CN1909402A - Self-adapting transmission method and apparatus used in spacing related MIMO system - Google Patents

Abstract

According to the present invention, an adaptive transmission method used in a space-correlated multiple-input multiple-output communication system is proposed, including: the receiving end estimates the channel statistical characteristics and feeds it back to the sending end; the sending end estimates the channel statistical characteristics according to the feedback , determine the power allocation matrix and preprocessing matrix; the receiving end determines the modulation and coding parameters according to the channel characteristics and sends them to the transmitting end; and the transmitting end performs preprocessing and power allocation on the transmitted signal according to the determined power allocation matrix and preprocessing matrix, And modulate and code the transmission signal according to the modulation and coding parameters received from the receiving end, so as to transmit the transmission signal through the antenna.

Description

Adaptive transmission method and device used in spatially correlated MIMO system

technical field

The present invention relates to an adaptive transmission technology in a spatially correlated MIMO system, and more specifically, to an adaptive transmission method and device used in a spatially correlated MIMO system, which can improve the performance of adaptive transmission in a correlated MIMO system without Bring excessive overhead and increase in complexity to the system.

Background technique

Increasingly high information transmission rate is one of the main problems faced by future wireless communication systems. In order to achieve this goal on limited spectrum resources, multi-antenna technology (MIMO, Multiple Input Multiple Output) has become one of the indispensable means used in future wireless communications. In the MIMO system, the transmitting end uses multiple antennas to transmit signals, and the receiving end uses multiple antennas to receive spatial signals. Studies have shown that compared with traditional single-antenna transmission methods, MIMO technology can significantly increase channel capacity, thereby increasing information transmission rates.

In addition to MIMO technology, adaptive transmission technology can also effectively improve the information transmission rate in fading channels. Adaptive modulation and coding (AMC) technology is an important adaptive transmission technology. Its basic idea is to transmit the modulation and coding parameters according to the current channel characteristics adaptive changes, and transmit more when the channel conditions are good. Information, when the channel condition is poor, less information is transmitted to improve the average throughput of the system, that is, the average spectrum utilization.

Therefore, if you combine MIMO and AMC technology, you will be able to obtain a higher information transmission rate than using only one technology.

FIG. 1 is a schematic diagram of the structure of a common MIMO system using the AMC technology. For example, the PARC (Antenna-by-Antenna Rate Control) scheme proposed by Lucent Company in 3GPP.

In this structure, the transmitting end and the receiving end respectively use n _T and n _R antennas to transmit and receive signals. At the sending end, the data to be sent is first processed by the serial-to-parallel conversion device 101 and divided into n _T data substreams, and each data substream corresponds to a transmitting antenna. Before transmission, these data sub-streams are adaptively modulated and coded in the adaptive modulation and coding device 102 according to the current channel transmission characteristics corresponding to each transmitting antenna. The modulation and coding parameters M ₁ , _{M 2} , .

At the receiving end, firstly, the n _R receiving antennas 104 receive all the spatial signals, and then the channel estimation device 105 performs channel estimation according to the pilot signal in the received signals or by using other methods to estimate the current channel characteristic matrix H (For a MIMO system, its channel characteristics can be described by an n _R ×n _T matrix). Then, the AMC parameter selection unit 107 determines the modulation and coding parameters used by each data substream at the sending end according to H, and sends the selected modulation and coding parameters of each data substream back to the sending end through the feedback channel. Here, in order to reduce the system feedback overhead, the antennas at the transmitting end are generally used to transmit with equal power, and when feeding back the AMC parameters of each substream, only the sequence numbers corresponding to the modulation and coding parameters are returned. Finally, the MIMO detection device 106 detects each transmission data sub-stream using a general MIMO detection method according to the channel characteristic matrix H and the modulation and coding parameters of each sub-stream output by the AMC parameter selection device 107, and obtains the original transmission data.

In PARC, the operation in the AMC parameter selection device 107, that is, the AMC parameter selection process can be divided into two steps:

(1) According to the current channel characteristic matrix H, pre-calculate the equivalent SINR (signal to interference and noise ratio) of each substream 1, 2, ..., n _T after MIMO detection: SINR(1), SINR(2), ... , SINR(n _T ).

