HK1057949B - Receiving station with interference signal suppression - Google Patents

Description

Receiving station with interference signal suppression

Technical Field

Methods and apparatus related to wireless communication systems are disclosed, particularly with respect to selection mechanisms for signal combining methods.

Background

In a conventional wireless communication system, mobile stations (cellular phones, portable computers, etc.) are served by base stations. Such a base station acts as a communication relay station for the mobile station. Therefore, whenever a mobile station is to communicate with other parts of the communication system, the mobile station must communicate wirelessly with at least one base station. A mobile station sometimes leaves the service area of one base station and enters the service area of another base station. The base stations perceive this and switch communications from the first base station to the second base station by "handoff". It is common for a mobile station to communicate with both a first and second base station over a period of time. A mobile station communicating with more than one base station is referred to as a "soft handoff". At any one time in some cases, the mobile station may be in soft handoff with more than two base stations.

Soft handoff is desirable because it reduces dropped calls. In addition. Soft handoff allows the mobile unit to receive the same information from more than one source and use all of its received information (or energy) to help decode the information sent by each base station to the mobile station. Using information transmitted from multiple base stations means a reduction in the power level required by any one base station.

One type of wireless communication system is known as Code Division Multiple Access (CDMA). CDMA systems provide greater capacity than other systems. That is, there are more information channels that can be communicated simultaneously in a CDMA system than in other systems, such as Time Division Multiple Access (TDMA) or Frequency Division Multiple Access (FDMA) systems.

In a CDMA system that communicates voice and data simultaneously, a base station transmits simultaneously to a plurality of mobile stations in the base station's active area on the same frequency. In addition, each such base station transmits at the same frequency as each other base station in the network. Signals transmitted to a particular mobile station are distinguished from signals transmitted to other mobile stations only by the different codes with which they are transmitted. In contrast, in a TDMA system, transmissions are made to a first mobile station during a first time period and to a second mobile station during a second time period, without overlapping time periods. In an FDMA system, transmissions to a first mobile station are made on a first frequency and transmissions to a second mobile station are made on a second frequency. Because a CDMA receiver can receive more than one channel at a time tuned to a certain signal frequency, a CDMA receiver can perform a soft handoff more conveniently than a TDMA receiver or an FDMA receiver.

While CDMA systems theoretically have the advantage of being well suited for soft handoff, signals transmitted to a first mobile station using a first code appear as noise to a second mobile station attempting to receive signals transmitted to the second mobile station using a second code. A preferred method of reducing this interference is by making the codes assigned to signals transmitted from a base station orthogonal to the codes assigned to all other signals transmitted from that base station. However, the code used by the first base station to transmit signals cannot be orthogonal to the code used by the second base station to transmit signals. Therefore, the base station must carefully control the amount of power used to transmit signals to the mobile station. The power must be high enough to allow the signal to pass, but it is preferable that the power not exceed the required value because the presence of additional power is additional interference to other mobile stations and reduces the number of mobile stations that the base station can serve.

Because conventional CDMA communication systems must process both voice and data, certain performance requirements must be met. One such requirement is that the delay from the time information is transmitted at one end of the communication system until the time information is received at the other end of the communication system must be relatively short. That is, when two people are talking, any perceptible delay between the time that one end of the line speaks and the time that the other end of the line hears that speech will be experienced as annoying by both the talker and the listener.

In contrast, many data communication systems can tolerate a long delay between the time of sending information and the time of receiving information. The recently introduced CDMA systems specifically designed to process data have the advantage of being able to tolerate relatively long delays. Such systems are referred to herein as High Data Rate (HDR) systems. In an HDR system, a base station is dedicated to communicating with only one mobile station at any one time. The capacity advantage of CDMA is realized by HDR systems. However, the system may have difficulty or may not be satisfactory in performing a soft handoff for the following reasons. First, in an HDR system, transmissions from a base station are directed to a mobile station at any particular time. Thus, while the number of code channels transmitted from the HDR base station is substantially the same, at any one time all of the code channels are intended to be received by one mobile station. Therefore, the adjustment of the transmission time between two base stations becomes complicated in order to allow a soft handover between the two base stations. Second, in order to perform soft handoff, it is necessary to distribute the same data among more than one base station. This can greatly increase the amount of data transferred between base stations, especially for high data rate applications. Third, assuming relatively fixed channel conditions, the capacity of the system increases for an HDR system if the mobile unit can always connect to the best serving base station without soft handoff. This is true because HDR base stations typically transmit at maximum power to achieve the best data rate. That is, the rate at which data can be transmitted is directly proportional to the amount of power received. Thus, in order to maximize the data rate, the maximum power is transmitted. However, this increases the amount of interference to the first base station signal received by the mobile station attempting to receive signals from the second base station.

Accordingly, there is a need for a method and apparatus for reducing the amount of interference carried by a first base station to a mobile station attempting to receive signals from one or more other base stations.

In addition, if the interference is non-directional, such as noise, the optimal signal combining algorithm may cause degradation in receiver performance due to estimation errors. Thus, a selection mechanism for the signal combination method based on the interference characteristics is also needed.

Disclosure of Invention

The presently disclosed method and apparatus reduces the amount of interference imposed on a first base station by transmissions of other base stations in the same communication system. The presently disclosed method and apparatus allows for soft handoff to be less easily implemented or undesirable in certain communication systems, such as High Data Rate (HDR) communication systems that transmit multiple code channels at a time to a receiving station, such as a mobile station. That is, in a typical HDR system, each base station transmits to one receiving station at a time. In order to allow a soft handover between two base stations, the adjustment of the transmission and data transfer times between the two base stations becomes complicated. Also, under normal channel conditions, the capacity of the HDR system can be increased if soft handoff is not used. Thus, the presently disclosed method and apparatus changes the conventional method of performing soft handoff and is in accordance with techniques for using two or more antennas for reducing interference between transmissions from a first base station and transmissions from one or more other base stations.

In accordance with the presently disclosed method and apparatus, two antennas are used in a receiving station to receive transmissions. A rake receiver is coupled to each antenna. The rake receiver has a plurality of fingers, each finger having the ability to identify and independently decode signals arriving with different propagation delays (i.e., the delay that occurs between the time the signal is transmitted and the time the signal is received). By optimally combining the signals received by each individual finger of the rake receiver, the interference associated between the finger associated with the first antenna and the finger associated with the second antenna can be reduced to a minimum relative to the desired signal. The optimal combination requires the determination of optimal combination coefficients as follows.

The optimal combining coefficient for each signal received by the rake receiver fingers is determined by first pairing an output from a first finger associated with a first antenna with an output from a second finger associated with a second antenna. The first finger receives the desired signal at substantially the same propagation delay as the second finger. That is, the signal path decoded by a first finger differs from the signal path decoded by a second finger only in that the first finger is associated with a first antenna and the second finger is associated with a second antenna. The autocorrelation matrix is evaluated. In one presently disclosed method and apparatus, the autocorrelation matrix is an estimate of the autocorrelation of the received signal. Alternatively, the autocorrelation matrix is an estimate of the autocorrelation of the received noise plus interference.

