WO2008000119A1 - Communication method for adaptive multiple antennas and device thereof - Google Patents

Abstract

A communication method and a device for adaptive multiple antennas are characterized in that the channel is classified by the characteristics on the relative matrix of array receiving signal and the relative matrix of transmitting signal, then different modulation manners and space-time processing manners are selected for different kinds of channels. Said method comprises the following steps: receiving terminal evaluates channels, feeds back the state information of channels, evaluates and identifies the environment characteristics of communication channels, selects modulation manners and Space-Time processing manners based on the characteristics of channels, processes in various manners among multiple antennas vertical Bell Laboratories Layered Space-Time coding, Space-Time trellis coding or beamforming. The device comprises a Space-Time processing manner selecting unit, a vertical Bell Laboratories Layered Space-Time Coding unit, a Space-Time trellis coding unit, an adaptive beamforming unit, transmitting antenna array unit, a receiving antenna array unit, a Space-Time receiving unit, and a channel state information unit. Present invention enlarges the availability scope of the communication environment for multiple antenna system, and promotes the performance value ratio.

Description

TECHNICAL FIELD The present invention relates to the field of communications technologies, and in particular, to an adaptive multi-antenna communication method and apparatus based on spatial channel characteristics. BACKGROUND In order to meet the requirements of future communication, it is necessary to continuously improve the system capacity and frequency utilization of wireless transmission. Therefore, the pursuit of the highest possible spectrum utilization has become a challenging issue. Multi-antenna technology has been around for a long time. Many cellular systems have adopted multiple antennas, such as the Personal Handy-phone System (PHS) and Time Division-Synchronous Code Division Multiple. Access ) (hereinafter referred to as TD-SCDMA) systems use smart antenna technology, and the new broadband wireless communication system, Mobile Broadband Wireless Access (hereinafter referred to as MBWA), supports spatial multiplexing. Multiple-input Multiple-Output (MIMO) antenna technology. At present, the next generation mobile communication system industry recognized that the MIMO antenna 4 and multi-carrier technology should be used to achieve a significant increase in the transmission rate of the system. A layered space-time code system model proposed by Bell's chamber, Vertical Bell Laboratories Layered Space-Time Coding (V-BLAST), is able to effectively improve wireless One of the main schemes of spectrum utilization. The system is a space division multiplexing system that decomposes a single data stream into multiple sub-data streams by serial-to-parallel conversion, each sub-data stream is independently encoded/modulated, and At the receiving end, the data is also received by multiple antennas, and then the V-BLAST algorithm is used to detect the data stream sent by each antenna, and then the parallel data is converted to recover the original data stream. At the threshold, the data rate increases with the number of transmit and receive antennas of the BLAST system, and the channel capacity provided can approach the theoretical upper limit. It should be noted that the MIMO wireless channel in the V-BLAST structure is always assumed to be flat fading. However, in the case of high-speed wireless multimedia communication, since the number of transmitting antennas in the V-BLAST structure cannot be unrealistically increased, it relies solely on V_BLAST. The spatial domain demultiplexing operation in the structure will not maintain the flat fading of the MIMO wireless channel, ie the MIMO wireless channel will inevitably have frequency selectivity. Without doubt, for the frequency selective V-BLAST structure, the flat fading V -BLAST The structure will be difficult to use. In addition, the V-BLAST algorithm is greatly affected by the channel correlation. After the correlation coefficient is greater than 0.6, the error performance of the system deteriorates drastically. However, in an actual wireless transmission system, the correlation coefficient of the channel is determined by many factors, such as the physical parameters of the transmitting and receiving antennas, the distance between the antennas, the distribution of the spatial scatterers, and the like. Therefore, the channel is often relevant in practical applications, which also limits the application of the V-BLAST structure to some extent. In the adaptive beamforming technique, the beamforming technique is used to determine the direction of arrival of the useful signal, and the appropriate main combining weight is used to form the main beam of the antenna in this direction, and at the same time reduce the direction of the gain sidelobe or the 4 bar zero-spot alignment interference signal. At the time of transmission, the desired user's received signal power can be maximized while minimizing or even zeroing the undesired users outside the narrow beam illumination range. The current multi-antenna technology research results show that there is no multi-antenna technology that can achieve higher transmission rate and better communication quality under the condition of strong channel spatial correlation and weak channel spatial correlation. Due to the use of transmit diversity and receive diversity, MIMO technology can achieve high channel capacity and better performance when the channel spatial correlation is weak, especially when the channels are independent of each other. With the increase of channel spatial correlation, MIMO communication The bit error rate increases and the data transmission rate decreases. However, smart antennas use adaptive beamforming technology to make the beam direction point to the signal direction and suppress interference in other directions. Therefore, the spatial correlation in the channel is strong, especially in the case of line of sight, which tends to be the best algorithm. As the spatial correlation of the channel is weakened, the bit error rate of smart antenna communication increases and the data transmission rate decreases. Overcoming the shortcomings of the structure of a single multi-antenna system and making it more widely applicable has been the direction of researchers' efforts. A feasible method is to orthogonal frequency division multiplexing (Orthogonal