Here, there are many MIMO detection methods that can be used, such as: linear detection methods, including zero-forcing (ZF) and minimum mean square error (MMSE) detection; serial interference cancellation (SIC) method; maximum likelihood detection method, etc. . The calculation methods of the equivalent SINR after sub-stream detection under various MIMO detections have been given in the existing literature. For example: when using ZF detection, the SINR of the kth sub-flow after detection is: $\mathrm{SINR}\left(k\right)=\frac{{\mathrm{E.}}_{\mathrm{the\; s}}}{{\mathrm{no}}_T{N}_0{\left[h^{*}h\right]}_{\mathrm{kk}}^{-1}};$ When using MMSE detection, the SINR of the kth sub-flow after detection is: $\mathrm{SINR}\left(k\right)=\frac{{\mathrm{E.}}_{\mathrm{the\; s}}}{{\mathrm{no}}_T{N}_0{[h^{*}h+N_0/{\mathrm{E.}}_{\mathrm{the\; s}}{I}_n]}_{\mathrm{kk}}^{-1}}-1,$ Among them, E _s is the total transmission power, N ₀ is the noise power, and I _nT is the unit matrix of n _T ×n _T.

(2) According to the obtained SINR(1), SINR(2), ..., SINR(n _T ), select modulation and coding parameters M ₁ , M ₂ , ..., M _nT for each substream.

A variety of methods can be used to determine the modulation and coding parameters by SINR, for example, the following method can be used: firstly select several combinations of modulation and coding parameters, and estimate the AWGN (additive white Gaussian noise) channel by theoretical derivation or numerical method BER performance for various parameters. Then, according to the detected SINR value of each sub-stream, the modulation and coding parameters that can meet a certain BER requirement and have the largest throughput are selected as the modulation and coding parameters on the transmission sub-stream.

The research shows that the adaptive transmission method of PARC can obtain better throughput performance in the MIMO system. However, the acquisition of this better throughput performance requires a prerequisite, that is, the MIMO channels are independent.

However, in practical MIMO systems, MIMO channels are often correlated. There are many reasons for MIMO channel correlation, such as not placing the antennas far enough apart, not having enough scattered objects around the antennas, and there is a direct line of sight (LOS) between the transceiver and so on. When the MIMO channel is correlated, its channel characteristic matrix H can be described by the following formula: $h={\mathrm{<figure-callout\; id="1"\; label="R\; r"\; filenames="A20051008939500151.png"\; state="\{\{state\}\}">R</figure-callout>}}_r^{}{h}_{\mathrm{<figure-callout\; id="1"\; label="w\; R\; t"\; filenames="A20051008939500151.png"\; state="\{\{state\}\}">w</figure-callout>}}{R}_t^{}{}_{1/2}^{}$

where H _w is the n _R ×n _T independent MIMO channel characteristic matrix, R _r and R _t are the n _R ×n _R and n _T ×n _T receive and transmit correlation matrices, respectively.

Compared with stand-alone MIMO systems, the PARC method suffers a large performance loss in practical correlated MIMO systems. In order to overcome the performance loss caused by channel correlation, the prior art provides a method based on performing singular value decomposition (SVD) on the instantaneous channel characteristic matrix H. The idea of this method is as follows: at each sending moment, the receiving end sends the estimated current channel characteristic matrix H back to the sending end through the feedback channel, and then the sending end performs SVD on H, and uses the SVD result to The signal is preprocessed. Although this method can improve the performance under correlated MIMO to a certain extent, it is unacceptable in terms of system feedback overhead and implementation complexity to perform feedback and SVD on H at every moment in the actual system.

Therefore, how to design a new adaptive transmission method for spatially correlated MIMO systems, on the one hand, it can obtain better adaptive performance than the traditional PARC method, and on the other hand, it will not bring too much overhead and cost to the system. This increase in complexity is one of the main objectives of the invention.

