Optimal weights for mmse space-time equalizer of multicode cdma system

The optimal weight for the minimum mean square error space-time equalizer in multiple proximity systems with multiple codes and codes

The large system of the present invention relates to a coded multiple proximity (CDMA) communication system, and more specifically, relates to a linear minimum mean square error (MMSE) space-time equalizer for multiple input multiple output (MIMO) multiple code CDMA systems .

In a wireless communication system, several users share a channel in a common frequency spectrum. In order to avoid conflicts caused by several users transmitting information on the communication channel at the same time, some rules regarding the allocation of available channel capacity to users are required. The rules for users to access the communication channel have been reached through various forms of multi-directional proximity agreements. One form of agreement is called code multiple proximity (CDMA). In addition to providing multi-directional proximity allocation to channels with limited capacity, an agreement can also provide other functions. For example, a protocol can provide isolation between users, limit interference between users, and provide security by increasing the difficulty of interception and decoding of unintended receivers (also known as low probability of interception) .

In the CDMA system, each signal is isolated from the signals of other users by encoding the signal. The information signal is specially encoded into a transmission signal. The intended receiver that knows the user's coding sequence can decode the transmission signal to receive the information. The frequency spectrum of the information signal is expanded by encoding, so that the frequency bandwidth of the encoded transmission signal is much larger than the original frequency bandwidth of the information signal. For this reason, CDMA is a kind of "spread spectrum" coding. The energy of each user's signal is spread across the channel bandwidth, so that each user's signal appears as noise to other users. As long as the decoding process can achieve a sufficient signal-to-noise ratio, the information in the signal can be recovered (the "noise" isolation between the user's signal and other users' signals is expected). Other factors that affect the information recovery of the user's signal are different for each user in this environment, such as fading, shadowing, and multipath. Shading is interference caused by physical objects that interrupt the signal transmission path between the transmitter and the receiver, such as larger buildings. Multipath is signal distortion, which occurs because the signal crosses multiple paths of different lengths and arrives at the receiver at different times. Multipath is also called "time dispersion" of communication channels. Signals received in phase strengthen each other and produce a stronger signal at the receiver, while signals received out of phase produce weaker or fading signals. Multipath fading can also change over time. For example, in a communication device carried by a moving vehicle, the amount of multipath fading can change rapidly.

To provide diversity against harmful path effects and improved performance, multiple transmit and receive antennas can be used. If the transmission path between the transmitting and receiving antennas is linearly independent (that is, the transmission on one path is not formed as a linear combination of the transmissions on the other paths, to some extent it is generally true) , When the number of antennas increases, the probability of receiving the transmitted signal correctly also increases. Generally, as the number of transmitting and receiving antennas increases, diversity increases and performance improves. The use of multiple antennas at the transmitter and receiver in multiple input multiple output (MIMO) systems.

If multiple antennas are available at the transmitter or receiver, techniques such as spatial multiplexing and code reuse can be used to increase peak traffic. Through code reuse, each channel allocated for transmission can be adjusted to up to M independent data streams, where M is the number of transmission antennas. Data streams sharing the same code are distinguished based on their spatial characteristics, which requires a receiver with at least M antennas. In principle, the peak traffic using code reuse is M times the rate that can be achieved with a single antenna.

In a MIMO multiple code CDMA system, if the space-time equalizer uses the minimum mean square error (MMSE) weight vector that minimizes the mean square error of the output chip sequence of the equalizer, the same spreading code in different transmission antennas Repeated use will degrade equalization performance. Unlike multipath interference and background noise components, the CDMA despreader distorts the inter-stream interference components. This will degrade the performance of the prior art MIMO system.

Therefore, an enhanced chip-level linear space-time equalizer for multiple-input multiple-output (MIMO) multiple code CDMA systems is required in this technology, in which spreading codes can be reused in different transmission antennas.

In one aspect, the CDMA receiver includes a space-time equalizer that can be operatively connected to the receiving antenna, wherein the space-time equalizer applies a weighting vector that is included as a function of a spreading factor The coefficient.

In another aspect, the CDMA receiver includes a space-time equalizer with an equalization coefficient and a despreader, wherein the equalization coefficient is at least partly a function of a spreading factor.

In yet another aspect, a method includes receiving a plurality of signals via a plurality of receiving antennas, wherein the received signal from each receiving antenna includes one or a combination of signals transmitted from a transmission device; and by A weighted vector with coefficients is used to process the signal to generate a plurality of bit streams, wherein the coefficients are at least partly a function of the repeated use of the spreading code.

In a further aspect, the CDMA receiver includes an equalization component that is operatively connected to the receiving antenna, wherein the equalization component applies a weighting vector, the weighting vector including coefficients that are a function of a spreading factor; and the solution The spreading component is operatively connected to the equalization component, wherein the de-spreading component divides the equalization metric sequence into a plurality of modulation symbol sequences.

The word "exemplary" is used herein to mean "serving as an example, instance, or illustration." Any embodiment described herein as "exemplary" is not necessarily construed as better or advantageous than other embodiments.

FIG. 1A is a diagram of a communication system 10, which supports many users and can implement at least some aspects and embodiments of the present invention. The system 10 provides communication for many units 2a to 2g each served by the corresponding base station 4. The units are organized in such a way as to achieve coverage of the expected area. For example, the coverage area can be defined as an area where the user of the terminal 6 can achieve a specific service level (GOS). The terminal 6 in the coverage area can be fixed or mobile, and is generally served by a main base station. For each active terminal, transmissions from other base stations and terminals represent potential interference.