In addition, the cross-correlation between the received signal and the transmitted signal is estimated by estimating the elements of the attenuation coefficient vector. Each element of the attenuation coefficient vector is an attenuation coefficient associated with one of the signal paths traversed by the signal received by the rake receiver. The fading coefficient vector is preferably estimated based on the pilot received by each finger.

The noise plus interference autocorrelation matrix is an estimate of the received noise component of each signal received by the rake receiver fingers. The received noise component for a particular finger is calculated by subtracting the finger received signal dependent attenuation coefficient from all signals received by that finger in the pilot pulse. In another method and apparatus now disclosed, noise and interference are estimated by subtracting signal y (m) (i.e., subtracting adjacent samples) from signal y (m +1) one chip later in time. In yet another method and apparatus now disclosed, an autocorrelation matrix R of noise plus interference_nnBy means of an autocorrelation matrix R from the received signal y (m)_yyThe attenuation coefficient vector multiplied by the transposed conjugate of the attenuation coefficient vector is subtracted to estimate.

Once the attenuation coefficient vector and autocorrelation matrix have been calculated for each pair of finger received signals, these values are used to calculate the optimal combining coefficients. Alternatively, once the attenuation coefficient vector and autocorrelation matrix for the noise plus interference for each pair of fingers have been calculated, they are also used to calculate the best combining coefficients.

From combining the received signals of each finger of the rake receiver using the optimal combining coefficients, it is desirable to compute the signal-to-interference plus noise ratio from the output of the optimal combiner. This ratio is calculated using the transposed conjugate of the best combined coefficient and the vector of attenuation coefficients. The result is a system in which interference from sources other than the desired signal source is suppressed relative to the desired signal to improve decoding. The resulting signal to noise plus interference calculation allows the receiving station to be configured to determine the data rate that the channel can sustain.

In another aspect, the present invention is directed to a selection mechanism for a method for signal combining in dependence on interference characteristics. Therefore, in one aspect of the present invention, the method of calculating optimal combining coefficients for a plurality of received signals is advantageous in that it comprises the steps of estimating autocorrelation matrices for a plurality of signals in dependence on received noise and a plurality of signal-related interferences; comparing the sizes of the off-diagonal elements of the autocorrelation matrix and the diagonal elements of the autocorrelation matrix; a step of setting the off-diagonal elements equal to zero if the off-diagonal elements of the autocorrelation matrix are significantly smaller than the size of the diagonal elements; and calculating a plurality of optimal combination coefficients according to the autocorrelation matrix.

Drawings

Fig. 1 is a schematic diagram of a wireless network in accordance with the presently disclosed method and apparatus.

A simplified block diagram of a receiving station incorporating the presently disclosed method and apparatus is shown in fig. 2a and 2 b.

FIG. 3 is a simplified block diagram of an optimal combiner of the presently disclosed method and apparatus.

FIG. 4a is a functional block diagram of a combined coefficient processor of the presently disclosed method and apparatus.

Fig. 4b is a functional block diagram of the combined coefficient processor 224' of an alternative method and apparatus.

Fig. 5 is an illustration of a data field in which three successive attenuation coefficient estimates are averaged and then inserted between successive averages.

Fig. 6a is a functional block diagram of a combined coefficient processor 600 of an alternative method and apparatus.

Fig. 6b is a functional block diagram of a combined coefficient processor 600' of another alternative method and apparatus.

Fig. 7 is a flow chart illustrating the steps of an algorithm for selecting the optimal signal combining algorithm based on interference characteristics.

Detailed Description

Fig. 1 is a schematic diagram of a wireless network 100 in accordance with the presently disclosed method and apparatus. In one presently disclosed method and apparatus, the wireless network 100 is a Code Division Multiple Access (CDMA) high speed data rate (HDR) system. The first base station 102 has an antenna 104. The first base station 102 transmits signals intended for reception by a receiving station 110, e.g. a mobile station, having two antennas 112, 114. The signal transmitted by the first base station 102 shows two distinct paths traversed from the first base station to the two antennas 112, 114 of the receiving station 110. Thus, 4 desired signals (y) from base station 102 are received at receiving station 110₁₁、y₁₂、y₂₁、y₂₂). Each desired signal has a different degree of delay (e.g., has a different propagation delay) due to the different paths traversed. The first subscript associated with the desired signal represents the antenna receiving the signal. The second subscript associated with the desired signal represents the propagation delay experienced by the signal.

It should be noted that although y₁₁、y₂₁Not all traversing the same path (e.g. they are received by different antennas as shown in the first subscript), but when y is compared₁₂And y₂₂Are substantially equal in delay. That is, in the signal y₁₁And y₂₁The difference in propagation delay between is much smaller thany₁₁And y₁₂Or y₂₂Difference in propagation delay between because of the signal y₁₂And y₂₂Traversed path far-ratio signal y₁₁The path traversed is long. Also, in signal y₁₂And y₂₂The propagation delay difference between will also be much smaller than in signal y₁₂And y₂₁Or y₁₁Difference in propagation delay between because of y₂₁And y₁₁Traversed path far-ratio signal y₁₂The path traversed is short.

The second signal source also transmits signals that are received by receiving station 110. For simplicity, the second signal source is described herein as a second base station 108 having an antenna 106. However, those skilled in the art will certainly appreciate that the second signal source may be a second antenna associated with the same or another base station, or a different portion of the same antenna transmitted from the same base station. However, the signal transmitted by the second base station 108 is not intended to be received by the receiving station 110. Both base stations 102 and 108 transmit wideband signals on the same frequency band. Thus, the signal received by the receiving station 110 from the second base station 108 interferes with the receiving station 110 receiving the signal transmitted by the first base station 102.

For ease of understanding, only two base stations 102, 108 are shown. And those skilled in the art will appreciate that there may be more than two base stations transmitting. Also, receiving station 110 is shown with only two antennas 112, 114. In one presently disclosed method and apparatus, however, receiving station 110 may be equipped with additional antennas.

In the presently disclosed method and apparatus, receiving station 110 uses signals received by two antennas 112, 114 to help suppress interference from signal sources transmitted from different antennas, or different portions of the antennas, to obtain a desired signal

A simplified block diagram of a receiving station incorporating the presently disclosed method and apparatus is shown in fig. 2a and 2 b. As mentioned above, the incoming signals are received on both antennas 112, 114 of the receiving station 110. The receiving station 10 preferably comprises two receiver modules 201a, 201 b. Each receiver module 201 includes: radio frequency/intermediate frequency (RF/IF) converters 200, 202; analog-to-digital (a/D) converters 204, 206; rake receivers 208, 210, pilot/data signal demultiplexers (demux)212, 214; and a plurality of Walsh decover modules 216a, 216b, 216c, 216d, 216e, 216 f.