Contents of the invention

The purpose of the present invention is to propose an adaptive transmission method and device used in a spatially correlated MIMO system, which can improve the performance of adaptive transmission in a correlated MIMO system without bringing too much overhead and complexity to the system Increase. In the method according to the present invention, the transmitting end performs preprocessing and power allocation on the transmitted signal according to the long-term statistical characteristics of the MIMO related channel, and the receiving end performs bit allocation on each transmission sub-stream according to the short-term channel characteristics. In this method, the long-term statistical characteristics and short-term characteristics of MIMO-related channels are used in the selection of adaptive parameters. Compared with traditional methods, better self-adaptation can be obtained under the premise of similar implementation complexity and feedback overhead. performance.

In order to achieve the above object, according to the present invention, an adaptive transmission method used in a space-correlated multiple-input multiple-output communication system is proposed, including: the receiving end estimates the channel statistical characteristics and feeds it back to the sending end; According to the statistical characteristics of the incoming channel, the power allocation matrix and preprocessing matrix are determined; the receiving end determines the modulation and coding parameters according to the channel characteristics and sends them to the transmitting end; and the transmitting end performs preprocessing on the transmitted signal according to the determined power allocation matrix and preprocessing matrix processing and power allocation, and modulate and code the transmission signal according to the modulation and coding parameters received from the receiving end, so as to transmit the transmission signal through the antenna.

Preferably, the step of determining the power allocation matrix and the preprocessing matrix according to the fed back statistical characteristics of the channel includes: determining the power allocation matrix and the preprocessing matrix according to the long-term statistical characteristics of MIMO related channels.

Preferably, the long-term statistical characteristics of the MIMO correlation channel is R=E{H ^H R _n ^-1 H}, where H is the MIMO channel characteristic matrix, and R _n is the additive noise in the received signal The covariance matrix of n, and E is the mathematical expectation operator.

Preferably, the receiving end determines the modulation and coding parameters according to the channel characteristics and sends them to the sending end including: determining the modulation and coding parameters according to the short-term channel characteristics.

Preferably, before the step of the transmitting end performing preprocessing and power allocation on the transmitted signal according to the long-term statistical characteristics of the multiple-input multiple-output related channel, it also includes the step of first determining the number of parallel sub-streams L, L transmitted by the transmitting end Selected by the system, its value should not be greater than the rank of R.

Preferably, the step of determining the preprocessing matrix according to the long-term channel statistical characteristics fed back by the sending end includes: performing eigenvalue decomposition on R; and selecting an L× n _T matrix.

Preferably, the step of determining the power allocation matrix at the transmitting end according to the fed back statistical characteristics of the channel includes: determining the power allocation matrix in the manner of equal power allocation and water injection power allocation.

Preferably, the method further includes: using the current channel characteristic matrix and the statistical characteristics of the channel at the receiving end to eliminate the effect of preprocessing at the sending end through filtering, whitening the noise through filtering, and performing matched filtering on the received signal.

According to the present invention, an adaptive transmission device used in a space-correlated multiple-input multiple-output communication system is also proposed, including: a statistical characteristic calculation device at the receiving end, used to estimate channel statistical characteristics and feed it back to the sending end; The long-term control device at the sending end determines the power allocation matrix and the preprocessing matrix according to the channel statistical characteristics fed back; the modulation and coding parameter selection device at the receiving end is used to determine the modulation and coding parameters according to the channel characteristics and send them to the sending end ; The power distribution device at the sending end is used for preprocessing and power distribution of the sent data according to the determined power distribution matrix; the preprocessing device at the sending end is used for preprocessing the sent data according to the determined preprocessing matrix; And the modulation and coding device at the sending end modulates and codes the sending signal according to the modulation and coding parameters fed back from the receiving end.

Preferably, the above-mentioned adaptive transmission device further includes: filtering means for eliminating the effect of preprocessing at the sending end, whitening the noise, and performing matched filtering on the received signal by using the current channel characteristic matrix and the statistical characteristics of the channel.

Description of drawings

The above objects, advantages and features of the present invention will become apparent by referring to the following detailed description of preferred embodiments employed in conjunction with the accompanying drawings, wherein:

FIG. 1 is a schematic diagram showing the structure of a traditional MIMO system using AMC technology;

FIG. 2 is a schematic diagram showing a structure of a MIMO system according to an embodiment of the present invention;

FIG. 3 is a detailed schematic diagram showing a filtering device according to an embodiment of the present invention;

FIG. 4 is a flowchart illustrating an adaptive transmission method according to an embodiment of the present invention;

FIG. 5 is a graph showing the performance comparison of the method adopted by the present invention and the conventional method.