As shown in Fig. 1A, various terminal machines 6 are scattered throughout the system. The terminal 6 includes a processing device 8. Examples of the processing device 8 include (but are not limited to) a processor, program logic, or other basic configuration that represents data and instructions. In other embodiments, the processors may include controller circuits, processor circuits, processors, general-purpose single-chip or multi-chip microprocessors, digital signal processors, embedded microprocessors, microcontrollers, and so on.

At any given moment, on the downlink and uplink, each terminal 6 communicates with at least one and possibly multiple base stations 4, depending on, for example, whether "soft handover" is used or whether the terminal is designed And depending on the operation to receive multiple transmissions from multiple base stations simultaneously or sequentially. The downlink means the transmission from the base station to the terminal, and the uplink means the transmission from the terminal to the base station.

In FIG. 1A, the base station 4a transmits data to the terminals 6a and 6j on the downlink, the base station 4b transmits the data to the terminals 6b and 6j, and the base station 4c transmits the data to the terminal 6c and so on. In Figure 1A, the solid line with arrows indicates data transmission from the base station to the terminal. A dotted line with an arrow indicates that the terminal is receiving pilot signals from the base station, but there is no data transmission. For brevity, uplink communication is not shown in Figure 1A.

Can be based on the communication system titled "HIGH EFFICIENCY, HIGH PERFORMANCE COMMUNICATIONS SYSTEM EMPLOYING MULTI-CARRIER MODULATION" (filed on March 22, 2000) disclosed in the US patent application serial number 09/532,492; or the US patent application serial number The system entitled "METHOD AND APPARATUS FOR HIGH RATE PACKET DATA TRANSMISSION" was disclosed in 08/963,386 to design the system 10. These patent applications have been assigned to the assignee of the present invention and are incorporated herein by reference middle. The system 10 can also be designed as a CDMA system that supports one or more CDMA standards (such as IS-95 standard, Wideband CDMA (W-CDMA) standard, other standards, or a combination thereof).

In the system 10, many terminals share a common resource, that is, the total operating bandwidth W. In order to achieve the expected performance level at a specific terminal, it is necessary to reduce the interference from other transmissions to an acceptable level. Similarly, in order to perform reliable transmission at a high data rate under a specific operating bandwidth, it is necessary to operate on a specific carrier to noise plus interference ratio (C/I) level or higher. Conventionally, by dividing the total available resources into small parts each allocated to a specific unit, interference reduction and required C/I are achieved.

For example, the total operating bandwidth W can be divided into N equal operating frequency bands (that is, B=W/N), and each unit can be allocated to one of the N frequency bands. These frequency bands are repeatedly used periodically to achieve higher spectral efficiency. For a 7-unit reusable mode such as the reusable mode supported by FIG. 1A, unit 2a can be allocated a first frequency band, unit 2b can be allocated a second frequency band, and so on.

Communication systems are usually designed to meet many system requirements, which may include, for example, quality of service (QOS), coverage, and performance requirements. The quality of service is usually defined as the ability of each terminal in the coverage area to achieve a specified minimum average bit rate within a specified percentage of time.

By using multiple antennas in both the transmitter and the receiver, the recent development of multiple input multiple output (MIMO) transmission technology heralds large traffic gains in future wireless communication systems. MIMO technology can be incorporated into various modulation and multi-directional proximity schemes, such as MIMO-CDMA, multiple input multiple output-orthogonal frequency division multiplexing (MIMO-OFDM) and so on.

In the 3G CDMA standard, high-speed packet data channels (such as high-speed downlink shared channel (HS-DSCH) and forward link packet data channel (F-PDCH), etc.) usually use multi-channel coding, such as Huaxu ( Walsh code, which has a fixed spreading factor (SF) to transmit and receive a larger amount of information data within a shorter frame interval. Depending on the data rate of the current packet, the base station (BS) can select many codes from the available channelization codes to provide a corresponding number of modulation symbols. When the MIMO-CDMA system supports multiple transmission streams via multiple transmission antennas, the corresponding BS usually reuses the same channelization code for different antennas. Unless otherwise designed in the MIMO-CDMA scenario, repeated use of codes in the transmission antenna will cause serious damage to the mobile station (MS) space-time equalizer.

System Model of MIMO Multiple Code CDMA

1B is a block diagram of an embodiment of a MIMO multiple code CDMA system 100 including a transmitter part 102 and a receiver part 104. In the following discussion, the spreading factor is expressed as<i>SF</i>。

The transmitter part 102 includes an encoder 106, a mapper 108, a demultiplexer 110, a plurality of spreaders 112, and a plurality of transmission antennas 114. The number of transmission antennas 114 is<i>M</i>, And the number of orthogonal spreading codes allocated to each transmission antenna 114 is<i>J</i>(<i>J</i><img file="TWI345904B_D0001.tif" /><i>SF</i>)。

The receiver part 104 includes a plurality of receiving antennas 116, a minimum mean square error (MMSE) space-time equalizer 118, a plurality of despreaders 120, a multiplexer 122, a demapper 124, and a decoder 126. The number of receiving antennas 116 is N, and the number of despreaders 120 allocated to each receiving antenna 116 is<i>J</i>(<i>J</i><img file="TWI345904B_D0002.tif" /><i>SF</i>), which corresponds to the number of spreaders 112 allocated to each transmission antenna 114. Those who are generally familiar with this technology understand that the space-time equalizer 118 discussed in this article can be applied to a general MIMO-CDMA system.

The terms encoder, decoder, rate matcher, interleaver, deinterleaver, mapper, demapper, spreader, despreader, and space-time equalizer are broad terms intended to have their general meanings. In addition, the encoder may be a device or method for encoding a signal (such as a bit stream) or data from one form into another form (such as encoding into a form suitable for transmission, storage, or processing). Generally, the encoder can be constructed in software or hardware, for example, by a program, algorithm, method, or in a circuit. The decoder may be a device that performs the reverse operation of the encoder, which cancels the encoding so that the original information can be retrieved.