Each of the two RF/IF converters 200, 202 is coupled to an associated one of the two antennas 112, 114. Thus, the signals received on each of the two antennas 112, 114 are coupled to a respective radio RF/IF converter 200, 202.

Each RF/IF converter 200, 202 is coupled to a respective one of two a/D converters 204, 206. a/D converters 204, 206 convert the outputs of the RF/IF converters 200, 202 into digital form. Alternatively, a single a/D converter may be used to convert the received analog signals on both antennas to digital form. Each a/D converter 204, 206 is coupled to a respective one of two rake receivers 208, 210.

Each rake receiver 208, 210 is capable of distinguishing each signal transmitted from a desired source base station and arriving at receiving station 110 experiencing a different propagation delay. Rake receivers for use in CDMA systems are well known in the art for receiving and identifying CDMA signals. Because of the signal y₁₁、y₁₂、y₂₁、y₂₂Different delays are encountered so that conventional rake receivers can distinguish between these signals. Each signal y having a distinguishable delay from the desired source (i.e., base station 102)₁₁、y₁₂、y₂₁、y₂₂Unique "fingers" 213a, 213b, 213c, 213d, 213e, 213f are assigned to rake receivers 208, 210. Each such finger 213 outputs a signal despread by the pseudo-random noise generated by the PN generator 211. The PN code output from the generator 211 is delayed by one of a plurality of delay blocks 209a, 209b, 209c, 209d, 209e, 209 f. The amount of delay applied by each delay module 209 is set so that the PN code output from each delay module 209 is synchronized with the PN code originally spread by the signal received from the desired source base station 102, plus the delay encountered in transmission from the base station 102 to the receiving station 110.

It is noted that the signals transmitted by each base station 102, 108 may be spread (i.e., coded) with the same PN code. While a substantially different delay is applied with respect to the start of the PN sequence used to encode each base station 102,108 signal. The difference in delay is substantially greater than the delay between any two signals transmitted from the same base station 102 that arrive at the receiving station 110 via different paths. Thus, by spreading signals transmitted from different base stations with the same PN code but substantially different delays, the signal from the first base station 102 can be distinguished from the signal from the second base station 108. Also, signals transmitted from the first base station 102 to the receiving station 110 have different propagation delays than signals transmitted from the second base station 108 to the receiving station 110. Therefore, these signals can be distinguished from each other. It is noted that the delay module 209 is not configured to facilitate reception of the transmitted signal from the second base station 108.

In a base station that generates a received signal by the presently disclosed method and apparatus, the pilot signal is time multiplexed with the data. In one such base station 102, the pilot and each data stream are covered (i.e., encoded) with different Walsh codes. The pilot is preferably covered with a Walsh code having a constant value, which can reduce the difficulty in decovering the pilot channel. During the time when the pilot channel is transmitted (i.e., the pilot burst), no data is transmitted. Two such pilot pulses occur in each forward link slot. The forward connection slot is a predetermined time period during which signals are transmitted from the base station to the receiving station. During the time when data is being transmitted (i.e., the data field), no pilot channel is transmitted. The data is code division multiplexed. That is, the data is divided into separate data streams. Each data stream is covered with a different Walsh code. All data streams are then transmitted at the same time. For example, a first portion of data is covered with a first Walsh code, a second portion of data is covered with a second Walsh code, and a third portion of data is covered with a third Walsh code. The first, second, and third portions are all transmitted simultaneously by the base station during the data field.

Since data and pilots are transmitted in a time multiplexed format, a method and apparatus are now disclosedIn the arrangement, receiving station 110 includes a pilot/data signal demultiplexer 212, 214 associated with each antenna 112, 114. While a single signal demultiplexer may be used to demultiplex the signals received by both antennas 112, 114. The output of the first demultiplexer 212 is a plurality of pilot streams y_p11(m)、y_p12(m)、…y_p1N(m) and a plurality of data streams y_d11(m)、y_d12(m)，…y_d1N(m) wherein y_p11(m) denotes pilot sample sequences, each obtained on a pilot channel received from the antenna 1 at a time "mT" and having a propagation delay of 1, and y_d11(m) denotes a sequence of data samples, each obtained on a data channel received from the antenna 1 at a time "mT" and having a delay of 1, while "m" is an integer and "T" is a time equal to one data chip.

Each pilot and data stream having the same numerical index is associated with the same finger 213 of the rake receiver 208, 210. Each data stream is coupled to a Walsh decover module 216a, 216b, 216c, 216d, 216e, 216 f. Each Walsh decover module 216 separates the code channels that were code-multiplexed into data fields prior to transmission from the base station 102. The outputs from the Walsh decover modules 216a, 216b, 216c, 216d, 216e, 216f separate the decovered data streams as is well known to those skilled in the art. These pilot and data signals are then coupled to an optimal combining processor 218.

The optimal combination processor 218 as shown in fig. 2b comprises 3 optimal combiners 220a, 220b, 220c and a combination coefficient processor 224. It is noted that each optimal combiner 220 is associated with a respective code channel (i.e., a Walsh code that covers data transmitted on that code channel). That is, if each Walsh decover module 216 outputs 3 data streams, each associated with a different code channel and already decovered by a different Walsh code, then 3 optimal combiners 220 are used. However, it should be understood that in alternative methods and apparatus, the number of code channels and optimal combiners 220 may differ from the 3 shown in FIG. 2 b. Furthermore, it is noted that a single module can perform the functions of more than one optimal combiner. Each optimal combiner 220 is coupled to all Walsh decover modules 216 and provides data transmitted on one code channel to each optimal combiner 220 via multiple paths received on two antennas. The output from each optimal combiner 220 is a stream of data symbols representing the signal data transmitted by the modulated base station 102. Due to the processing by the optimal combiner processor 218, less interference is encountered when decoding the symbols than would be produced by combining the inputs of the optimal combiner 220 as is conventional. That is, the SINR of the signal modulated with the output symbols is greater than the SINR of any data stream input to the optimal combiner 220. Also, the SINR of the output symbols is greater than the SINR resulting from combining the inputs of the conventional optimal combiner 220.

Fig. 3 is a simplified block diagram of an optimal combiner 220a of the presently disclosed method and apparatus. Because each optimal combiner 220 is effectively identical, only one optimal combiner 220a will be discussed. The optimal combiner 220a includes a plurality of two-input multiplication modules 302. The multiplication module 302 multiplies the first input signal and the second input signal and provides a product at the output of the multiplication module 302. It is noted that the multiplication module 302 may be implemented as a programmable device such as a general purpose processor or DSP or as dedicated hardware or programmable logic circuitry, for example, or any other execution function that allows the multiplication function method (e.g., a processing function in a circuit or Application Specific Integrated Circuit (ASIC)) to be performed.