Detailed ways

The specific implementation manner of the present invention will be described below in conjunction with the accompanying drawings.

FIG. 2 is a schematic diagram of a MIMO system using the technology of the present invention.

Wherein, the sending end and the receiving end respectively use n _T and n _R antennas to send and receive signals. At the sending end, the data to be sent is first divided into L parallel data sub-streams through the serial-to-parallel conversion device 101, L≤min{n _T , n _R }. Before transmission, the L data substreams need to be converted into n _T channels of signals through the adaptive modulation and coding device 102, the power allocation device 201 and the preprocessing device 202, and then sent out from the n _T antennas at the transmitting end. Here, the _modulation and coding parameters M={M ₁ , M ₂ , . Short-term feedback refers to feedback once at each transmission moment, such as each time slot). In addition, the power distribution device 201 and the preprocessing device 202 are realized by a power distribution matrix P and a preprocessing matrix T respectively. The purpose of the power allocation matrix P is to weight the power of the L transmission sub-streams. P is a real positive definite diagonal matrix of L×L, which can be expressed as $P=\mathrm{diag}\{\sqrt{{\mathrm{<figure-callout\; id="1"\; label="P"\; filenames="A20051008939500151.png"\; state="\{\{state\}\}">P</figure-callout>}}_1},\sqrt{{\mathrm{<figure-callout\; id="2"\; label="P"\; filenames="A20051008939500151.png"\; state="\{\{state\}\}">P</figure-callout>}}_2},...,\sqrt{P_L}\},$ and satisfied $\underset{i=1}{\overset{L}{\mathrm{\&Sigma;}}}{P}_i\&le;P_{\mathrm{total}},$ Wherein P _i is the transmission power allocated on the i-th substream, i=1, 2, ..., L, and P _total is the total transmission power limit. The preprocessing matrix T is an L×n _T matrix, and its purpose is to combat the deterioration of system performance caused by the correlation of MIMO channels. Different from the adaptive modulation and coding parameter M, the parameters P and T both come from the long-term feedback of the receiving end (the so-called long-term feedback here refers to feedback once in a long time, and the feedback time interval is much larger than the short-time Feedback time interval in the feedback), and the result calculated by the sending end through the long-term control device 203 according to the feedback result.

At the receiving end, the spatial signals are first received by n _R receiving antennas 104, and then the following three operations are completed:

(1) The channel estimation device 105 performs channel estimation according to the received signal, and estimates the current channel characteristic matrix H.

(2) Calculate the parameters required by the sending end for adaptive transmission at the next sending time, and send the calculated parameters back to the sending end through the feedback channel 108 . Specifically, this step includes:

(2.1) According to the result of the channel estimation, calculate the channel statistical characteristic in the means 213 for calculating the channel statistical characteristic, and feed it back to the sending end. Since the channel statistical characteristics change very slowly over time, the calculation and feedback of the channel statistical characteristics here are long-term processes, that is, they are calculated and fed back once in a long period of time.

(2.2) According to the current channel characteristic matrix H obtained by channel estimation and the calculated channel statistical characteristics, the modulation and coding parameters M={M ₁ , M ₂ ,  … , M _L } to select, which is executed in the AMC parameter selection device 214, and the selected result M is fed back to the sending end. Since the channel characteristic matrix H itself changes quickly, the calculation and feedback of the parameter M here is a short-term process, that is, the selection and feedback of M must be performed within a short period of time, for example, in each time slot.

(3) Detecting the currently received signal, including the functions of the filtering device 211 and the MIMO detecting device 212 . Here, the purpose of the operation of the filtering device 211 is to eliminate the role of preprocessing at the sending end, where the current channel characteristic matrix H and the statistical characteristics of the channel need to be used. In addition to H and channel statistical characteristics, the MIMO detection device 212 also needs to use the modulation and coding parameter information adopted by each substream at the sending end. Here, the filtering device 211 may be refined as shown in FIG. 3 .

Figure 3 shows a detailed schematic diagram of the filtering device.

