HK1117974A - Fine timing acquisition - Google Patents

Description

Precision timing acquisition

Claiming priority according to 35U.S. § 119

This patent application claims priority from provisional application 60/660,901 filed on 10.3.2005 and assigned to the present applicant, which is expressly incorporated herein by reference.

Background

The present invention relates generally to data communications, and more particularly to synchronization in information transmission systems using Orthogonal Frequency Division Multiplexing (OFDM).

In the OFDM system, a transmitter acquires a modulation symbol by processing data, and further generates an OFDM symbol by performing modulation on the modulation symbol. The transmitter then conditions and transmits the OFDM symbols over the communication channel. An OFDM system may use some sort of transmission structure to transmit data in superframes, where each superframe has a duration. Different types of data (e.g., traffic/packet data, overhead/control data, pilot, etc.) may be sent in different portions of each super-frame. Each superframe may be split into multiple frames. The term "pilot" generally refers to data and/or transmissions that are known a priori to the transmitter and receiver.

The receiver typically needs to acquire accurate frame and symbol timing in order to properly recover the data transmitted by the transmitter. For example, the receiver may need to know about each superframe and the beginning of a frame in order to properly recover the different types of data transmitted in the superframe. In general, the receiver does not know the time at which each OFDM symbol was transmitted by the transmitter, nor the propagation delay introduced by the communication channel. Therefore, the receiver needs to determine the timing of each OFDM symbol received via the communication channel in order to perform correct complementary OFDM demodulation on the received OFDM symbols.

In this disclosure, the term synchronization refers to the process performed by the receiver to acquire frame and symbol synchronization. The receiver may also perform other tasks such as frequency error estimation and channel estimation. Synchronization may be performed at different times to improve timing and correct for variations in the channel. Performing synchronization quickly makes it easier to acquire the signal.

Disclosure of Invention

In one aspect, the present disclosure provides a method for synchronizing timing of a receiver with a received Orthogonal Frequency Division Multiplexing (OFDM) signal. In one step, a first timing acquisition is performed with a first received Time Division Multiplexed (TDM) pilot to determine a coarse timing (coarse timing) estimate of a received OFDM signal. A second timing acquisition is performed with a second TDM pilot to determine a fine timing estimate for an OFDM symbol in the received OFDM signal. In a second timing acquisition process, the accumulated energy of the channel taps over a detection window is determined and the trailing edge of the accumulated energy curve is detected. In an alternative embodiment, one or both of the leading edge and the trailing edge may be determined during the second timing acquisition. Adjusting a Fourier Transform (FT) collection window position for a subsequent OFDM symbol in accordance with the second timing acquisition step.

In one aspect, an OFDM system for synchronizing the timing of a receiver with a received OFDM symbol is disclosed. The OFDM system includes: means for performing a first timing acquisition, means for performing a second timing acquisition, and means for adjusting a DFT collection window position. The means for performing a first timing acquisition with the first received TDM pilot determines a coarse timing estimate of the received OFDM signal. The means for performing a second timing acquisition with a second TDM pilot determines a fine timing estimate of the received OFDM signal. The means for performing the second timing acquisition includes means for determining and means for detecting. The means for determining the accumulated energy of the plurality of channel taps within the detection window for the plurality of start positions forms an accumulated energy curve. The means for detecting finds the trailing edge of the accumulated energy curve. Means for adjusting the FT collection window position for a subsequent OFDM symbol operates in accordance with the output of the means for performing the second timing acquisition.

In one aspect, a method for synchronizing timing of a receiver with a received signal is disclosed. In one step, a first timing acquisition is performed to determine a coarse timing estimate of the received signal. A second timing acquisition is performed with the TDM pilot to determine a fine timing estimate for a symbol of the received signal. The second timing acquisition determines an accumulated energy for a plurality of channel taps within a detection window for a plurality of start positions to form an accumulated energy curve. In addition, the second timing acquisition also detects a trailing edge of the accumulated energy curve. The step of determining the accumulated energy and the step of detecting the trailing edge are performed at least partially simultaneously in time for a particular channel tap of the plurality of channel taps. The FT collection window position is adjusted for subsequent symbols according to the step for performing the second timing acquisition.

In one aspect, a communication device for synchronizing timing of a receiver with a received signal is disclosed. The communication device includes a processor and a memory coupled to each other. The processor is configured to perform at least the following steps:

1. a first timing acquisition is performed with a first received Time Division Multiplexed (TDM) pilot to determine a coarse timing estimate of the received OFDM signal.

2. A second timing acquisition is performed with a second TDM pilot to determine a fine timing estimate of the received OFDM signal. The step for performing a second timing acquisition further comprises the sub-steps of: the accumulated energy of a plurality of channel taps within the detection window is determined for a plurality of start positions to form an accumulated energy curve, and a trailing edge of the accumulated energy curve is detected.

3. A Fourier Transform (FT) collection window position is adjusted for a subsequent OFDM symbol in accordance with the step for performing the second timing acquisition.

Brief Description of Drawings

The disclosure is described in conjunction with the following figures:

FIG. 1 is a block diagram of an embodiment of a base station and a wireless receiver in an Orthogonal Frequency Division Multiplexing (OFDM) system;

fig. 2A and 2B are block diagrams of embodiments of a superframe structure for an OFDM system;

FIG. 3 is a diagram of an embodiment of a frequency domain representation of Time Division Multiplexed (TDM) pilot 2;

FIG. 4 is a block diagram of an embodiment of a Transmit (TX) data and pilot processor;

FIG. 5 is a block diagram of an embodiment of an OFDM modulator;

FIG. 6 is a diagram of an embodiment of a time domain representation of a TDM pilot 2;

FIG. 7 is a block diagram of an embodiment of a synchronization and channel estimation unit;

FIG. 8 is an illustration of an embodiment of an operational timeline for Fine Timing Acquisition (FTA);

FIG. 9 is a block diagram of an embodiment of a symbol timing detector;

FIGS. 10A-10D are diagrams illustrating processing of pilot-2 OFDM symbols;

FIG. 11 is a diagram of an embodiment of a pilot transmission scheme incorporating TDM and FDM pilots;

FIG. 12 is a block diagram of an embodiment of logic circuitry for removing pilot symbol modulation;

FIG. 13 is a block diagram of an embodiment of normal operation for implementing timing synchronization;

fig. 14 is a block diagram of an embodiment of a fixed point implementation of first phase FAP detection in the FTA;

fig. 15 is a flow diagram of an embodiment of a process for illustrating three phases of a FAP detection algorithm;

fig. 16 is a block diagram of an embodiment of the update step in the third phase of FAP detection;

FIG. 17 is a block diagram of an embodiment for initializing Data Mode Time Tracking (DMTT);

FIG. 18 is a block diagram of an embodiment of an OFDM system that synchronizes receiver timing to a received OFDM signal; and

fig. 19 is a flow diagram of an embodiment of a process for synchronizing receiver timing with a received OFDM signal.

In the drawings, the same reference numerals will be used for the same components and/or features.

Detailed Description

The following description is merely provided as one or more preferred exemplary embodiments, and is not intended to limit the scope, applicability, or configuration of the invention. In contrast, the following description of one or more preferred embodiments provides those skilled in the art with an enabling description for implementing a preferred exemplary embodiment of the invention. It being understood that various changes may be made in the function and arrangement of elements without departing from the spirit and scope of the invention as set forth in the appended claims.

Specific details are set forth in the following description in order to provide a thorough understanding of the embodiments. It will be understood by those of ordinary skill in the art that the embodiments may be practiced without these specific details. For example, circuits are shown in block diagrams in order not to obscure the embodiments in unnecessary detail. In other instances, well-known circuits, processes, algorithms, structures, and techniques have not been shown in unnecessary detail in order to avoid obscuring the embodiments.

It is also noted that the embodiments may be described as a process which is depicted as a flowchart, a flow diagram, a structure diagram, or a block diagram. Although a flowchart may describe the operations as a sequential process, many of the operations can be performed in parallel or concurrently. In addition, the order of the operations may be rearranged. The process terminates when the operation ends, but the process may have additional steps not included in the figure. A process may correspond to a method, a function, a procedure, a routine, a subroutine, etc. When a process corresponds to a function, its termination will correspond to the return result of the function being provided to the calling function or the main function.

In addition, the term "storage medium" disclosed herein may represent one or more devices for storing data, including Read Only Memory (ROM), Random Access Memory (RAM), magnetic RAM, core memory, magnetic disk storage media, optical storage media, flash memory devices, and/or other machine-readable media for storing information. The term "machine-readable medium" includes, but is not limited to portable or fixed storage devices, optical storage devices, wireless channels and various other mediums capable of storing, containing or carrying one or more instructions and/or data.

Furthermore, the embodiments may be implemented by hardware, software, firmware, middleware, microcode, hardware description languages, or any combination thereof. When implemented in software, firmware, middleware or microcode, the program code or code segments to perform the necessary tasks may be stored in a machine-readable medium such as a storage medium. These necessary tasks may be performed by one or more processors. A code segment or machine-executable instruction may represent a procedure, a function, a subprogram, a program, a routine, a subroutine, a module, a software package, a class, or any combination of instructions, data structures, or program statements. A code segment may be coupled to another code segment or a hardware circuit by passing and/or receiving information, data, arguments, parameters, or memory contents. Information, arguments, parameters, data, etc. may be passed, forwarded, or transmitted via any suitable means including memory sharing, message passing, token passing, network transmission, etc.