The number of multiplication modules 302 in optimal combiner 220 is preferably equal to the total number of Walsh decover modules 216 in receiving station 110. A first input 302 to each multiplication module 302 is coupled to a uniquely corresponding Walsh decover module 216. Thus, each data stream received by one of the code channels associated with the optimal combiner 220a is coupled to a first input of a particular dual-input multiplication module 302a associated with the Walsh decover module 216 a. In the method and apparatus shown in FIG. 2, there are 6 decover modules 216a, 216b, 216c, 216d, 216e, 216 f. Thus, as shown in the optimal combiner 220a in FIG. 3, there are 6 multiplication modules 302a, 302b, 302c, 302d, 302e, 302 f. The second input of each multiplication module 302 is through a signal line223 is coupled to a combining coefficient processor 224. The combining coefficient processor 224 calculates an optimal combining coefficient (w)_ij ^*(m)). The subscript "i" indicates the particular antenna with which the finger 213 is associated, while the subscript "j" indicates the particular delay experienced by the signal transmitted from the desired source. As mentioned above, when signals are received by both antennas 112, 114 (e.g., signal y shown in FIG. 1)₁₁And y₂₁) Where the paths traversed are not equal due to the fact that the signals are received by different antennas, the same second subscript is used to indicate that the signals actually experience equal propagation delays. Similarly, the second index of the best combining coefficient indicates which propagation delay the signal multiplied by that best combining coefficient encounters.

The multiplication performed by the multiplication module 302 allows each received signal to be weighted and rotated (i.e., the phase and amplitude of the received signal can be adjusted). By rotating the phase, the signal-to-interference-plus-noise ratio (SINR) of the combined signal output from the summing block is optimized. That is, the SINR will have the highest probability. Thus, the interference caused by the undesired signal will be reduced. That is, the power received from base station 102 attempting to communicate with receiving station 110 may be maximized relative to the power received from base station 108 not attempting to communicate with receiving station 110.

A coefficient inclusion processor 224 (shown in figure 2 b) is coupled to a second input of each multiplication module 302 via signal line 223. It is noted that in order to reduce the number of lines in fig. 2b, a single signal line 223 is shown from the combining coefficient processor 224 to each optimal combiner 220. And this line 223 represents a connection over which each multiplication module 302 is provided with a plurality of wij (m) values in each optimal combiner (6 in the case shown in fig. 2b and 3). These values allow the multiplication module 302 to optimally adjust the received signals before combining in the summation module 304. The summing module 304 adds the products to combine each received rotated signal. Thus, the optimal combiner 220a performs a dot product operation. The output of the summing block 304 (i.e., the output from the optimal combiner 220 a) is provided as input samples to a conventional decoding or detection block or processor, such as the error correction decoder 226, that performs conventional decoding or detection functions. The output from the summing module 304 may be expressed as:

wherein H represents a conjugate substitution; y (m) ═ y₁₁(m)、y₁₂(m)、…y_ij(m)、…〕^TIs a vector containing the sampled received signal at each rake finger associated with each communication 112, 114 at time mT after Walsh decovering; y is_i，jIs the mT time and i after Walsh decovering^thJ of antenna coupling^thThe signal received in rake fingers 213; w (m) ═ w₁₁(m)、w₁₂(m)、…w_ij(m)、…〕^TIs the vector containing the best combination coefficient at time mT.

Note j associated with the first antenna^thThe rake finger has j associated with the second antenna^thRake refers to the same delay. For example, 2, receiving signals from the first antenna 112^ndThe delay imposed by the delay module 209b associated with the rake finger 213b is related to the 2 received signal from the second antenna 114^ndThe delay applied by the delay module 209e associated with the rake finger 213e is the same. Thus, in the presently disclosed non-summing device, each delay module 209 associated with the first antenna 112 preferably has a counterpart delay module 209 associated with the second antenna 114. Each module 209 of such a pair of paired delay modules 209 preferably has the same delay.

After Walsh decovering, at i^thJ of the antenna^thThe signal received at the rake fingers 213 may be expressed as:

y_ij(m)＝c_ij(m)·x(m)+n_ij(m) (2)

where x (m) is the symbol transmitted at time mT; c. C_ij(m) is the attenuation coefficient at time mT; and n is_ij(m) is the sum of thermal noise plus mT time and i^thJ of antenna coupling^thThe complex number of the interference at the rake fingers 213. Coefficient of attenuation c_ij(m) is the instantaneous path gain at time "mT" including the effects of propagation loss, shadowing, and fast fading. In binary phase shift keying, the value of the symbol x (m) is either +1 or-1. Whereas in quadrature phase shift keying, quadrature amplitude modulation, or other such modulation techniques, the symbol x (m) belongs to a modulation constellation.

Generation of optimal combining coefficients

A disclosed method and apparatus for determining optimal combining coefficients will now be described in detail. Fig. 4a is a functional block diagram of the combiner coefficient processor 224 of one presently disclosed method and apparatus. It is noted that the operation of each optimal combiner 220 is the same. Thus, for simplicity, the operation of only one such optimal combiner 220 will be described.

Each function performed by the modules shown in fig. 4a may be implemented by a programmable device such as a general purpose processor or a DSP, or special purpose hardware or programmable logic circuitry, or any other method for allowing the function to be performed, such as a processing function in a circuit or Application Specific Integrated Circuit (ASIC). These functions may be performed by one module or by a plurality of modules. Moreover, each such module may be physically combined with or physically separated from one or more other modules.

The combining coefficient processor 224 is coupled to both pilot/data signal demultiplexers 212,214. The pilot/data multiplexer 212 will direct the pilot signal y_p1(m) are provided to a combining coefficient processor 224, where y_p1(m)＝〔y_p11(m)，y_p12(m)…y_p1N(m). Pilot/data multiplexer214 will direct signal y_p2(m) are provided to a combining coefficient processor 224, where y_p2(m)＝〔y_p21(m)，y_p22(m)…y_p2N(m)〕。

In a presently disclosed method and apparatus, the values of the optimal combining coefficients are taken as the cross-correlation r of the autocorrelation matrix and the received signal y (m) with the transmitted symbols x (m)_yx(m) is calculated as a function of (m). In one presently disclosed method and apparatus, the autocorrelation matrix is an estimate of the autocorrelation of the received signal. Therefore, the combination coefficient processor 224 calculates the optimal combination coefficient as follows:

wherein: r_yy(m) is a code sequence contained in the Walsh decover (i.e., R)_yy(m)＝E[y(m)·y^H(m)]) Then, at time mT, the vector y (m) of the sampled received signal at each finger 213 coupled to each antenna is [ y ]₁₁(m)，y₁₂(m)，…y_ij(m)，…]^TThe autocorrelation matrix of (a); and r_yx(m) is the vector y (m) and the transmission symbols x (m) (i.e., r)_yx＝E[y(m)·x*(m)]) Where E represents the expected value defined in statistical mathematics, and x (m) represents the complex conjugate of x (m).