From the realization function, it includes two parts of operation: whitening the noise, and matching filtering to the signal. Wherein, the noise whitening and filtering device 301 can be described by a weighting matrix R _n ^−1/2 , where R _n is the covariance matrix of the additive noise at the receiving end, and R _n can be estimated by 213 . The matched filtering process of the received signal can be realized by the devices 302, 303, 304 and 305, that is, the matrices R _n -H ^{/ 2} , H ^H , TH and P ^H are used to perform matched filtering on the input signal, wherein () ^H Represents the conjugate transpose of a matrix.

Compared with the traditional MIMO-AMC structure in Fig. 1, ie PARC, the main differences of the MIMO-AMC system using the technology of the present invention are:

(1) At the sending end, the number of sending sub-streams is L, L≤min{n _T , n _R }, and the value of L can vary at different times, instead of the fixed n _T sending sub-streams used in the traditional PARC system flow. In addition, the transmitting end performs power allocation 201 and preprocessing 202 operations on the signal before transmission, and a long-term control device 202 is also added for this purpose. The information required by the power allocation device 201 and the preprocessing operation device 202 comes from the feedback of the receiving end, but because it is a long-term process, that is, the power allocation matrix P and the preprocessing T remain unchanged for a long time, so it will not It brings too much complexity and an increase in feedback overhead to the system.

(2) At the receiving end, a filtering device 211 and a statistical characteristic computing device 213 are added. The purpose of the statistical characteristic calculating means 213 is to provide the transmitting end with the statistical characteristic of the channel for power allocation and preprocessing operations of the transmitting end. At the same time, since the transmitting end has added a preprocessing operation, before performing MIMO detection on the received signal, a matched filtering operation must be performed on the receiving signal in the filtering device 211 to cancel the effect of the transmitting end preprocessing on the signal.

Specifically, the method proposed in the present invention can be described by FIG. 4 .

Figure 4 shows the adaptive transmission method adopted by the invention.

Specifically, the implementation of this method mainly includes the following steps:

Step 1: The receiving end estimates the channel statistical characteristic R, and feeds it back to the sending end, such as S401. Specifically include:

(1.1) Estimate the covariance matrix R _n of the additive noise n in the received signal, where n is an n _R- dimensional column vector, R _n =E{nn ^H }, and E{} is the expectation function. In practice, the long-term average of time can be used to obtain R _n .

(1.2) Estimate the channel statistical characteristic R, where R=E{H ^H R _n ⁻¹ H}.

(1.3) Send the estimated R back to the sending end through the feedback channel.

In the second step, the sending end determines the power allocation matrix P and the preprocessing matrix T according to R, such as S411. This step is completed by the long-term control device 203 at the sending end, and its specific implementation steps include:

(2.1) Determine L, where L is the number of parallel sub-streams sent by the sender. In this method, it is required that L≤rank(R), where rank() represents the rank of the matrix. In this step, rank(R) is calculated first, and then the system selects a positive integer not greater than rank(R) as the value of L.

(2.2) Determine the preprocessing matrix T. The specific method is to perform eigenvalue decomposition (EVD) on R, and then select the eigenvectors corresponding to the largest L eigenvalues, and form the L eigenvectors into an L×n _T preprocessing matrix T.

(2.3) Determine the power allocation matrix P, where two power allocation methods can be used:

(2.3.1) Equal power distribution of each sub-flow, namely $P=\mathrm{diag}\{\sqrt{P_{\mathrm{total}}/L},\sqrt{P_{\mathrm{total}}/L},....,\sqrt{P_{\mathrm{total}}/L}\},$ Wherein P _total is the total transmission power limit.

