CN100370709C - Multiple code receiver of multiple input and multiple output system - Google Patents

Abstract

The present invention discloses a multiple-code receiver of a multiple-input multiple-output (MIMO) system, which comprises a de-spread unit group and at least two-stage joint detection units, wherein each component of the vector of a received signal is decomposed and spread by the de-spread unit group; the obtained de-spread signal vector is input to a first-stage joint detection unit of the at least two-stage joint detection units; all symbols of a code channel corresponding to each de-spread signal vector are detected by the first-stage joint detection unit in groups; then, the correction of code-domain interference cancellation and the de-spread signal vector are carried out for the signal estimation of all the obtained symbols by detection of each code channel and the estimation of a resultant signal; the corrected de-spread signal vector of each symbol of each code channel is output to a next-stage joint detection unit; processing is carried out step by step until a final-stage joint detection unit receives the corrected de-spread signal vector and obtains the estimation of all the symbols of each code channel. The receiver improves the demodulation performance of multiple-code transmission downlink receiver; therefore, the processing time delay of the receiver is reduced, and the system capacity and peak rate are improved.

Description

Multi-code receiver of multi-input multi-output system

Technical Field

The present invention relates to the field of wireless communication technologies, and in particular, to a multi-code receiver of a Multiple Input Multiple Output (MIMO) system.

Background

Mimo (Multiple Input Multiple output) technology is a major breakthrough in the field of wireless mobile communication, and can improve the capacity and spectrum utilization of a communication system by multiples without increasing bandwidth. The MIMO technology simultaneously transmits and receives signals using multiple antennas (antenna arrays) at a transmitting end and a receiving end. Because the signals transmitted by the transmitting antennas simultaneously occupy the same frequency band, the communication bandwidth is not increased. There is one spatial channel between each transmit antenna and each receive antenna. The MIMO system can create a plurality of parallel independent spatial channels between a transmitting end and a receiving end through a plurality of transmitting antennas and a plurality of receiving antennas if channel impulse responses of each spatial channel are independent. By transmitting information independently through these parallel spatial channels, the transmission data rate of the MIMO system must be multiplied. The above conclusions were fully demonstrated by g.j.fosschini and m.j.gans in 1998 and quantitatively indicated: assuming that the MIMO system has M transmitting antennas and N receiving antennas, an nxm order channel matrix can be established under a flat slow fading channel. The elements of the matrix are complex Gaussian random variables which are independently and identically distributed. The channel capacity that can be obtained by the MIMO system will be min (M, N) (min (M, N) represents the minimum of M and N) times that of the siso (Single Input Single output) system, and the total transmission power remains unchanged.

In 1996 G.J.Foshini proposed a D-BLAST (Diagnonal BLAST, BLAST: Bell Laboratories layerspaced-time Architecture) method for MIMO systems, which can achieve demodulation of MIMO systems and can achieve a capacity close to 90% of the theoretical capacity. However, the D-BLAST method is complicated and not easy to implement in real time.

In 1999, g.d. golden, j.foshini, r.a. valenzuela and p.w.wolfiansky proposed a simplified BLAST method-V-BLAST method (Vertical BLAST). The method has been implemented in real time in the laboratory. The MIMO system with 8 antennas and 12 antennas at the transmitting end and the receiving end respectively can obtain the spectral efficiency of 20-40bps/Hz in the indoor environment when the average SNR is changed in the range of 24 dB-34 dB. Although this spectral efficiency is obtained in an indoor environment, spectral efficiency of this order is unprecedented. The conventional techniques of the prior art cannot achieve spectral efficiency of this order regardless of the propagation environment and SNR. However, the V-BLAST method also has its disadvantages: the processing delay is large.

To overcome the shortcomings of the V-BLAST method, i proposed a MIMO receiver in another application. Referring to fig. 1, the receiver includes a grouped vertical layered space-time detection unit 101 and a parallel vertical layered space-time detection unit 102. The grouped vertical layered space-time detection unit 101 receives signal vectors, divides symbols to be detected in all the received signal vectors into Z groups, completes grouped vertical layered space-time detection, outputs corrected received signal vectors of all the symbols to the parallel vertical layered space-time detection unit 102, and the parallel vertical layered space-time detection unit 102 receives the received signal vectors of all the symbols corrected by the grouped vertical layered space-time detection unit 101, performs parallel vertical layered space-time detection on the received signal vectors, and outputs estimation of each symbol.

The working principle of the receiver is as follows: the grouped vertical layered space-time detection unit 101 divides symbols to be detected in all received signal vectors into Z groups, performs detection of all symbols in parallel in a first group to obtain estimation of all symbols and signal estimation of all symbols, and eliminates interference of all symbols in the first group from the received signal vectors to obtain corrected received signal vectors, which are used as input signals of the first group to a second group for detection of the second group of symbols, so that detection of the second group of symbols is not interfered by the first group of symbols; and taking the received signal vector output by the first group as the received signal vector adopted in the detection of each symbol in the second group, carrying out the detection of all symbols in parallel to obtain the estimation of each symbol in the second group and the signal estimation of all symbols, eliminating the interference of all symbols in the second group from the received signal vector output by the first group to obtain a further corrected received signal vector, and taking the received signal vector as the input signal vector of the second group to the third group for the detection of the third group of symbols. And detecting other subsequent groups of symbols by analogy. After the detection of the Z groups of symbols is finished, the signal estimation of each group of symbols in the Z groups and the received signal vector corrected by the Z group are respectively obtained, and then the corrected received signal vectors of all the symbols are obtained through the signal estimation of each group of symbols and the received signal vector corrected by the Z group. Finally, the corrected received signal vectors of all symbols are output and sent in parallel to the parallel vertical layered space-time detection unit 102.

The MIMO receiver integrates the advantages of parallel implementation and grouping implementation, reduces the processing delay of the V-BLAST method and improves the detection performance. However, the MIMO receiver is a MIMO receiver under single-code transmission, and if the transmitting end employs multi-code transmission, the receiving end can demodulate to obtain a symbol sequence transmitted per code channel by using a plurality of such single-code MIMO receivers. However, the MIMO receiver can only effectively eliminate the interference between symbols in the code channels, but cannot eliminate the mutual interference between the multiple codes.

Considering that the V-BLAST method is a method for detecting serial interference Cancellation in the space Domain under a single Code, and cannot eliminate mutual interference between Code channels under multi-Code transmission, it has also been proposed to apply Code Domain serial interference Cancellation to the V-BLAST method, and to propose a CD-SIC V-BLAST (Code Domain serial interference Cancellation V-BLAST) method and a corresponding receiver. The CD-SIC V-BLAST receiver structure is shown in FIG. 2.

As can be seen from fig. 2: code channel k₀N correlator sets for device channel code k₀De-spread the received signals of N receiving antennas to obtain code channel k₀And (4) despreading signals of the lower N antennas. V-BLAST detector pair code channel k₀Next N daysDetecting the de-spread signal of the line and serial space domain interference cancellation to obtain a code channel k₀Estimation of the transmitted symbol vector for the lower N antennas

Then, the detected code channel k is processed in the CD-SIC device₀Is cancelled from the received signal to obtain a modified received signal vector. And then demodulating the symbol vector transmitted by the N antennas under the next code channel. By the CD-SICV-BLAST method, the performance of the MIMO system under multi-code transmission can be greatly improved.

The code field V-BLAST method adopts the thought of serial cancellation to eliminate the mutual interference between different code channels under the condition of multi-code transmission, and can improve the demodulation performance of a receiver under the condition of multi-code transmission. However, serial interference between code channels eliminates serial detection of each symbol in the code, which causes the processing delay of the method to be large, and the system capacity and peak rate are affected. Therefore, this method also cannot guarantee the performance of the receiver under multi-code transmission.

Disclosure of Invention

It is therefore an object of the present invention to provide a receiver, which can effectively eliminate the mutual interference between multiple codes and reduce the processing delay.