The rate matcher may be a device or method that adjusts the rate or bit rate of the data stream to an expected rate. For example, in a transmitter, the rate matcher can adjust the bit rate to match the capabilities of the transmitter. In a receiver, the rate matcher can perform the reverse process.

The interleaver can be a device or method that arranges data in a non-contiguous manner to increase performance. Generally, the deinterleaver can perform the reverse operation of the interleaver and arrange the interleaved data in a contiguous manner to make it easier to process.

The mapper can be a device or method that collects a set of bits and converts it into a single modulation symbol. The demapper may be a device or method that generally implements the reverse operation of the mapper (such as converting a single modulation symbol into a group of bits).

The spreader can be a device or method that increases the bandwidth of the transmitted signal by a factor to exceed the bandwidth of the information signal. The despreader can be a device or method that generally implements the reverse operation of the spreader and reduces the bandwidth of the received signal. For example, the despreader can reduce the bandwidth of the received signal to its information bandwidth.

The space-time equalizer can be a device or method that provides a fixed ratio and combination of space and time to a signal. For example, the space-time equalizer can ratio and combine a received signal in space and time to restore the original signal.

Referring to FIG. 1B, the encoder 106 receives a source bit sequence 128. In the encoder 106, the source bit sequence 128 in each frame is encoded, rate-matched (that is, punctured or repeated), and interleaved, and mapped to a modulation symbol sequence (e.g., four-phase sequence) in the mapper 108 Shift keying (QPSK), 16-point quadrature amplitude modulation (16QAM), etc.). Then, in the demultiplexer 110, the modulation symbol sequence is demultiplexed into<i>J</i>Liuzhi<i>M</i>Groups in which the<i>m</i>Transmission antenna 114 to transmit the<i>m</i>Group. In the spreader 112 by<i>J</i>Spreading codes in each group<i>J</i>Streams, of which<i>j</i>The spreading code is equal to the<i>j</i>Channelized codes (for example, spreading factor<i>SF</i>The product of the orthogonal code, quasi-orthogonal code, or Walsh code) and the pseudo-random stirring code of the BS. Each group usually reuses the same set of<i>J</i>There are two spreading codes, and each transmission antenna 114 usually uses the same transmission power, but the present invention is not limited to these specific situations.

After passing through the multi-dimensional multi-path fading channel, the transmitted signal reaches the<i>N</i>Receiving antennas 116, where the MMSE space-time chip equalizer 118 corresponds to the<i>M</i>Transmission antenna 114, which divides the received signal into<i>M</i>The equalized soft metric sequence of each group. Then, in the despreader 120, it is equal to the<i>J</i>The conjugate of a spreading code<i>J</i>A despreading code divides each group of equalized soft metric sequences into<i>J</i>A sequence of soft demodulation symbols, each of which corresponds to an orthogonal Walsh channel in the group. Will be formed in the multiplexer 122<i>JXM</i>The demodulated symbol sequence is multiplexed into a single stream, and is demapped into a sequence such as a log-likelihood ratio (LLR) sequence in the demapper 124. The sequence is de-interleaved, inverted rate-matched, and decoded in the decoder 126 to restore the original source bit sequence to the decoded bit 130.

2A is a block diagram of an embodiment of a MIMO multiple code CDMA system 200 including a transmitter part 202 and a receiver part 204. In the following discussion, the spreading factor is expressed as<i>SF</i>。

The transmitter part 202 includes a plurality of encoders 206, a plurality of mappers 208, a plurality of demultiplexers 210, a plurality of spreaders 112, and a plurality of transmission antennas 114. The number of transmission antennas 114 is<i>M</i>, And the number of spreading codes allocated to each transmission antenna 114 is<i>J</i>(<i>J</i><img file="TWI345904B_D0003.tif" /><i>SF</i>)。

The receiver part 204 includes a plurality of receiving antennas 116, a minimum mean square error (MMSE) space-time equalizer 118, a plurality of despreaders 120, a plurality of multiplexers 222, a plurality of demappers 224, and a plurality of Decoder 226. The number of receiving antennas 116 is<i>N</i>, And the number of despreaders 120 allocated to each receiving antenna 116 is<i>J</i>(<i>J</i><img file="TWI345904B_D0004.tif" /><i>SF</i>), which corresponds to the number of spreaders 112 allocated to each transmission antenna 114.

Each encoder 206 receives the source bit sequence 128 for the encoder 206. The source bit sequence 128 in each frame is encoded, rate-matched (that is, punctured or repeated), and interleaved in its corresponding encoder 206, and is mapped to a modulation symbol sequence in its corresponding mapper 208 (For example, QPSK, 16QAM, etc.). Then, the modulation symbol sequence is demultiplexed into its corresponding demultiplexer 210<i>J</i>A group of flows, in which through the first<i>m</i>The transmission antenna 114 transmits the<i>m</i>Group. In the spreader 112, by<i>J</i>Spreading codes to expand the<i>J</i>Streams, of which<i>j</i>The spreading code is equal to the<i>j</i>Channelized codes (for example, spreading factor<i>SF</i>Orthogonal code, quasi-orthogonal code, or Walsh code) and BS pseudo-random stirring code. Each group usually reuses the same set of<i>J</i>There are two spreading codes, and each transmission antenna 114 usually uses the same transmission power, but the present invention is not limited to these specific situations.

