmission channel. The demodulated data from demodulation element 712 is then deinterleaved by a deinterleaver 714 in a manner complementary to that performed by channel interleaver 214, and the deinterleaved data is further decoded by a decoder 716 in a manner complementary to that performed by encoder 212. For example, a Turbo decoder or a Viterbi (Viterbi) decoder may be used for decoder 716 if Turbo or convolutional coding, respectively, is performed at the transmitter. The decoded data stream from decoder 716 represents an estimate of the transmitted data being recovered.

Referring back to FIG. 6A, the estimated modulation symbolsx' and/or combined modulation symbolsx"are also provided to a CSI processor 626, which estimates the SNR for each transmission channel. For example, CSI processor 626 may estimate a noise covariance matrix φ from received pilot signals_nnThen, the SNR of the ith transmission channel is calculated according to equation (9) or (11). The SNR may be estimated similar to conventional pilot-assisted single or multi-carrier systems, as is known in the art. The SNR for all transmission channels may include the CSI reported back to the transmitter system for that transmission channel. CSI processor 626 further provides control signals identifying the selected transmission channel to RX data processor 620 or combiner 618.

Estimated modulation symbolsx' is further provided to a channel estimator 622 and a matrix processor 624, which estimate a coefficient matrix, respectivelyHAnd derive a square matrixR. Possibly as estimated channel coefficient matrixHAnd estimated modulation symbols corresponding to pilot data and/or traffic data are used.

Referring back to fig. 5, the input signal of the first stage 510a includes all of the transmit signals, while the input signal of each subsequent stage includes one transmit signal (i.e., one term) cancelled by the previous stage. Thus, the channel MIMO/data processor 520a in the first stage 510a may be designed and used to estimate the channel coefficient matrixHAnd provides the matrix to all subsequent stages.

The CSI information reported back to transmitter system 110 by receiver system 150 may include the SNR of the transmission channel, as determined by stages within RX MIMO/data processor 156.

MMSE technique

For MMSE spatial processing techniques, the receiver system first combines the received symbol vectorsrAnd a weighting coefficient matrixMMultiplying to derive a vector of transmitted symbolsxInitial MMSE estimation ofThis can be expressed as:

img id="idf0017" file="C0280965500312.GIF" wi="67" he="16" img-content="drawing" img-format="GIF"/

img id="idf0018" file="C0280965500313.GIF" wi="159" he="26" img-content="drawing" img-format="GIF"/formula (12)

Wherein

M＝ H ^H( HH ^H+ φ _nn)^-1Formula (13)

Selection matrixMSo that the initial MMSE estimateAnd transmitting the symbol vectorxInter error vectore(i.e., the amount of the acid,img id="idf0020" file="C0280965500316.GIF" wi="61" he="17" img-content="drawing" img-format="GIF"/) The mean square error of (a) is minimal.

To determine the SNR of the transmission channel of the MMSE technique, the SNR of the transmission channel can be determined according to the knowledgexTime of flightFirst determines the signal component, the mean is averaged over the additive noise, as follows:

img id="idf0022" file="C0280965500318.GIF" wi="149" he="18" img-content="drawing" img-format="GIF"/

img id="idf0023" file="C0280965500319.GIF" wi="181" he="25" img-content="drawing" img-format="GIF"/

img id="idf0024" file="C02809655003110.GIF" wi="173" he="26" img-content="drawing" img-format="GIF"/

img id="idf0025" file="C02809655003111.GIF" wi="37" he="16" img-content="drawing" img-format="GIF"/

wherein the matrixVCan be expressed as:

img id="idf0026" file="C02809655003112.GIF" wi="383" he="25" img-content="drawing" img-format="GIF"/

initial MMSE estimationThe ith element ofCan be expressed as:

img id="idf0029" file="C02809655003115.GIF" wi="242" he="20" img-content="drawing" img-format="GIF"/formula (14)

If all the elements areAre all uncorrelated and have a zero mean value, thenThe expected value of the ith element may be expressed as:

img id="idf0032" file="C02809655003118.GIF" wi="102" he="18" img-content="drawing" img-format="GIF"/formula (15)

As shown in the formula (15),is x_iAnd the bias can be removed to obtain improved performance. By means of a handleDivided by v_iiCan obtain x_iUnbiased estimation of (d). Thus, it is possible to estimate the bias by combiningAnd diagonal matrixD _V ^-1Obtained by left multiplicationxUnbiased estimation of (d):the following are described:

img id="idf0037" file="C02809655003123.GIF" wi="65" he="21" img-content="drawing" img-format="GIF"/formula (16)

Wherein

img id="idf0038" file="C02809655003124.GIF" wi="255" he="24" img-content="drawing" img-format="GIF"/Formula (17)

And v is_iiIs a matrixVThe diagonal elements of (a).

Unbiased estimation for noise plus interference determinationAnd transmitting the symbol vectorxError betweenêCan be expressed as:

img id="idf0040" file="C0280965500322.GIF" wi="100" he="20" img-content="drawing" img-format="GIF"/

img id="idf0041" file="C0280965500323.GIF" wi="213" he="25" img-content="drawing" img-format="GIF"/

for the MMSE technique, the SNR of the received symbol vector after processing (i.e.,the ith element) can be expressed as:

img id="idf0043" file="C0280965500325.GIF" wi="117" he="50" img-content="drawing" img-format="GIF"/formula (18)

Wherein u is_iiIs an error vectorêAnd the matrix U may be represented as:

img id="idf0044" file="C0280965500326.GIF" wi="236" he="23" img-content="drawing" img-format="GIF"/formula (19)

If the ith transmission symbol x_iVariance of (2)Average equal to one (1.0) and from equation (19)img id="idf0046" file="C0280965500328.GIF" wi="84" he="42" img-content="drawing" img-format="GIF"/The SNR of the received symbol vector after processing may be expressed as:

img id="idf0047" file="C0280965500329.GIF" wi="101" he="38" img-content="drawing" img-format="GIF"/formula (20)

Estimated modulation symbols as described above for the CCMI techniqueMay be similarly combined to obtain combined modulation symbols.

Fig. 6B is a block diagram of an embodiment of a channel MIMO/data processor 520y, which can implement the MMSE technique described above. Channel MIMO/data processor 520y includes a processor 610y (which performs MMSE processing) coupled to an RX data processor 620.

Within processor 610y, multiplier 634 adds the received vector of modulation symbolsrAnd matrix MLeft-multiplying to form a vector of transmitted symbolsxIs estimated byAs shown in the above equation (8). Similar to the CCMI technique, it is possible to estimate matrices from received pilot signals and/or data transmissionsHAnd phi_nn. Then, the matrix M is calculated according to equation (9). Multiplier 636 evaluatesFurther diagonal matrixD _V ^-1Left-multiplying to form an unbiased estimate of the transmitted symbol vectorAs shown in the above equation (12).

Likewise, for some transmit processing schemes, it may be possible to provide a number of estimated symbols to combiner 638Streams, which correspond to the number of transmit antennas used to transmit the data streams, combiner 638 combines the redundant information in time, space, and frequency. Then, the combined modulation symbolsIs provided to an RX data processor 620. For some other transmit processing schemes, the estimated modulation symbolsMay be provided directly (not shown in fig. 6B) to RX data processor 620. RX data processor 620 demodulates, deinterleaves, and decodes the modulated symbol streams for the data stream being recovered as described above.