In order to achieve the above object, the present invention provides a multicode receiver of a mimo system, the receiver comprising:

a set of despreading units 301 and at least two stages of joint detection units 305, wherein,

each despreading unit in the despreading unit group 301 despreads each component in the received signal vector, and inputs one or more resulting despread signal vectors to a first-stage joint detection unit 302 in the at least two-stage joint detection unit 305; the first-stage joint detection unit 302 performs packet detection on all symbols in code channels corresponding to each received despread signal vector to obtain signal estimation and total signal estimation of all symbols in each code channel, then performs code domain interference cancellation and despread signal vector correction on the obtained signal estimation and total signal estimation of all symbols in each code channel, outputs the corrected despread signal vector of each symbol in each code channel to the next-stage joint detection unit for processing until the last-stage joint detection unit 304, and the last-stage joint detection unit 304 performs detection on the received corrected despread signal vector to obtain estimation of all symbols in each code channel.

The despreading unit group comprises more than one despreading unit, wherein the c-th despreading unit despreads each component in a received signal vector by using an OVSF code c and a scrambling code to obtain a c-th despread signal vector.

The first-stage joint detection unit includes: c first-stage detection units 401 and a code-domain interference cancellation and despreading signal vector correction unit 402, wherein,

the c first-stage detection unit 401 obtains signal estimates of all symbols of the code c and total signal estimates from the c despreading signal vector, and each first-stage detection unit 401 inputs the obtained signal estimates of all symbols and total signal estimates to the code domain interference cancellation and despreading signal vector correction unit 402; a code domain interference cancellation and despreading signal vector correction unit 402 completes the correction of chip-level interference cancellation and despreading signal vectors to obtain corrected despreading signal vectors of all symbols corresponding to codes 1 to C; and then outputting the despread signal vector corrected by each symbol to a next-stage joint detection unit.

The c-th first-stage detection unit 401 includes:

a symbol grouping and rearranging unit 501 for dividing the M symbols into Z symbols according to a certain criterion by the input despread signal vector and the estimate of the channel matrix_cGrouping and rearranging the M symbols according to the grouping order and the symbol order in the groupArranging and rearranging column vectors forming the channel matrix to form a new channel matrix, inputting the despread signal vector and the estimation of the new channel matrix to the first new intra-group symbol detection and interference cancellation unit 502, and simultaneously outputting the estimation of the new channel matrix to the remaining new intra-group symbol detection and interference cancellation units 502 and outputting the despread signal vector to the new preprocessing unit 503;

Z_ca new intra-group symbol detection and interference cancellation unit 502, wherein the first new intra-group symbol detection and interference cancellation unit performs symbol detection and parallel interference cancellation of one group from the received despread signal vectors to obtain signal estimates of all symbols in the group and corrected despread signal vectors, and sends the signal estimates of all symbols to the code domain interference cancellation and despread signal vector correction unit 402 and outputs the obtained corrected despread signal vectors to the next new intra-group symbol detection and interference cancellation unit; second to Z_cThe symbol detection and interference cancellation unit in the new group completes symbol detection and parallel interference cancellation of one group by the received corrected despread signal vector to obtain signal estimation of all symbols in the group and the corrected despread signal vector, and sends the signal estimation output of all symbols to the code domain interference cancellation and despread signal vector correction unit 402, and outputs the obtained corrected despread signal vector to the next symbol detection and interference cancellation unit in the new group;

a new preprocessing unit 503 for receiving the data from Z_cThe modified despread signal vectors output by the symbol detection and interference cancellation unit and the despread signal vectors output by the symbol grouping and reordering unit estimate a total signal estimate and send it to the code domain interference cancellation and despread signal vector modification unit 402.

Z_cThe ith intra-group symbol detection and interference cancellation unit in the intra-group symbol detection and interference cancellation unit 502 includes:

a weight vector calculation unit 601 for calculating more than one weight vector according to the estimation of the input channel matrix, and sequentially inputting each weight vector to a decision statistic calculation unit 602;

each decision statistic calculation unit 602, configured to calculate a decision statistic according to the modified received signal vector ri and the weight vector output by the (i-1) th new group symbol detection and interference cancellation unit, and send the calculated decision statistic to a transmitted symbol estimation unit 603 connected thereto;

each transmitted symbol estimating unit 603 is configured to estimate an estimated value of a corresponding symbol in the group according to the received decision statistic, and send the estimated value to a signal estimating and interference canceling unit 604;

the signal estimation and interference cancellation unit 604 completes the signal estimation and interference cancellation of all symbols in the group by the estimated values of all symbols in the received group and the corrected received signal vector, outputs the signal estimation of all symbols to the code domain interference cancellation and despreading signal vector correction unit 402, and outputs the corrected despreading signal vector to the (i +1) th new group symbol detection and interference cancellation unit.

The code domain interference cancellation and despreading signal vector correction unit 402 includes: c spreading sections 701, code domain interference cancellation sections 702, C despreading sections 703 and despread signal vector correction sections 704,

each spreading unit estimates the received total signal of the corresponding code channel to perform spreading, then outputs the spreading signal to the code domain interference cancellation unit 702, the code domain interference cancellation unit 702 performs interference cancellation on the received spreading signal, then outputs the residual signals obtained through the interference cancellation to all the despreading units 703 in parallel, the despreading unit 703 despreads each component of the received signal, outputs the obtained despread signal vector to the despread signal vector correction unit 704, and the despread signal vector correction unit 704 finishes the correction of the despread signal vector and outputs the corrected despread signal vectors of all the symbols of all the code channels.

The ith (1 < i < S) stage joint detection unit 303 in the at least two stages of joint detection units 305 includes: c mid-stage detection units 801 and a code domain interference cancellation and despreading signal vector correction unit 802,

wherein, the c-th intermediate detection unit estimates all the symbol estimates of the code c according to the de-spread signal vector corrected by all the symbols of the code c output by the previous-stage joint detection unit, and obtains the signal estimate and the total signal estimate of each symbol according to the estimation of each symbol, and then outputs the signal estimate of each symbol and the total signal estimate of the code c to the code domain interference cancellation and de-spread signal vector correction unit 802; the code domain interference cancellation and despreading signal vector correction unit 802 performs interference cancellation on the received signal vector, and outputs the obtained corrected despreading signal vector of each symbol of each code channel to the (i +1) th cascade combination detection unit after completing despreading of residual signals obtained by the interference cancellation and correction of despreading signal vectors.

The c-th intermediate stage detection unit 801 includes:

a weight vector calculation unit 901 which calculates more than one weight vector according to the estimation of the channel matrix and inputs each weight vector to a decision statistic calculation unit 902 in sequence;

each decision statistic calculation unit 902 calculates a decision statistic according to the modified received signal vector of one symbol of the code c output by the i-1 th cascade combination detection unit and the weight vector of the symbol, and sends the calculated decision statistic to the sending symbol estimation unit 903 connected with the decision statistic calculation unit;

each transmitted symbol estimation unit 903 estimates an estimation value of a corresponding symbol according to the received decision statistic and sends the estimation value to a signal estimation unit 904;

signal estimation section 904 performs signal estimation of all symbols and total signal estimation of code channel c from the estimated values of all symbols of received code channel c, and outputs the signal estimation and total signal estimation of all symbols to code domain interference cancellation and despreading signal vector correction section 802.

The last-stage joint detection unit 304 includes: c last-stage detection units are arranged in the circuit,

wherein, the C (C is more than or equal to 1 and less than or equal to C) last-stage detection unit estimates the estimation of all symbols of the code C according to the de-spread signal vector corrected by all symbols of the code C output by the previous-stage joint detection unit, and the estimation is used as the final detection result.

The C-th last stage detection unit 1001 of the C last stage detection units includes:

a weight vector calculation unit 1101 that calculates M weight vectors and outputs the M weight vectors to 1 to M decision statistic calculation units 1102, respectively;

each decision statistic calculation unit 502 calculates a decision statistic according to the received weight vector and the received modified despread signal vector of the corresponding symbol, and then outputs the calculated decision statistic to the transmission symbol estimation unit 1103 connected thereto;

the transmit symbol estimation unit 1103 estimates a symbol according to the decision statistic, and then directly outputs the symbol estimation value as a final detection result.