After passing through the multi-dimensional multi-path fading channel, the transmitted signal reaches the<i>N</i>Receiving antennas 116, where the MMSE space-time chip equalizer 118 corresponds to the<i>M</i>A transmission antenna 114 divides the received signal into M groups of equalized soft metric sequences. Then, in the despreader 120, it is equal to the<i>J</i>The conjugate of a spreading code<i>J</i>A despreading code divides each group of equalized soft metric sequences into<i>J</i>A sequence of soft demodulation symbols, each of which corresponds to an orthogonal Walsh channel in the group. Earned<i>M</i>Piece<i>J</i>Each of the demodulated symbol sequences is multiplexed into a single stream in its corresponding multiplexer 222, and demapped into a sequence such as a log-likelihood ratio (LLR) sequence in its corresponding demapper 224. Each of the M sequences is de-interleaved, inversely rate-matched, and decoded in its corresponding decoder 226 to restore the original source bit sequence to the decoded bit 230.

In an embodiment, after the MMSE space-time equalization, the soft metric sequence of the MIMO CDMA system 100, 200 includes five components: expected signal; one or more punctual inter-stream interference (or crosstalk in signals from different transmission antennas) ), which reuses the same spreading code as the expected signal; one or more punctual inter-stream interference, which does not reuse the same spreading code as the expected signal; one or more multipath interference (meaning, The signal component of the total service unit, which is not on time); and background noise (interference from other units, thermal noise, etc.).

The on-time inter-stream interference: when the spreading code of the expected signal is repeatedly used by the despreading process, it remains intact; or when the spreading code of the expected signal is not repeatedly used by the despreading process In the case, it was discarded. The SF factor roughly suppresses multipath interference and background noise.

FIG. 2B is a block diagram of an embodiment of the space-time equalizer 118. As shown in FIG. The space-time equalizer 118 includes M equalization memory banks 250 (memory bank m, where m=0, 1,..., M-1) corresponding to M transmission antennas 114. Each memory bank 250 includes N filters 252 (filter n, where n=0, 1,..., N-1) corresponding to N receiving antennas 116, and an adder 254. The filters 252 have a filter coefficient V^H_m_,_n_O_P_T, Where m=0,1,2,...,M-1, and n=0,1,2,...,N-1, and each filter 252 generates a filtered output signal. Each memory bank 250 receives a signal from each of the N receiving antennas 116 and processes the signal in the corresponding filter 252. The adder 254 adds the filtered output signals from each filter 252 in each memory bank 250 to generate an equalization metric sequence 256.

Pay attention to the equalization memory bank 0250a. For the jth filter (in memory bank 0, j=0, 1,..., N-1, it has a filter coefficient V^H₀_,_j_O_P_T), the input end of the filter j is connected to the j-th receiving antenna, and the output end of the filter j is connected to the input end of the adder 254a.

For example, when the filter coefficient V^H₀_,₀_O_P_TThe input terminal of the filter 0 252a in the equalization memory bank 0 250a is connected to the receiving antenna 0 116a, and the output terminal of the filter 0 252a is connected to the input terminal of the adder 254a. Similarly, with filter coefficient V^H₀_,_N_－₁_O_P_TThe input terminal of the filter N-1 252b is connected to the receiving antenna N-1 116b, and the output terminal of the filter N-1 252b is connected to the input terminal of the adder 254a.

In the adder 254a, the outputs from the filter n (n=0, 1,..., N-1) in the block 0 250a are added to generate an equalization metric sequence, that is, the sequence 0 256a.

Similarly, the N filtered outputs of the N filters 252 in each block m 250 (here m=0, 1,..., M-1) are added to produce M etc.Chemical metric sequence 256.

The self-pilot signal calculates the channel coefficient h as further described in Equation 8._<i>i</i>And noise covariance R_n. Use the calculated channel coefficient h_<i>i</i>And noise covariance R_nTo calculate the filter coefficient V^H_m_,_n_O_P_T, Where m=0,1,2...,M-1 and n=0,1,2...,N-1.

In another embodiment, the equalizer 118 is constructed as software in the processor 8.

FIG. 3 is a flowchart 300 illustrating the operation of an embodiment of the multiple code CDMA receiving system 104, 204. In one embodiment, the multiple code CDMA receiving systems 104, 204 operate in a continuous loop, which starts at the "start" block and ends at the "end" block. In block 310, the equalizer 118 receives a sequence of pilot symbols. In block 312, the equalizer 118 uses the pilot symbols to calculate the equalizer coefficients.

In block 314, the receiving systems 104 and 204 receive a signal via the antenna 116. In block 316, the received signal is equalized in the equalizer 118 using the equalization coefficient. The equalizer 118 processes the received signals to generate an equalization metric sequence 256.

In block 318, the equalization metric sequence 256 is processed by the despreader 120 to generate a demodulation symbol sequence.

The existence of on-time inter-stream interference makes the traditional wafer-level MMSE equalizer non-optimal because it does not take into account the de-spreading effect. In MIMO CDMA applications, the traditional chip-level MMSE weights are controlled in the non-optimal direction of the noise space, which degrades the decoding performance. In addition, the despreading effect of MMSE weight optimization in single-input single-output (SISO) multiple code CDMA does not change the weight (or control direction), except for different scaling factors. Assuming that the demappers 124 and 224 re-ratio soft demodulate the symbols, the decoding performance in SISO multiple code CDMA is not affected.

Generally, when the number of spreading codes used for each stream increases, the gap between the optimal MMSE weight (which takes the despreading effect into account) and the non-optimal MMSE weight decreases due to the on-time inter-stream interference The de-spreading gain will be roughly discounted by the number factor of the used spreading code, as discussed below.

Linear MMSE equalizer weight for MIMO multiple code CDMA

<b>Optimization of traditional chip-level MMSE weights in MIMO multiple code CDMA</b>

The traditional MMSE space-time chip equalizer corresponds to M transmission antennas 114, and divides the received signal into equalized soft metric sequences.<i>M</i>Groups. The multiplexers 122, 222, demappers 124, 224, and decoders 126, 226 then process the sequences to generate decoded bits 130, 230, respectively.