(2.3.2) Power allocation method based on "water injection". In this method, the largest L eigenvalues of R are first calculated, expressed as {λ ₁ ,λ ₂ ,…,λ _L }, and get $P_i={(\mathrm{\&mu;}-\frac{{\mathrm{L\&sigma;}}_{\mathrm{no}}^2}{P_{\mathrm{total}}{\mathrm{\&lambda;}}_i})}_{+},$ where μ is a constant, the function ${\left(x\right)}_{+}=\left\{\begin{array}{cc}x& x\&Greater\; Equal;0\\ 0& x<0\end{array}\right..$ Thus, the power allocation matrix is obtained as

$\mathrm{<span\; class="notranslate">\; P</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; diag</span>}<span\; class="notranslate">\; \{</span>\sqrt{{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="1"\; label="P"\; filenames="A20051008939500151.png"\; state="\{\{state\}\}">P</figure-callout></span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}<span\; class="notranslate">\; ,</span>\sqrt{{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="P"\; filenames="A20051008939500151.png"\; state="\{\{state\}\}">P</figure-callout></span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>\sqrt{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; L</span>}}}<span\; class="notranslate">\; \}</span><span\; class="notranslate">\; .</span>$

Step 3: The receiving end selects the _modulation and coding parameters M={M ₁ , M ₂ , . Specifically include:

(3.1) According to the current channel characteristic matrix H, pre-calculate the equivalent SINR values SINR(1), SINR(2),..., SINR(L) of each substream 1, 2, ..., L after filtering and MIMO detection. The specific method is the same as the process of calculating the equivalent SINR in the PARC introduced in the background art, and will not be repeated here. However, it should be noted that the equivalent channel matrix is changed due to the addition of preprocessing and filtering devices in this method. Specifically, the equivalent channel matrix at this time is changed from H in traditional PARC to H ₀ =P ^H T ^H H ^H R _n ^-1 HTP. That is to say, all the H in the original SINR calculation formula are replaced by H ₀ , and all n _T are replaced by L. For example, when MMSE detection is used, the SINR of the kth sub-flow after detection is: $\mathrm{SINR}\left(k\right)=\frac{{\mathrm{E.}}_{\mathrm{the\; s}}}{L{N}_0{[{h_0}^{*}{h}_0+N_0/{\mathrm{E.}}_{\mathrm{the\; s}}{I}_L]}_{\mathrm{kk}}^{-1}}-1.$

(3.2) According to the obtained SINR(1), SINR(2), ..., SINR(L), select modulation and coding parameters M={M ₁ , M ₂ , ..., M _L } for each substream. This step is the same as the traditional PARC method introduced in the background art, and will not be repeated here.

(3.3) The receiving end sends the estimated modulation and coding parameters M={M ₁ , M ₂ , . . . , M _L } of each substream back to the sending end through the feedback channel.

Step 4: The transmitting end performs corresponding processing on the transmitted data according to the obtained power allocation matrix P, preprocessing matrix T and modulation and coding parameters M, and sends it out, such as S412. Specifically, the transmitting end first performs adaptive modulation and coding on the L transmission sub-streams according to the modulation and coding parameters M={M ₁ , M ₂ ,..., M _L }, and then uses the matrices P and T to perform power allocation and Preprocessing, and finally sending out from n _T antennas.

Step 5: After receiving the signal, the receiving end performs filtering and MIMO detection on the received signal according to the corresponding parameters P, T and M, such as S403. Specifically include:

(5.1) Filter processing is realized by the filter device 211 . Assuming that the received signal is y, then the output signal after filtering is z=P ^H T ^H H ^H R _n ⁻¹ y, where y is an n _R -dimensional column vector, and z is an L-dimensional column vector.

(5.2) MIMO detection is performed on the filtered output signal, and any conventional MIMO detection method can be used here.

Step 6: Judging whether it is necessary to re-estimate the channel statistical characteristics, if so, go to the first step, otherwise go to the third step, such as S404.

Since the channel statistical characteristics remain unchanged for a long time, the estimation and feedback of the channel statistical characteristics in this method is a long-term process, that is, it is performed once in a long time, and the specific time length can be initially set by the system. As for the third step, which is to select modulation and coding parameters for sending substreams, because H changes quickly, this operation is a short-term process, that is, within a short period of time, such as within a time slot, the To execute once.

Figure 5 shows the performance comparison between the method used in the present invention and the traditional method.