The multi-code receiver of the MIMO system performs parallel detection among all symbols in code channels, and performs parallel processing and parallel interference cancellation among all code channels. The receiver can effectively solve the demodulation problem of the MIMO system under multi-code transmission, not only improves the demodulation performance of the receiver under multi-code transmission, but also greatly reduces the processing delay of the receiver by parallel demodulation and parallel cancellation among a plurality of code channels and parallel demodulation of each symbol in the code channels, and simultaneously improves the system capacity and the peak rate.

Drawings

Fig. 1 is a schematic structural diagram of a single-code MIMO receiver in the prior art;

FIG. 2 is a schematic diagram of a prior art CD-SIC V-BLAST receiver;

the modulation greatly reduces the processing delay of the receiver and simultaneously improves the system capacity and the peak rate.

Drawings

Fig. 1 is a schematic structural diagram of a single-code MIMO receiver in the prior art;

FIG. 2 is a schematic diagram of a prior art CD-SIC V-BLAST receiver;

FIG. 3 is a schematic diagram of a receiver according to the present invention;

FIG. 4 is a schematic diagram of the first-level joint detection structure 302 in FIG. 3;

FIG. 5 is a diagram of the first stage detection unit 401 of code c in FIG. 4;

fig. 6 is a schematic diagram of a symbol detection and interference cancellation unit 502 in the ith new group in fig. 5;

fig. 7 is a diagram of the code-domain interference cancellation and despread signal vector correction unit 402 in fig. 4;

FIG. 8 is a schematic diagram of the ith (1 < i < S) cascade binding detection structure 303 of FIG. 3;

FIG. 9 is a schematic diagram of the mid-level detection unit 801 of FIG. 8;

FIG. 10 is a schematic diagram of the S-th cascade junction detection structure 304 of FIG. 3;

fig. 11 is a schematic structural diagram of the last stage detection unit 1001 in fig. 10.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the accompanying drawings.

The invention carries out parallel detection among all symbols in the code channels, and carries out parallel processing and parallel interference cancellation among all code channels. As shown in fig. 3, the MIMO receiver of the present invention includes a despreading unit group 301 and a joint detection structure 305. Wherein, the despreading unit group 301 includes C despreading units, and the joint detection structure 305 includes a first-stage joint detection structure 302 to an S-th-stage joint detection structure 304. The despreading unit group 301 receives signal vectors, despreads the received signal vectors by using C despreading units, and inputs the obtained despread signal vectors of C code channels to the first-stage joint detection structure 302; the first-stage joint detection structure 302 performs parallel packet detection and interference cancellation on the despread signal vector of each code channel, then performs code domain interference cancellation and correction on the despread signal vector on the obtained signal estimation and total signal estimation of all symbols of each code channel, outputs the corrected despread signal vector of each symbol of each code channel to the second-stage joint detection structure, and the second-stage joint detection structure performs similar processing: the corrected despread signal vector of each symbol of each code channel is detected to obtain an estimate of each symbol of each code channel, then the signal estimate of each symbol and the total signal estimate of each code channel are obtained from the estimate of each symbol of each code channel, then the signal estimate of each symbol of each code channel, the total signal estimate of each code channel and the received signal vector are used to correct the despread signal vector by code-domain interference, and the updated corrected despread signal vector of each symbol of each code channel is output to the next stage of joint detection structure, and the subsequent stages of joint detection structure perform the same processing until the (S-1) th stage of joint detection structure 303 outputs the updated corrected despread signal vector of each symbol of each code channel to the S-th stage of joint detection structure 304. The S-th joint detection structure 304 detects the received modified despread signal vectors for each symbol of each code channel, respectively, to obtain estimates of all symbols of each code channel.

The signal detection process of the MIMO receiver of the invention is as follows:

the number of transmitting antennas is set to be M, and the number of receiving antennas of the receiver is set to be N. The N-dimensional chip-level received signal vector is r (t). The transmitting end adopts C OVSF codes to carry out code multiplexing transmission. The C OVSF codes are respectively: OVSF code C, C is 1, 2, 3.

Under the assumption, the C-th despreading unit in the despreading unit group despreads each component in the N-dimensional baseband received signal vector r (t) by using the OVSF code C and the scrambling code, so as to obtain 1N-dimensional despread signal vector, i.e., Rc, C is 1, 2. And inputs the resulting C N-dimensional despread signal vectors into the first-stage joint detection structure 302.

As shown in fig. 4, the first-stage joint detection structure 302 includes C first-stage detection units 401 and a code-domain interference cancellation and despread signal vector modification unit 402. Wherein the c-th first stage detection unit 401 obtains the signal estimates of all symbols of code c and the total signal estimate from the c-th despread signal vector. Each first-stage detection unit 401 inputs the obtained signal estimates of all symbols and the total signal estimate to a code-domain interference cancellation and despreading signal vector correction unit 402. The code domain interference cancellation and despreading signal vector correction unit 402 completes the chip-level interference cancellation and the correction of the despreading signal vector, and obtains the corrected despreading signal vectors of all symbols corresponding to codes 1 to C. The despread signal vectors corrected for each symbol are then output to the second stage joint detection structure 303.

The second-stage joint detection structure 303 to the (S-1) -th joint detection structure 303 have the same internal structure, as shown in FIG. 8. When 1 < i < S, the (i-1) th cascade combination detection structure comprises C mid-stage detection units 801 and 1 code domain interference cancellation and despreading signal vector correction unit 802. The C middle-stage detection units 801 respectively receive the modified despread signal vectors of all symbols of code 1 to the modified despread signal vectors of all symbols of code C output by the previous-stage joint detection structure. Each inter-stage detection unit 801 obtains signal estimates and total signal estimates for all symbols of the corresponding code channel from the corrected despread signal vectors for all symbols of the received code channel, and outputs the signal estimates and the total signal estimates for all symbols of the code channel to a code domain interference cancellation and despread signal vector correction unit 802. The unit completes chip-level interference cancellation from the received signal estimates and total signal estimates of all symbols of the 1 to C code channels, and obtains corrected despread signal vectors of all symbols of the codes 1 to C. These output signals are fed in parallel to the (i +1) th cascade combination detection structure. Wherein the c-th inter-stage detection unit 801 obtains the signal estimates and the total signal estimate of all symbols of the code c by detection from the corrected despread signal vectors of all symbols of the code c.

And finishing the signal processing from the second-level joint detection structure to the (S-1) th joint detection structure according to the processing procedure.

As shown in fig. 10, the S-th cascade bonding detection structure includes C last-stage detection units 1001. The c-th last stage detection unit 1001 derives estimates of all symbols of the code c from the corrected despread signal vectors of all symbols of the code c from the (S-1) -th joint detection structure.

The first-stage detection unit 401 in fig. 4, the intermediate-stage detection unit 801 in fig. 8, the last-stage detection unit 1001 in fig. 10, and the code-domain interference cancellation and despread signal vector correction unit 402 will be described in detail.

Referring to fig. 5, the first stage detection unit 401 includes symbol grouping and reordering units 501, Z_cA new intra-group symbol detection and interference cancellation unit 502 and a new pre-processing unit 503.

The symbol grouping and rearranging unit 501 divides M symbols into Z groups according to a certain criterion from the input despread signal vector and the estimation of the channel matrix, rearranges the M symbols according to the grouping order and the intra-group symbol order, rearranges the column vectors correspondingly forming the channel matrix to form a new channel matrix, and inputs the despread signal vector and the estimation of the new channel matrix to the first new intra-group symbol detection and interference cancellation unit 502. At the same time, the despread signal vectors are output to a new preprocessing unit 503, and the new estimates of the channel matrix are sent to other new intra-group symbol detection and interference cancellation units 502.

The new intra-group symbol detection and interference cancellation unit 502 performs symbol detection and parallel interference for one group from the received despread signal vectors to obtainThe signal estimates to all symbols in the group and the modified despread signal vectors, and the signal estimates for all symbols are sent as output from the first stage detection unit to the code domain interference cancellation and despread signal vector modification unit 402. The despread signal vectors modified by the new intra-group symbol detection and interference cancellation unit 502 are used for the detection of the next group of symbols. For the c-th first stage detection unit 401, it contains Z_cThe symbol detection and interference cancellation units in the new group respectively complete the first group to the Z < th > group_cThe detection and interference cancellation of the group symbols and the obtained signal estimation of all symbols in the corresponding group are output to the code domain interference cancellation and de-spread signal vector correction unit 402.