In the following discussion on the optimization of traditional chip-level MMSE weights, the span of multipath delay spread is<i>L</i>Chip length, the span of the equalizer is<i>E</i>Each chip is long and the receiver uses each chip<i>P</i>Samples (that is, the oversampling factor is<i>P</i>). in addition,<i>h</i>_<i>n</i>_,_<i>m</i>_,_<i>p</i>(<i>l</i>)(<i>l</i>＝0、1、...、<i>L</i>－1；<i>n</i>＝0、1、...、<i>N</i>－1；<i>m</i>＝0、1、...、<i>M</i>－1；<i>p</i>＝0、1、...、<i>P</i>-1) is the first<i>m</i>Transmission antenna 114 and the first<i>n</i>The channel coefficient between the two receiving antennas 116, which corresponds to the first<i>l</i>Chip delay and chips<i>p</i>Samples. In chip time<i>k</i>No.<i>m</i>The chip signal of each transmission antenna 114 is determined by σ_<i>x</i><i>x</i>_<i>m</i>(<i>k</i>) Means where<i>E</i>[|<i>x</i>_<i>m</i>(<i>k</i>)|²] = 1 and<img file="TWI345904B_D0005.tif" />Is the average chip energy of each transmission antenna 114.

definition

<b>x</b>_<i>m</i>(k)σ_<i>x</i>[<i>x</i>_<i>m</i>(<i>k</i>)<i>x</i>_<i>m</i>(<i>k</i>＋1)…<i>x</i>_<i>m</i>(<i>k</i>＋<i>E</i>＋<i>L</i>－2)]^T(1) is the first<i>m</i>Of the transmission antennas 114 (<i>E</i>＋<i>L</i>－<i>1</i>)-Dimensional chip vector, its self-exponent<i>k</i>Span to<i>k</i>＋<i>E</i>＋<i>L</i>-2. Similarly, let<i>y</i>_<i>n</i>_,_<i>p</i>(<i>k</i>)and<i>n</i>_<i>n</i>_,_<i>p</i>(<i>k</i>) Is the first<i>n</i>The 116th receiving antenna<i>k</i>Chip number<i>p</i>The received samples and their background noise components of each sample.

In addition, define<b>y</b>_<i>n</i>(k)[<i>y</i>_<i>n</i>_,₀(<i>k</i>)…<i>y</i>_<i>n</i>_,_<i>P</i>_－₁(<i>k</i>)…<i>y</i>_<i>n</i>_,₀(<i>k</i>＋<i>E</i>－1)…<i>y</i>_<i>n</i>_,_<i>P</i>_－₁(<i>k</i>＋<i>E</i>－1)]^T(2) and<b>n</b>_<i>n</i>(k)[<i>n</i>_<i>n</i>_,₀(<i>k</i>)…<i>n</i>_<i>n</i>_,_<i>P</i>_－₁(<i>k</i>)…<i>n</i>_<i>n</i>_,₀(<i>k</i>＋<i>E</i>－1)…<i>n</i>_<i>n</i>_,_<i>P</i>_－₁(<i>k</i>＋<i>E</i>－1)]^T(3) is the first<i>n</i>116 receiving antennas<i>PE</i>Dimension of the received sample vector and the corresponding background noise vector, then<maths><img file="TWI345904B_D0006.tif" /></maths>In Equation 4,<b>H</b>_<i>n</i>_,_<i>m</i>Said in the first<i>m</i>Transmission antenna 114 and the first<i>n</i>Between the receiving antennas 116<i>PE</i>×(<i>E</i>＋<i>L</i>-1) Multipath channel matrix, and given as<maths><img file="TWI345904B_D0007.tif" /></maths>

In addition, define<b>y</b>(<i>k</i>)≡[<b>y</b>₀(<i>k</i>)^<i>T</i><i>y</i>₁(<i>k</i>)^<i>T</i>…<b>y</b>_<i>N</i>_－₁(<i>k</i>)^<i>T</i>]^<i>T</i>for<i>NPE</i>The vector of samples received by the population of dimensions, the definition<b>n</b>(<i>k</i>)≡[<b>n</b>₀(<i>k</i>)^<i>T</i><b>n</b>₁(<i>k</i>)^<i>T</i>…<b>n</b>_<i>N</i>_－₁(<i>k</i>)^<i>T</i>]^<i>T</i>for<i>NPE</i>Dimensional overall background noise vector, definition<b>R</b>_<i>n</i>≡<i>E</i>[<b>n</b>(<i>k</i>)<b>n</b>(<i>k</i>)^<i>H</i>]for<i>NPE</i>×<i>NPE</i>Noise covariance matrix, and its definition<maths><img file="TWI345904B_D0008.tif" /></maths>for<i>NPE</i>×<i>M</i>(<i>E</i>＋<i>L</i>－<i>1</i>) Of the overall multipath channel matrix. Then for the target delay of D chips<i>m</i>Transmission antenna chip flow, minimizing<img file="TWI345904B_D0009.tif" />The best chip-level linear MMSE weighting vector<img file="TWI345904B_D0010.tif" />Becomes<maths><img file="TWI345904B_D0011.tif" /></maths>As mentioned above, the channel matrix coefficients are calculated from a pilot signal.