In an alternative method and apparatus, the optimal combining coefficients are calculated by the combining coefficient processor 224' as follows (as shown in FIG. 4b and described in detail below):

wherein R is_nn(m) is the thermal noise plus interference vector n (m) ═ n₁₁(m)，n₁₂(m)，…n_ij(m)…]^TOf (i.e. R)_nn(m)＝E[n(m)·n^H(m)]) (ii) a And r_yx(m) is as defined above. In a system where the guide symbol is represented by | x | ═ 1;

r_yx＝E[y(m)x*(m)]＝c＝[c₁₁(m)，c₁₂(m)，…，c_ij(m)…]^T(4a) wherein c is_ij(m) is the attenuation coefficient at time mT.

The variable W described by equation (3) and W' described in equation (4) differ only by a scalar coefficient. That is to say that the first and second electrodes,

W’＝(1+h)W (5)

wherein:

cross correlation r_yxValuation

Cross-correlation r between received signal y (m) and transmitted symbol x (m)_yxThe estimate is determined by the vector c of attenuation coefficients during the pilot pulse in the forward connecting bin, since the cross-correlation r is_yxEqual to the attenuation coefficient vector c ═ c as described above₁₁，c₁₂，…c_ij…]^T。

Estimation of cThe values are performed by the attenuation coefficient estimation block 401 using the pilot pulses of the forward link bin as follows. The attenuation factor module 401 receives each pilot signal y from each of the two pilot/data demultiplexer 212, 214 outputs_p11(m)，y_p12(m)，…y_p1N(m)，y_p21(m)，y_p22(m)，…y_p2N(m) of the reaction mixture. For simplicity reasons, FIG. 4a shows vector y_p(m)＝y_p11(m)，y_p12(m)，…y_p1N(m)，y_p21(m)，y_p22(m)，…y_p2N(m) of the reaction mixture. In one presently described method and apparatus, the transmitted signal during the pilot pulse is equal to a constant 1 (i.e., x-1). Thus, the attenuation coefficient vector c_ijEach element of (m) can be estimated as:

whereinIs used for i^thJ of the antenna^thAttenuation coefficient c of the pilot pulse at the rake fingers_ij(m) evaluation, y_pij(m) is i^thJ of the antenna^thM of received signal in pilot pulse at rake finger^thSamples and M is the number of symbols in the pilot pulse.

Attenuation coefficient determined in equation (6)The cross-correlation r is given only in the pilot pulse_yxAn estimate of (2). Therefore, to perform coherent detection and determine the optimal combining coefficient using equation (3)Attenuation coefficient in data sliceThe estimate is calculated in a first interpolation module 403.

In one presently disclosed method and apparatus, attenuation coefficients in data slicesIs estimated by two attenuation coefficients determined in successive pilot pulses by the linear interpolation block 403Interpolation between estimates. Or, attenuation coefficient in data sheetIs estimated by averaging a plurality of attenuation coefficients in successive pilot pulses in the block 405(e.g., two or three) are averaged. Interpolation is then performed between two consecutive calculated averages by interpolation module 403.

Fig. 5 is an illustration of a data field in which three successive attenuation coefficient estimates are averaged and then inserted between successive averages. Fig. 5 shows two forward attachment slots 500, 502. Each forward coupler slot 500, 502 has two pilot pulses 504, 506, 508, 510. The attenuation coefficient for each pilot pulse 504, 506, 508, 510 is estimated. A first average attenuation coefficient estimate c (k) is calculated by summing the three estimates of the first three consecutive pilot pulses 504, 506, 508 and dividing by 3. Next, a second average attenuation coefficient c (k +1) is calculated by adding the attenuation coefficients of each pilot pulse 506, 508, 510 and dividing by 3. A linear interpolation is performed between the first and second average attenuation coefficients. To estimate the attenuation coefficient for the portion of the data at a distance a from the pilot pulse 506, the following equation is used:

c(m)＝(1-a)·c(k)+a·c(k+1)，0＜a＜1 (7)

are mutually connectedThe vector r_yxBy repeating this process for each rake finger for each antenna. Autocorrelation matrix R_yyEvaluation of

The autocorrelation matrix of the received signal may be represented as:

where M is the number of samples used to perform the estimation, and y (M) ═ y₁₁(m)，y₁₂(m)，y_1N1(m)，y₂₁(m)，y₂₂(m)，…y_2N2(m)]^TIs a vector containing the received signal; y is_ij(m) is the i sampled at mT time after Walsh decover^thAn antenna and j^thA received signal of a rake receiver finger; n is a radical of₁Is the rake index associated with antenna 1, and N₂Is the rake index associated with antenna 2.

As can be seen from equation (8), at N₁＝N₂In the case of (i.e. receiving input signals using the same number of fingers on each of the two antennas), R_yyIs a 2N × 2N matrix, comprising 2 × 2 sub-matrices. Thus, the number of 2 × 2 sub-matrices is equal to N²。

Those skilled in the art will recognize that the interference of the different rake fingers for each antenna is uncorrelated due to differences in the propagation delays of the signals received by the different rake fingers. Thus, the autocorrelation matrix R is derived from rake fingers having different delays_yyCan be assumed to be 0: only the calculations need to be made with the same delay (i.e., j)₁＝j₂) Is determined by the 2 x 2 autocorrelation matrix estimate of the rake finger signal.

In the presently disclosed method and apparatus using two antennas, each 2 x 2 sub-matrix is located in a matrix R_yy(as shown above) diagonal R^(s)Can be expressed as follows:

wherein y is_1s(m) is the passage of s at time mT^thRake refers to the signal received by the first antenna, and y_2s(m) is the passage of s at time mT^thThe rake refers to the signal received by the second antenna. Each zero in the matrix R shown above represents an assumed value of a non-diagonal 2 x 2 sub-matrix.

In the presently disclosed method and apparatus, the autocorrelation matrix of the received signal is used to estimate R by using pilot pulses in the data slice_yyValue R_yyThe estimation module 407 estimates. And in alternative methods and apparatus, R_yyThe evaluation module uses the data slice directly, or both the pilot pulse and the data slice to determine R in the data slice_yyThe value is obtained.

Estimating R in a presently disclosed system using pilot pulses in a data slice_yyIn the process and apparatus of (1), R_yyThe interpolation module 411 will determine the value of R from the pilot pulse_yyAnd (4) inserting. In alternative methods and apparatus, R_yyThe estimation block 407 is used to calculate the R determined in two or three pilot pulses_yyR of the mean value_yyThe averaging module 409 is coupled. R_yyAveraging output of averaging module 409 and determination of R in data using interpolated averages_yyThe interpolation module 411 of values is coupled. The averaging and interpolation performed by the averaging and interpolation blocks 409, 411 is substantially the same as that done in the averaging and interpolation blocks 405, 403.