The system throughput performance comparison between the traditional PARC method and the method of the present invention is given in the simulation. Wherein, the number n _T of transmitting antennas and the number n _R of receiving antennas are both 4, and the channel adopts a flat Rayleigh fading channel. Among them, the AMC adopts adaptive modulation without coding, and the modulation parameters are: "no transmission", BPSK, QPSK, 8PSK and 16QAM, and the target BER=10 ^-3 . In addition, the sending correlation matrix is

${\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; t</span>}}<span\; class="notranslate">\; =</span>\left[\begin{array}{cccc}\mathrm{<span\; class="notranslate">\; 1</span>}& {\mathrm{<span\; class="notranslate">\; 0.76</span>}\mathrm{<span\; class="notranslate">\; e</span>}}^{\mathrm{<span\; class="notranslate">\; 0.17</span>}\mathrm{<span\; class="notranslate">\; \&pi;j</span>}}& {\mathrm{<span\; class="notranslate">\; 0.43</span>}\mathrm{<span\; class="notranslate">\; e</span>}}^{\mathrm{<span\; class="notranslate">\; 0.35</span>}\mathrm{<span\; class="notranslate">\; \&pi;j</span>}}& {\mathrm{<span\; class="notranslate">\; 0.25</span>}\mathrm{<span\; class="notranslate">\; e</span>}}^{\mathrm{<span\; class="notranslate">\; 0.53</span>}\mathrm{<span\; class="notranslate">\; \&pi;j</span>}}\\ {\mathrm{<span\; class="notranslate">\; 0.76</span>}\mathrm{<span\; class="notranslate">\; e</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 0.17</span>}\mathrm{<span\; class="notranslate">\; \&pi;<figure-callout\; id="1"\; label="j"\; filenames="A20051008939500151.png"\; state="\{\{state\}\}">j</figure-callout></span>}}& \mathrm{<span\; class="notranslate">\; 1</span>}& {\mathrm{<span\; class="notranslate">\; 0.76</span>}\mathrm{<span\; class="notranslate">\; e</span>}}^{\mathrm{<span\; class="notranslate">\; 0.17</span>}\mathrm{<span\; class="notranslate">\; \&pi;j</span>}}& {\mathrm{<span\; class="notranslate">\; 0.43</span>}\mathrm{<span\; class="notranslate">\; e</span>}}^{\mathrm{<span\; class="notranslate">\; 0.35</span>}\mathrm{<span\; class="notranslate">\; \&pi;j</span>}}\\ {\mathrm{<span\; class="notranslate">\; 0.43</span>}\mathrm{<span\; class="notranslate">\; e</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 0.35</span>}\mathrm{<span\; class="notranslate">\; \&pi;j</span>}}& {\mathrm{<span\; class="notranslate">\; 0.76</span>}\mathrm{<span\; class="notranslate">\; e</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 0.17</span>}\mathrm{<span\; class="notranslate">\; \&pi;<figure-callout\; id="1"\; label="j"\; filenames="A20051008939500151.png"\; state="\{\{state\}\}">j</figure-callout></span>}}& \mathrm{<span\; class="notranslate">\; 1</span>}& {\mathrm{<span\; class="notranslate">\; 0.76</span>}\mathrm{<span\; class="notranslate">\; e</span>}}^{\mathrm{<span\; class="notranslate">\; 0.17</span>}\mathrm{<span\; class="notranslate">\; \&pi;j</span>}}\\ {\mathrm{<span\; class="notranslate">\; 0.25</span>}\mathrm{<span\; class="notranslate">\; e</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 0.53</span>}\mathrm{<span\; class="notranslate">\; \&pi;j</span>}}& {\mathrm{<span\; class="notranslate">\; 0.43</span>}\mathrm{<span\; class="notranslate">\; e</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 0.35</span>}\mathrm{<span\; class="notranslate">\; \&pi;j</span>}}& {\mathrm{<span\; class="notranslate">\; 0.76</span>}\mathrm{<span\; class="notranslate">\; e</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 0.17</span>}\mathrm{<span\; class="notranslate">\; \&pi;<figure-callout\; id="1"\; label="j"\; filenames="A20051008939500151.png"\; state="\{\{state\}\}">j</figure-callout></span>}}& \mathrm{<span\; class="notranslate">\; 1</span>}\end{array}\right]$

Corresponding to the situation in ITU that the antenna spacing is λ/2, the transmission direction is 10°, and the angle extension is 15°. Also assume that the receiver is uncorrelated. It can be seen from the results in FIG. 5 that, compared with the traditional method, the method proposed in the application of the present invention can obtain better throughput performance. In addition, it can be seen from the previous analysis that the power allocation and preprocessing parameters added by the sending end in the method of the present invention all come from the long-term feedback and calculation of the system, so it will not bring too much feedback overhead and implementation complexity to the system on the increase.