Estimated modulation symbolsAnd/or combined modulation symbolsAnd also to a CSI processor 626, which estimates the SNR for each transmitted signal. For example, CSI processor 626 may estimate the SNR of the ith transmitted signal according to equation (18) or (20). The SNR of the selected transmitted signal may be reported back to the transmitter system. CSI processor 626 also provides control signals identifying the selected transmit signals to RX data processor 620 or combiner 618.

Estimated modulation symbolsIs further provided to an adaptive processor 642 which derives matrices from equations (13) and (17), respectivelyMAnd diagonal matrixD _V ^-1。

Space-time processing technique for time-dispersive channels

As described above, many space-time processing techniques may be used to process signals received over time-dispersive channels. These techniques include the use of time domain channel equalization techniques such as MMSE-LE, DFE, MLSE, and spatial processing techniques for the non-dispersive channels described above. Within each channel MIMO/data processor 520 for N_RThe input signals are space-time processed.

MMSE-LE technique

When time dispersion exists, channel coefficient matrixHExhibit a time-delay scale and are matrixHEach element of (a) corresponds to a linear transfer function rather than a coefficient. In this case, the channel coefficient matrixHCan be expressed as a channel transfer function matrixHThe form of (τ) is written, which can be expressed as:

H(τ)＝{h_ij(τ) } for 1. ltoreq. i.ltoreq.N_RAnd j is 1. ltoreq. N_TFormula (21)

Wherein h is_ij(τ) is a linear transfer function from the jth transmit antenna to the ith receive antenna. Due to the fact thatLinear transfer function h_ij(τ), received vectorr(t) is the channel transfer function matrixH(tau) with the transmitted signal vectorx(t), which can be expressed as:

r(t)＝∫ H(τ) x(1- τ) d τ equation (22)

As part of the demodulation function (performed by demodulator 154 in fig. 5), the received signal is sampled to provide received samples. In general, the time-dispersive channel and the received signal may be represented in a discrete-time representation in the following description. First, a delay k channel transfer function vector associated with the jth transmit antennah _j(k) Can be expressed as:

img id="idf0058" file="C0280965500334.GIF" wi="246" he="22" img-content="drawing" img-format="GIF"/for k is more than or equal to 0 and less than or equal to L formula (23)

Wherein h is_ij(k) Is the kth tap weight of the channel transfer function associated with the path between the jth transmit antenna and the ith receive antenna, and L is the maximum degree of channel dispersion (in sample intervals). Second, N with a delay of k_R×N_TThe channel transfer function matrix may be expressed as：

img id="idf0059" file="C0280965500341.GIF" wi="234" he="19" img-content="drawing" img-format="GIF"/For 0 ≦ k ≦ L equation (24)

Vector of received signals at sampling time nr(n) may be expressed as:

img id="idf0060" file="C0280965500342.GIF" wi="323" he="41" img-content="drawing" img-format="GIF"/equation (25)

WhereinIs N_R×(L+1)N_TA matrix of block structure representing the sampled channel matrix transfer functionH(k) And can be expressed as:

img id="idf0062" file="C0280965500344.GIF" wi="198" he="22" img-content="drawing" img-format="GIF"/

whileimg id="idf0063" file="C0280965500345.GIF" wi="30" he="18" img-content="drawing" img-format="GIF"/Is a sequence of L +1 vectors of received samples captured for L +1 sampling intervals, each vector comprising N_RN of root receiving antenna_RAnd may be expressed as:

img id="idf0064" file="C0280965500346.GIF" wi="126" he="102" img-content="drawing" img-format="GIF"/

MMSE linear space-time processor by combining a sequence of received signal vectors at time nr(N) and 2K +1, N_R×N_TSequence of weighting matricesM(k) Convolution to calculate an estimate of the transmitted symbol vectorAs follows:

img id="idf0066" file="C0280965500348.GIF" wi="242" he="50" img-content="drawing" img-format="GIF"/formula (26)

Whereinimg id="idf0067" file="C0280965500349.GIF" wi="232" he="42" img-content="drawing" img-format="GIF"/K is a parameter that determines the degree of delay of the equalizer filter,

and is

img id="idf0068" file="C02809655003410.GIF" wi="128" he="128" img-content="drawing" img-format="GIF"/

Selecting a sequence of weighting matricesM(k) Minimizing the mean square error, which can be expressed as:

ε＝E{ e ^H(k) e(k) equation (27)

Wherein the error ise(k) Can be expressed as:

img id="idf0069" file="C02809655003412.GIF" wi="129" he="18" img-content="drawing" img-format="GIF"/formula (28)

The MMSE solution can then be expressed as a sequence of weighting matrices that satisfy a linear constraintM(k)：

img id="idf0070" file="C02809655003413.GIF" wi="333" he="75" img-content="drawing" img-format="GIF"/Formula (29)

WhereinR(k) Is N_R×N_TA sequence of space-time correlation matrices, which can be expressed as:

formula (30)

Wherein* _zz(k) Is the noise autocorrelation function, expressed as:

* _zz(k)＝E{ z(λ-k) z ^H(lambda) } formula (31)

For white (temporally uncorrelated) noise,* _zz(k)＝ * _zzδ (K) wherein* _zzOnly the spatial correlation matrix is represented. For spatially and temporally uncorrelated noise with equal power at each receive antenna,* _zz(k)＝σ² Iδ(k)。

equation (29) may be expressed as:

img id="idf0072" file="C0280965500354.GIF" wi="205" he="27" img-content="drawing" img-format="GIF"/formula (32)

WhereinIs a block-Toeplitz (Toeplitz) wherein the block j, k consists ofR(j-k) is given, and

img id="idf0074" file="C0280965500356.GIF" wi="137" he="157" img-content="drawing" img-format="GIF"/

wherein0 _m×nIs a zero matrix.

According to the MMSE spatial processing described above, an unbiased minimum mean square error estimate is derived for determining the SNR associated with the symbol estimates. First, for the MMSE-LE estimate derived above,

img id="idf0075" file="C0280965500357.GIF" wi="216" he="30" img-content="drawing" img-format="GIF"/

img id="idf0076" file="C0280965500358.GIF" wi="457" he="20" img-content="drawing" img-format="GIF"/

formula (33)

Where it is expected on the noise. If it is assumed that the modulation symbols are uncorrelated in time and that the expectation is obtained over all the above-mentioned intersymbol interference (all transmitted signal components not transmitted at time n), the expectation can be expressed as:

img id="idf0077" file="C0280965500361.GIF" wi="223" he="30" img-content="drawing" img-format="GIF"/

img id="idf0078" file="C0280965500362.GIF" wi="358" he="18" img-content="drawing" img-format="GIF"/

formula (34)

Wherein

img id="idf0081" file="C0280965500365.GIF" wi="158" he="25" img-content="drawing" img-format="GIF"/

Finally, after averaging the interference from the other spatial subchannels, the average of the signal from the ith transmit antenna at time n may be expressed as:

img id="idf0082" file="C0280965500366.GIF" wi="179" he="19" img-content="drawing" img-format="GIF"/formula (35)

Wherein v is_iiIs thatVThe ith diagonal element of (v)_iiIs a scalar quantity), andis the ith element of the MMSE-LE estimate.