Determination of the optimal combining coefficient w:

optimum combination coefficientw is determined by the combining coefficient estimation module 415 using equation (3) as described above. Once R is present_yyFrom R_yyThe estimation block 407 estimates and inserts R's in the representative slice_yy，R_yyIs compared with the inverse matrix R_yyCoupled to the inversion module 413. Equation (3) is then used to determine the optimal combining coefficient.

In another alternative method and apparatus, R_yyDetermined by the pilot pulse. Equation (3) is used to determine the optimal combining coefficient in the pilot pulse. Linear interpolation is then used to determine the optimal combining coefficients in the data slice.

Due to the matrix R_yyAll off-diagonal submatrices R of^(s)The complexity of the inversion calculation required in equation (3) is low for zero, i.e., the present invention can be used without applying the entire matrix R_yyIn the case of inversion. Furthermore, the sub-matrix R^(s)May be inverted separately.

The above algorithm is common and can be applied to M × M classes of autocorrelation matrices, where for autocorrelation matrix inversion we can use direct inversion or the well-known Recursive Least Squares (RLS) algorithm.

Estimation of autocorrelation matrices

Fig. 4b is a functional block diagram of the combined coefficient processor 224' of an alternative method and apparatus. As shown in fig. 4b, the module included in the combining coefficient processor 224' is actually the same as the module in the combining coefficient processor 224 shown in fig. 4 a. While the processor 224' of fig. 4b includes calculating the noise plus interference R_nnR of an estimate of the autocorrelation matrix of (m)_nnThe estimator module 407' replaces R of FIG. 4a_yyAn evaluation module 407. To estimate noise plus interference R_nn(m) preferably using a pilot pulse. The first formula is expressed as follows:

where n (M) is the noise estimated at time mT plus the interference of the receiver in the samples and M is the number of samples (i.e., the number of symbols in the pilot pulse) used to perform the estimation. Each component in the vector n (m) is represented by R_nn(m) the estimation block 407' receives the pilot pulse y by_ij(m) minus the channel gain c_ij(m)：

n_ij(m)＝y_pij(m)-c_ij(m) (10)

Wherein y is_pij(m) is i after Walsh decovering^thJ of the antenna^thM of the received signal in the pilot pulse at the rake fingers^thSampling, and c_ij(m) is estimated by using the attenuation coefficient estimation block 401' of equation (6) as described above. The vector n (m) is generated by repeating each rake finger processing for each antenna 112, 114.

Or, R_nnThe estimate can be estimated by the following equation:

wherein each component of the vector n (m) is represented by R_nn(m) the estimator module 407' determines using the following equation:

n_ij(m)＝y_pij(m)-y_pij(m+1) (12)

where M is the number of samples used to perform the estimation (i.e., the number of symbols in the pilot pulse, minus 1).

One pilot symbol of a rake finger of the first antenna 112 is subtracted from the next pilot symbol of the same rake finger of the same antenna, according to equation (12). M is the number of samples used to perform the evaluation, i.e. (number of symbols in pilot pulse-1).

It is noted that calculating R_nnAnother alternative of (a) is:

R_nn＝R_yy-cc^H (13)

wherein cc is^HIs the product of the attenuation coefficient vector and the transposed conjugate of the attenuation coefficient vector.

In one presently disclosed method and apparatus, the second interpolation module 411 inserts R in the pilot_nnValue to determine R in a slice_nnThe value is obtained. Or the mean module 409 for a plurality of R determined by the steering_nnValue (i.e. two or three R)_nnPilot value) is averaged. The mean and interpolated mean are then used to determine R in the slice_nnThe second interpolation module 411 of values is incorporated as described above in connection with fig. 5. It is noted that the functions of the combining coefficient processors 224, 224 ' are substantially the same, except that the processing performed by the evaluation modules 407, 407 ' and the processing performed by the combining coefficient estimation modules 415, 415 '.

In the method and apparatus now disclosed as shown in FIG. 4b, once R is determined in a slice_nnValue of (1), matrix R_nnIs coupled to the inversion module 413. Inverted R is output from the inversion module 413_nnCoupled to the combining coefficient estimation module 415'. Equation (4)) And then applied to a combining coefficient estimation block 415 'to determine the own combining coefficient w'.

Post-estimation interpolation

Fig. 6a is a functional block diagram of a combined coefficient processor 600 of an alternative method and apparatus. As shown in fig. 6a, the combining coefficient processor 600 comprises substantially the same modules as those of the combining coefficient processor 224 shown in fig. 4 a. Whereas in the processor 600 of fig. 6a the interpolation is performed after the best combination coefficient estimation. Therefore, the attenuation coefficient estimation block 401 is directly coupled to the estimation block 415. Likewise, R_yyThe evaluation module 407 is directly coupled to the matrix inversion module 413. To determine the value of w during the data portion of each forward connected slot, the estimation block 415 is then coupled to an interpolation block 601 which performs a linear interpolation between the output w values from the estimation block 415.

Similarly, fig. 6b is a functional block diagram of a combined coefficient processor 600' in another alternative method and apparatus. As shown in fig. 6b, the combination coefficient processor 600 'includes substantially the same modules as those of the combination coefficient processor 224' shown in fig. 4 b. Whereas in the processor 600' of fig. 6b the interpolation is performed after the best combination coefficient estimation. Therefore, the attenuation coefficient estimation block 401 is directly coupled to the estimation block 415. Likewise, R_yyThe evaluation module 407' is directly coupled to the matrix inversion module 413. To determine the value of w 'during the data portion of each forward connected slot, the estimation block 415 is then coupled to an interpolation block 601 which performs a linear interpolation between the output w' values from the estimation block 415.

Signal-to-interference-plus-noise (SINR) ratio estimation

In one presently disclosed method and apparatus, c from the combining coefficient processor 224_ij(m) and w_ijThe (m) value is coupled to the SINR estimation module 228 (as shown in fig. 2 b). In one presently disclosed method and apparatus, the SINR may be calculated as follows:

where w is determined from equation (3) and c is determined from equation (6).

In another method and apparatus for determining w 'using the combining coefficient processor 224', the SINR may be calculated as follows:

SINR＝w’^Hc (15)

wherein c is determined from (6).

In one presently disclosed method and apparatus, SINR is used to determine the rate at which data is transmitted from base station 102 to receiving station 101. As shown in fig. 2a and 2b, the combining coefficient processors 224, 224' are coupled to the SINR estimation module 228 via signal lines 230, 231. The signal line 230 provides the SINR estimation block 228 with the w or w 'values calculated by the combining coefficient processors 224, 224', and the signal line 231 provides the SINR estimation block 228 with c_ij(m) value. The SINR estimation module 228 determines the SINR according to equation (14) or (15). The SINR estimation module 228 is coupled to the DRC module 232. The DRC module 232 determines the rate at which data is received from the base station 102 taking into account the SINR of the data received from the base station 102. This rate is then communicated to the base station.