The synchronization techniques described herein may be used for various multicarrier systems as well as the downlink and uplink. The downlink (or forward link) refers to the communication link from the base stations to the radio receivers, and the uplink (or reverse link) refers to the communication link from the radio receivers to the base stations. For clarity, these techniques are described below with respect to the downlink in an Orthogonal Frequency Division Multiplexing (OFDM) system. The pilot detection structure is well suited for broadcast systems, but it can also be used for non-broadcast systems.

An improved method and system for timing synchronization after initial acquisition in an OFDM system is disclosed herein. For initial synchronization acquisition based on the processing of Time Division Multiplexed (TDM) pilot 1, the result is a coarse timing estimate. This coarse timing estimate provides information about the start of the superframe and gives a coarse estimate of the start of TDM pilot 2. By using the TDM pilot-2 structure to perform further timing estimation, the receiver will estimate the exact start position for the subsequent OFDM symbol. This step is called Fine Timing Acquisition (FTA). A byproduct of this calculation is the channel estimate, which can be used to initialize the channel estimation component.

In one embodiment, such an algorithm is designed to successfully process each channel with a delay spread of up to 1024 chips or samples. In one embodiment, the initial coarse timing estimate inaccuracy is corrected to correct for coarse timing errors between-K and +1024-K chips. In another embodiment, errors based on between-256 and +768 chips are corrected. The FTA process is designed in such a way that timing corrections can be made when needed for application. In other words, the FTA ends before the next symbol is received.

In one embodiment, the TDM pilot-2 symbol includes oneA cyclic prefix followed by two identical pilot 2 sequences in the time domain. The receiver collects N at least in the sampling window_CN/2 or 2048 samples, where N may have different values in different embodiments, and the sampling window begins at a position determined from the coarse timing and an initial intentional offset introduced to avoid collecting data from neighboring symbols. These 2048 samples correspond to a cyclic shift with a TDM pilot 2 sequence period convolved with the channel. After L-point FFT, pilot demodulation, and IFFT, what remains is the cyclic shift of the channel impulse response.

Next, the beginning of the channel impulse response in this circularly shifted image of length 2048 is determined. The entire channel energy is contained within a detection window of length 1024. If the channel is shorter than 1024 chips, there will be several consecutive positions in the energy window that result in the maximum energy. In this case, the algorithm will choose the last position in the cumulative energy curve, since this position usually corresponds to the First Arriving Path (FAP) of the channel. This is done by considering the running energy sum and the order N_DIs achieved by convex combination of local finite differences. Once the location of the FAP is in the shifted channel estimate of length 2048, this information is easily translated into a timing offset for application in subsequent OFDM symbol sampling.

Another product of the algorithm is a time domain channel estimate of length 1024. The means for channel estimation uses three consecutive time domain channel estimates of length 512 and combines them within a time filtering operation to produce a channel estimate of length 1024 that opposes timing variations. We initialize the channel estimation component with "clean", i.e., filtered, channel estimates of length 1024 obtained during FTA. This is done by aliasing it into a 512 length form compatible with the channel estimation components. This processing is then used to generate an effective channel estimate for the first symbol of interest.

The accuracy in timing synchronization is achieved by linking it with the channel estimate and introducing both the accumulated energy curve and its first derivative in the process of detecting the FAP. At the same time, this process makes the method robust against excessive delay spread. The repeated structure of TDM pilot 2 will produce a cyclic shift of the channel estimate. There is a simple one-to-one correspondence between these cyclic shifts and the timing offsets. The structure of the TDM pilot-2 symbol and the deliberately introduced initial offset may make the system more robust against errors in the coarse timing acquisition estimate. Finally, the novel architecture of FTA operation in the symbol timing searcher block and its inter-combination with the IFFT block will make it very computationally efficient and, in one embodiment, provide strict computation time requirements to be met.

Referring initially to fig. 1, a block diagram of an embodiment of a base station 110 and a wireless receiver 150 in an OFDM system 100 is shown. The base station 110 is typically a fixed station and may also be referred to as a Base Transceiver System (BTS), an access point, or some other terminology. The radio receiver 150 may be fixed or mobile and may also be referred to as a user terminal, a mobile station or some other terminology. In addition, the wireless receiver 150 may also be a portable unit, such as a cellular telephone, a handheld device, a wireless module, a Personal Digital Assistant (PDA), a television receiver, and so forth.

At base station 110, a TX data and pilot processor 120 receives different types of data (e.g., traffic/packet data and overhead/control data) and generates data symbols by processing (e.g., encoding, interleaving, and symbol mapping) the received data. As used herein, a "data symbol" is a modulation symbol for data, a "pilot symbol" is a modulation symbol for pilot, and a modulation symbol is a complex value of a point in a signal constellation (signal constellation) in a modulation scheme (e.g., M-PSK, M-QAM, etc.). In addition, pilot processor 120 also generates pilot symbols by processing pilot data and provides the data and pilot symbols to OFDM modulator 130.

The OFDM modulator 130 multiplexes the data and pilot symbols onto the correct subbands and symbol periods, as described below, and further generates OFDM symbols by performing OFDM modulation on the multiplexed symbols. A transmitter (TMTR) unit 132 converts the OFDM symbols into one or more analog signals and further conditions (e.g., amplifies, filters, upconverts, etc.) the one or more analog signals to generate a modulated signal. The modulated signals are then transmitted by the base station 110 from the antenna 134 to the radio receivers in the OFDM system 100.

At radio receiver 150, signals transmitted from base station 110 are received by antennas 152 and provided to receiver units 154. Receiver unit 154 conditions (e.g., filters, amplifies, downconverts, etc.) the received signal and performs digital processing on the conditioned signal to obtain a stream of input samples. An OFDM demodulator 160 performs OFDM demodulation on the input samples to obtain received data and pilot symbols. The OFDM demodulator 160 also uses the channel estimates (e.g., frequency response estimates) to detect (e.g., match filter) the received data symbols to obtain detected data symbols, which are estimates of the data symbols transmitted by the base station 110. The OFDM demodulator 160 may provide detected data symbols to a Receive (RX) data processor 170.

As described below, a synchronization/channel estimation unit (SCEU)180 receives input samples from the receiver unit 154 and determines frame and symbol timing by performing synchronization. The SCEU180 also derives channel estimates using the pilot symbols received from the OFDM demodulator 160. SCEU180 provides the symbol timing and channel estimates to OFDM demodulator 160 and may provide the frame timing to RX data processor 170 and/or controller 190. The OFDM demodulator 160 uses the symbol timing to perform OFDM demodulation and uses the channel estimate to detect the received data symbols.

An RX data processor 170 processes (e.g., symbol demaps, deinterleaves, decodes, etc.) the detected data symbols from OFDM demodulator 160 and provides decoded data. RX data processor 170 and/or controller 190 may use the frame timing to recover different types of data transmitted by base station 110. In general, the processing by OFDM demodulator 160 and RX data processor 170 is complementary to the processing by OFDM modulator 130 and TX data and pilot processor 120, respectively, at base station 110.

Controllers 140, 190 direct operation at base station 110 and wireless receiver 150, respectively. The controllers may be processors and/or state machines. Memory units 142, 192 provide storage for program codes and data used by controllers 140 and 190, respectively. These storage units 142, 192 may use various types of storage media to store information.

The base station 110 may send point-to-point transmissions to a single wireless receiver, multicast transmissions to a group of wireless receivers, broadcast transmissions to all wireless receivers within its coverage area, and any combination of these transmissions is possible. For example, base station 110 may broadcast pilot and overhead/control data to all wireless receivers within its coverage area. In various situations and embodiments, base station 110 may also unicast transmission of user-specific data to particular wireless receivers, multicast data to a group of wireless receivers, and/or broadcast data to all receivers.

Referring to fig. 2A, one embodiment of a superframe structure 200 that may be used in the OFDM system 100 is shown. The data and pilot may be transmitted in units of superframes, each having a predetermined duration. A superframe may also be referred to as a frame, a slot, or some other terminology. In this embodiment, each super-frame includes a TDM pilot 1 field 212 for the first TDM pilot, a TDM pilot 2 field 214 for the second TDM pilot, an overhead field 216 for overhead/control data, and a data field 218 for traffic/packet data.

The four fields 212-218 are multiplexed in a time-division manner in each superframe such that only one field is transmitted at any given time. In addition, the four fields are arranged in the order shown in fig. 2, thereby providing convenience for synchronization and data recovery. The pilot OFDM symbols in pilot fields 212 and 214 are transmitted first in each super-frame and these symbols may be used to detect the overhead OFDM symbols in field 216 that are transmitted next in the super-frame. The overhead information retrieved from field 216 is then used to recover the traffic/packet data in data field 218 that was last transmitted in the super-frame.