According to the definition

img id="idf0084" file="C0280965500368.GIF" wi="252" he="24" img-content="drawing" img-format="GIF"/Formula (36)

The unbiased MMSE-LE estimate of the transmit signal vector at time n can be expressed as:

formula (37)

The error covariance matrix associated with the unbiased MMSE-LE can be expressed as:

img id="idf0087" file="C02809655003611.GIF" wi="210" he="21" img-content="drawing" img-format="GIF"/formula (38)

Finally, the SNR associated with the estimate of the transmitted symbol on the ith transmit antenna may be expressed as:

img id="idf0088" file="C02809655003612.GIF" wi="147" he="42" img-content="drawing" img-format="GIF"/formula (39)

The MMSE-LE technique may be implemented with the channel MIMO/data processor 520y of fig. 6B. In this case, multiplier 634 may be designed to sequence the received signal vectorr(n) sequence of weighting matricesM(k) Convolution, as shown in equation (26). Multiplier 636 may be designed to estimateAnd diagonal matrixD _V ^-1Left-multiplying to derive unbiased MMSE-LE estimatesAs shown in equation (37). Adaptive processor 642 may be designed to derive a sequence of weighting matrices as shown in equation (32)M(k) And a diagonal matrix shown in equation (36)D _V ^-1. The subsequent processing may be implemented in a manner similar to that described above for the MMSE technique. CSI processor 626 may estimate the SNR for the symbol stream from the jth transmit antenna based on equation (39).

DFE techniques

Fig. 6C is a block diagram of an embodiment of a channel MIMO/data processor 520z, which can implement the DFE space-time equalization technique. Channel MIMO/data processor 520z includes a space-time processor 610z coupled to RX data processor 620 and that performs DFE processing.

For DFE techniques, a forward receive processor 654 receives and processes the received vector of modulation symbolsr(n) to provide estimated modulation symbols for the data stream to be recovered. Forward receive processor 654 may implement the CCMI or MMSE technique described above, or some other linear spatial equalization technique. Summer 656 then feeds back the estimated values provided by processor 658The distortion component is combined with the estimated modulation symbols to provide "equalized" modulation symbols with the distortion component removed. Initially, the estimated distortion component is zero and the equalized modulation symbols are only estimated modulation symbols. The equalized modulation symbols from summer 656 are then demodulated and decoded by an RX data processor 620 to provide a decoded data stream.

The decoded data stream is then re-encoded and re-modulated by channel data processor 210x to provide re-modulated symbols, which are estimates of the modulation symbols at the transmitter. The channel data processor 210x performs the same processing as that performed at the transmitter for the data stream, as shown in fig. 2. The remodulated symbols from channel data processor 210x are provided to a feedback processor 658, which processes the symbols to derive estimated distortion components. Feedback processor 658 may implement a linear spatial equalizer (e.g., a linear transverse equalizer).

The resulting estimate of the vector of transmitted symbols at time n can be expressed as:

img id="idf0091" file="C0280965500371.GIF" wi="341" he="53" img-content="drawing" img-format="GIF"/formula (40)

Whereinr(n) is the vector of received modulation symbols, given in equation (25) above,is a symbol decision vector provided by the channel data processor 210x,M _f(k) (wherein-K)₁K ≦ 0) is used by the Forward receive processor 654 (K₁+1)-(N_T×N_R) A matrix of feedforward coefficients, ofM _b(k) (wherein 1. ltoreq. K. ltoreq.K₂) Is K used by feedback processor 658₂-(N_T×N_R) A feedback coefficient matrix. Equation (40) can also be expressed as:

img id="idf0093" file="C0280965500373.GIF" wi="186" he="23" img-content="drawing" img-format="GIF"/formula (41)

Whereinimg id="idf0094" file="C0280965500374.GIF" wi="264" he="20" img-content="drawing" img-format="GIF"/img id="idf0095" file="C0280965500375.GIF" wi="205" he="18" img-content="drawing" img-format="GIF"/

img id="idf0096" file="C0280965500376.GIF" wi="146" he="103" img-content="drawing" img-format="GIF"/And animg id="idf0097" file="C0280965500377.GIF" wi="154" he="101" img-content="drawing" img-format="GIF"/

If the coefficient matrix is found using the MMSE criterion, a chinese dictionary can be used in which the mean square error ∈ E ═ E-e ^H(k) e(k) At the smallest }Andwherein the error ise(k) Can be expressed as:

img id="idf0100" file="C0280965500383.GIF" wi="129" he="17" img-content="drawing" img-format="GIF"/

the MMSE solution of the feedforward filter can then be expressed as:

img id="idf0101" file="C0280965500384.GIF" wi="106" he="29" img-content="drawing" img-format="GIF"/formula (42)

Wherein

img id="idf0102" file="C0280965500385.GIF" wi="106" he="31" img-content="drawing" img-format="GIF"/

And isIs formed by N_R×N_RMade up of blocks (K)₁+1)N_R×(K₁+1)N_R) And (4) matrix.The (i, j) th block in (a) is given by:

img id="idf0105" file="C0280965500388.GIF" wi="347" he="44" img-content="drawing" img-format="GIF"/formula (43)

The MMSE solution for the feedback filter is:

img id="idf0106" file="C0280965500389.GIF" wi="302" he="53" img-content="drawing" img-format="GIF"/formula (44)

As in MMSE-LE above, an unbiased estimate is first determined by finding a conditional average of the transmitted symbol vectors:

formula (45)

Whereinimg id="idf0108" file="C02809655003811.GIF" wi="189" he="27" img-content="drawing" img-format="GIF"/. Secondly, the first step is to carry out the first,mean of ith elementIs represented as:

img id="idf0111" file="C02809655003814.GIF" wi="193" he="20" img-content="drawing" img-format="GIF"/

wherein v is_dfe，iiIs thatV _dfeThe ith diagonal element of (1). To form an unbiased estimate, the elements are, similar to the above-described mannerV _dfeThe orthogonal matrix of the orthogonal element negation of (a) is first defined as:

img id="idf0112" file="C02809655003815.GIF" wi="278" he="24" img-content="drawing" img-format="GIF"/formula (46)

The unbiased estimate is then expressed as:

img id="idf0113" file="C02809655003816.GIF" wi="351" he="27" img-content="drawing" img-format="GIF"/formula (47)

The resulting error covariance matrix is given by:

img id="idf0115" file="C02809655003818.GIF" wi="321" he="22" img-content="drawing" img-format="GIF"/ formula (48)

The SNR associated with the estimate of the symbol sent on the ith transmit antenna may be expressed as:

img id="idf0116" file="C0280965500391.GIF" wi="172" he="42" img-content="drawing" img-format="GIF"/formula (49)

According to the DFE technique, the decoded data stream is used to derive an estimate of the distortion produced by the information bits that have been decoded. If the data stream is decoded without error (or with little error), the distortion component can be accurately estimated and it is possible to effectively cancel the intersymbol interference generated by the already decoded information bits. The processing by forward receive processor 654 and feedback processor 658 are typically modulated at the same time to minimize the Mean Square Error (MSE) of the intersymbol interference in the equalized modulation symbols. The DFE processing is further detailed in the above-mentioned paper by Ariyavistakul et al.