As shown in fig. 6, the new intra-group symbol detection and interference cancellation unit 502 includes a weight vector calculation unit 601, a decision statistic calculation unit 602, a transmitted symbol estimation unit 603, and a signal estimation and interference cancellation unit 604.

The new pre-processing unit 503 obtains a total signal estimate from the corrected despread signal vector from the last group and the despread signal vector output by the symbol grouping and rearranging unit 501, and outputs the total signal estimate to the code domain interference cancellation and despread signal vector correction unit 402.

Let the input signal of the c-th first-stage detection unit be R_c，c＝1，2，3…，C。R_cThe following can be written:

R_c＝Ha_c+v_c (1a)

here, R_cThe signal vector after N-dimensional despreading is obtained by the c-th despreading unit in the despreading unit group 301; a is_c＝[a_c1，a_c2，…，a_cM]^TThe vector is an M-dimensional transmitting symbol vector, and the vector is transmitted through M transmitting antennas after being spread by OVSF code c and scrambling codes; h is an N multiplied by M dimensional channel matrix; v. of_cRepresenting an N-dimensional noise vector whose components are independently identically distributed gaussian white noise.

Order to

R_1，c，1＝R_c (1b)

In the invention, R is_1，c，1The function of each subunit in the c-th first-stage detection unit 401 in fig. 4 is explained for the input signal vector of the unit.

The symbol grouping and reordering unit 501 has the following functions:

a symbol grouping and rearranging unit 501 despreads the signal vector R_1c，1And the estimation of the channel matrix H is based on a certain criterion on M symbols a_c1，a_c2，…，a_cMIs divided into Z_cGroup i (i ═ 1, 2, …, Z)_c) The number of the group symbols is L_c，i. Ith group L_c，iA symbol is b_ci1，b_ci2，…，b_ciLc，i. Wherein, $b_{\mathrm{cij}}=a_{c{M}_{c,i,j}}$ (j varies with the group number i and ranges from 1, 2, …, L_c，i) This means that the ith group of jth symbols under code channel c has a sequence number M in the original symbol set_c，i，j(M_c，i，j∈[1，2，…，M]). Now the symbol sequence a_c1，a_c2，…，a_cMRearrangement is carried out, and the rearrangement result is: b_c11，b_c12，…，b_c1Lc，1，b_c21，…，b_c2Lc，2，…，b_cZc1，…，b_cZcLc，Zc. The corresponding transmitted symbol vector is converted to:. The channel matrix H in equation (1a) can be written in the form of a column vector:H＝[H₁，H₂，…，H_M]. For sequence H₁，H₂，…，H_MIs subjected to the sequence a_c1，a_c2，…，a_cMThe same rearrangement gives a sequence Q_c1，Q_c2，…，Q_cM. Note Q_c＝[Q_c1，Q_c2，…，Q_cM]. Through H and a_cThe equations (1a) and (1b) can be updated as follows:

R_1，c，1＝Q_cb_c+v_c (1f)

the symbol grouping and reordering unit 501 updates the channel matrix Q_cIs sent to Z_cA new intra-group symbol detection and interference cancellation unit 502. Will despread the signal vector R_1，c，1To a first new intra-group symbol detection and interference cancellation unit 502 and a new pre-processing unit 503.

The new intra-group symbol detection and interference cancellation unit 502 functions as follows:

ith (i e [1, 2, …, Z)_c]) The symbol detection and interference cancellation unit in the new group de-spread the signal vector at the input as R_1，ciTotal number of symbols in the group is L_c，iThen complete the intra-group L_c，iA symbol b_ci1，b_ciw，…，b_ciLc，iDetecting and obtaining a corrected despread signal vector R_1，c，i+1。

Referring to fig. 6, the symbol detection and interference cancellation unit in the ith new group performs the following calculations:

weight vector calculation unit 601 calculates each zero-forcing weight vector w according to a formula_{1，c，i，j}，j＝1，2，…，L_c，i。

L_c，iThe decision statistic calculation unit 602 calculates L in parallel_c，iA decision statistic. The jth decision statistic calculation unit 602 calculates a jth decision statistic from the jth zero-forcing vector:

${\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="C20041005005400281.png,C20041005005400291.png"\; state="\{\{state\}\}">y</figure-callout>}}_{1,c,i,j}={\mathrm{<figure-callout\; id="1"\; label="w"\; filenames="C20041005005400281.png,C20041005005400291.png"\; state="\{\{state\}\}">w</figure-callout>}}_{1,c,i,\mathrm{<figure-callout\; id="1"\; label="j\; T\; R"\; filenames="C20041005005400281.png,C20041005005400291.png"\; state="\{\{state\}\}">j</figure-callout>}}^T{R}_{}{}_{1,c,i},$ j＝1，2，…，L_c，i(2a)

L_c，ithe symbol estimation units 603 calculate L in parallel_c，iEstimation of individual symbols. The jth symbol estimation unit 603 obtains an estimate of the jth transmitted symbol from the jth decision statistic:

${\hat{b}}_{c,i,j}=D\left({\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="C20041005005400281.png,C20041005005400291.png"\; state="\{\{state\}\}">y</figure-callout>}}_{1,c,i,j}\right),$ j＝1，2，…，L_c，i(2b)

the signal estimation and interference cancellation unit 604 obtains a signal estimate for each symbol and a modified despread signal vector from the estimates for all symbols in the group. The signal estimates for each symbol are fed to a code domain interference cancellation and despread signal vector correction unit 402 in fig. 4, and the corrected despread signal vectors are used for detection of the (i +1) th group of symbols.

The signal estimate for each symbol is:

$s_{1,c,M_{c,i,j}}={\hat{b}}_{c,i,j}{\left({\hat{Q}}_c\left(i\right)\right)}_{J+j},$ j＝1，2，…，L_c，i， <math><mrow> <mi>J</mi> <mo>=</mo> <munderover> <mi>&Sigma;</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <mrow> <mi>i</mi> <mo>-</mo> <mn>1</mn> </mrow> </munderover> <msub> <mi>L</mi> <mrow> <mi>c</mi> <mo>,</mo> <mi>k</mi> </mrow> </msub> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>2</mn> <mi>c</mi> <mo>)</mo> </mrow> </mrow></math>

here, M_c，i，j(M_c，i，j∈[1，2，…，M]) Indicating the sequence number of the ith group of jth symbols in the original symbol set.

Is an estimate of the updated channel matrix.

The following can be obtained:

front of <math><mrow> <mi>J</mi> <mo>=</mo> <munderover> <mi>&Sigma;</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <mrow> <mi>i</mi> <mo>-</mo> <mn>1</mn> </mrow> </munderover> <msub> <mi>L</mi> <mrow> <mi>c</mi> <mo>,</mo> <mi>k</mi> </mrow> </msub> </mrow></math> The column is a 0 column vector, and the (J +1) th to Mth columns require Q to be paired by the pilot sequence after the (i-1) th interference cancellation_cThe channel estimation is performed again in the (J +1) th column to the Mth column of the matrix. If the channel estimation is not performed again, then

The following can be obtained: make the matrix

Middle front <math><mrow> <mi>J</mi> <mo>=</mo> <munderover> <mi>&Sigma;</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <mrow> <mi>i</mi> <mo>-</mo> <mn>1</mn> </mrow> </munderover> <msub> <mi>L</mi> <mrow> <mi>c</mi> <mo>,</mo> <mi>k</mi> </mrow> </msub> </mrow></math> The column vector of the column is 0 columnVectors, and other column vectors are Q's obtained from the last channel estimate_cThe estimated (J +1) th to mth columns of the matrix. If the latest channel estimation is performed in the kth (k < i) new group symbol detection and interference cancellation unit

From the (J +1) th column to the Mth column of

The (J +1) th to M-th columns of (a).