Frequency-Division Multiplexing X (hereinafter referred to as OFDM) technology introduces limitations such as V-BLAST structure, vacancy or demultiplexing operation. Such as the literature [WU Xiao-jun, LI Xing, WANG Ji-long.

Blind estimation of frequency-selective channels for V-BLAST OFDM system.

空间相关矩阵为 JOURNAL OF CHINA INSTITUTE OF COMMUNICATIONS, 2004, 25(10)] The structure of a tagged delay diversity V-BLAST OFDM system is presented by joint space frequency demultiplexing operation. This structure is capable of imparting rotational invariance properties to V-BLAST OFDM systems. However, this structure is extremely complicated and difficult to implement in an actual system. In addition, in the Chinese Patent Application (Application No. 200510021539.6) "Multi-antenna communication method and communication system adaptively adjusted according to channel spatial correlation", an existing MIMO technology and a smart antenna technology are combined to propose a kind A multi-antenna communication method adaptively adjusted according to channel spatial correlation. However, the wireless channel state is time-varying, due to the movement of the mobile station and the barrier of the building, different The channel spatial correlation at the moment changes greatly. If the simple correlation coefficient is used as the judgment basis, the communication mode of the mobile communication system is frequently switched. Therefore, the method is difficult to use in the actual mobile communication system. SUMMARY OF THE INVENTION The object of the present invention is to overcome the shortcomings of the existing multi-antenna transmission mode that it is difficult to continuously perform high-rate, high-quality communication in the case of channel spatial correlation change, and an adaptive multi-antenna communication device and method are proposed to solve the problem. The use of a single multi-antenna system architecture cannot cope with the problems of various complex communication environments. To achieve the above object, the present invention provides an adaptive multi-antenna communication method, which is characterized in that channel classification is performed according to a correlation matrix of an array received signal and a correlation matrix characteristic of a transmitted signal, and different modulations are selected for different types of channels. The method and the space-time processing manner include the following steps: Step 1: Generate measurement information of the channel quality by using the channel estimation of the receiving end, and give the signal-to-noise ratio (hereinafter referred to as SNR) of the link At the same time, a channel matrix R _H reflecting the channel characteristic value is given; the channel matrix contains the channel response relationship between each pair of transmitting and receiving antennas; Step 2: the link SNR obtained by the above step 1 and the channel reflecting the channel characteristic value The matrix performs feedback; Step 3: # Represents the channel matrix R _H reflecting the channel eigenvalues, estimates the condition number Cd of the channel matrix, and compares Cd with the threshold values c0, cl; according to the condition number of the spatial correlation matrix of the channel The range of the interval, and then the channel classification; Step 4: According to the result of the channel classification determined in step 3, select the above The same modulation scheme and space-time processing mode; Step 5: modulation scheme according to the channel classification result and the selected space-time treatment. In the above step 1, the information of the channel matrix R _H is as follows; if the multi-antenna input/output system is configured with the M #-transmit antenna and the N # 妾 receive antenna, the channel impulse response is:

The spatial correlation matrix is

R _H =E[H ^l] H] where ^H represents the conjugate transpose; the spatial correlation matrix of the channel is completed according to the following formula

The clothing shows the Kronecker product, R _RX represents the spatial correlation matrix at the transmitting end, and R _TX represents the spatial correlation matrix of the transmitting end. In the above step 3, the condition number Cd of the channel matrix is estimated, and is completed according to the following formula:

Cd R _H )= RR: R

Wl ^AM _ 其中， ^/7灣表示第 ^η根发射天线与第 m根接收天线之间的信道响应。 同时 将上述信道矩阵重新描述为 "~"'" ¹¹ where ¹¹ represents the p-th order norm of R. The condition number of the 4-word matrix is the ratio of the largest singular value to the smallest singular value of the channel matrix, that is, the condition number of p=2 norm; The classification method is as follows: If the condition number Cd<c0 of the spatial correlation matrix belongs to the uncorrelated fading channel, the channel coefficient is an independent and identically distributed random complex Gaussian random variable, and the correlation between the channel coefficients is weak; for example, the condition number cO of the spatial correlation matrix ≤Cd<cl, which belongs to the semi-correlated fading channel. At this time, there are only a large number of scatterers in the near field at the receiving end or the starting end, and the angular spread at the other end is small or there is a line-of-sight component; for example, the condition number of the spatial correlation matrix Cd ≥ cl, It belongs to the fully correlated fading channel. At this time, the scatterer is located in the far field of the receiving end and the transmitting end, and the multi-input and output channels are fully correlated channels. For the IEEE802.20 MIMO multi-output channel model, the corresponding threshold parameters are c0=10, cl =40. In the above step 4, according to the channel classification result determined in the above step 3, different modulations are selected. The method of the mode and the space-time processing mode is: for the uncorrelated fading channel, the corresponding space-time coding mode is selected as the V-BLAST unit; for the semi-correlated fading channel, only one of the transmission channel correlation matrix and the reception channel correlation matrix is In the case of a unit matrix, it is also necessary to examine the value of the SNR reflecting the channel quality information: For a higher SNR, such as a normalized SNR of 1 to 10, the channel quality increases as the SNR increases, and when SNR>5, the selection is empty. Time grid coding mode; when SNR ≤ 5, select multi-antenna V-BLAST unit mode; for fully correlated fading channel, when the transmit channel correlation matrix and the receive channel correlation matrix are non-unit arrays, select adaptive beamforming mode . An adaptive multi-antenna communication device, comprising: a transmitting end and a receiving end: the transmitting end comprises: a space-time processing mode selecting unit, a V-BLAST unit, a space-time trellis coding unit, an adaptive beam forming unit, a transmitting antenna Array unit; according to the channel state information fed back by the channel estimation of the receiving end, the transmitting data is sent to the V-BLAST unit or the space-time trellis coding unit or the adaptive beam forming unit by adaptive selection after the space-time processing mode selection unit And then channel coding, the modulation is sent by the transmitting antenna array unit; the receiving end comprises: a receiving antenna array unit, a space time receiving unit, a channel state information unit; and transmitting the array information received by the receiving antenna array unit to the space-time receiving unit, After demodulation, the estimated output data is decoded, and the channel state information of the channel estimation of the receiving end is fed back to the transmitting end. The invention improves the limitation of the application environment of the single multi-antenna processing method, expands the application range of the multi-antenna system to the communication environment, and can adapt to different communication environments such as line-of-sight and non-line-of-sight; and the system structure based on software radio Compared with the conventional multi-antenna system, it does not require a large hardware cost, which greatly improves the performance-price ratio of the multi-antenna system. The invention adopts a multi-antenna system implementation method based on spatial feature recognition, and has many advantages compared with the existing single multi-antenna processing method. First, it solves the problem that the single multi-antenna transmission mode has a decrease in the data transmission rate due to the change of the channel space characteristics, and the communication quantity is reduced. Different signal transmission modes are adopted for different channel conditions, so that the spatial correlation characteristics in the channel are obtained. In the case of change, communication can maintain a higher transmission rate and better communication quality, thereby making the multi-antenna system more widely adaptable. Second, adaptive communication methods and modulation based on spatial channel characteristics The code selection method uses the condition number of the spatial correlation matrix of the channel as the switching basis of the transmission mode, and can sensitively reflect the channel change, thereby ensuring the accuracy and timeliness of the transmission mode switching, and the method switching is simple, the calculation is fast, and the method is reduced. The difficulty of hardware implementation is easy to implement. The specific implementation of the present invention will be further described in detail below with reference to the accompanying drawings. The above and other objects, features and advantages of the present invention will become apparent to those skilled in BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a structural view of a device of the method of the present invention; FIG. 2 is a V-BLAST space-time transmission and reception structure; FIG. 3 is a flow chart of the method of the present invention; Detection and channel classification flow chart; Figure 5) is a relationship diagram of the spatial channel correlation matrix condition number corresponding to the number of operations N in the wireless communication environment; Figure 5 (b) is the spatial channel correlation matrix condition number corresponding to the wireless communication environment and Distribution probability relationship diagram. The specific manners of the present invention will be further described in detail below with reference to the accompanying drawings, which can be easily implemented by those skilled in the art. BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a block diagram of the apparatus of the method of the present invention. The space-time processing selecting unit T101 transmits the data to the space-time processing mode selection unit according to the channel state information CSI R102 value, and then adaptively selects the V-BLAST T102, the space-time network code STTC T103, the adaptive beamforming T104, and the like. The space-time processing mode, after channel coding, modulation, etc., is finally sent by the transmitting antenna array unit T105.1-T105.M; at the receiving end, the receiving antenna array unit R103.1-R103.N transmits the received array information. The receiving space processing unit R101 is demodulated, decoded, etc. to obtain estimated