Interference cancellation

Fig. 8 is a block diagram of an interference canceller 530x, which is an embodiment of interference canceller 530 of fig. 5. Within interference canceller 530x, the decoded data streams from channel MIMO/data processor 520 at each stage are re-encoded, interleaved, and re-modulated by channel data processor 210y to provide re-modulated symbols, which are estimates of the modulation symbols at the transmitter prior to MIMO processor and channel distortion. The channel data processor 210y performs the same processing (i.e., encoding, interleaving, and modulation) as was performed for the data stream at the transmitter. The demodulated symbols are then provided to a channel simulator 810, which processes the symbols with an estimated channel response to provide an estimate of the interference caused by the decoded data stream.

For non-dispersive channels, channel simulator 810 correlates the stream of remodulated symbols associated with the ith transmit antenna with a vectorMultiplication, vector ofIs the ith transmit antenna and N for which the data stream is being recovered_REstimation of the channel response between each of the root receive antennas. Vector quantityMay be expressed as:

img id="idf0120" file="C0280965500395.GIF" wi="84" he="109" img-content="drawing" img-format="GIF"/formula (50)

And is the estimated channel responseMatrix arrayIn the above-mentioned column (c),can be expressed as:

img id="idf0123" file="C0280965500398.GIF" wi="235" he="111" img-content="drawing" img-format="GIF"/ formula (51)

Matrix arrayPossibly provided by a channel MIMO/data processor 520 within the same stage.

If the stream of remodulated symbols corresponding to the ith transmit antenna is represented asThe estimated interference component generated by the recovered transmitted signalMay be represented as:

formula (52)

Interference vectorInner N_REach element corresponding to N_RAt each of the root receive antennas is a component of the received signal produced by the symbol stream transmitted on the ith transmit antenna. Each element of the vector represents an estimated component produced by a decoded data stream within a corresponding received modulation symbol stream. These components are for N_RA stream of received modulation symbols (i.e., vectors)r ^k) Interference with the remaining (undetected) transmitted signal and vector from the received signal with summer 812r ^kSubtracted (i.e., eliminated) to provide a modified vectorr ^k+1Wherein components from the decoded data stream have been removed. This cancellation can be expressed as shown above in equation (5). Modified vectorr ^k+1As input vectors are provided to the next receiver processing stage as shown in fig. 5.

For dispersive channels, the estimation of the channel transfer function vector as defined in equation (23) is used(where 0. ltoreq. k. ltoreq.L) instead of the vector. Then, the estimated interference vector at time n may be represented as:

formula (53)

WhereinIs a symbol that is retuned at time n. Equation (54) effectively convolves the demodulated symbols with the channel response estimates for each transmit-receive antenna pair.

For simplicity, the receiver architecture shown in fig. 5 provides a stream of (received or modified) modulation symbols to each receiver processing stage 510, with the interference components resulting from the previously removed (i.e., cancelled) decoded data stream. In the embodiment shown in fig. 5, each stage removes the interference component generated by the data stream decoded by that stage. In some other designs, the received modulation symbol stream may be provided to all stages, each stage possibly canceling interference components from all previously decoded data streams (which may have been provided from previous stages). It is also possible to skip interference cancellation for one or more stages (e.g., if the SNR of the data stream is high). Various modifications to the receiver architecture shown in fig. 5 may be made and are within the scope of the invention.

Deriving and reporting CSI

For simplicity, various aspects and embodiments of the present invention have been described above, wherein the CSI comprises SNR. In general, the CSI may comprise any type of information characterizing the communication link. Various information types may be provided as CSI, some examples of which are described below.

In one embodiment, the CSI includes a signal-to-noise-plus-interference ratio (SNR), which is obtained as a ratio of signal power to noise-plus-interference power. Each transmission channel typically used for data transmission (e.g., each transmitted data stream) is estimated and provided with an SNR, which may then provide an aggregate SNR for a number of transmission channels. The SNR estimate may be quantized to a value having a certain number of bits. In an embodiment, the SNR estimate is mapped to an SNR index using, for example, a look-up table.

In another embodiment, the CSI includes a signal power and an interference plus noise power. These two components may be obtained and provided separately for each transmission channel used for data transmission.

In yet another embodiment, the CSI includes signal power, interference power, and noise power. These three components may be obtained and provided separately for each transmission channel used for data transmission.

In yet another embodiment, the CSI includes the signal-to-noise ratio plus a list of interference powers for each observable interference term. This information may be obtained and provided for each transmission channel used for data transmission.

In yet another embodiment, the CSI comprises signal components in the form of a matrix (e.g., N for all transmit-receive antenna pairs)_T×N_RComplex term) and a matrix-form noise plus interference component (e.g., N)_T×N_RA plurality of terms). The transmitter unit may then appropriately combine the signal components and the noise-plus-interference components for the appropriate transmit-receive antenna pairs to derive a quality for each transmission channel used for data transmission (e.g., a post-processed SNR for each transmitted data stream received at the receiver unit).

In yet another embodiment, the CSI includes a data rate indicator for the transmitted data stream. It is first possible to determine the quality of the transmission channel used for data transmission (e.g., based on an estimated SNR for the transmission channel) and then identify the data rate corresponding to the determined channel quality (e.g., based on a look-up table). The identified data rate represents the maximum data rate that may be transmitted on the transmission channel at the desired performance level. The data rate is then mapped to and represented by a Data Rate Indicator (DRI), which can be efficiently encoded. For example, if the transmitter of each transmit antenna supports (up to) seven possible data rates, it is possible to represent DRI with a 3-bit value, where zero may represent data rate 0 (i.e., no transmit antenna is used) and 1 through 7 are used to represent seven different data rates. In typical implementations, the quality measure (e.g., SNR estimate) is directly mapped to the DRI according to, for example, a look-up table.

In yet another embodiment, the CSI includes an indication of a particular processing scheme to be used at the transmitter unit for each transmitted data stream. In this embodiment, the indicator may identify a particular coding scheme and a particular modulation scheme to be used to transmit the data stream in order to achieve a desired level of performance.

In yet another embodiment, the CSI includes a differential indicator for transmitting a particular measure of channel quality. First, the SNR or DRI or some other quality measure of the transmission channel is determined and reported as a reference measurement value. Thereafter, the quality of the transmission channel continues to be monitored and the difference between the last reported measurement and the current measurement is determined. The difference may then be quantized into one or more bits, and the quantized difference is mapped to and represented by the difference indicator and then reported. The differential indicator may indicate that the last reported measurement was increased or decreased by a particular step size (or that the last reported measurement was maintained). For example, the differential indicator may indicate that (1) the SNR observed for a particular transmission channel has increased or decreased by a particular step, or (2) the data rate should be adjusted by a particular amount, or some other change. Reference measurements may be periodically issued to ensure that errors in the differential indicators and/or erroneous receptions of these indicators do not accumulate.