The modified received signal vector is:

<math><mrow> <msub> <mi>R</mi> <mrow> <mn>1</mn> <mo>,</mo> <mi>c</mi> <mo>,</mo> <mi>i</mi> <mo>+</mo> <mn>1</mn> </mrow> </msub> <mo>=</mo> <msub> <mi>R</mi> <mrow> <mn>1</mn> <mo>,</mo> <mi>c</mi> <mo>,</mo> <mi>i</mi> </mrow> </msub> <mo>-</mo> <munderover> <mi>&Sigma;</mi> <mrow> <mi>j</mi> <mo>=</mo> <mn>1</mn> </mrow> <msub> <mi>L</mi> <mrow> <mi>c</mi> <mo>,</mo> <mi>i</mi> </mrow> </msub> </munderover> <msub> <mi>s</mi> <mrow> <mn>1</mn> <mo>,</mo> <mi>c</mi> <mo>,</mo> <msub> <mi>M</mi> <mrow> <mi>c</mi> <mo>,</mo> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> </mrow> </msub> <mo>,</mo> </mrow></math> j＝1，2，…，L_c，i(2d)

the new pre-processing unit 503 functions as follows:

the unitFrom the Z th_cThe modified despread signal vectors and the code-to-c despread signal vectors obtained by the new intra-group symbol detection and interference cancellation unit 502 result in a total signal estimate for code channel c as follows:

${\mathrm{<figure-callout\; id="1"\; label="S"\; filenames="C20041005005400281.png,C20041005005400291.png"\; state="\{\{state\}\}">S</figure-callout>}}_{1,c}={\mathrm{<figure-callout\; id="1"\; label="R"\; filenames="C20041005005400281.png,C20041005005400291.png"\; state="\{\{state\}\}">R</figure-callout>}}_{1,c,1}-{\mathrm{<figure-callout\; id="1"\; label="R"\; filenames="C20041005005400281.png,C20041005005400291.png"\; state="\{\{state\}\}">R</figure-callout>}}_{1,c,Z_c+1}---\left(3\right)$

the internal structure of the code-domain interference cancellation and despread signal vector correction unit 402 is shown in fig. 7, and the structure of the other stages of joint detection structure is the same as that of the unit. In the ith cascade combination detection structure, the code domain interference cancellation and despread signal vector correction unit 402 includes C spreading units 701, a code domain interference cancellation unit 702, C despreading units 703, and a despread signal vector correction unit 704. Wherein, each spreading unit carries out spreading on the received total signal estimation of the corresponding code channel, and then outputs the spread signal to a code domain interference cancellation unit 702, the code domain interference cancellation unit 702 carries out code domain interference cancellation on the received spread signal to obtain residual signals, and then respectively outputs the residual signals to a corresponding despreading unit 703, the c-th despreading unit 703 despreads each of the received signals by using OVSF code c scrambling code to obtain a c-th despread signal vector, and outputs the c-th despread signal vector to a despread signal vector correcting unit 704, and the despread signal vector correcting unit 704 finishes the correction of the despread signal vector by using the c despread signal vectors and the signal estimation of all symbols of all code channels, and outputs the despread signal vectors of all symbols of all code channels.

The specific working principle is as follows:

first, the c-th spreading unit 701 estimates s from the c-th OVSF code, scrambling code, and total signal of the c-th code channel_i，cObtaining the regenerated signal g of the c code channel_i，c。

The regenerated signals of C code channels obtained by the C spreading units 701 are sent to the interference cancellation unit 702. The unit completes interference cancellation to obtain a residual signal. The residual signal is calculated as follows:

<math><mrow> <msub> <mi>R</mi> <mrow> <mi>i</mi> <mo>+</mo> <mn>1</mn> </mrow> </msub> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>=</mo> <mi>R</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>-</mo> <munderover> <mi>&Sigma;</mi> <mrow> <mi>c</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>C</mi> </munderover> <msub> <mi>g</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>c</mi> </mrow> </msub> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>4</mn> <mi>a</mi> <mo>)</mo> </mrow> </mrow></math>

the residual signals are fed in parallel to C despreading units 703. The c-th despreading unit 703 despreads each component of the input N-dimensional residual signal with the c-th OVSF code and the scrambling code to obtain a despread signal vector R_i+1，c。

The C despread signal vectors obtained by the C despreading units 703 are all fed to a despread signal vector correction unit 704. The unit derives a corrected despread signal vector for each symbol of each code track from the C despread signal vectors and the signal estimates for all symbols of code 1 through code C. The corrected despread signal vector for the jth symbol of the c code channel obtained by the ith cascade combination detection structure is as follows:

R_i+1，c，j＝R_i+1，c+s_i，c，j＝1，2，…，M，c＝1，2，…，C (4b)

the corrected despread signal vectors for all symbols are fed into the (i +1) th joint detection structure.

The structure of the c-th intermediate-level detection unit 801 in the i-th (1 < i < S) th cascade-combination detection unit in fig. 8 is shown in fig. 9. The unit completes the ith detection of all symbols of the c code channel. The unit comprises a weight vector calculation unit 901, a decision statistic calculation unit 902, a transmitted symbol estimation unit 903 and a signal estimation unit 904. The weight vector calculation unit 901 calculates more than one weight vector according to the estimation of the H matrix, and inputs each weight vector to the decision statistic calculation unit 902 in sequence; each decision statistic calculation unit 902 calculates a decision statistic for a symbol of the code track c to which the (i-1) th joint detection structure is applied, based on the modified despread signal vector for the symbol and the weight vector for the symbol, and sends the calculated decision statistic to its connected transmitted symbol estimation unit 903. Each transmitted symbol estimation unit 903 estimates the estimation value of the corresponding symbol according to the received decision statistic and sends the estimation value to the signal estimation unit 904, and after the signal estimation unit 904 completes the signal estimation of all symbols of the code channel c and the total signal estimation of the code channel c according to the received estimation values of all symbols, the signal estimation and the total signal estimation of all symbols of the code channel c are output to the code domain interference cancellation and despreading signal vector correction unit 802.

The c-th intermediate detection unit 801 performs the following functions:

the weight vector calculation unit 901 calculates each zero-forcing weight vector w according to a formula_i，c，j，j＝1，2，…，M。

The M decision statistic calculation units 902 calculate M decision statistics in parallel. IthDecision statistic calculation section 902 calculates a jth decision statistic from the jth zero-forcing vector and the modified despread signal vector of the jth symbol: $y_{i,c,j}=w_{i,c,j}^T{R}_{i,c,j},$ j＝1，2，…，M(5a)

m symbol estimation units 903 compute estimates of the M transmitted symbols in parallel. The jth symbol estimation unit 903 obtains an estimate (obtained by soft decision) of the jth transmitted symbol from the jth decision statistic: ${\hat{a}}_{i,c,j}=D\left(y_{i,c,j}\right),$ j＝1，2，…，M(5b)

signal estimation unit 904 derives signal estimates for the M symbols and a total signal estimate from the estimates for the M transmitted symbols.

The signal estimate for each symbol is:

$s_{i,c,j}={\hat{a}}_{i,c,j}{\left(\hat{H}\left(i\right)\right)}_j,$ j＝1，2，…，M，(5c)

for the updated estimation of the channel matrix H, it is obtained by re-performing channel estimation on the matrix H at the i-th stage.

The following can be obtained:

after the interference cancellation of the previous stage joint detection structure, the channel estimation is performed again on each column of H by the pilot sequence. If channel estimation is not to be performed anew at the ith stage

The following can be obtained:

equal to the last estimate of the channel matrix H. If the most recent channel estimation is performed in the kth (k < i) th joint detection structure $\hat{H}\left(i\right)=\hat{H}\left(k\right).$

The total signal is estimated as:

<math><mrow> <msub> <mi>S</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>c</mi> </mrow> </msub> <mo>=</mo> <munderover> <mi>&Sigma;</mi> <mrow> <mi>j</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>M</mi> </munderover> <msub> <mi>s</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>c</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>,</mo> </mrow></math> j＝1，2，…，M，(5d)

the structure of the last stage of the detection unit is shown in fig. 11. Let i be S, the c-th last-stage detection unit 1001 in the S-th cascade combination detection structure 304 in fig. 10 includes a weight vector calculation unit 1101, a decision statistic calculation unit 1102, and a transmitted symbol estimation unit 1103. The weight vector calculation unit 1101 calculates M weight vectors, and outputs the M weight vectors to the 1-M decision statistic calculation units 1102, each decision statistic calculation unit 1102 calculates a decision statistic according to the received weight vector and the received modified received signal vector of the corresponding symbol, and then outputs the calculated decision statistic to the transmission symbol estimation unit 1103 connected thereto. After the transmit symbol estimation unit 1103 estimates a symbol, it directly outputs the symbol estimation value as the final detection result.