In one embodiment, TDM pilot 1 field 212 conveys an OFDM symbol for TDM pilot 1 and TDM pilot 2 field 214 conveys an OFDM symbol for TDM pilot 2. In general, each field may have any duration and the fields may be arranged in any order. TDM pilots 1 and 2 are periodically broadcast in each super-frame to facilitate synchronization performed by the wireless receiver. The overhead field 216 and/or the data field 218 may also contain pilot symbols that are frequency division multiplexed with the data symbols, as described below.

OFDM system 100 has a total system bandwidth of BW MHz, which is split into N orthogonal sub-bands using OFDM. The spacing between adjacent sub-bands is BW/N MHz. Of the total N subbands, M subbands may be used for pilot and data transmission, where M < N, and the remaining N-M subbands may be unused and serve as guard subbands. In one embodiment, the OFDM system uses an OFDM structure in which the total sub-band N is 4096, the usable sub-band M is 4000, and the guard sub-band N-M is 96. In general, any OFDM structure with any number of total subbands, usable subbands, and guard subbands may be used for an OFDM system.

TDM pilots 1 and 2 may be designed to facilitate synchronization processing by wireless receivers in the system. The wireless receiver can use TDM pilot 1 to detect the beginning of each super-frame, obtain a coarse estimate of the symbol timing, and estimate the frequency error. The wireless receiver may use TDM pilot 2 to obtain more accurate OFDM symbol timing.

Referring to fig. 2B, another embodiment of a superframe structure 200 that may be used in the OFDM system 100 is shown. This embodiment has TDM pilot 2214 following TDM pilot 1212 with overhead OFDM symbol 216 inserted in between. The number and duration of overhead symbols are known, and thus, by synchronizing with the TDM pilot 1 symbol 212, it can be estimated where the TDM pilot 2 symbol begins.

Referring next to fig. 3, shown is an embodiment of the TDM pilot 2214 in the frequency domain. For this embodiment, TDM pilot 2214 includes L pilot symbols transmitted on L subbands that are uniformly distributed across a total of N subbands and equally spaced apart by S subbands, where S is N/L. For example, N is 4096, L is 2048, and S is 2. Likewise, other values may be used for N, L, S. This structure of the TDM pilot 2214 may provide accurate symbol timing in different types of channels, including poor multipath channels. The wireless receiver 150 is also capable of: (1) TDM pilot 2214 is processed in an efficient manner to obtain the symbol timing before the next OFDM symbol, which in one embodiment is obtained just after TDM pilot 2, and (2) the symbol timing is applied to the next OFDM symbol, as described below. The L bands for TDM pilot 2 are selected to generate S identical pilot 2 sequences for TDM pilot 2214.

Referring to fig. 4, a block diagram of one embodiment of TX data and pilot processor 120 of base station 110 is shown. Within pilot processor 120, a TX data processor 410 performs receive, encode, interleave, and symbol map processing on the traffic/packet data to generate data symbols.

In one embodiment, a pseudo-random number (PN) generator 420 is used to generate data for the pilots 212, 214. For example, PN generator 420 may be implemented with a 15-tap Linear Feedback Shift Register (LFSR), and the shift register implements a generator polynomial g (x) x¹⁵+x¹⁴+1. In this case, the PN generator 420 includes: (1)15 are provided withDelay elements 422 a-422 o coupled in series, and (2) an adder 424 coupled between delay elements 422n and 422 o. Delay element 422o provides pilot data that is also fed back to the input of delay element 422a and to one input of summer 424. PN generator 420 may be initialized with different initial states for pilots 212, 214, e.g., for TDM pilot 1, to "011010101001110"; for TDM pilot 2, it is initialized to "010110100011100" and for Frequency Division Multiplexed (FDM) pilot, it is initialized to "010110101011101". Generally, either data may be used for the pilots 212, 214. By selecting the pilot data, the difference between the peak amplitude and the average amplitude of the pilot OFDM symbol can be reduced (i.e., peak-to-average variation in the time domain waveform of the TDM pilot is minimized). The pilot data for TDM pilot 2 may also be generated using the same PN generator used to scramble the data. The wireless receiver has knowledge of the data for TDM pilot 2 but does not need to have knowledge of the data for TDM pilot 1.

The bit-to-symbol mapping unit 430 receives pilot data from the PN generator 420 and bit maps the pilot data to pilot symbols according to a modulation scheme. The same or different modulation schemes may be used for pilots 212, 214. In one embodiment, QPSK is used for both TDM pilots 1 and 2. In this case, mapping unit 430 would group the pilot data into 2-bit binary values and further map each 2-bit value to a particular pilot modulation symbol. Each pilot symbol is a complex value in the signal constellation for QPSK. If QPSK is used for the TDM pilot, mapping unit 430 would use 2L for TDM pilot 1₁Mapping of pilot data bits to L₁One pilot symbol, and further 2L for TDM Pilot 2₂Mapping of pilot data bits to L₂And a pilot symbol. A multiplexer (Mux)440 receives the data symbols from the TX data processor 410, the pilot symbols from the mapping unit 430, and the TDM _ Ctrl data from the controller 140. As shown in fig. 2A and 2B, the multiplexer 440 may provide the OFDM modulator 130 with pilot symbols for the pilots 212, 214 and overhead for each superframeAnd data symbols of the data field.

Referring next to fig. 5, a block diagram of one embodiment of OFDM modulator 130 of base station 110 is shown. The symbol-Subband mapping unit 510 receives data and pilot symbols from the TX data and pilot processor 120 and maps the symbols onto the correct subbands according to the Subband _ Mux _ Ctrl signal from the controller 140. In each OFDM symbol period, mapping unit 510 provides one data or pilot symbol on each subband for data and pilot transmission and a "zero symbol" (with a signal value of 0) for each unused subband. The TDM pilot symbols 212, 214 designated for those unused subbands are replaced with zero symbols. Mapping unit 510 may provide N "transmit symbols" for a total of N subbands for each OFDM symbol period, where each transmit symbol may be a data symbol, a pilot symbol, or a zero symbol.

In each OFDM symbol period, an Inverse Discrete Fourier Transform (IDFT) unit 520 receives N transmit symbols, transforms the N transmit symbols to the time domain using an N-point IDFT, and provides a "transformed" OFDM symbol that contains N time-domain samples. Each sample is a complex value to be transmitted in one sample period. In the general case, N is a power of 2, and if so, the N-point IDFT may also be replaced by performing an N-point Inverse Fast Fourier Transform (IFFT).

A parallel-to-serial (P/S) converter 530 serializes the N samples in each transformed symbol. Cyclic prefix generator 540 then repeats a portion (i.e., C samples) of each transformed symbol to form an OFDM symbol that includes N + C samples. For example, the cyclic prefix is the last 512 samples of an OFDM symbol. The cyclic prefix is used to combat inter-symbol interference (ISI) and inter-carrier interference (ICI) due to long time delay spread in a communication channel. Typically, the delay spread is the time difference between the FAP and the Last Arriving Path (LAP) at the receiver. An OFDM symbol period (or simply "symbol period") is the duration of one OFDM symbol and is equal to N + C sample periods.

Referring to fig. 6, one embodiment of a time domain representation of TDM pilot 2 is shown. The OFDM symbol for TDM pilot 2 (i.e., the "pilot-2 OFDM symbol") also contains a transformed symbol of length N and a cyclic prefix of length C. The transformed symbols for pilot 2 comprise S identical pilot-2 sequences, where each pilot-2 sequence comprises L time-domain samples. The cyclic prefix for TDM pilot 2 contains the rightmost C samples in the transformed symbol and is inserted in front of the transformed symbol. For example, if N4096, L2048, S2, and C512, the pilot-2 OFDM symbol will include two complete pilot-2 sequences, each pilot-2 sequence including 2048 time-domain samples. The cyclic prefix for TDM pilot 2 would contain only a portion of the pilot-2 sequence.

Referring next to fig. 7, a block diagram embodiment of the SCEU180 at the wireless receiver 150 is shown. Within the SCEU180, a superframe detector 710 receives input from the receiver unit 154, detects the beginning of each superframe by processing the input samples, and provides superframe timing. A symbol timing detector 720 receives the input samples and the superframe timing, detects the beginning of a received OFDM symbol by processing the input samples, and provides symbol timing. The frequency error estimator 712 estimates the frequency error in the received OFDM symbol. A channel estimator 730 receives the output from the symbol timing detector 720 and derives a channel estimate. In the following, the detector and estimator in the SCEU180 will be described.

Superframe detector 710 performs superframe synchronization by detecting TDM pilot 1 in the input samples from receiver unit 154. For the present embodiment, superframe detector 710 is implemented using a delayed correlator that uses the periodic nature of the pilot-1 OFDM symbol for superframe detection.