Other forms of CSI may also be used and are within the scope of the invention. Generally, the CSI includes sufficient information, in whatever form, that can be used to adjust the processing at the transmitter so that a desired level of performance is obtained for the transmitted data stream.

The CSI may be derived from signals transmitted from the transmitter unit and received at the receiver unit. In an embodiment, the CSI is derived from a pilot reference included within the transmitted signal. Alternatively or additionally, the CSI may be derived from data included within the transmitted signal.

In yet another embodiment, the CSI includes one or more signals transmitted on the reverse link from the receiver unit to the transmitter unit. In some systems, there may be some degree of correlation between the forward and reverse links (e.g., in a Time Division Duplex (TDD) system in which the uplink and downlink share the same frequency band in a time division multiplexed manner). In these systems, the quality of the forward link may be estimated (to the requisite accuracy) based on the quality of the reverse link, which may be estimated based on a signal (e.g., a pilot signal) transmitted from the receiver unit. The pilot signal would then represent a means for the transmitter to estimate the CSI observed by the receiver unit.

The signal quality may be estimated at the receiver unit according to various techniques. Some of these techniques are described in the following patents, which are assigned to the assignee of the present application and are incorporated herein by reference:

U.S. Pat. No. 5799005 entitled "SYSTEM AND METHOD FOR detecting recording POWER AND PATH LOSS IN A CDMA COMMUNICATION SYSTEM" published 25/8 IN 1998,

U.S. Pat. No. 5903554 entitled "METHOD AND APPARATUS FOR MEASURING LINKQUALITY IN A SPREAD SPECTRUM COMMUNICATION SYSTEM" published 5/11/1999,

US patent Nos. 5056109 AND 5265119, both entitled "METHOD AND APPATUS FOR TRANSMISSION POWER IN A CDMA CELLULAR MOBILE TELEPHONE SYSTEM", published respectively on 10/8/1993 AND 11/23/1993, AND

U.S. Pat. No. 6097972 entitled "METHOD AND APPARATUS FOR PROCESSING POWER SYSTEM SIGNALS IN A CDMA MOBILE TELEPHONE SYSTEM" published 8/1/2000.

Various information types FOR CSI AND various CSI reporting mechanisms are also described in U.S. patent application serial No. 08/963386 entitled "METHOD AND APPARATUS FOR HIGH RATE PACKET datatansransisiton", filed on 3.11.1997 AND assigned to the assignee of the present application; also described in "TIE/EIA/-856 cdma2000 High Rate Packet Data Air Interface Specification", both of which are incorporated herein by reference.

The CSI may be reported back to the transmitter using various CSI transmission mechanisms. For example, the CSI may be transmitted in full, differential, or a combination of both. In an embodiment, the CSI is reported periodically and a differential update is sent based on previously sent CSI. In another embodiment, the CSI is sent only when there is a change (e.g., if the change exceeds a certain threshold), which may reduce the effective rate of the feedback channel. For example, the SNRs may be sent back only when they change (e.g., differentially). For OFDM systems (with or without MIMO), correlation in the frequency domain may be employed to allow a reduction in the amount of CSI to be fed back. For an OFDM system, for example, if the SNR of a particular spatial subchannel corresponding to NM frequency subchannels is the same, the SNR may be reported along with the first and last frequency subchannels for which this condition holds. Other compression and feedback channel error recovery techniques for reducing the amount of CSI to be fed back are also possible and are within the scope of the invention.

Referring back to fig. 1, the CSI (e.g., channel SNR) determined by RX MIMO processor 156 is provided to a TX data processor 162, which processes the CSI and provides processed data to one or more modulators 154. The modulator also conditions the processed signal and transmits CSI back to transmitter system 110 over the reverse channel.

At system 110, the transmitted feedback signal is received by antenna 124, demodulated by demodulator 122, and provided to a RX data processor 132. RX data processor 132 performs the inverse of TX data processor 162 and recovers the reported CSI, which is then provided to and used by TX data processor 114 and TX MIMO processor 120 to adjust their processing.

Transmitter system 110 may adjust (i.e., adapt) its processing based on the CSI (e.g., SNR information) from receiver system 150. For example, the coding of each transmission channel may be adjusted so that the information bit rate matches the transmission performance supported by the channel SNR. Furthermore, the modulation scheme of the transmission channel may be selected according to the channel SNR. Other processing (e.g., interleaving) may also be accommodated and is within the scope of the invention. Adjusting the processing of each transmission channel according to the SNR determined by the channel enables the MIMO system to achieve high performance (i.e., high throughput or bit rate for a particular level of performance). The adaptation process may be applied to a single carrier MIMO system or a multi-carrier based MIMO system (e.g., a MIMO system using OFDM).

The adjustment within the code and/or selection of the modulation scheme may be accomplished at the receiver system according to a number of techniques, one of which is described in the aforementioned U.S. patent application serial No. 09/776975.

MIMO system operation scheme

Various operating schemes may be implemented for MIMO systems that employ the adaptive transmitter processing (depending on available CSI) and successive cancellation receiver processing techniques described above. Some such operating schemes are described in further detail below.

In one operating scheme, the coding and modulation scheme for each transmission channel is selected based on the transmission performance of the channel as determined by the SNR of the channel. As described in detail below, this scheme may provide improved performance when used in conjunction with successive cancellation receiver processing techniques. When there is a large difference between the worst-case and best-case transmission channels (i.e., transmit-receive antenna pairs), the codes may be selected to introduce sufficient redundancy to allow the receiver system to recover the original data stream. For example, the worst transmit antenna may have a poor SNR associated with it at the receiver output. A Forward Error Correction (FEC) code is then selected that is strong enough to allow the symbols transmitted from the worst-case transmit antennas to be correctly received at the receiver system. In practice, the improved error correction performance comes at the cost of increased redundancy, which means a loss of overall throughput. There is therefore a trade-off between reduced throughput and increased redundancy when using FEC coding.

When the SNR of each recovered transmitted signal is provided to the transmitter, a different coding and/or modulation scheme may be used for each transmitted signal. For example, a particular coding and modulation scheme may be selected for each transmitted signal based on its SNR such that error rates associated with the transmitted signals are approximately equal. Thus, throughput is not dictated by the SNR of the worst case transmitted signal.

Consider, for example, a 4 x 4 MIMO system with 4 transmit antennas and 4 receive antennas and employing the above-described successive cancellation receiver processing technique. The SNR of the four transmitted signals is 5dB, 8.5dB, 13dB, and 17.5dB for this example. If the same coding and modulation scheme were used for all four transmit signals, the selected scheme would be dominated by transmit signals with 5dB SNR. Using the information given in table 1, each transmit antenna will employ a coding rate of 3/4 and QPSK modulation, with a total modulation efficiency of 6 information bits/symbol, or 1.5 information bits/symbol/transmitted signal.

Depending on the available CSI, the transmitter may select the following coding and modulation schemes for the four transmitted signals, as shown in table 2.