LLR calculation

Most communication systems require log-likelihood ratio (LLR) estimates of the coded bits in order to perform decoding at the receiver (e.g., iterative or "turbo" decoding, conventional viterbi decoding, etc.). One advantage of the presently received method and apparatus is that LLR values can be easily calculated from the soft decision values denoted by w or w'.

For example, assume Quadrature Phase Shift Keying (QPSK) or 4-Quadrature amplitude modulation (4-QAM) at the transmitterAnd (4) end use. Also, assume d₀And d₁Respectively representing and modulating the signal y_d，ijThe associated first and second coded bits map coded bit 0 at the modulator input to a modulation value of +1 and coded bit 1 at the modulator input to a modulation value of-1, such that LLR values can be calculated using equations (16) and (17) or equations (18) and (19):

LLR(d₀|y_d(m))＝4·Re(w’(m)^H·y_d(m)) (16)

LLR(d₁|y_d(m))＝4·Im(w’(m)^H·y_d(m)) (17)

LLR(d₀|y_d(m))＝4·(1+h)·Re(w(m)^H·y_d(m)) (18)

LLR(d₁|y_d(m))＝4·(1+h)·Im(w(m)^H·y_d(m)) (19)

wherein H represents a transposed conjugate; re (-) and Im (-) denote real and imaginary parts of the complex number, respectively; y is_d(m)＝[y_d，11(m)，y_d，12(m)，…y_d，ij(m)，…]^TIs a vector containing the sampled received data signal at each finger 213 associated with each antenna 112, 114 at time mT after Walsh decovering; y is_d，ijIs at time mT after Walsh decover, and i^thJ of antenna coupling^thData received in rake fingers 213; w (m) ═ w₁₁(m)，w₁₂(m)，…w_ij(m)…]^TIs a vector containing the optimum combination coefficient estimated using equation (3), w '(m) ═ w'₁₁(m)，w’₁₂(m)，…w’_ij(m)…]^TIs the vector at time mT containing the best combination coefficient estimated using equation (4), and (1+ h) is defined as equation (5).

For a BPSK signal, only equation (16) or (18) need be used because the imaginary part is 0.

Thus, in one presently disclosed method and apparatus, the error correction decoder 226 performs the calculations as shown in equations (16) and (17), or equations (18) and (19).

Combinatorial approach selection

According to one embodiment, a selection mechanism is employed to determine whether to use the best combination algorithm, the maximum ratio combination algorithm, or a combination of the maximum ratio combination algorithm and the best combination algorithm. The difference between the maximum ratio combining algorithm and the optimal combining algorithm described herein is the calculation of the combining coefficients. The maximum ratio combining algorithm has an advantage in that the off-diagonal elements of the received signal autocorrelation matrix, which are used to calculate the optimal combining coefficients as described above, are set to 0. This selection mechanism takes advantage of the fact that there is no directional interference, such as noise, making the maximum ratio combining algorithm optimal.

The method steps in the flow chart shown in fig. 7 may be used to implement the selection mechanism. Such steps may be conveniently performed by a processor in combination with the combination coefficient processor 600 as described above. Alternatively, such steps may be performed by a DSP, an ASIC, discrete gate or transistor logic, discrete hardware components such as registers, FIFO and comparators, or a processor executing a set of micro-program firmware instructions, or any conventional programmable software module. The processor is preferably a microprocessor. But in the alternative, the processor may be implemented as a conventional processor, controller, microcontroller, or state machine. The software modules may reside in RAM memory, flash memory, registers, or any other form of writeable storage medium known in the art.

In step 700, the autocorrelation matrix of the received signals formed by the processor performs the optimal combination as described above. The processor then proceeds to step 702. In step 702, the processor compares the size of the off-diagonal elements of the autocorrelation matrix to the size of the diagonal elements of the autocorrelation matrix. If the size of the off-diagonal elements is significantly smaller than the size of the diagonal elements, the processor proceeds to step 704. On the other hand, if the size of the off-diagonal elements is not significantly smaller than the size of the diagonal elements, the processor proceeds to step 706. Some or all of the off-diagonal elements are set to 0 in step 704 and the maximum ratio combining coefficient or a combination of the maximum ratio combining coefficient and the optimal combining coefficient is calculated by a modified autocorrelation matrix, for example, according to equation (3) above. In step 706, the optimal combination coefficients are calculated from the unmodified autocorrelation matrix, e.g., according to equation (3) above.

The comparison of step 702 is performed for each off-diagonal element of the autocorrelation matrix. Off-diagonal element c_ij(non-diagonal elements of row i and column j) and then associated diagonal element c_iiAnd c_jj(the diagonal element of the ith row and ith column and the diagonal element of the jth row and jth column). For example, given off-diagonal element c₂₃Squared and then squared with diagonal element c₂₂And c₃₃The product of (a) is compared.

In a particular embodiment, in step 704, the size of all off-diagonal elements in the autocorrelation matrix is set to 0 if and only if, for each off-diagonal element, the square of the off-diagonal element is significantly less than the product of the associated diagonal elements. In an alternative embodiment, the size of a given off-diagonal element is set to 0 only if the square of the off-diagonal element is significantly less than the product of the associated diagonal elements in step 704.

In a particular embodiment, an autocorrelation matrix comprising a plurality of 2 x 2 sub-matrices with diagonal lines is employed as described above, and the determination made in step 702 of whether the size of the two off-diagonal elements of each 2 x 2 sub-matrix is significantly smaller than the size of the two diagonal elements of the same matrix is performed by squaring the size of each of the two off-diagonal elements and obtaining the product of the sizes of the two diagonal elements. The obtained ratio α of the square size of each of the two off-diagonal elements to the product of the sizes of the two diagonal elements is compared. If the squared magnitude of one of the two off-diagonal elements is less than α multiplied by the product of the magnitudes of the two diagonal elements, the two off-diagonal elements are set to 0 in step 704 and the maximum ratio combining coefficient is calculated using the modified autocorrelation matrix. Otherwise (e.g., if none of the two off-diagonal elements has a squared magnitude less than a multiplied by the product of the two diagonal element magnitudes), the optimal combining coefficient is computed using the uncorrected autocorrelation matrix in step 706. In one embodiment, the ratio α is significantly less than 1. In another embodiment the split ratio alpha is equal to 1/50.

Industrial applications

The disclosed methods and apparatus have potential for development in the industry and may be manufactured and used when wireless data transmission is desired. The various parts of the apparatus and methods described herein, listed separately from the others, may be entirely conventional and are incorporated in the present claims.

While various apparatus and method models have been discussed, the true spirit and scope of the invention is not limited thereto but only by the claims set forth below and their equivalents.