The c-th last-stage detecting unit 1001 performs the following functions:

weight vector calculation section 1101 calculates each zero-forcing weight vector w according to the formula_i，c，j，j＝1，2，…，M。

The M decision statistic calculation units 1102 calculate M decision statistics in parallel. The jth decision statistic calculation unit 1102 calculates a jth decision statistic from the jth zero-forcing vector according to equation (5 a).

M symbol estimation units 1103 compute estimates of M transmitted symbols in parallel. The ith symbol estimation unit 1103 obtains an estimate of the jth transmitted symbol from the jth decision statistic according to formula (5 b).

The despreading units in fig. 3 and 7 perform despreading of the input signal vectors. Let the c 0VSF code be o_c(t) scrambling code is s (t). In fig. 3 and 7, the c-th despreading unit inputs an N-dimensional signal vector v (t) ═ v₁(t)，v₂(t)，…，v_N(t)]^TThe output U of the c-th despreading unit_c(t)＝[u_c1(t)，u_c2(t)，…，u_cN(t)]^TCalculated according to the following formula:

<math><mrow> <msub> <mi>u</mi> <mi>cn</mi> </msub> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mn>1</mn> <msub> <mi>T</mi> <mi>s</mi> </msub> </mfrac> <munder> <mo>&Integral;</mo> <msub> <mi>T</mi> <mi>s</mi> </msub> </munder> <msub> <mi>v</mi> <mi>n</mi> </msub> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <msub> <mi>o</mi> <mi>c</mi> </msub> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <msup> <mi>s</mi> <mo>*</mo> </msup> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>,</mo> </mrow></math> n＝1，2，…，N，c＝1，2，…，C (6)

wherein, T_sIs symbolNumber period.

In practice, the input signal vectors of the c-th despreading unit in fig. 3 and 7 are different and are the received signal vector and the residual signal, respectively.

The spreading unit 701 in fig. 7 functions as follows:

let the input of the c-th spreading unit be a signal vector U of dimension N_c(t)＝[u_c1(t)，u_c2(t)，…，u_cN(t)]^TOutput V of the unit_c(t)＝[v_c1(t)，v_c2(t)，…，v_cN(t)]^TCalculated according to the following formula:

v_cn(t)＝u_cn(t)o_c(t)s(t)，n＝1，2，…，N，c＝1，2，…，C (7)

the calculation of the weight vector is crucial in the first to the S-th cascade joint detection structures.

Under ZF criterion, in the first modified symbol detection and interference cancellation unit in the c-th first-stage detection unit in the first-stage joint detection structure, the estimation of the channel matrix Qc can be usedThe zero forcing weight vector is calculated according to the formula (8a) and the formula (8 b):

$G_{c,1}={\left({\hat{Q}}_c\left(1\right)\right)}^{+}---\left(8a\right)$

$w_{c,1,j}={\left(G_{c,1}\right)}_j^T,$ j＝1，2，…，L_c，1(8b)

wherein (G)_c，1)_jRepresentation matrix G_c，1Line j of (1), symbol T denotes transpose, L_c，1Indicating the number of group 1 elements under code track c.

Under the ZF criterion, in the ith (1 < i ≦ Z_c) When detecting the group symbol, the channel matrix can be used

The zero forcing weight vector is calculated according to the formula (8c) and the formula (8 d):

$G_{c,i}={\left({\hat{Q}}_c\left(i\right)\right)}^{+}---\left(8c\right)$

$w_{c,i,j}={\left(G_{c,i}\right)}_{J+j}^T,$ j＝1，2，…，L_c，i， <math><mrow> <mi>J</mi> <mo>=</mo> <munderover> <mi>&Sigma;</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <mrow> <mi>i</mi> <mo>-</mo> <mn>1</mn> </mrow> </munderover> <msub> <mi>L</mi> <mrow> <mi>c</mi> <mo>,</mo> <mi>k</mi> </mrow> </msub> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>8</mn> <mi>d</mi> <mo>)</mo> </mrow> </mrow></math>

wherein L is_c，iIndicating the number of elements of the ith group.

Indicating the channel matrix Q at the time of the i-th group detection_cIs estimated.

Channel matrix without re-channel estimation

Front of <math><mrow> <mi>J</mi> <mo>=</mo> <munderover> <mi>&Sigma;</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <mrow> <mi>i</mi> <mo>-</mo> <mn>1</mn> </mrow> </munderover> <msub> <mi>L</mi> <mrow> <mi>c</mi> <mo>,</mo> <mi>k</mi> </mrow> </msub> </mrow></math> The columns are 0-column vectors, and the (J +1) -th to Mth columns are the most recent Q_cThe (J +1) th to M-th columns of the estimation. If the latest channel estimation is performed in the kth (k < i) new group symbol detection and interference cancellation unit

From the (J +1) th column to the Mth column of

The (J +1) th to M-th columns of (a).

In case of channel estimation anew, the channel matrix

Front of <math><mrow> <mi>J</mi> <mo>=</mo> <munderover> <mi>&Sigma;</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <mrow> <mi>i</mi> <mo>-</mo> <mn>1</mn> </mrow> </munderover> <msub> <mi>L</mi> <mrow> <mi>c</mi> <mo>,</mo> <mi>k</mi> </mrow> </msub> </mrow></math> The column is a 0-column vector, and the (J +1) -th to mth columns are obtained by performing channel estimation anew at the time of the i-th group detection. Wherein, J is the [ J +1, M ]]Column passing pair Q_cThe jth e [ J +1, M ] of the matrix]And the column is obtained by performing channel estimation again.

Under the MMSE criterion, in the first modified symbol detection and interference cancellation unit in the group, the channel can be selectedMatrix array

The weight vector is calculated according to equation (9a) and equation (9 b):

<math><mrow> <msub> <mi>G</mi> <mrow> <mi>c</mi> <mo>,</mo> <mn>1</mn> </mrow> </msub> <mo>=</mo> <msubsup> <mi>&sigma;</mi> <mi>s</mi> <mn>2</mn> </msubsup> <msup> <mi>F</mi> <mi>H</mi> </msup> <msup> <mrow> <mo>[</mo> <msubsup> <mi>&sigma;</mi> <mi>s</mi> <mn>2</mn> </msubsup> <msup> <mi>FF</mi> <mi>H</mi> </msup> <mo>+</mo> <msup> <mi>&sigma;</mi> <mn>2</mn> </msup> <msub> <mi>I</mi> <mi>M</mi> </msub> <mo>]</mo> </mrow> <mrow> <mo>-</mo> <mn>1</mn> </mrow> </msup> <mo>,</mo> </mrow></math> $F=\left({\hat{Q}}_c\left(1\right)\right)---\left(9a\right)$

$w_{c,1,j}={\left(G_{c,1}\right)}_j^T,$ j＝1，2，…，L_c，1(9b)

under the MMSE criterion, in the ith (1 < i ≦ Z_c) When detecting the group symbol, the channel matrix can be used

The weight vector is calculated according to equation (9c) and equation (9 d):

<math><mrow> <msub> <mi>G</mi> <mrow> <mi>c</mi> <mo>,</mo> <mn>1</mn> </mrow> </msub> <mo>=</mo> <msubsup> <mi>&sigma;</mi> <mi>s</mi> <mn>2</mn> </msubsup> <msup> <mi>F</mi> <mi>H</mi> </msup> <msup> <mrow> <mo>[</mo> <msubsup> <mi>&sigma;</mi> <mi>s</mi> <mn>2</mn> </msubsup> <msup> <mi>FF</mi> <mi>H</mi> </msup> <mo>+</mo> <msup> <mi>&sigma;</mi> <mn>2</mn> </msup> <msub> <mi>I</mi> <mi>M</mi> </msub> <mo>]</mo> </mrow> <mrow> <mo>-</mo> <mn>1</mn> </mrow> </msup> <mo>,</mo> </mrow></math> $F=\left({\hat{Q}}_c\left(i\right)\right)---\left(9c\right)$