TABLE 2

SNR(dB)	Coding rate	Modulation code element	Information bits/symbols #
				5	3/4	QPSK	1.5
8.5	5/8	16-QAM	2.5
				13	7/12	64-QAM	3.5
17.5	5/6	64-QAM	5

By adjusting the coding and modulation scheme at the transmitter according to the available CSI, the effective modulation efficiency obtained is twice as much as 12.5 bit symbols over 6 bits/symbol without CSI. Since the coding and modulation schemes are selected to achieve this performance level, the decoded error rate for each transmitted signal will be approximately equal.

By adaptive processing at the transmitter system based on available CSI, successive cancellation receiver processing techniques may be altered to take advantage of the fact that the bit error rates of the transmitted signals are approximately equal. If the coding and modulation schemes used on each transmitted signal provide an equivalent decoded error rate, the ordering step (i.e., highest to lowest SNR) may be omitted from the receiver processing, which may simplify the processing. In practical implementations, there may be slight differences in the decoded error rate of the transmitted signal. In this case, it is possible to align the SNR of the transmitted signal (after linear or non-linear processing) and to select the best post-processing SNR for first detection (i.e., demodulation and decoding), as described above.

Depending on the CSI available at the transmitter, the throughput is no longer dominated by the longest case transmitted signal, since the coding and modulation scheme is selected to provide a particular level of performance (e.g., a particular BER) on each transmission channel depending on the SNR of the channel. Since FEC coding is applied independently to each transport channel, the minimum amount of redundancy required to meet the target performance level is used and throughput is maximized. The performance that can be achieved with CSI-based adaptive transmitter processing and successive cancellation receiver processing competes with the performance of the full-CSI processing scheme (thereby fully characterizing the available transmit-receive antenna pairs) under certain operating conditions, as described in detail below.

In another operating scheme, the transmitter is not provided with the SNR achieved for each transmission channel, but is provided with a single value representing the average SNR for all transmission channels, or perhaps some information indicating which antennas are to be used for data transmission. In this scheme, the transmitter may use the same coding and modulation scheme for all transmit antennas used for data transmission, which may be N antennas_TThe root may use a subset of the transmit antennas. When the same coding and modulation scheme is used on all transmit antennas, performance may be compromised. This is because the overall performance of the successive cancellation receiver processing technique depends on the ability to decode each transmitted signal without error. This correct detection is important for effectively canceling the interference generated by the recovered transmitted signal.

By using the same coding and modulation scheme for all transmitted signals, the recovered transmitted signal with the worst SNR will have the highest decoded error rate. This ultimately limits the performance of the MIMO system, since the coding and modulation schemes are chosen such that the error rate associated with the worst-case transmitted signal meets the overall error rate requirements. To improve efficiency, additional receive antennas may be used to provide improved error rate performance on the first recovered transmitted signal. By using more receive antennas than transmit antennas, the diversity order of the error rate performance of the first recovered transmit signal is (N)_R-N_T+1) and improved reliability.

In yet another operating scheme, the transmit data stream is "cycled" through all available transmit antennas. This scheme provides SNR statistics for each recovered transmitted signal since the transmitted data does not belong to the worst case transmission channel but to all transmission channels. The decoder associated with a particular data stream is effectively given by a "soft decision" that represents the average over all possible transmit-receive antenna pairs. This protocol is described in detail in European patent application Ser. No. 99302692.1 entitled "WIRELESS COMMUNICATIONS SYSTEMHAVING A SPACE-TIME ARCHITECTURE EMPLOYING MULTI-ELEMENT ANTENNAS AT BOTHEE TRANSMITTER AND RECEIVER" and incorporated herein by reference.

Successive cancellation receiver processing techniques enable the MIMO system to use the additional dimensions established by using multiple transmit and receive antennas, which is a major benefit of using MIMO. Depending on the characteristics of the MIMO channel, the received signal may be processed with a linear spatial equalization technique (e.g., CCMI or MMSE) or a space-time equalization technique (e.g., MMSE-LE, DFE, or MLSE). When used in conjunction with adaptive transmitter processing based on available CSI, successive cancellation receiver techniques may allow the same number of modulation symbols to be transmitted for each slot, as for MIMO systems that employ full CSI.

Other linear and non-linear receiver processing techniques may also be used in conjunction with the successive cancellation receiver processing technique and the adaptive transmitter processing technique, which are within the scope of the present invention. Similarly, fig. 6A through 6C illustrate embodiments of three receiver processing techniques that can process a MIMO transmission and determine the characteristics (i.e., SNR) of the transmission channel. Other receiver designs based on the techniques presented herein, as well as other receiver processing techniques, are also contemplated as falling within the scope of the present invention.

Linear and non-linear receiver processing techniques (e.g., CCMI, MMSE-LE, DFE, MLSE, and others) can also be used in a straightforward manner without adaptive processing at the transmitter when only the total received signal SNR, or the achievable overall throughput estimated from such SNR, is fed back. In an implementation, the modulation format is determined based on the received SNR estimate or estimated throughput, and the same modulation format is used for all transmission channels. This approach may reduce the overall system throughput, but may also greatly reduce the amount of information sent back on the reverse link.

System performance

Improvements in system performance may be achieved by using successive cancellation receiver processing techniques and adaptive transmitter processing techniques based on available CSI. The system throughput with CSI feedback may be calculated and compared to the throughput of full CSI feedback. System throughput may be defined as:

img id="idf0133" file="C0280965500471.GIF" wi="135" he="43" img-content="drawing" img-format="GIF"/formula (54)

Wherein gamma is_iIs the SNR per received modulation symbol. The SNR of some receiver processing techniques has been summarized above.

Fig. 9A illustrates the SNR improvement of a 4 x 4 MIMO channel using a successive cancellation receiver processing technique. Results the results were obtained from the computer simulation. In the simulation, the following assumptions were made: (1) independent Rayleigh (Rayleigh) fading channels between receive-transmit antenna pairs (i.e., no array correlation), (2) total interference cancellation (i.e., no decision error in the decoding process and accurate channel estimates available at the receiver). In practical implementations, the channel estimates are not completely accurate and a compensation factor may be used in the modulation scheme selected for each transmitted data stream. In addition, certain decision errors may occur in detecting each transmitted data stream. This probability can be reduced if the independently transmitted data is separately encoded, which would then allow the receiver to independently decode the data streams, thereby reducing the probability of decision errors. In this case, the decoded data is re-encoded to construct an interference estimate for use in successive interference cancellation.

As shown in fig. 9A, the first recovered transmitted signal has the worst SNR distribution. Each subsequent recovered transmitted signal has an improved SNR profile and the last recovered transmitted signal (i.e., the fourth in this example) has the best overall SNR profile. Also shown is the distribution of the average SNR formed by summing the SNRs of the individual transmit signals and dividing by four. The SNR distribution of the first recovered transmitted signal gives an SNR distribution that is achieved without successive spatial equalization and interference cancellation. When comparing the SNR distribution of the first recovered transmitted signal with the average SNR distribution, it can be seen that the spatial equalization and interference cancellation techniques improve the effective SNR at the receiver.

Fig. 9B shows the average throughput for several receive processing techniques, including (1) linear spatial equalization techniques (without interference cancellation), (2) spatial equalization and interference cancellation techniques, and (3) full-CSI techniques. For each of these schemes, all or a portion of the CSI for all of the transmitted signals is provided to the transmitter, and the data for each transmitted signal is encoded and modulated according to the SNR. For the graph shown in fig. 9B, the CCMI and MMSE techniques are used for the linear spatial equalization technique.