Although the present invention has been illustrated in conjunction with the preferred embodiments thereof, those skilled in the art will understand that various modifications, substitutions and alterations can be made to the present invention without departing from the spirit and scope of the invention. Accordingly, the invention should not be limited by the above-described embodiments, but by the appended claims and their equivalents.

Claims (10)

1. An adaptive transmission method used in a space-correlated multiple-input multiple-output communication system, comprising: The receiving end estimates channel statistics and feeds them back to the sending end; The transmitting end determines the power allocation matrix and the preprocessing matrix according to the channel statistical characteristics fed back; The receiving end determines the modulation and coding parameters according to the channel characteristics and sends them to the sending end; and The transmitting end performs preprocessing and power allocation on the transmitted signal according to the determined power allocation matrix and preprocessing matrix, and modulates and encodes the transmitted signal according to the modulation and coding parameters received from the receiving end, so as to transmit the transmitted signal through the antenna Signal. 2. The method according to claim 1, characterized in that: the step of determining the power allocation matrix and the preprocessing matrix according to the channel statistical characteristics fed back by the transmitting end comprises: according to the long-term statistical characteristics of the MIMO correlation channel To determine the power allocation matrix and preconditioning matrix. 3. The method according to claim 2, characterized in that: the long-term statistical characteristic of the MIMO correlation channel is R=E{H ^H R _n ^-1 H}, where H is the MIMO channel characteristic matrix, R _n is the covariance matrix of additive noise n in the received signal, and E is the mathematical expectation operator. 4. The method according to claim 1, wherein the step of determining the modulation and coding parameters at the receiving end according to channel characteristics and sending them to the sending end comprises: determining the modulation and coding parameters according to short-term channel characteristics. 5. The method according to claim 3, characterized in that: before the step of performing preprocessing and power allocation on the transmitted signal at the transmitting end according to the long-term statistical characteristics of the MIMO related channel, further comprising the step of: first Determine the number L of parallel substreams sent by the sender, L is selected by the system, and its value should not be greater than the rank of R. 6. The method according to claim 5, characterized in that: the step of determining the preprocessing matrix according to the long-term channel statistical characteristics fed back by the transmitting end comprises: after performing eigenvalue decomposition on R; and selecting the largest L features An L×n _T matrix consisting of the eigenvectors corresponding to the values. 7. The method according to any one of claims 1 to 6, characterized in that: the step of determining the power allocation matrix at the transmitting end according to the channel statistical characteristics fed back comprises: following equal power allocation and water injection power allocation Determine the power allocation matrix. 8. The method according to claim 1, further comprising: the receiving end uses the current channel characteristic matrix and the statistical characteristics of the channel to eliminate the effect of preprocessing at the sending end through filtering, and whiten the noise through filtering, And match filtering is performed on the received signal. 9. An adaptive transmission device for use in a spatially correlated multiple-input multiple-output communication system, comprising: A statistical characteristic calculation device arranged at the receiving end for estimating channel statistical characteristics and feeding it back to the sending end; The long-term control device arranged at the sending end determines the power allocation matrix and the preprocessing matrix according to the channel statistical characteristics fed back; A modulation and coding parameter selection device arranged at the receiving end, used to determine the modulation and coding parameters according to the channel characteristics and send them to the sending end; A power distribution device arranged at the sending end, used for preprocessing and power distribution of the transmitted data according to the determined power distribution matrix; A preprocessing device arranged at the sending end, configured to preprocess the sent data according to the determined preprocessing matrix; and The modulation and coding device installed at the sending end modulates and codes the sending signal according to the modulation and coding parameters fed back from the receiving end. 10. The apparatus of claim 9, further comprising: The filter device is used to eliminate the effect of preprocessing at the sending end, whiten the noise, and perform matched filtering on the received signal by using the current channel characteristic matrix and the statistical characteristics of the channel.

Images