$w_{c,i,j}={\left(G_{c,i}\right)}_{J+j}^T,$ j＝1，2，…，L_c，i， <math><mrow> <mi>J</mi> <mo>=</mo> <munderover> <mi>&Sigma;</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <mrow> <mi>i</mi> <mo>-</mo> <mn>1</mn> </mrow> </munderover> <msub> <mi>L</mi> <mrow> <mi>c</mi> <mo>,</mo> <mi>k</mi> </mrow> </msub> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>9</mn> <mi>d</mi> <mo>)</mo> </mrow> </mrow></math>

channel matrix without re-channel estimation

Front of <math><mrow> <mi>J</mi> <mo>=</mo> <munderover> <mi>&Sigma;</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <mrow> <mi>i</mi> <mo>-</mo> <mn>1</mn> </mrow> </munderover> <msub> <mi>L</mi> <mrow> <mi>c</mi> <mo>,</mo> <mi>k</mi> </mrow> </msub> </mrow></math> The columns are 0-column vectors, and the (J +1) -th to Mth columns are the most recent Q_cThe (J +1) th to M-th columns of the estimation. If the latest channel estimation is performed in the kth (k < i) new group symbol detection and interference cancellation unit

From the (J +1) th column to the Mth column of

The (J +1) th to M-th columns of (a).

In case of channel estimation anew, the channel matrix

Front of <math><mrow> <mi>J</mi> <mo>=</mo> <munderover> <mi>&Sigma;</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <mrow> <mi>i</mi> <mo>-</mo> <mn>1</mn> </mrow> </munderover> <msub> <mi>L</mi> <mrow> <mi>c</mi> <mo>,</mo> <mi>k</mi> </mrow> </msub> </mrow></math> The column is a 0-column vector, and the (J +1) -th to mth columns are obtained by channel estimation. Wherein, J is the [ J +1, M ]]Column passing pair Q_cThe jth e [ J +1, M ] of the matrix]And the column is obtained by performing channel estimation again.

Since the computation of the weight vector in the second-stage to S-stage combined detection structure is independent of the OVSF code channel number, the following structure simplification can be performed in the MIMO receiver described above:

in the ith (1 < i < S) cascade combination detection structure, the weight vector calculation unit in each intermediate stage detection unit can be eliminated. And adding a weight vector calculation unit in each cascade combination detection structure, and calculating according to a formula (10) to obtain M weight vectors. The M weight vectors are simultaneously supplied to the C mid-stage detection units.

In the final-stage joint detection structure, the weight vector calculation unit in each final-stage detection unit can be eliminated. And adding a weight vector calculation unit in the last-stage joint detection structure, and calculating to obtain M weight vectors by using i as S according to a formula (10). The M weight vectors are simultaneously supplied to the C last stage detection units.

Under the ZF criterion, in the ith (i is more than or equal to 1 and less than or equal to S) cascade combination detection structure, the channel matrix can be used for estimating the channel again

And (3) calculating a weight vector of the jth symbol of the code channel c according to the formula (10 a):

$W_{i,j}=\frac{{\left(\hat{H}\left(i\right)\right)}_j^{*}}{{\left|\right|{\left(\hat{H}\left(i\right)\right)}_j\left|\right|}^2},$ i＝1，…，S；j＝1，2，…，M (10a)

in the above formula, the first and second carbon atoms are,to represent

The jth row of (1), | B | | non-woven phosphor²＝B^HB, B is a column vector,is an estimate of the channel matrix H in the ith combined detection structure.

Under the ZF criterion, in the ith (i is more than or equal to 1 and less than or equal to S) cascade combination detection structure, the estimation of the latest channel matrix can be carried out without carrying out channel estimation again

(k < i) the weight vector of the jth symbol of the code channel c is calculated according to the formula (10 b):

$W_{i,j}=\frac{{\left(\hat{H}\left(k\right)\right)}_j^{*}}{{\left|\right|{\left(\hat{H}\left(k\right)\right)}_j\left|\right|}^2},$ i＝2，…，S；j＝1，2，…，M (10b)

here, ,

indicating that the most recent channel estimation was performed in the kth (k < i) cascade joint detection configuration.

Under the MMSE criterion, in the ith (1 ≦ i ≦ S) cascade joint detection structure, the estimation of the channel matrix can be performed under the condition of channel estimation again

Calculating the jth symbol weight vector of the code channel c according to the formula (10 c):

<math><mrow> <msub> <mi>w</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>=</mo> <msup> <mrow> <mo>(</mo> <msubsup> <mi>&sigma;</mi> <mi>s</mi> <mn>2</mn> </msubsup> <msup> <mi>F</mi> <mi>H</mi> </msup> <msup> <mrow> <mo>[</mo> <msubsup> <mi>&sigma;</mi> <mi>s</mi> <mn>2</mn> </msubsup> <msup> <mi>FF</mi> <mi>H</mi> </msup> <mo>+</mo> <msup> <mi>&sigma;</mi> <mn>2</mn> </msup> <msub> <mi>I</mi> <mi>M</mi> </msub> <mo>]</mo> </mrow> <mrow> <mo>-</mo> <mn>1</mn> </mrow> </msup> <mo>)</mo> </mrow> <mi>T</mi> </msup> <mo>,</mo> </mrow></math> $F={\left(\hat{H}\left(i\right)\right)}_j---\left(10c\right)$

wherein σ²For receiving the power of the noise component in the signal vector, σ_s ²Is the power of the transmitted symbol. Under the MMSE criterion, in the ith (1 ≦ i ≦ S) cascade joint detection structure, the estimation of the channel matrix can be carried out without re-channel estimation

Calculating a weight vector of the jth symbol of the code channel c according to the formula (10 d):

<math><mrow> <msub> <mi>w</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>=</mo> <msup> <mrow> <mo>(</mo> <msubsup> <mi>&sigma;</mi> <mi>s</mi> <mn>2</mn> </msubsup> <msup> <mi>F</mi> <mi>H</mi> </msup> <msup> <mrow> <mo>[</mo> <msubsup> <mi>&sigma;</mi> <mi>s</mi> <mn>2</mn> </msubsup> <msup> <mi>FF</mi> <mi>H</mi> </msup> <mo>+</mo> <msup> <mi>&sigma;</mi> <mn>2</mn> </msup> <msub> <mi>I</mi> <mi>M</mi> </msub> <mo>]</mo> </mrow> <mrow> <mo>-</mo> <mn>1</mn> </mrow> </msup> <mo>)</mo> </mrow> <mi>T</mi> </msup> <mo>,</mo> </mrow></math> $F={\left(\hat{H}\left(k\right)\right)}_j---\left(10d\right)$

wherein σ²For receiving the power of the noise component in the signal vector, σ_s ²Is the power of the transmitted symbol.

In the MIMO receiver, the value range of the series S is that S is more than or equal to 2 and less than M. Generally, the value of S can be determined according to specific performance and latency requirements. Generally, when S is 2 or 3, the performance should be good, and the performance is not improved much by increasing S.

The number of groups Z in the first stage structure of the MIMO receiver_cCan be preset; the SNR of each symbol can be calculated from the channel matrix and then the symbols can be grouped according to the SNR of each symbol, and a specific grouping method can be usedSo as to borrow the grouping method in the GSIC method of the grouping serial interference cancellation in the multi-user detection.

In short, the above description is only a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention.

Claims (10)

1. A multicode receiver for a multiple-input multiple-output system, the receiver comprising:

a set of despreading units (301) and at least two stages of joint detection units (305), wherein,

each despreading unit in the group of despreading units (301) despreads each component in the received signal vector and inputs one or more resulting despread signal vectors to a first-stage joint detection unit (302) in the at least two-stage joint detection unit (305); the first-stage joint detection unit (302) performs grouping detection on all symbols in code channels corresponding to each received despread signal vector to obtain signal estimation and total signal estimation of all symbols in each code channel, then performs code domain interference cancellation and despread signal vector correction on the obtained signal estimation and total signal estimation of all symbols in each code channel, outputs the corrected despread signal vector of each symbol in each code channel to the next-stage joint detection unit for processing until the last-stage joint detection unit (304), and the last-stage joint detection unit (304) detects the received corrected despread signal vector to obtain estimation of all symbols in each code channel.