Fig. 9B shows the theoretical capacity achieved with CSI processing when decomposing the MIMO channel into eigenmodes (curve 920). Fig. 9B also shows the throughput for both the CCMI technique (curve 924) and the MMSE technique (curve 922), both techniques having partial-CSI without interference cancellation, with lower throughput than the upper capacity limit (curve 920).

Since capacity is proportional to SNR, as shown in equation (20), and SNR is improved by using successive interference cancellation, the average capacity is improved using spatial equalization and interference cancellation techniques. By using spatial equalization (with CCMI) and interference cancellation techniques and partial CSI, the throughput (curve 926) is improved over the scheme with spatial equalization only (curves 922 and 924), the more the performance improves as the SNR increases. By using spatial equalization (with MMSE) and interference cancellation techniques and partial CSI, the throughput (curve 928) is equal to the upper capacity limit (curve 920), which represents significant system performance. Curve 920 assumes an ideal channel estimate and no decision error. The throughput estimation shown in fig. 9B is for successive spatial equalization and interference cancellation techniques with partial-CSI processing, which may degrade in practical implementations due to inadequate interference cancellation and detection errors.

Fig. 9C shows the average throughput of successive space-time equalization (with MMSE-LE) and interference cancellation techniques, where the adaptive transmitter processes CSI based on a 4 x 4 MIMO system. These curves are obtained by averaging over a large number of statistical realizations of the dispersive channel model (i.e., VehA). Fig. 9C shows the upper capacity limit (curve 930) and performance of the MMSE-LE technique with interference cancellation (curve 934) and without successive interference cancellation (curve 932). The throughput performance of MMSE-LE without successive interference cancellation technique (curve 932) degrades at higher SNR values. The throughput performance of MMSE-LE with successive interference cancellation techniques (curve 934) is close to the channel capacity, which represents a high performance level.

Elements of the transmitter and receiver system may be implemented with one or more of the following: a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a processor, a microprocessor, a controller, a microcontroller, a Field Programmable Gate Array (FPGA), a programmable logic device, other electronic units, or any combination thereof. Some of the functions and processes described above may also be implemented in software executing on a processor.

Certain aspects of the present invention may be implemented in a combination of software and hardware. The calculation of the symbol estimates for linear spatial equalization, space-time equalization, and channel SNR deviation may be performed, for example, in accordance with program code executing on a processor (controller 540 in fig. 5).

For simplicity, the receiver architecture shown in fig. 5 includes a number of receive processing stages, one for each data stream to be decoded. In some implementations, these multiple stages may be implemented with a single hardware unit or may be a single software module that executes repeatedly for each stage. In this way, the hardware or software may be time-shared to simplify the receiver design.

Subheadings are included herein for reference and to aid in locating particular sections. These headings are not intended to limit the scope of the concepts described therein under, and these concepts may have applicability in other sections throughout the entire specification.

The previous description of the preferred embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without the use of the inventive faculty. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims (57)

1. A method for processing data at a receiver unit in a multiple-input multiple-output (memo) communication system, comprising:

processing a plurality of input signals to provide a decoded data stream for one of the one or more symbol streams, the input signals including one or more symbol streams corresponding to the one or more data streams;

deriving a plurality of modified signals from the input signal and having components resulting from substantially removing the decoded data stream;

performing processing and selectively deriving each of one or more iterations, and wherein the input signal for each iteration after the first iteration is a modified signal from a previous iteration; and

channel state information indicative of characteristics of a multiple-input multiple-output channel used to transmit the data stream is determined, wherein the data stream is adaptively processed at the transmitter unit based in part on the channel state information.

2. The method of claim 1, wherein the deriving is omitted for a last iteration.

3. The method of claim 1, wherein the processing comprises

Processing an input signal according to a receive processing scheme to provide one or more symbol streams, an

Processing the selected one of the one or more symbol streams to provide a decoded data stream.

4. The method of claim 3, further comprising:

for each of the iterations it is desirable to,

estimating a quality of one of the one or more unprocessed symbol streams included within the input signal; and

one unprocessed symbol stream is selected for processing based on the estimated quality of the one or more unprocessed symbol streams.

5. The method of claim 4, wherein the quality of each unprocessed symbol stream is estimated based on a signal-to-noise-and-interference ratio.

6. The method of claim 4, wherein the unprocessed symbol stream with the best estimation quality is selected for processing.

7. The method of claim 3, wherein the receive processing scheme performs linear spatial processing on the input signal.

8. The method of claim 7, wherein the receive processing scheme employs a channel correlation matrix inversion technique.

9. The method of claim 7, wherein the receive processing scheme employs a minimum mean square error technique.

10. The method of claim 7, wherein the receive processing scheme employs a full-channel state information processing technique.

11. The method of claim 3, wherein the receive processing scheme performs space-time processing on the input signal.

12. The method of claim 11, wherein the receive processing scheme employs a minimum mean square error linear controlled equalizer technique.

13. The method of claim 11, wherein the receive processing scheme employs a decision feedback space-time equalizer technique.

14. The method of claim 11, wherein the receive processing scheme employs a maximum likelihood sequence estimator technique.

15. The method of claim 1, wherein the deriving comprises

Generating a remodulated symbol stream from the decoded data stream;

forming a plurality of interference signals from the remodulated symbol stream; and

the interfering signal is removed from the input signal to derive a modified signal as the input signal for a subsequent iteration.

16. The method of claim 15, wherein the interference signal is formed based on a channel coefficient matrix H representing characteristics of a multiple-input multiple-output channel.

17. The method of claim 1, further comprising:

channel state information is transmitted from the receiver unit to the transmitter unit.

18. The method of claim 1, wherein the channel state information comprises a signal-to-noise-and-interference ratio estimate for each of one or more transmission channels making up a multiple-input multiple-output channel.

19. The method of claim 1, wherein the channel state information includes characteristics of one or more transmission channels that make up a multiple-input multiple-output channel.

20. The method of claim 1, wherein the channel state information comprises an indication of a data rate, a particular data rate being supported by each of one or more transport channels used for data transmission.

21. The method of claim 1, wherein the channel state information comprises an indication of a processing technique to be used for each of one or more transport channels.

22. The method of claim 1, wherein the channel state information comprises signal measurements and noise-plus-interference measurements for one or more transmission channels.

23. The method of claim 1, wherein the channel state information comprises signal measurements, noise measurements, and interference measurements for one or more transmission channels.

24. The method of claim 1, wherein the channel state information comprises signal-to-noise ratio and interference measurements for one or more transmission channels.

25. The method of claim 1, wherein the channel state information comprises signal components and noise-plus-interference components of one or more transmission channels.

26. The method of claim 1, wherein the channel state information comprises an indication of a change in a characteristic of one or more transmission channels.

27. The method of claim 1, wherein channel state information is determined at the receiver and reported to the transmitter unit.

28. The method of claim 1, wherein the channel state information is determined at the transmitter unit based on one or more signals transmitted by the receiver unit.

29. The method of claim 1, wherein each data stream is encoded at the transmitter unit according to a coding scheme, wherein the coding scheme is selected based on channel state information for a transmission channel used to transmit the data stream.