Referring to fig. 8, a block diagram illustrates a timeline 800 for one embodiment of the FTA. FAP detection or channel position search at the end of FTAAnd (6) executing the stages. In the illustrated portion of the process, the length N is aggregated in block 812_CThe sampling window of (1). Next, in block 814, N is performed on the sample window_CPoint FFT of which N_CIn this example 2048. The FFT is done in a cascaded 512 point FFT by using interleaved sequences 6, 4, 2 and 0. In block 816, the pilot information is demodulated and the frequency channel is derived from the subcarriers in the same interleaving sequence. At block 818, N is performed on the demodulated pilot signal using the same interleaving sequence_CPoint IFFT, where N_CThe point IFFT is performed as a cascaded 512-point IFFT. The twiddle multiplications for 6, 4, and 2 interlaces begin after block 816 ends. At block 820, an FTA search is initiated to begin the process of finding a FAP. This pipeline processing will be further described below, and allows for faster precision timing fetches.

Referring to fig. 9, a block diagram of an embodiment of a symbol timing detector 720 for one embodiment that performs timing synchronization based on a pilot-2 OFDM symbol is shown. Within symbol timing detector 720, a sample buffer 912 receives input samples from receiver unit 154 and holds a window of "samples" of L input samples for the pilot-2 OFDM symbol. The beginning of the sampling window is determined by offset calculation unit 910 based on the superframe timing from superframe detector 710.

Referring to fig. 10A, a timing diagram for processing a pilot-2 OFDM symbol in one embodiment is shown. Even if pilot-1 is at some later point (denoted as T)_D) As detected above, superframe detector 710 also provides coarse symbol timing (denoted as T) based on pilot-1 OFDM symbols_C). Offset calculation section 910 determines T by determining T_WTo locate the sampling window 1012. The pilot-2 OFDM symbol contains S identical pilot-2 sequences, each of length L (e.g., if N4096 and L2048, then the two pilot-2 sequences are 2048 in length). N is a radical of_CA sample window 1012 of input samples is set by sample buffer 912 to be at position T_WPilot-2 OFDM symbol collection from.

The beginning of the sampling window 1012 will be timed T from the coarse symbol_CIs delayed by an initial offset OS_initI.e. T_W＝T_C+OS_init. This initial offset need not be particularly accurate and by choosing the initial offset, it is ensured that a complete pilot-2 sequence is collected in sample buffer 912 regardless of whether an error occurs in the coarse timing estimate. The initial offset may also be selected to be small enough to end processing the pilot-2 OFDM symbol before the next OFDM symbol so that the symbol timing obtained from the pilot-2 OFDM symbol may be applied to the next OFDM symbol. In this embodiment, the concept of symbol boundaries is tracked by an OFDM sample counter. At the beginning of the OFDM symbol cyclic prefix, the OFDM sample counter takes the value 0 and will count up to the value N_OFDM-1, wherein N_OFDMIs the total duration of the OFDM symbol after which it is flipped back to zero. In processing a regular OFDM symbol, the samples are sent to FFT engine 914 to arrive at a value of N at an OFDM sample counter_CPDemodulation follows C. The symbol timing correction determined by symbol timing searcher 920 is applied by changing the current value of the OFDM sample counter by an amount corresponding to the calculated timing offset. After the coarse acquisition, at time T_DThe rough concept of symbol boundaries at the receiver can be achieved by combining the value T_D-T_CWritten to the OFDM sample counter to be captured. Then, the initial offset OS is applied in two steps_init. First, the OFDM sample counter value is incremented by K and reduced by the window duration between OFDM symbols (e.g., 17 in this embodiment) in offset calculation block 910. The constant K corresponds to the algorithm's ability to correct coarse timing errors, in this embodiment K256. When the OFDM sample counter reaches the count value 1024 in the present embodiment, this will be considered as the beginning T of the sampling period_WAnd the sampling window 1012 begins. Other embodiments may use other values for the first and second constants and the meterNumerical values.

Referring back to FIG. 9, a Discrete Fourier Transform (DFT) unit 914 samples N collected by the buffer 912_CAn L-point DFT or FTT is performed on the L input samples and L frequency-domain values are provided for the L received pilot symbols. If the beginning of the sampling window 1012 is not aligned with the beginning of the pilot-2 OFDM symbol (i.e., T)_W≠T_S) Then the channel impulse response will be cyclically shifted, which means that the front part of the channel impulse response will be turned back to the rear part.

For the present embodiment, pilot-2 OFDM symbol 214 is continuous with cyclic prefix 1004 and two pilot-2 sequences 1008. In one frequency domain embodiment, pilot-2 symbol 214 comprises 2000 non-zero QPSK subcarriers or subbands, which are each separated by a zero-valued subcarrier with guard subcarriers 304 as shown in fig. 3 at each end. By inserting a zero-valued subcarrier between two non-zero subcarriers, it is ensured that TDM pilot-2 contains a period of two 2048 samples, and where each sample is in the time domain. At the receiver, only 2048 or N of TDM pilot 2 will be acquired in sampling window 1012_CAnd (4) sampling.

After the initial L-point FFT 914 is performed, these initial 2000 non-zero subcarriers and 48 guard subcarriers will be available for L2048 after the channel is traversed. The non-zero subcarriers will be modulated by the information on the channel and add noise. To recover the channel information, i.e., estimate the channel impulse response for up to 2048 taps, we need to "undo" the scrambling of the non-zero subcarriers and zero-out those subcarriers that have been omitted (i.e., guard subcarriers) before the L-point IFFT component 918. This operation is referred to as TDM pilot-2 symbol demodulation and extrapolation, which is performed in pilot demodulation unit 916.

Referring next to fig. 12, shown is an embodiment of pilot demodulation logic for demodulating non-zero pilot sequences in any interlace. In this embodiment, interlace represents an interlace having N_IA subset of subcarriers, and the subcarriers are evenly spaced in an initial set of N subcarriers. For example, as in the present embodiment, N may be 4096, and if 8 interlaces are used, each interlace I is one with N_IA set of subcarriers separated by seven subcarriers not belonging to interlace I. At the input of the demodulation block 916, the in-phase and quadrature-phase components of the pilot observations are given by 9 signed bits, while after demodulation the bit width will remain 9.

Referring back to fig. 9, each output sample of the L-point FFT section 914 is a complex number, in this embodiment, where both the real and imaginary numbers are 9-bit signed numbers. The process of removing the pilot modulation essentially multiplies each pilot carrier by the reference value corresponding to that subcarrier, which can be done at the receiver. This operation is performed four times using four different reference sequences since four different interlaces (i.e., 6, 4, 2, 0) are collected from the output of FFT section 914. The observation of the pilot in interlace i (i ═ 0, 2, 4, 6) on carrier k (k ═ 0, 1, …, 499) is represented by Y_i，kGiven, and the corresponding reference symbols (from QPSK modulation) are passed through S at the receiver_i，k＝[b_2k+1 b_2k]Given the scrambling operation. The process of removing the modulation on the pilot subcarriers is performed as one rotation operation (rotation by 0,90, 180, 270 degrees) followed by multiplication by (1-j). The rotation amount is based on the reference symbol S_i，kAnd (4) determining. This rotation operation is followed by a process of adding and subtracting the real and imaginary components. In Table I below, the dependence on scrambler output bit (b) is given_2k+1 b_2k) Y of (A) is_i，kThe table is based on gray mapping (gray mapping) of bits to QPSK constellation symbols.

Table I: rotation angle as a function of bits from a scrambler

(b2k+1 b2k) (from scrambler)	Rotation angle (degree)
		00	0
01	90
		11	180
10	270

In this respect, it should be noted that Y in the ith interleaving buffer_i，0Beginning at memory location 262. Thus, 500 pilot observations are taken in sequence by starting at memory location 262, passing through memory location 511 and wrapping around to memory location 0, and then passing through memory location 249. It should be noted that the storage units 250-261 correspond to guard sub-carriers and in the present embodiment they are set equal to zero. Interlace 0 for FTA follows the convention for data, namely writing pilots from memory locations 262 through 511, skipping memory location 0 (corresponding to DC) and outputting it as zeros, while filling memory locations 1-250. At this time, the guard carriers reside in the storage units 251 to 261.

Referring next to fig. 10B, shown is an L-tap channel impulse response from IDFT unit 918 in one embodiment. The impulse response shows the cyclic shift in the channel estimation. Each of the L taps is associated with a complex channel gain on the tap delay. The channel impulse response may be cyclically shifted, meaning that the tail of the channel impulse response may be rotated around and may appear in an early portion of the output of IDFT unit 918.

Referring back to fig. 9, the symbol timing searcher 920 may determine the symbol timing by detecting the onset of channel energy as shown in fig. 10B. The fixed point function of the symbol timing searcher 920 is divided into two subsections: means for channel positioning and means for fine timing correction. This detection of the onset of channel energy (also known as the "first arrival path", or FAP) can be performed by sliding a "detection" window 1016 on the channel impulse response by a length N, as shown in fig. 10B_WTo be implemented. The detection window size may be determined as follows. At the start of each window, the energy of all taps falling within the detection window is calculated to find the cumulative energy shown as a curve in FIG. 10C.

Referring to FIG. 10C, a graph of cumulative energy at different window starting positions is shown, in one embodiment. The detection window is shifted to the right in a cyclic manner, whereby the arrival at the right edge of the detection window is at the index N_CAfter the last tap on, the window will turn around to the first tap at index 1. This allows the same number of channel taps to be collected for each detection window start position.