2. The receiver of claim 1, wherein the set of despreading units includes more than one despreading unit, and wherein the c-th despreading unit despreads each component of the received signal vector with an OVSF code c and a scrambling code to obtain a c-th despread signal vector.

3. The receiver of claim 1, wherein the first-stage joint detection unit comprises: c first-stage detection units (401) and a code-domain interference cancellation and despread signal vector modification unit (402), wherein,

the c first-stage detection unit (401) obtains signal estimation and total signal estimation of all symbols of the code c from the c despreading signal vector, and each first-stage detection unit (401) inputs the obtained signal estimation and total signal estimation of all symbols to a code domain interference cancellation and despreading signal vector correction unit (402); a code domain interference cancellation and despreading signal vector correction unit (402) completes the correction of chip level interference cancellation and despreading signal vectors to obtain corrected despreading signal vectors of all symbols corresponding to codes 1 to C; and then outputting the despread signal vector corrected by each symbol to a next-stage joint detection unit.

4. A receiver according to claim 3, characterized in that the c-th first stage detection unit (401) comprises:

symbol grouping withA rearrangement unit (501) for dividing the M symbols into Z symbols according to a predetermined criterion from the input despread signal vector and the estimate of the channel matrix_cThe M symbols are rearranged according to the grouping sequence and the symbol sequence in the group, column vectors forming the channel matrix are also rearranged to form a new channel matrix, the despread signal vector and the estimation of the new channel matrix are input to a first new intra-group symbol detection and interference cancellation unit (502), meanwhile, the estimation of the new channel matrix is output to the rest new intra-group symbol detection and interference cancellation units (502) and the despread signal vector is output to a new preprocessing unit (503);

Z_ca new intra-group symbol detection and interference cancellation unit (502), wherein the first new intra-group symbol detection and interference cancellation unit performs symbol detection and parallel interference cancellation of one group from the received despread signal vectors, obtains despread signal vectors of signal estimation and correction of all symbols within the group, and sends the signal estimation outputs of all symbols to a code domain interference cancellation and despread signal vector correction unit (402), and outputs the obtained corrected despread signal vectors to the next new intra-group symbol detection and interference cancellation unit; second to Z_cThe symbol detection and interference cancellation unit in the new group completes symbol detection and parallel interference cancellation of one group by the received corrected despread signal vector to obtain signal estimation of all symbols in the group and the corrected despread signal vector, and sends the signal estimation output of all symbols to a code domain interference cancellation and despread signal vector correction unit (402), and outputs the obtained corrected despread signal vector to the next symbol detection and interference cancellation unit in the new group;

a new pre-processing unit (503) received from the Z-th_cThe modified despread signal vectors output by the symbol detection and interference cancellation unit and the despread signal vectors output by the symbol grouping and reordering unit in the new group estimate a total signal estimate and send it to a code domain interference cancellation and despread signal vector modification unit (402).

5. The receiver of claim 4, wherein Z is_cAn ith intra-group symbol detection and interference cancellation unit of the plurality of intra-group symbol detection and interference cancellation units (502) comprises:

a weight vector calculation unit (601) for calculating one or more weight vectors from the estimation of the input channel matrix and sequentially inputting each weight vector to a decision statistic calculation unit (602);

each decision statistic calculation unit (602) is used for calculating decision statistics according to the corrected received signal vector ri and the weight vector output by the symbol detection and interference cancellation unit in the (i-1) th new group, and sending the calculated decision statistics to a sending symbol estimation unit (603) connected with the decision statistics calculation unit;

each transmitted symbol estimation unit (603) is used for estimating the estimation value of the corresponding symbol in the group according to the received decision statistic and sending the estimation value to a signal estimation and interference cancellation unit (604);

and a signal estimation and interference cancellation unit (604) for outputting the signal estimates of all symbols to a code domain interference cancellation and despread signal vector modification unit (402) after the signal estimation and interference cancellation of all symbols in the group are completed by the estimated values of all symbols in the received group and the modified received signal vector, and outputting the modified despread signal vector to an i +1 th new intra-group symbol detection and interference cancellation unit.

6. The receiver according to claim 3, wherein said code-domain interference cancellation and despread signal vector correction unit (402) comprises: c spreading units (701), a code domain interference cancellation unit (702), C despreading units (703) and a despread signal vector correction unit (704),

wherein, each spread unit carries out spread spectrum to the received signal estimation corresponding to the code channel, then outputs the spread spectrum signal to a code domain interference cancellation unit (702), the code domain interference cancellation unit (702) carries out interference cancellation to the received spread spectrum signal, then outputs the residual signal obtained by the interference cancellation to all the de-spread units (703) in parallel, the c-th de-spread unit de-spreads each component in the input signal vector by OVSF code c and scrambling code to obtain the c-th de-spread signal vector, outputs the obtained de-spread signal vector to a de-spread signal vector correction unit (704), and the de-spread signal vector correction unit (704) finishes the correction of the de-spread signal vector and outputs the corrected de-spread signal vector of all the symbols of all the code channels.

7. The receiver according to claim 1, wherein the ith joint detection unit (303) of the at least two stages of joint detection units (305) comprises: c middle-stage detection units (801) and code domain interference cancellation and de-spread signal vector correction units (802), wherein i is more than 1 and less than S, and S is the total number of a joint detection structure;

wherein, the c-th intermediate detection unit estimates the estimation of all symbols of the code c according to the corrected despread signal vectors of all symbols of the code c output by the previous-stage joint detection unit, and obtains the signal estimation and the total signal estimation of each symbol according to the estimation of each symbol, and then outputs the signal estimation of each symbol and the total signal estimation of the code c to a code domain interference cancellation and despread signal vector correction unit (802); a code domain interference cancellation and de-spread signal vector correction unit (802) performs interference cancellation on the received signal vector, and outputs the obtained corrected de-spread signal vector of each symbol of each code channel to an (i +1) th cascade combination detection unit after de-spreading the residual signal obtained by the interference cancellation and correcting the de-spread signal vector.

8. The receiver according to claim 7, characterized in that the c-th intermediate stage detection unit (801) comprises:

a weight vector calculation unit (901) which calculates more than one weight vector according to the estimation of the channel matrix and inputs each weight vector to a decision statistic calculation unit (902) in sequence;

each decision statistic calculation unit (902) calculates a decision statistic according to the corrected received signal vector of one symbol of the code c output by the i-1 th cascade combination detection unit and the weight vector of the symbol, and sends the calculated decision statistic to a sending symbol estimation unit (903) connected with the decision statistic calculation unit;

each transmitted symbol estimation unit (903) estimates an estimation value of a corresponding symbol according to the received decision statistic and sends the estimation value to a signal estimation unit (904);

a signal estimation unit (904) performs signal estimation of all symbols and total signal estimation of the code channel c from the estimated values of all symbols of the received code channel c, and outputs the signal estimation of all symbols and the total signal estimation to a code domain interference cancellation and despreading signal vector correction unit (802).

9. The receiver according to claim 1, wherein the last stage joint detection unit (304) comprises: c last-stage detection units are arranged in the circuit,

the C last stage detection unit estimates the estimation of all symbols of the code C according to the de-spread signal vectors corrected by all symbols of the code C output by the previous stage joint detection unit, and as the final detection result, C is more than or equal to 1 and less than or equal to C, wherein C is the total number of the last stage detection units.

10. The receiver according to claim 9, characterized in that the C-th last detection unit (1001) of said C last detection units comprises:

a weight vector calculation unit (1101) which calculates M weight vectors and outputs the M weight vectors to 1-M decision statistic calculation units (1102);

each decision statistic calculation unit (1102) respectively calculates a decision statistic according to the received weight vector and the received corrected despread signal vector of the corresponding symbol, and then outputs the calculated decision statistic to a sending symbol estimation unit (1103) connected with the decision statistic;

and a sending symbol estimation unit (1103) which directly outputs the symbol estimation value as the final detection result after estimating the symbol according to the decision statistic.

Images