30. The method of claim 29, wherein each data stream is independently coded according to a coding scheme, wherein the coding scheme is selected based on channel state information for a transmission channel used to transmit the data stream.

31. The method of claim 29, wherein each data stream is further modulated according to a modulation scheme, wherein the modulation scheme is selected based on channel state information for a transmission channel used to transmit the data stream.

32. The method of claim 31, wherein the coding and modulation scheme are selected at a transmitter unit based on channel state information.

33. The method of claim 31, wherein the coding and modulation scheme are indicated by channel state information.

34. The method of claim 3, wherein processing the selected symbol stream comprises

Demodulating the stream of symbols to provide demodulated symbols, an

The demodulated symbols are decoded to provide a decoded data stream.

35. The method of claim 34, wherein processing the selected symbol stream further comprises

Deinterleaves the demodulated symbols, wherein the deinterleaved symbols are decoded to provide a decoded data stream.

36. The method of claim 1, wherein the multiple-input multiple-output system employs Orthogonal Frequency Diversity Modulation (OFDM).

37. The method of claim 36, wherein the processing at the receiver unit is performed independently for each of a plurality of frequency subchannels.

38. A method of processing data at a receiver unit in a multiple-input multiple-output communication system, comprising:

receiving a plurality of signals through a plurality of receiving antennas;

processing the received signal in accordance with a receive processing scheme to provide a plurality of symbol streams corresponding to a plurality of transmitted data streams;

processing the selected one of the symbol streams to provide a decoded data stream;

forming a plurality of interference signals from the decoded data stream;

deriving a plurality of modified signals from the received signal and the interfering signal;

processing the received signal and the selected symbol stream and selectively forming and deriving one or more iterations, one for each transmitted data stream to be decoded, with a first iteration being performed on the received signal and each subsequent iteration being performed on a modified signal from a previous iteration; and

channel state information indicative of characteristics of a multiple-input multiple-output channel used to transmit the data stream is determined, wherein the data stream is adaptively processed at the transmitter unit based in part on the channel state information.

39. A method of processing data at a receiver unit in a multiple-input multiple-output communication system, comprising:

at the location of the receiver unit(s),

receiving a plurality of signals via a plurality of receive antennas, wherein each received signal comprises a combination of one or more signals transmitted from a transmitter unit,

the received signal is processed in accordance with a successive cancellation receiver processing technique to provide a plurality of decoded data streams transmitted from the transmitter unit,

determining channel state information indicative of characteristics of a multiple-input multiple-output channel used for transmitting the data stream, and

transmitting the channel state information back to the transmitter unit; and

at the location of the transmitter unit, it is,

each data stream is adaptively processed in accordance with received channel state information prior to transmission on the multiple-input multiple-output channel.

40. The method of claim 39, wherein the successive cancellation receiver processing technique performs a plurality of iterations to provide decoded data streams, one iteration for each decoded data stream.

41. The method of claim 40, wherein each iteration comprises

The plurality of input signals are processed in accordance with a linear or non-linear processing scheme to provide one or more symbol streams,

processing the selected one of the one or more symbol streams to provide a decoded data stream, an

A plurality of modified signals are derived from the input signal and have components due to substantial removal of the decoded data stream, wherein the input signal for a first iteration is the received signal and the input signal for each subsequent iteration is the modified signal from the previous iteration.

42. The method of claim 39, wherein the channel state information comprises a signal-to-noise-and-interference ratio for each of one or more transmission channels making up a multiple-input multiple-output channel.

43. The method of claim 39, wherein the channel state information comprises an indication of a data rate supported by each of one or more transport channels comprising a multiple-input multiple-output channel.

44. The method of claim 39, wherein the channel state information comprises an indication of a processing technique to be used for each of one or more transport channels comprising a multiple-input multiple-output channel.

45. The method of claim 39, wherein the adaptive processing at the transmitter unit comprises

The data stream is encoded in accordance with a coding scheme selected based on channel state information associated with the data stream.

46. The method of claim 45, wherein the adaptive processing at the transmitter unit further comprises

The encoded data stream is modulated in accordance with a coding scheme selected based on channel state information associated with the data stream.

47. A multiple-input multiple-output communication system, comprising:

a receiver unit comprising

A plurality of front-end processors for processing a plurality of received signals to provide a plurality of symbol streams,

at least one receive processor coupled to the front-end processor for processing the symbol stream in accordance with a successive cancellation receiver processing scheme to provide a plurality of decoded data streams, and for deriving channel state information indicative of characteristics of a multiple-input multiple-output channel used to transmit the data streams, an

A transmit data processor operatively coupled to the receive processor for processing the channel state information for transmission back to the transmitter unit; and

a transmitter unit comprising

At least one demodulator for receiving and processing one or more signals from the receiver unit to recover transmitted channel state information, an

A transmit data processor for adaptively processing the data for transmission back to the receiver unit based on the recovered channel state information.

48. A receiver in a multiple-input multiple-output communication system, comprising:

a plurality of front-end processors for processing a plurality of received signals to provide a plurality of received symbol streams;

at least one receive processor coupled to the front-end processor for processing the stream of receive symbols to provide a plurality of decoded data streams, each receive processor comprising a plurality of processing stages, each stage for processing an input symbol stream to provide a corresponding decoded data stream and channel state information associated with the decoded data stream, and for selectively providing a modified symbol stream for a subsequent stage, wherein the input symbol stream for each stage is either the stream of receive symbols or the modified symbol stream from a preceding stage; and

a transmit processor for receiving and processing channel state information associated with the decoded data stream for transmission from the receiver unit, wherein the data stream is adaptively processed based on the channel state information prior to transmission.

49. The receiver unit of claim 48, wherein each processing stage other than the last stage comprises

A channel processor for processing the input symbol stream to provide a decoded data stream, an

An interference canceller to derive a modified symbol stream from the decoded data stream and the input symbol stream.

50. The receiver unit of claim 49, wherein each channel processor comprises

An input processor for processing the input symbol stream to provide a recovered symbol stream, an

A data processor is operative to process the recovered symbol stream to provide a decoded data stream.

51. The receiver unit of claim 50, wherein each input processor comprises

A first processor for processing the input symbols in accordance with a linear or non-linear reception processing scheme to provide a stream of recovered symbols, an

A channel quality estimator for estimating a quality of the recovered symbol stream.

52. The receiver unit of claim 51, wherein the estimated quality comprises a signal-to-noise-and-interference ratio.

53. The receiver unit of claim 51, wherein the channel quality estimator is further operative to provide an indication of supported data rates for the recovered symbol streams based on the quality estimates.

54. The receiver unit of claim 51, wherein the channel quality estimator is further operative to provide an indication of a processing technique to be used at the transmitter unit for the recovered symbol stream based on the quality estimate.

55. The receiver unit of claim 51, wherein the estimated quality comprises an error signal indicative of a noise plus interference level detected at an output of the receiver unit.

56. The receiver unit of claim 51, wherein the first processor performs linear spatial processing on the input symbol stream.

57. The receiver unit of claim 51, wherein the first processor performs space-time processing on the input symbol stream.