Detecting the Window size N_WThe selection may be made according to the expected system delay spread. The delay spread of a wireless receiver is the time difference between the earliest and latest arriving signal components at the wireless receiver. The delay spread of the system is the maximum delay spread among all receivers in the system. If the detection window size is equal to or greater than the system delay spread, then when properly aligned, the detection window will capture all of the energy of the signal impulse response. In one embodimentIn (2), the detection window size N_WCan also be selected to be not greater than N_CHalf (i.e. N)_W≤N_C/2) in order to avoid uncertainty in the detection of the onset of the channel impulse response. Thus, only N is required_CChosen to be greater than or equal to the maximum expected channel delay spread, the FTA can detect the OFDM symbol timing unambiguously regardless of the channel implementation.

Referring next to fig. 10D, an example of the negative derivative of the accumulated energy curve is shown. The onset of the channel impulse response, i.e., the FAP, may be detected as follows: (1) determine the peak energy of the starting positions of all detection windows 1016 in the cumulative energy curve shown in fig. 10C, and (2) identify the starting position of the rightmost detection window 1016 having the peak energy if multiple window starting positions have the same or similar peak energy. A score can be derived from the weighted sum of the tap energies of the detection window 1016 and the finite difference from the cumulative energy curve. By effectively maximizing this score, the trailing edge of the largest region of the cumulative energy curve can be found. The energy at the start of the different windows may also be averaged or filtered to obtain a more accurate estimate of the corresponding onset of the channel pulse in a noisy channel. In any case, the beginning of the channel impulse response is denoted by FAP in fig. 10D. Once the channel impulse response T is determined_BThe fine symbol timing correction can then be uniquely calculated. These corrections are designed to drive the FAB position, position T in FIG. 10B_BClose to the position zero during the next OFDM symbol or any expected position in the channel estimate.

In a different embodiment, the fine timing correction may depend on both the FAP location and the estimated channel delay spread D. This delay spread D can be determined by looking for the leading and trailing edges of the accumulated energy curve. Similar to finding the trailing edge, the leading edge can be found by scoring a weighted sum of the accumulated energy and its positive finite difference. In a different embodiment, a fine timing searcher headerFirstly finding the position T where the maximum accumulated energy appears_MAnd store this maximum value E_M. Next, check that T is located_MLeft and right cumulative energy curves to determine those cumulative energy values below a value (1-b) E for a predetermined value b less than 1_MThe position of (a). In other words, the leading and trailing edges of the accumulated energy curve are defined at locations where the accumulated energy is below a certain percentage (e.g., 5% or 3%) of its maximum value in the detection window 1016. This percentage defines a band around the maximum of the cumulative energy position. Entering this band defines the leading edge T of the flat part of the band_LWhile leaving this band defines the trailing edge T of the flat part of this band_T. The trailing edge coincides with the position of the first arriving path, while the leading edge is equal to the last arriving path minus N_W. The difference between the leading edge and the trailing edge is then equal to N_WThe delay spread D is subtracted. Thus, the delay spread D can be calculated as D ═ N_W-T_T-T_L. Once D is calculated, a fine timing correction can be determined so that the channel content remains centered in the cyclic prefix region in the channel estimate during the next OFDM symbol.

Referring back to fig. 10A, where the fine symbol timing indicates the beginning of the received OFDM symbol. Precision symbol timing T_SCan be used to accurately and correctly place the DFT collection window for each subsequent received OFDM symbol (i.e., all subsequent OFDM symbols conveying data and FDM pilots). The DFT collection window represents a particular number N of input samples (from the N + C input samples) for each OFDM symbol received for collection. The N input samples within the DFT collection window are then transformed using an N-point DFT to obtain N received data/pilot symbols for the received OFDM symbol. By accurately placing the DFT collection window for each OFDM symbol received, it may help avoid (1) inter-symbol interference (ISI) from previous or next OFDM symbols, (2) degradation of channel estimation (e.g., incorrect placement of the DFT collection window may result in erroneous channel estimates), (3) cyclic prefix dependent errors in the processing(e.g., frequency tracking cycles, etc.), and (4) other deleterious effects. The pilot-2 OFDM symbol may also be used to obtain a more accurate frequency error estimate by using the periodic characteristics of TDM pilot-2.

The channel impulse response from IDFT unit 918 may also be used to derive a frequency response estimate for the communication channel between base station 110 and wireless receiver 150. Unit 922 receives the L-tap channel impulse response, cyclically shifts the channel impulse response so that the beginning of the channel impulse response is at index 1, inserts an appropriate number of zeros after the cyclically shifted channel impulse response, and provides an N-tap channel impulse response. A DFT unit 924 then performs an N-point DFT on the N-tap channel impulse response to provide a frequency response estimate consisting of the N complex channel gains for the N total subbands. OFDM demodulator 160 may use the frequency response estimate to detect a received data symbol in a subsequent OFDM symbol. In other embodiments, this initial channel estimate may also be derived in other ways.

Referring to fig. 11, one embodiment of a pilot transmission scheme that combines TDM and FDM pilots is shown. Base station 110 may transmit TDM pilots 1 and 2 in each super-frame to facilitate initial acquisition by the wireless receiver. The overhead for the TDM pilot is two OFDM symbols, which is very small compared to the size of the super-frame. The base station may also transmit FDM pilots in all, most, or some of the remaining OFDM symbols of each superframe. For the embodiment shown in fig. 11, the FDM pilot is sent in alternating interlaces, such that pilot symbols are sent on one of the even-numbered symbol periods and on the other of the odd-numbered symbol periods. Each interlace contains a sufficient number of subbands to provide support for channel estimation and possibly frequency and time tracking for the wireless receiver. In general, any number of interlaces may be used for the FDM pilot.

The wireless receiver may use TDM pilots 1 and 2 for initial synchronization such as superframe synchronization, frequency offset estimation, and fine symbol timing acquisition (for proper placement of the DFT collection window for subsequent OFDM symbols). For example, the wireless receiver may perform initial synchronization when accessing the base station for the first time, when receiving or requesting data for the first time or after a long period of inactivity, when first powering on, and so on.

The wireless receiver may detect the presence of a pilot-1 OFDM symbol by performing a delayed correlation on the pilot-1 sequence and thereby determine the start of a superframe as described above. Thereafter, the wireless receiver may use the pilot-1 sequence to estimate the frequency error in the pilot-1 OFDM symbol and correct this frequency error before receiving the pilot-2 OFDM symbol. The pilot-1 OFDM symbol allows for a larger frequency error to be estimated and the sampling window 1012 to be placed more reliably for the next pilot-2 OFDM symbol than conventional methods using a cyclic prefix structure for the data OFDM symbol. Thus, the pilot-1 OFDM symbol may provide improved performance for terrestrial radio channels with a large multipath delay spread.

The wireless receiver may use the pilot-2 OFDM symbol to obtain fine symbol timing to more accurately place the DFT collection window for subsequent received OFDM symbols. The DFT collection window is a portion of the time domain signal that will capture the necessary information for use in decoding data transmitted by a particular OFDM signal. The wireless receiver may also use the pilot-2 OFDM symbol for channel estimation and frequency error estimation. The pilot-2 OFDM symbol allows for fast and accurate determination of fine symbol timing and proper placement of the DFT collection window.

The wireless receiver may use the FDM pilot for channel estimation and time tracking, and possibly also for frequency tracking. As described above, the wireless receiver may obtain an initial channel estimate from the pilot-2 OFDM symbol. As shown in fig. 11, the wireless receiver may use the FDM pilot to obtain more channel estimates, especially when the FDM pilot is transmitted via a superframe. The wireless receiver may also use the FDM pilot to update a frequency tracking loop that corrects for frequency errors in the received OFDM symbols. The wireless receiver may also use the FDM pilot and the channel estimate obtained therefrom to update a time tracking loop that can account for timing offsets in the input samples (e.g., due to changes in the channel impulse response of the communication channel)

Channel position and FAP detection algorithm

The output of IFFT component 918 may be considered a time domain channel estimate, which is 2048 taps in length and may be cyclically shifted by T as shown in fig. 10B_B. The task of the algorithm for channel estimation detection is to determine this cyclic shift T_BThe number of the cells. This process may be implemented by combining the accumulated energy within the sliding detection window with the negative difference calculation described in fig. 10D. This form of channel location detection algorithm is known as first arrival path or FAP detection because the metric described is designed to have a peak at the FAP location. In other embodiments, channel location detection may be performed using another alternative algorithm, where the FAP and LAP locations are determined by detecting the edges of a flat area using a percentage method, as previously described. For the sake of simplicity, only an embodiment of the FAP detection algorithm is described in detail below. N is a radical of_CAnd N_WDefined as the length of the channel estimate sampling window 1012 and the sliding energy detection window 1016, respectively. To avoid ambiguity in FAP detection, the embodiment will generally satisfy the relation N_W＝N_C/2. In IFFT section 918, this is done by letting N_C2048 and N_W1024. These values are selected assuming that the maximum delay spread is no more than 1024 taps (or about 185 mus in one embodiment), and the total channel energy can be captured in a sliding detection window 1016 of length equal to half of the channel estimation sampling window 1012.