By applying the matrix inverse lemma, Equation 7 can be rewritten as:<maths><img file="TWI345904B_D0012.tif" /></maths>The SNR of the equalizer output chip is<maths><img file="TWI345904B_D0013.tif" /></maths>

In addition, the equalizer output soft chip metric becomes<maths><img file="TWI345904B_D0014.tif" /></maths>

When by<i>C</i>_<i>j</i>(<i>k</i>) (Where |<i>C</i>_<i>j</i>(<i>k</i>)|²=1)) represents the j-th spreading code (or the product of the j-th Huaxu code and the common stirring code), the despreader 120 (where the spreading factor is<i>SF</i>)'S output soft symbol becomes<maths><img file="TWI345904B_D0015.tif" /></maths>Where A^＊Represents the conjugate complex number of A. The demapper 124 then scales and converts the output soft symbol into a symbol exponent<i>n</i>, Coding index<i>j</i>And transmission antenna index<i>m</i>The bit value.

Under the observation of decoders 126 and 226, the MMSE weighting vector of equation (7) is not the best, because it is performed without considering the significant nature of the on-time inter-stream interference in the despreader 120Upoptimization.

<b>Enhanced chip-level MMSE weighting vector for MIMO multiple code CDMA</b>

The following discusses a MIMO multiple code CDMA system that equalizes the received signal before de-spreading. The space-time equalizer applies a weight vector with coefficients that are a function of the spreading factor.

Consider transferring chip value<i>x</i>_<i>m</i>(<i>k</i>)Depend on<i>J</i>Consisting of two orthogonal channel components, which means,<maths><img file="TWI345904B_D0016.tif" /></maths>in<img file="TWI345904B_D0017.tif" />(<i>k</i>) Corresponds to the first<i>m</i>No. 114 of the transmission antennas<i>j</i>Chip components of a spread spectrum code (where<img file="TWI345904B_D0018.tif" />), the output symbol metric of the despreader in equation (11)<img file="TWI345904B_D0019.tif" />(<i>n</i>) The SNR can be<maths><img file="TWI345904B_D0020.tif" /></maths>

Note that the orthogonal solution spread spectrum should be introduced relative to the chip SNR<i>SF</i>Gain factor and<i>J</i>Loss factor.

However, in the MIMO multiple code CDMA system 100 where the code is reused, the actual SNR of the output symbol of the despreader becomes lower than that of equation (13). This is because the on-time inter-stream interference in the despreading process is expressed as It is different from multipath interference or background noise. In addition, under the observation of decoders 126 and 226, the MMSE weighting vector of equation (7) is not the best. This is because the significant nature of the on-time inter-stream interference in the despreader 120 is not considered. Optimized under. Therefore, as discussed further below, the SNR of equation (13) is difficult to achieve in practice.

Refer to equations (4)-(6) and equations (10)-(12), by the weighted vector<img file="TWI345904B_D0021.tif" />(For the first<i>m</i>Transmission antenna streams) are equalized and divided by the first<i>j</i>Unspread code<img file="TWI345904B_D0022.tif" />(<i>k</i>) The soft demodulation symbol for despreading can be written as:<maths><img file="TWI345904B_D0023.tif" /></maths>The first and second terms represent signal and interference components respectively. More specifically, in equation (14)<img file="TWI345904B_D0024.tif" />(<i>n</i>)、<img file="TWI345904B_D0025.tif" />(<i>n</i>)and<img file="TWI345904B_D0026.tif" />(<i>n</i>) Respectively represent the expected symbol components after de-spreading, and the<i>j</i>The on-time inter-stream interference component of each spreading code, and the multi-path interference component. Do not use the first<i>j</i>The on-time inter-stream interference component of each spreading code disappears during the de-spreading process. Instead, use the first<i>j</i>The on-time inter-stream interference component of each spreading code has due to the de-spreading<i>SF</i>Spread-spectrum gain, as expected for signal components. The despreading operation will not change the multipath interference component and the background noise component (in equation (14)<img file="TWI345904B_D0027.tif" />(<i>n</i>) Represents the common variance of).

Under the observation of decoders 126 and 226, the best MMSE weighting vector<img file="TWI345904B_D0028.tif" />Should be minimized<img file="TWI345904B_D0029.tif" />(That is, it should be minimized relative to the target symbol), and therefore it becomes<maths><img file="TWI345904B_D0030.tif" /></maths>

By applying the matrix inversion lemma, the MMSE weighting vector that depends on the spreading factor can be rewritten as<maths><img file="TWI345904B_D0031.tif" /></maths>

NS<i>m</i>No. 114 of the transmission antennas<i>j</i>The output symbol SNR of the coded despreader becomes<maths><img file="TWI345904B_D0032.tif" /></maths>

Equations (13) and (17) are shown by the SF/J factor, that the variance of the on-time interference component of equation (17) is larger than that of equation (13). Therefore, the achievable SNR in equation (17) is lower than the expected SNR in equation (13), unless independent SF codes are allocated to data transmission, and the transmission antennas 114 fully reuse them (that is, J =SF). In practice, the number of codes that are allocated and reused is usually due to data rate-dependent spreading code allocation (e.g., a smaller number of codes at lower data rates, and a larger number of codes at higher data rates. Encoding), control channel, audio channel existence, etc., and less than SF. Equations (8) and (16) show that the difference in power factor SF/J due to the interference component between on-time streams, in the soft symbol level to be used by the demapper 124, 224 and the decoder 126, 226, The traditional wafer-level optimized MMSE weighting vector is not optimal. The traditional wafer-level MMSE weighting vector underestimates the on-time inter-stream interference component because it does not take the despreading effect into account, and therefore controls in the non-optimal direction. Therefore, in an embodiment, using the weighting vector of equation (8), the actual symbol SNR becomes even lower than equation (17), which is far from the upper limit of equation (13). When we reduce the number of spreading codes for multiple antenna reuse, the performance gap between the MIMO-CDMA optimized MMSE weight vector in equation (16) and the traditional weight vector in equation (8) becomes more Big.