Claims (18)

1. A method for calculating optimal combining coefficients for a plurality of received signals, the method comprising the steps of:

estimating an autocorrelation matrix of a plurality of signals as a function of received noise and correlated interference of the plurality of signals;

comparing the off-diagonal element size of the autocorrelation matrix to the diagonal element size of the autocorrelation matrix;

setting the off-diagonal element to 0 if the size of the off-diagonal element is significantly smaller than the size of the diagonal element; and

and calculating a plurality of optimal combination coefficients according to the autocorrelation matrix.

2. The method of claim 1, wherein the autocorrelation matrix comprises a plurality of diagonal 2 x 2 sub-matrices, wherein each sub-matrix comprises two diagonal elements and two off-diagonal elements, and wherein the step of comparing comprises the steps of squaring the size of each off-diagonal element of each of the sub-matrices and multiplying the size of the diagonal elements of the same sub-matrix to obtain a product, and wherein the step of setting comprises setting the two off-diagonal elements of each sub-matrix to 0 if the squared size of each off-diagonal element is less than a predetermined fraction of the product of the diagonal elements of the same sub-matrix.

3. The method of claim 2, wherein the predetermined score is less than 1.

4. The method of claim 2 wherein the predetermined fraction is equal to 1/50.

5. The method of claim 1, wherein the comparing step includes the step of squaring each off-diagonal element size and, for each off-diagonal element, multiplying two diagonal elements located in the same row or column as the off-diagonal element to obtain a product, and the setting step includes the step of setting each off-diagonal element to 0 if the squared size of each off-diagonal element is significantly less than the product of two diagonal elements located in the same row or column as the off-diagonal element.

6. The method of claim 1, wherein the comparing step includes the step of squaring each off-diagonal element size and, for each off-diagonal element, multiplying two diagonal elements located in the same row or column as the off-diagonal element to obtain a product, and the setting step includes the step of setting 0 for each off-diagonal element having a squared size significantly smaller than the product of two diagonal elements located in the same row or column as the off-diagonal element.

7. An optimally combined processor, comprising:

a) a combined coefficient processor comprising

i) An attenuation coefficient estimation module;

ii) a first interpolation module coupled to the attenuation coefficient estimation module;

iii) an autocorrelation estimation module;

iv) a second interpolation module coupled to the autocorrelation estimation module;

v) an inversion module coupled with the second interpolation module; and

vi) a combining coefficient estimation module coupled to the inversion module and the first interpolation module; and

b) at least one optimal combiner, each optimal combiner coupled to the combining coefficient processor and comprising:

i) a plurality of two-input multiplication modules, each module having an output, a first input of each of said multiplication modules configured to receive a signal from one of a plurality of fingers of a rake receiver, a second input of each of said multiplication modules coupled to a combining coefficient estimation module to receive said combining coefficient; and

ii) an addition module having the same number of inputs as the number of said two-input multiplication modules, each of said inputs being coupled to a corresponding one of the multiplication module outputs.

8. The optimal combining coefficient processor of claim 7 wherein the autocorrelation estimation module is a Ryy autocorrelation module of the received signal.

9. The optimal combining coefficient processor of claim 7 wherein the autocorrelation estimation module is a noise plus interference Rnn autocorrelation module.

10. An optimally combined processor, comprising:

a) a combined coefficient processor comprising

i) An attenuation coefficient estimation module;

ii) an autocorrelation estimation module;

iii) an inversion module coupled to the autocorrelation estimation module; and

iv) a combining coefficient estimation module coupled to the inversion module;

v) an interpolation module coupled to the estimation module; and

b) at least one optimal combiner, each optimal combiner coupled with the interpolation module and comprising:

i) a plurality of two-input multiplication modules, each module having an output, a first input of each of said multiplication modules configured to receive a signal from one of a plurality of fingers of a rake receiver, a second input of each of said multiplication modules coupled to said interpolation module to receive said combining coefficients; and

ii) an addition module having the same number of inputs as the number of said two-input multiplication modules, each of said inputs being coupled to a corresponding one of the multiplication module outputs.

11. The optimal combining coefficient processor of claim 10 wherein the autocorrelation estimation module is the R of the received signal_yyAnd an autocorrelation module.

12. The optimal combining coefficient processor of claim 10 wherein said autocorrelation estimation module is noise plus interference R_nnAnd an autocorrelation module.

13. A receiving station, comprising

a) A first antenna;

b) a first receiving module coupled with the first antenna;

c) a second antenna;

d) a second receiving module coupled with the second antenna;

e) a best combining processor coupled with the first and second receiving modules, the best combining processor comprising:

i) a combined coefficient processor, comprising:

(1) an attenuation coefficient estimation module;

(2) a first interpolation module coupled to the attenuation coefficient estimation module;

(3) an autocorrelation estimation module;

(4) a second interpolation module coupled to the autocorrelation estimation module;

(5) an inversion module coupled with the second interpolation module; and

(6) a combining coefficient estimation module coupled to the inversion module and the first interpolation module; and

ii) at least one optimal combiner, each optimal combiner coupled to the combining coefficient processor and comprising:

(1) a plurality of two-input multiplication modules, each module having an output, a first input of each of said multiplication modules configured to receive a signal from one of a plurality of fingers of a rake receiver, a second input of each of said multiplication modules coupled to a combining coefficient estimation module to receive said combining coefficient; and

(2) an addition module having as many inputs as the number of said two-input multiplication modules, each of said inputs being coupled to a corresponding one of the multiplication module outputs.

14. The receiving station of claim 13 wherein said autocorrelation estimation module is the R of the received signal_yyAnd an autocorrelation module.

15. The receiving station of claim 13 wherein said autocorrelation estimation module is noise plus interference R_nnAnd an autocorrelation module.

16. A receiving station, comprising:

a) a first antenna;

b) a first receiving module coupled with the first antenna;

c) a second antenna;

d) a second receiving module coupled with the second antenna;

e) a best combining processor coupled with the first and second receiving modules, the best combining processor comprising:

i) a combined coefficient processor, comprising:

(1) an attenuation coefficient estimation module;

(2) an autocorrelation estimation module;

(3) an inversion module coupled to the autocorrelation estimation module; and

(4) a combining coefficient estimation module coupled with the inversion module;

(5) an interpolation module coupled with the estimation module; and

ii) at least one optimal combiner, each optimal combiner coupled with the interpolation module and comprising:

1) a plurality of two-input multiplication modules, each module having an output, a first input of each of said multiplication modules configured to receive a signal from one of a plurality of fingers of a rake receiver, a second input of each of said multiplication modules coupled to said interpolation module to receive said combining coefficients; and

2) an addition module having as many inputs as the number of said two-input multiplication modules, each of said inputs being coupled to a corresponding one of the multiplication module outputs.

17. The receiving station of claim 16 wherein said autocorrelation estimation module is the R of the received signal_yyAnd an autocorrelation module.

18. The receiving station of claim 16 wherein said autocorrelation estimation module is noise plus interference R_nnAnd an autocorrelation module.