In the absence of noise, when (window start position + N)_W) Mode N_CGreater than the last channel tap position, the maximum energy in the window is reached and it will start at the windowThe starting location stays at the maximum until it moves and exceeds the FAP. Thus, detecting FAP corresponds to detecting only the trailing edge of the flat region near the maximum value of the accumulated energy curve shown in fig. 10C. This process can be implemented by combining the accumulated energy measurements within the detection window with a negative finite difference combination. The energy measurement En, and the order N_DNegative finite difference of D_nIs defined as follows:

and

wherein N is more than or equal to 0 and less than or equal to N_C-1 denotes the beginning of the detection window, h (N) is the channel estimate, and from the sum the upper and lower limits and the subscript shall be denoted by N_CIn the modular sense, the window will "spin". The location of the FAP will then be roughly determined by the index n that maximizes the number of scores. In other words, such that:

and

then, the FAP location may be found to be:

FAP＝(n^*-N_D)modN_C. (3)

in the above algorithm, the freely adjustable parameters are α and N_D. Value N_DAnd α is programmable, and (N)_DDifferent combinations of a) pairs will result in the algorithm having different importance for detecting weak leading edge taps in the channel impulse response. In other words, has a low value of N_DAnd embodiments with high values of alpha typically detect FAPs with very small amplitudes. However, larger N_DThe value will average out more noise in performing the FAP decision. In one embodiment, the value used for fine timing acquisition is N_D5 and α 0.9375.

FAP detection implementation mode

One of the cases for FAP detection implemented in FTA mode is with respect to a strict time line of computation, which occurs before the start of the next symbol. The time for the calculation (e.g., 300-400 milliseconds in one embodiment) ends before the next OFDM overhead symbol 216 is received as shown in fig. 10A. Thus, in this embodiment, the calculation of the initial windowed energy measurement in equation (1) is integrated with the last stage of the FFT block 918.

By optimizing the FFT and IFFT implementations for fine timing acquisition, strict timelines can be satisfied in the following manner:

an FFT architecture is used to compute the first stage in the FFT process in parallel with the input data. An exemplary FFT architecture is described in U.S. application 10/775,719 filed on 9/2/2004This application is incorporated herein by reference for all purposes. The FFT implementation is selected to match the number of subbands in each interlace (N)_I) And (4) matching. For example, if pilot-2 uses N_IThe FFT implementation is chosen to be a cascaded 4 x 512FFT, with 512 and 4 interlaces, and the 4-point FFT is computed without additional latency when the samples are received.

2. A 512-point FFT is calculated for the interlace in a particular order optimized for speed. For example, if TDM pilot 2 is transmitted on an even subcarrier, the FFT is performed in the order of 6, 4, 2, and 0 as follows.

3. Pilot demodulation is performed on an interlace-by-interlace basis.

4. Once pilot demodulation is complete, a 2048-point IFFT is calculated. With the present embodiment, this process is performed in three steps.

a. Interleaves 6, 4, 2, and 0 are processed with a 512-point IFFT.

b. The twiddle multiplication is applied to only interleaves 6, 4 and 2. Interleave 0 does not use any twiddle multiplication. Thus, the IFFT for interlace 0 can occur simultaneously with the twiddle multiplication for other interlaces, saving time.

The c.4-point IFFT will combine the outputs of the 512-point IFFTs.

The 5.4-point IFFT stage is combined with the initialization of the FAP detection algorithm. The 4-point IFFT provides the following samples:

h(n)，h(n+N_W/2)，h(n+N_W)，h(n+3N_W/2)，for 0≤n≤N_W/2-1.

it should be noted that to calculate the window energy in equation (1), i.e., E, from position 0₀We will wait until all N_WThe/2 4-point IFFT ends. However, where we have enough data to compute E_NWSimultaneously; whereby the two sliding window accumulators can be computed in parallel. Also, consider the energies for the two accumulatorsAnd (3) quantity updating step:

E_n+1＝E_n-{|h(n)|²-|h(n+N_W)|²}＝E_n-d(n)，for 0≤n≤N_W-2 and

since the same correction factor is used to update both accumulators, the values d (n) will be saved for future use. The first phase of FAP detection involves computing E₀、E_NWAnd a value d (N), where 0. ltoreq. n.ltoreq.N_W-1. The first phase is with N_WThe/2 4-point IFFTs are executed in parallel and thus can be as much time. One embodiment of such a calculation is shown in fig. 14. Each norm operation 1408 is identical and will produce 11 unsigned bits. A block diagram for norm operation 1408 is shown in fig. 13.

The channel estimate obtained using TDM pilot 2 may be "noisy" in cases where the SNR is low. Sometimes, noise may appear as artifact of the channel content, and the timing correction during FTA may erroneously account for this artifact when analyzing the channel estimate. Sometimes, the symbol timing calculated based on noise results in worse performance. In one embodiment, the channel tap energies are compared to a predetermined threshold to remove the tap energies when the tap energies are below the threshold. After norm operation 1408, some embodiments include a threshold component 1404 for removing tap energy. In one embodiment, the threshold limit may be chosen to be K times the expected variance of the noise, assuming that the input SNR is some predetermined lower value P. By choosing P and K appropriately, we can adjust the probability that an artificial tap will appear in the TDM 2 channel estimate due to noise with an input SNR of P or higher. In one embodiment, K may be selected to be 12, and P may be selected to be-2 dB. In any case, the threshold is programmable, and if it is set to zero, then the thresholding will not be effectively implemented in block 1404.

After the first phase is over, a second phase is performed in which the finite difference D is measured as used in equation (2)_nAnd score S_nThe value of (c) is initialized. Storage month E_nA plurality of boundaries ofThe value is obtained. The second stage is described before a series of operations are provided. According to equation (1), the first value of the calculated finite difference is D_2ND-1And for this calculation an energy value E will be found₀～E_2ND-1. These energy values are calculated using recursive equation (4). In the whole process, N is also formed along two groups_WParallel computing of the shifted trajectories otherwise; in other words, the energy value E_NW～E_NW+2ND-1To be calculated and used for initializing D_NW+2ND-1. At the same time, the energy value E₀～E_2ND-2And E_NW～E_NW+2ND-2Will be saved and they will be used to calculate the boundary values of the finite difference values and scores. For one embodiment, the sequence of operations in the second stage is of the form:

1) initializationS^*0. Finite difference is scaled to 2⁵14 signed number and maximum score S^*Is a 12-bit unsigned number (scaled to 2)⁴). By updatingAndthe same accuracy can be maintained. Will E₀And E_NWStored in memory.

2) For n ═ 1; n is less than or equal to 2N_D-1; n + +, the following operations are performed:

● update the value E according to equation (4)_nAnd E_n+NW(ii) a After each addition/subtraction, the result is saturated backwards to 12 unsigned bits (the result will be guaranteed to be positive).

● if N < N_DThen the difference is updated toAndotherwiseAndreverse saturation is 14 unsigned bits.

● mixing E_nAnd E_n+NWStored in a memory; they are used at the end of the FAP detection final phase.

3) Two run buffers are initialized:

it should be noted that E_2ND-1And E_NW+2ND-1Not used for calculating D_nThe boundary values, but are also stored by this embodiment, which may result in fewer hardware exceptions. The end of phase two marks the initialization of the FAP detection means. This detection is performed in stage three and will be described next.

In summary, the following variables are initialized:

●2N_Da running buffer E for each of the components_BUFF1And E_BUFF2。

● best score S^*＝0。

● energy value E₀，E₁，……，E_2ND-1And month E stored for future use_NWE_NW+1，……，E_NW+2ND-1。

● and is initialized to the programmable parameter a of 5-bit unsigned value.

● stored in memory, where 0 ≦ N ≦ N_W-1。

● also performs the following initialization:

and

stage three of the FAP detection algorithm may be summarized as a flow chart as shown in fig. 15, and as exemplified, the FAP location may take the value in the interval:

N_D+1≤n≤N_W-N_D，and N_W+N_D+1≤n≤N_C-N_D.

the missing point is located on the boundary of the two start window positions, i.e. around position 0 and position N_W. These extreme cases are handled by step 1508, entitled "update FAP", and depend on the stored energy value. In one embodiment, the sequence of operations of step 1508 take the form of 1 for n; n is less than or equal to 2N_D-1; n + +, the following operations are performed:

1) updatingAnd

D_TEMP2＝D_TEMP2-E_BUFF2[0]+2E_BUFF2[N_D]-E_n

2) will E_BUFF1And E_BUFF2Shift one element to the left, respectively E_n+NWAnd E_nAdded to the right side thereof.