In obtaining the enhanced wafer-level equalizer 118, the system model of FIG. 1 and FIG. 2 is used, in which multiple antennas 114 repeatedly use the same spreading code, and all antennas 114 and codes use approximately the same amount of transmission power.

Referring to equation (8) and equation (16), the component of changing the control direction of the weight vector is on-time inter-stream interference. Therefore, in the SISO multi-code CDMA system without inter-stream interference, the traditional wafer-level MMSE weight vector and the enhanced MMSE weight vector are controlled in the same direction (that is, they are aligned in the signal space). However, the fixed ratios of these weighting vectors may be different. The scaling factor is a function of SNR, and if the demappers 124 and 224 can accurately re-ratio input soft symbols to obtain a fair evaluation, the traditional chip-level MMSE weighting vector and the enhanced MMSE weighting vector have approximately the same decoding performance .

<b>Generalization of the Enhanced Equalizer to Arbitrary Power and Code Allocation in MIMO Multiple Code CDMA</b>

In generating the enhanced wafer-level MMSE weight vectors for the MIMO multiple code CDMA receiver 104, 204, in equations (12)-(17), it is assumed that all<i>M</i>Transmission antenna 114 reuses the same<i>J</i>Spread codes, and<img file="TWI345904B_D0033.tif" />The total transmission chip energy is divided equally and distributed to those separated by the transmission antenna 114 and the spreading code<i>JM</i>Streams. Equivalently, assuming<i>JM</i>Everyone in the stream has<img file="TWI345904B_D0034.tif" />/<i>J</i>Chip energy. In this section, arbitrary coding and power allocation scenarios take into account the existence of actual coded multiplexing pilots, control and audio channels, and unequal power allocation.

To this end,<img file="TWI345904B_D0035.tif" />Defined as assigned to the first<i>m</i>Transmission antenna 114 (<i>m</i>＝0、1、...、<i>M</i>-1) and the first of the spreading factor SF<i>j</i>Codes (<i>j</i>＝0、1、...、<i>SF</i>-1) chip energy, which includes the distribution to the first<i>j</i>All possible sub-code trees of each code (if it is being used in the first<i>m</i>The sum of the chip energy in each antenna 114). Jodie<i>m</i>Transmission antenna 114 does not use the<i>j</i>Codes, then<img file="TWI345904B_D0036.tif" />Equal to 0. As mentioned above, the result of the enhanced wafer-level MMSE weighting vector for MIMO multiple code CDMA is valid for special cases, where<maths><img file="TWI345904B_D0037.tif" /></maths>

And the transmission power is allocated to data transmission. In one embodiment, no control channel or pilot channel and the MIMO data stream share the transmission power at the same time. Depend on<img file="TWI345904B_D0038.tif" />Represents the first<i>m</i>The total transmission chip energy of each transmission antenna 114, which includes all channels, such as data, pilots, control channels, etc., and defines<img file="TWI345904B_D0039.tif" />,NS<i>j</i>Code and<i>m</i>The best MMSE weighting vector for each transport stream<img file="TWI345904B_D0040.tif" />Can be obtained in the manner used in equation (15), which becomes<maths><img file="TWI345904B_D0041.tif" /></maths>

In addition, applying the matrix inversion lemma, the equivalent weight vector becomes<maths><img file="TWI345904B_D0042.tif" /></maths>Among them<i>m</i>No. 114 of the transmission antennas<i>j</i>The output symbol SNR of the coded despreader becomes<maths><img file="TWI345904B_D0043.tif" /></maths>

As illustrated in Figure 4, for various chip SNR values, compare the block error rate (BLER) performance simulation values between the traditional equalizer (traditional EQ) and the enhanced equalizer (enhanced EQ) (<i>Ec</i>/<i>No</i>). The simulations are performed for the case of 4 transmission (or M=4) antennas 114 and 4 reception (or N=4) antennas 116. Configure coding, rate matching, interleaving, cluster mapping and receiver counterparts according to 3GPP HSDPA HS-DSCH specifications. In HS-DSCH, the chip rate is 3.84 Mcps, the frame length (or block length) is 2 ms,<i>SF</i>It is 16, and for each antenna 114, the number of modulation symbols per spreading code per frame is 480. In the simulation, the modulation cluster is fixed to QPSK. Therefore, use<i>J</i>The total number of coded bits transmitted by a spreading code in the frame via 4 antennas 114 is 3840<i>J</i>. The 4 transmission antennas 114 are set to use the same set of<i>J</i>A spread spectrum code, and the same amount of transmission chip energy<i>Ec</i>/<i>M</i>Divided equally and allocated to each antenna 114<i>J</i>Encoding channels.

For the sake of brevity, load channels (for example, common pilot channels, control channels, audio channels, etc.) are not modeled in this simulation. Therefore, the overall BS transmission chip energy<i>Ior</i>Equal to HS-DSCH chip energy<i>Ec</i>. The turbo code in the 3GPP HSDPA specification is used for encoding, and the encoding rate is maintained at about 1/3 through the simulation. Set the carrier frequency to 2GHz. Power spectral density<i>N</i>_<i>0</i>The space-independent white Gaussian stochastic process is used to imitate the background noise component of the four receiving antennas 116. A fully synchronized and fully evaluated chip-isolated equalizer 118 with channel coefficients and noise covariance is used in the simulation (meaning that the oversampling factor<i>P</i>Set to 1). When multipath delays span<i>L</i>Time span of the space-time equalizer<i>E</i>And target delay<i>D</i>set as<i>3L</i>Chips and<i>2L</i>-1 chip.

Figure 4 shows the 3 km/h vehicle A model described by the standard SCM link level (6 paths, 2 degrees of BS angular spread, 35 degrees of MS angular spread, 10 wavelengths of BS antenna spacing, 0.5 wavelength of MS antennas BLER performance of single code repeated use in interval). Set the corresponding information data rate to 640 kbps, and the number of coded bits is 3840. In the case of single encoding, as illustrated in Figure 4, at 10^－²Under the block error rate (BLER), there is a gain of about 3 dB.

It is observed that when the number of codes increases, the gain of the self-enhanced equalizer decreases. When the number of codes is close to<i>SF</i>At this time, the power balance between the on-time inter-stream interference component and the multipath interference and background noise component of (16) becomes closer to the power balance of the traditional equalizer of (8). Therefore, the improvement for 15 codes is smaller than the improvement for a single code.

The conventional chip-level MMSE weighting vector (8) provides a smaller signal-to-noise ratio than the enhanced MMSE weighting vector (16) of MIMO multiple code CDMA where the same code is repeatedly used in different transmission antennas 114. As we can see in the comparison between (8) and (16), the two weighting vectors are still controlled in different directions even after compensating for the fixed ratio factor. In one embodiment, on-time inter-stream interference is a key component. Therefore, the enhanced MMSE weight vector considering the despreading effect is better.

Those familiar with this technology will understand that any of a variety of different technologies and methods can be used to represent such information and signals. For example, voltage, current, electromagnetic waves, magnetic fields or particles, optical fields or particles, or any combination thereof can be used to represent data, instructions, commands, information, signals, bits, symbols, and chips that can be referenced in all the descriptions.

Those familiar with this technology will further understand that various descriptive logic blocks, modules, circuits, and algorithm steps described in the embodiments disclosed herein can be constructed as electronic hardware, computer software, or a combination of both. In order to clearly illustrate the interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been generally described above based on their respective functionality. The construction of this functionality as hardware or software will depend on the specific application and the design constraints of the entire system. Skilled technicians can construct the functionality in different ways for each specific application, but these construction decisions should not be construed as deviating from the scope of the present invention.

It can be used by general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices designed to perform the functions described in this article, Discrete gate or transistor logic, discrete hardware components, or any combination thereof, are used to construct or execute various illustrative logic blocks, modules, and circuits described in conjunction with the embodiments disclosed herein. The general-purpose processor may be a microprocessor, a conventional processor, a controller, a microcontroller state machine, etc. A processor can also be constructed as a combination of computing devices, for example, a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in combination with a DSP core, or any other such configuration .

The steps of the method or algorithm described in combination with the embodiments disclosed herein can be directly included in the hardware, in a software module executed by a processor, or in a combination of the two. The software module can reside in random access memory (RAM), flash memory, read-only memory (ROM), erasable programmable read-only memory (EPROM), electronic erasable programmable read-only memory ( EEPROM), register, hard disk, removable disc, compact disc-read only memory (CD-ROM), or any other form of storage medium known in the art. A storage medium is coupled to the processor so that the processor can read information from the storage medium and write information into the storage medium. In the alternative, the storage medium may be integrated in the processor. The processor and storage medium can reside in an ASIC. The ASIC can reside in a user terminal. In the alternative, the processor and storage medium may reside as discrete components in a user terminal.

These modules may include (but are not limited to) any of the following items: software or hardware components, such as software object-oriented software components, class components and task components, processes, methods, functions, attributes, Steps, procedures, program chips, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays or variables.

The previous description of the disclosed embodiments is provided to enable anyone familiar with the art to make or use the present invention. For those familiar with the art, various modifications to these embodiments are extremely obvious, and the general principles defined herein can be applied to other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown in this text, but should conform to the broadest scope consistent with the principles and novel features disclosed in this text.

<p>2A-2G. . . unit</p><p>4A-4G. . . Base station</p><p>6A-6J. . . Terminal</p><p>8. . . processor</p><p>10. . . Communication Systems</p><p>100. . . MIMO multiple code CDMA system</p><p>102. . . Transmitter part</p><p>104. . . Receiver part</p><p>106. . . Encoder</p><p>108. . . Mapper</p><p>110. . . Demultiplexer</p><p>112. . . Spreader</p><p>114. . . Transmission antenna</p><p>116. . . Receive antenna</p><p>116a. . . Receive antenna</p><p>116b. . . Receive antenna</p><p>118. . . Minimum Mean Square Error Space Time Equalizer</p><p>120. . . Despreader</p><p>122. . . Multiplexer</p><p>124. . . Demapper</p><p>126. . . decoder</p><p>128. . . Source bit sequence</p><p>130. . . Decoded bit</p><p>200. . . MIMO multiple code CDMA system</p><p>202. . . Transmitter part</p><p>204. . . Receiver part</p><p>206. . . Encoder</p><p>208. . . Mapper</p><p>210. . . Demultiplexer</p><p>222. . . Multiplexer</p><p>224. . . Demapper</p><p>226. . . decoder</p><p>230. . . Decoded bit</p><p>250. . . Equalization memory</p><p>250a. . . Equalization memory</p><p>252. . . filter</p><p>252a. . . filter</p><p>252b. . . filter</p><p>254. . . Adder</p><p>254a. . . Adder</p><p>256. . . Equalization metric sequence</p><p>256a. . . Equalization metric sequence</p>

Figure 1A is a diagram of a communication system that supports many users and can implement at least some aspects and embodiments of the present invention.

Figure 1B is a block diagram of an embodiment of a MIMO multiple code CDMA system.

Figure 2A is a block diagram of another embodiment of a MIMO multiple code CDMA system.

Figure 2B is a block diagram of an embodiment of the MMSE space-time equalizer.

Figure 3 is a flow chart illustrating the operation of an embodiment of a multiple code CDMA system.

4 is a graph of block error rates of various chips-signal-to-noise ratio (SNR) values of an embodiment of the present invention using 1 code reuse and 3 km/h vehicle A multi-path channel model.