3)S＝α·E_BUFF1[N_D-1]+(1-α)·D_TEMP1(ii) a If S > S^*Update S^*S and

FAP＝(n+N_W-N_D+1)modN_W。

4)S＝α·E_BUFF2[N_D-1]+(1-α)·D_TEMP2(ii) a If S > S^*Update S^*S and

FAP＝N_W+(n+N_W-N_D+1)modN_W。

at this point in the process, the FTA algorithm has completed phase three, the FAP has been detected, and the FAP location has been saved in the variable FAP. The last stage of the FTA algorithm is to compute a fine timing calibration from this information. Before we describe this stage, we will provide additional details regarding the implementation of stage three above. For this purpose, consider here fig. 16, which shows a fixed point implementation of the update step feature for phase three. This figure will be explained in conjunction with the flow chart of fig. 15, as the flow chart shows the sequence of operations. Once scores S have been calculated for both halves of the channel response (note: FIG. 16 shows only the first half), these values will be compared to the current maximum score value S^*The maximum credit value and FAP location are updated, if necessary, in the manner described above. The final output of the FAP detection algorithm is an integer FAP which can take 0 and N_C-1 ═ 2047. In the following we will describe how this integer value is used to calculate the fine offset and its effect on the OFDM sample counter.

Fine timing offset calculation and correction

For the gyratory channel estimation in fig. 10C, it has the indication of the FAP location T_BWill be converted to a fine timing offset as the final result of the FTA algorithm. This step is complicated by the fact that: i.e., in sampling the TDM pilot-2 symbol, we introduce an intentional delay of 1024-K samples, where K is 256 in the above embodiment, and the coarse offset provided by the coarse acquisition may differ by more than + -512 samples. This embodiment of the algorithm proceeds as follows：

If the FAP is > 512,

offset is FAP +512-_OFF；

If not, then,

offset FAP +512+17-B_OFF；

Here, the factor of 17 corresponds to a window of 17 samples inserted between two OFDM symbols in the present embodiment, and it should be understood that this corresponding factor may vary in different embodiments. Next, factor B_OFFIs a programmable parameter responsible for inserting a deterministic delay at perceived symbol boundaries or, equivalently, introducing a bias at the FAP location for future OFDM symbols. This parameter is typically chosen to be positive because it can be shown that producing a negative error in the symbol boundary estimate (referred to as "late symbol sampling") results in worse performance. In one embodiment, B is_OFFThe value of 127 is chosen, but other values may be used in other embodiments.

In various conditional trends, the first option occurs more frequently, thus assuming that the coarse acquisition error is less than ± 512 samples. In principle, the FTA algorithm can handle coarse timing errors of up to ± 1024 samples, but if the initial acquisition algorithm postpones more than 512 samples, it may not be possible to leave enough time to calculate the correct offset and apply it before the first symbol start of the overhead OFDM symbol 216 shown in fig. 2A and 2B.

By modifying the OFDM sample counter contents before the start of the next OFDM symbol as described above, fine timing correction can be applied using the integer value of the offset calculated above. Once the value 4625 is reached, the counter will roll back, but the point of this roll back can be effectively changed by updating the current value in the counter. In one embodiment, the value of the offset calculated as above may be first limited to ± 512 before application to facilitate a simpler transformation process for the frequency tracking component.

The last stage in the FTA algorithm uses the channel estimates obtained as above to initialize the time filter in the channel estimation means. This initialization process may help in correctly demodulating the next symbol. Next, a channel estimation initialization process will be described.

Pilot channel estimation

Hereinafter, an algorithm for guiding channel estimation for the channel estimator 730 will be described. One purpose of the channel estimator 730 is to provide a starting point for the channel estimation time filter. This time filter works on three consecutive channel estimates h (n-1), h (n +1), which are 512 samples in length and represent the past, present and future. All three positions are initialized to all zeros. At the end of the last phase of the FTA, the position h (n) corresponding to the present will be initialized with a 512-tap channel estimate derived from the estimate of length 1024 calculated as above [ we call this impulse response as the impulse responseTo is directed atThe modification of (a) is three-fold:

1)is a circularly shifted version of a correctly calibrated 1024 long channel estimate that has been acquired when the symbol timing is correct. This offset FAP is acquired at stage three of the FAP detection described above. Therefore, when guiding channel estimation, we consider the estimation at hand by cyclic shiftWhile the obtained channel estimate h₁₀₂₄(n) of (a). In other words:

2)h₁₀₂₄(n) is converted to a long 512 channel estimate, which would be obtained during TDM pilot 2 if it were replaced with a data symbol having 512 pilot tones on interlace 6. One reason for doing this is the time filtering operation of the channel estimation unit 730. That is, the channel estimates used for data demodulation are obtained in a "time filtering" unit of a channel estimation component that, in one embodiment, combines the estimates obtained from the FDM pilots in three consecutive OFDM symbols. For this component, the FDM pilots are staggered in the interlaces on consecutive OFDM symbols, as shown in fig. 11. It should be noted that after TDM pilot 2, the FDM pilot in the first symbol is located on interlace 2, so that for the corresponding FDM pilot, if it is a normal OFDM symbol, it will be placed on interlace 6 in TDM pilot 2. Thus, by carefully steering the channel estimation component using TDM pilot 2, it can be allowed to forge normal in the location of TDM pilot 2The presence of the symbols thereby accelerates the generation of a first channel estimate that can be used for data demodulation. This conversion to channel observations of length 512 is by converting h to₁₀₂₄(n) the second half is shifted to the first half; in other words, for 0 ≦ N < N_WFor/2:

(6)

3) obtained in equation (6)The factor * is enlarged relative to the channel estimate. Thus, the last step is to scale the channel estimate by the appropriate factor:

data pattern time tracking

In Data Mode Time Tracking (DMTT), the problem is similar to timing correction that can be done based on channel estimation, except that the channel estimation is now acquired using FDM pilots. In one embodiment, the algorithms for finding the timing correction (or timing offset as described above) from the channel estimates are very similar. In this case, most of the software used for the FTA can be reused for DMTT purposes.

In one embodiment, the TDM pilot 2 based channel estimate (e.g., 2048 taps in length) in FTA mode is longer than the channel estimate in DMTT (e.g., 1024 taps in length). For example, when the channel is longer than 512 taps but shorter than 1024 taps, longer channel estimates may help resolve the uncertainty in the OFDM symbol timing. Since the DMTT is performed on channel estimates of length 1024, any channel response longer than 512 taps can potentially cause problems for some DMTT algorithms. In one embodiment, however, the TDM pilot 2 based channel estimate in FTA mode is twice as long, thereby allowing the position of channels up to 1024 taps in length to be uniquely resolved.

With at least TDM pilot 2 transmitted in each superframe, the receiver may periodically acquire TDM pilot 2 once in N superframes to account for potential timing uncertainty present in some embodiments. N is programmable and may vary depending on delay spread or other factors. FTA processing may be performed on every N superframes to correct for upcoming DMTT processing.

Referring next to fig. 18, disclosed is an OFDM system 1800 for synchronizing the timing of a receiver to a received OFDM signal. The OFDM system includes means 1804 for performing a first timing acquisition, means 1808 for performing a second timing acquisition, and means 1820 for adjusting a DFT collection window position. An apparatus that performs a first timing acquisition using a first received TDM pilot determines a coarse timing estimate of the received OFDM signal. An apparatus that performs a second timing acquisition using a second TDM pilot determines an accurate timing estimate of the received OFDM signal. The first TDM pilot is received before the second TDM pilot and the fine timing estimate is a refinement of the coarse timing estimate. Means for performing a second timing acquisition includes means for determining 1816 and means for detecting 1812. The means for determining the accumulated energy of the plurality of channel taps within the detection window for a plurality of start positions forms an accumulated energy curve. The means for detecting finds the trailing edge of the accumulated energy curve. Means for adjusting the FT collection window position for subsequent OFDM symbols is done based on the output of the means for performing the second timing acquisition.

Referring to fig. 19, disclosed is an embodiment of a process 1900 for synchronizing the timing of a receiver to a received OFDM signal. A first timing acquisition is performed with the first received TDM pilot to determine a coarse timing estimate of the received OFDM signal, block 1904. A second timing acquisition is performed with the second TDM pilot to determine a precise timing estimate for the OFDM symbol in the received OFDM signal, block 1906. In a second timing acquisition block 1906, the accumulated energy of the channel taps over the detection window is determined in block 1908, and the trailing edge of the accumulated energy curve is detected in block 1912. At block 1916, the FT collection window position for the subsequent OFDM symbol is adjusted based on the trailing edge and/or leading edge information.

The synchronization techniques described herein may be implemented using different means. For example, these techniques may be implemented in hardware, software, or a combination thereof. For a hardware implementation, the processing units at the base station (e.g., TX data and pilot processor 120) used to support synchronization may be implemented within one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), processors, controllers, micro-controllers, microprocessors, other devices designed to perform the functions described herein, or a combination thereof. The processing units (e.g., SCEU180) on the wireless receiver for performing synchronization may also be implemented in one or more ASICs, DSPs, and the like.

For a software implementation, the synchronization techniques may be implemented with modules (e.g., procedures, functions, and so on) that perform the functions described herein. The software codes may be stored in a memory unit (e.g., memory unit 192 in fig. 1) and executed by a processor (e.g., controller 190). The memory unit may be implemented within the processor or external to the processor.

While the principles of the disclosure have been described above in connection with specific apparatus and methods, it is to be clearly understood that this disclosure is made only by way of example and not as a limitation on the scope of the invention.

Claims (38)

1. A method for synchronizing a timing of a receiver with a received Orthogonal Frequency Division Multiplexing (OFDM) signal, the method comprising the steps of:

performing a first timing acquisition with a first received Time Division Multiplexed (TDM) pilot to determine a coarse timing estimate of the received OFDM signal;

performing a second timing acquisition with a second TDM pilot to determine a fine timing estimate of the received OFDM signal, wherein said step for performing a second timing acquisition comprises the sub-steps of:

determining accumulated energy for a plurality of channel taps within a detection window for a plurality of starting positions to form an accumulated energy

A cumulative energy curve, and

detecting a trailing edge of the accumulated energy curve; and

adjusting a Fourier Transform (FT) collection window position for a subsequent OFDM symbol in accordance with the step for performing a second timing acquisition.

2. The method for synchronizing the timing of a receiver to a received OFDM signal as recited in claim 1, wherein the first TDM pilot is received before the second TDM pilot.

3. The method for synchronizing timing of a receiver with a received OFDM signal as recited in claim 1, wherein the fine timing estimate is a refinement of the coarse timing estimate.

4. The method for synchronizing timing of a receiver to a received OFDM signal of claim 1, wherein the trailing edge is located using a weighted sum of an accumulated energy at a particular one of a plurality of starting positions and a negative finite difference of the accumulated energy curve at the particular starting position.

5. The method for synchronizing the timing of a receiver with a received OFDM signal of claim 1, wherein the sub-step of detecting allows for detection of a First Arriving Path (FAP).

6. The method for synchronizing the timing of a receiver with a received OFDM signal of claim 1, wherein the leading edge and the trailing edge of a flat region in the cumulative energy curve are both detected from a flat region in the cumulative energy curve that differs from a maximum point by a percentage of energy.

7. The method for synchronizing the timing of a receiver to a received OFDM signal of claim 1, wherein at least one of a trailing edge and a leading edge of the accumulated energy curve is transformed into a timing correction.

8. The method for synchronizing timing of a receiver to a received OFDM signal of claim 7, wherein FAP is placed with respect to the trailing edge.

9. The method for synchronizing the timing of a receiver to a received OFDM signal of claim 1, wherein at least one of the leading edge and the trailing edge of the accumulated energy curve is transformed into a timing correction by placing a position of a channel profile relative to at least one of the trailing edge and the leading edge.

10. The method for synchronizing the timing of a receiver to a received OFDM signal as recited in claim 1, wherein each of the plurality of channel taps corresponds to a complex channel gain on a corresponding tap delay.

11. The method for synchronizing timing of a receiver with a received OFDM signal as claimed in claim 1, wherein the step for performing a second timing acquisition is done before the end of a second TDM pilot.

12. The method for synchronizing the timing of a receiver with a received OFDM signal as set forth in claim 1, wherein the determining sub-step and the detecting sub-step are performed at least partially simultaneously in time for a particular one of the plurality of channel taps.

13. The method for synchronizing a timing of a receiver with a received OFDM signal of claim 1, wherein the receiver is at least one of a wired receiver or a wireless receiver.

14. The method for synchronizing the timing of a receiver with a received OFDM signal as claimed in claim 1, further comprising the steps of: using the channel estimate obtained in said step for performing second timing acquisition, steering the channel estimate.

15. The method for synchronizing the timing of a receiver with a received OFDM signal as claimed in claim 1, wherein said step for performing a second timing acquisition further comprises the sub-steps of: performing a Fourier transform on the FT collection window, wherein the size of the FT collection window is twice the size of the detection window.

16. The method for synchronizing the timing of a receiver to a received OFDM signal of claim 1, wherein the accumulated energy curve is filtered to reduce false trailing edge detection.

17. The method for synchronizing the timing of a receiver with a received OFDM signal as claimed in claim 1, wherein said step for performing a second timing acquisition further comprises the sub-steps of: performing thresholding on each of the plurality of channel taps prior to the determining substep.

18. An OFDM system for synchronizing the timing of a receiver to a received OFDM signal, the OFDM system comprising:

means for performing a first timing acquisition with a first received TDM pilot to determine a coarse timing estimate of the received OFDM signal;

means for performing a second timing acquisition with a second received TDM pilot to determine a fine timing estimate of the received OFDM signal, wherein the means for performing a second timing acquisition comprises:

means for determining the accumulated energy of a plurality of channel taps within a detection window for a plurality of start positions to form an accumulated energy curve, and

means for detecting a trailing edge of the accumulated energy curve; and

means for adjusting an FT collection window position for a subsequent OFDM symbol based on an output of the means for performing a second timing acquisition.

19. The OFDM system for synchronizing timing of a receiver to a received OFDM signal as claimed in claim 18, wherein the first TDM pilot is received before the second TDM pilot.

20. The OFDM system for synchronizing timing of a receiver to a received OFDM signal as claimed in claim 18, wherein the fine timing estimate is a refinement of the coarse timing estimate.

21. The OFDM system for synchronizing timing of a receiver to a received OFDM signal as claimed in claim 18, wherein the trailing edge is located using a weighted sum of an accumulated energy at a particular one of a plurality of starting positions and a negative finite difference of the accumulated energy curve at the particular starting position.

22. The OFDM system for synchronizing timing of a receiver to a received OFDM signal as claimed in claim 18, wherein a leading edge and a trailing edge of a flat region in the cumulative energy curve are both detected from a flat region in the cumulative energy curve that differs from a maximum point by a percentage of energy.

23. The OFDM system for synchronizing receiver timing to a received OFDM signal as claimed in claim 18, wherein each of the plurality of channel taps corresponds to a complex channel gain on a respective tap delay.

24. The OFDM system for synchronizing timing of a receiver to a received OFDM signal as claimed in claim 18, wherein the second TDM pilot comprises a cyclic prefix and a plurality of identical pilot sequences.

25. The OFDM system for synchronizing timing of a receiver to a received OFDM signal as claimed in claim 18, wherein said means for determining and said means for detecting are used at least partially simultaneously in time for a particular one of said plurality of channel taps.

26. The OFDM system for synchronizing a timing of a receiver with a received OFDM signal as claimed in claim 18, wherein the receiver is at least one of a wired receiver or a wireless receiver.

27. The OFDM system for synchronizing timing of a receiver to a received OFDM signal as claimed in claim 18, wherein the cumulative energy curve is filtered to reduce false trailing edge detection.

28. A method for synchronizing the timing of a receiver with a received signal, the method comprising the steps of: performing a first timing acquisition to determine a coarse timing estimate of the received signal;

performing a second timing acquisition with the TDM pilot to determine a fine timing estimate for a symbol in the received signal, wherein said step for performing the second timing acquisition comprises the sub-steps of:

determining an accumulated energy for a plurality of channel taps within a detection window for a plurality of start positions to form an accumulated energy curve,

detecting a trailing edge of the cumulative energy curve, an

For a particular channel tap of said plurality of channel taps, performing said determining substep and said detecting substep at least partially simultaneously in time; and

adjusting an FT collection window position for a subsequent symbol according to the step for performing a second timing acquisition.

29. A method for synchronizing timing of a receiver with a received signal as defined in claim 28, wherein the fine timing estimate is a refinement of the coarse timing estimate.

30. The method for synchronizing timing of a receiver to a received signal of claim 28, wherein the trailing edge is located using a weighted sum of an accumulated energy at a particular one of a plurality of start positions and a negative finite difference of the accumulated energy curve at the particular start position.

31. The method for synchronizing timing of a receiver with a received signal of claim 30, wherein the subsequent symbol is an OFDM symbol and comprises:

a plurality of data symbols, an

A plurality of Frequency Division Multiplexed (FDM) pilots.

32. A method for synchronizing the timing of a receiver with the timing of a received signal as recited in claim 30, wherein the leading edge and the trailing edge of a flat region in the cumulative energy curve are both detected from a flat region in the cumulative energy curve that differs by a percentage of the energy of the maximum point.

33. The method for synchronizing timing of a receiver to a received signal as recited in claim 28, wherein each of the plurality of channel taps corresponds to a complex channel gain on a corresponding tap delay.

34. The method for synchronizing timing of a receiver with a received signal of claim 28, wherein the receiver is at least one of a wired receiver or a wireless receiver.

35. The method for synchronizing timing of a receiver to a received signal of claim 28, wherein the accumulated energy curve is filtered to reduce false trailing edge detection.

36. A communication device for synchronizing timing of a receiver with a received signal, the communication device comprising:

a processor configured to:

performing a first timing acquisition with a first received Time Division Multiplexed (TDM) pilot to determine a coarse timing estimate of the received OFDM signal;

performing a second timing acquisition with a second TDM pilot to determine a fine timing estimate of the received OFDM signal, wherein said step for performing a second timing acquisition comprises the sub-steps of:

determining an accumulated energy for a plurality of channel taps within a detection window for a plurality of start positions to form an accumulated energy curve,

detecting a trailing edge of the cumulative energy curve, an

Adjusting a Fourier Transform (FT) collection window position for a subsequent OFDM symbol in accordance with the step for performing a second timing acquisition; and

a memory coupled to the processor.

37. The communication device of claim 36, wherein the first TDM pilot is received before the second TDM pilot.

38. The communications device of claim 36, wherein said fine timing estimate is a refinement of said coarse timing estimate.