output data. And channel information of the channel state information CSI R102 of the receiving end (including signal to noise ratio, The output of the channel correlation matrix, etc.) is fed back to the transmitter. It should be noted that, since the functions of the scrambling/descrambling, interleaving/deinterleaving, synchronization, and the like in the adaptive multi-antenna system are not of interest to the present invention, these units are generally not described in this embodiment. Directly reference these units or their output signals. Figure 2 shows a V-BLAST space-time mode transmit and receive structure diagram. For the V-BLAST method, the transmitting end performs mapping processing using the output of the channel coding unit T201 (Reed Solomon) (hereinafter referred to as RS) code, etc., and the mapping unit T202 implements the conversion of the bit group to the signal constellation point according to the mapping scheme, which may be adopted. Gray-mapped m-quadrature Amplitude Modulation (QAM) (hereafter referred to as m-QAM) (including: Binary Phase Shift Keying (hereinafter referred to as BPSK), Orthogonal Reversal Key (Quarature Phase Shift Keying) (hereinafter referred to as QPS10, 16-QAM, 64-QAM, etc.). Then V-BLAST space-time coding unit T203 is used to decompose a single data stream into multiple sub-data streams by serial-to-parallel conversion. The sub-data streams are respectively subjected to multiple access modulation units T204.1-T204.M, and simultaneously transmitted from multiple transmit antenna arrays T205.1-T205.M. At the receiving end, any receiving antenna R200.1-R205.N The received signals are superpositions of signals transmitted by multiple transmit antenna arrays T205.1-T205.M. Assuming that the channels between the pairs of transmit and receive antennas are independent of each other, these transmitted signals have experienced The same channel fading. The receiver performs frequency offset estimation on the received signal through the channel estimation unit R201.10-R201.N0, and after obtaining the correct sampled data and carrier frequency, the multiple access demodulation unit R201.ll-R201.N1 The received data is parallel-serial converted, and then the result is sent to the V-BLAST decoding unit R202 for detection processing, and then sent to the inverse mapping unit R203 for symbol inverse mapping, and finally the channel is decoded by the channel R204, and the original data is estimated. In addition, according to the characteristics of the correlation matrix of the channel, the working mode of the multi-antenna system is determined. That is, the V-BLAST method is adopted or the space-time trellis coding method is adopted, or the beamforming method is adopted, and detailed steps are given in the subsequent sections. Referring to Figure 3, there is shown a flow chart of the method of the present invention. In an adaptive multi-antenna system, the communication transmission mode corresponding to the spatial features can be arbitrarily adjusted to maximize the possible matching of the current transmission environment. The channel situation is very complicated. At present, using a single multi-antenna communication method makes it in various environments. It has good performance and is difficult to implement. Therefore, by analyzing the characteristics of the communication channel, in the adaptive multi-antenna system, the corresponding suitable communication transmission mode is selected, so that the multi-antenna system has wider adaptability. Is the starting point of the present invention. The method set forth in the present invention can be implemented in the following steps: Step 31, receiving channel estimation. Channel quality measurement information is generated by channel estimation, and is generally given in the form of a total signal to noise ratio SNR of the link. At the same time, a channel matrix reflecting the channel characteristic values is provided, and the channel matrix includes a channel response relationship between each pair of the transmitting and receiving antennas. Step 32, feedback of communication channel state information. The total signal-to-noise ratio SNR of the link with the channel estimation feedback of the receiving end and the channel matrix reflecting the channel characteristic value are obtained, and the modulation mode is selected according to the SNR, and the system space-time processing mode is determined according to the channel matrix. Step 33: The communication channel environment feature implements selection of a modulation mode and a space time processing mode. Step 34: According to the information of step 33, select the modulation mode and the space-time processing mode according to the channel characteristics in real time; for the channel type 1, that is, the condition number Cd<the threshold c0, the transmit channel correlation matrix and the receive channel correlation matrix are both In the case of a unit array, the system can obtain a relatively large diversity order. It is assumed that there are M transmitting antennas and N receiving antennas in the system, and there are a total of available wireless links in the system. If all of these links are independent, then the order space diversity gain can be obtained. For wideband systems, the channel model can be thought of as consisting of L (L is the order of the channel pulses) flat independent channels. At this time, the MIMO system can be regarded as M transmit antennas and NxL root virtual receive antennas, and the diversity gain obtained by the system is MxNxL order. The corresponding space time coding mode is selected as V-BLAST. For channel type 2, that is, the condition number cO<Cd<cl, when only one of the transmit channel correlation matrix and the receive channel correlation matrix is a unit matrix, it is also necessary to simultaneously consider the value of the SNR reflecting the channel quality information: The SNR, such as the normalized SNR is 1 to 10, and the channel quality increases as the SNR increases. Then, when SNR>5, the STTC mode is selected; when SNR≤5, the multi-antenna V-BLAST mode is selected. For channel type 3, that is, the condition number Cd>cl, when the transmit channel correlation matrix and the receive channel correlation matrix are both non-unit arrays, the adaptive beamforming mode is selected. Step 35 is implemented in a multi-antenna V-BLAST space-time manner or a space-time trellis coding method or an adaptive beamforming method. The specific implementation methods of these several methods are well known in the related art and will not be described in detail herein. According to the space-time processing method determined by the above steps, the corresponding modulation and coding are performed. Step 36, end. 4 is a flow chart of wireless communication environment detection and channel identification. The process specifically includes the following steps; Step 41, to give spatial channel spatial correlation matrix ^{Κ Η.} Assuming that the system is configured with a root transmit antenna and a root receive antenna, the channel impulse response is:

Wl ^AM _ where ^/7 bay represents the channel response between the ^nth transmit antenna and the mth receive antenna. At the same time, the above channel matrix is re-described as

H = [/7, ■·· /7„ ... h where ' is the first line of H, ie =\ , ι ··· KM ] In order to calculate the spatial correlation, the following will be rewritten into the following IXMV dimension vector Form, ie

H = [/7, h ₂ ··· h _n ■■■ h _N ] Then, the spatial correlation matrix is R _H =E[H ^H H] where ^H represents the conjugate transpose. In order to analyze the cartridge, it can be considered that the correlation between the receiving antennas does not depend on the transmitting antenna, and vice versa. That is, only adjacent antenna environments produce correlations between array elements and no effect on array elements at the other end of the link. For MIMO channels, between different transmit (receive) antennas The inter-correlation property is represented by ^K Tx and ^K Rx , and the correlation coefficient between different antenna signals can be defined as

'

其中， ^和 ²表示两个天线的信号， 〈'〉表示时间平均， 也就是对某个样本 在一定时间范围内求取平均， 即

Where ^ and ² represent the signals of the two antennas, and 〈'> represents the time average, that is, the average of a certain sample over a certain time range, ie

<'' _m i ' „2 ) = }^~ [ _Γ r _m , (t) · r _m2 (t)dt For MIMO systems, assuming that the fading of the channel is ergodic, then its statistical average can be considered The time averages are equal, so that <'' ^wl , '^;' ² 〉 is to find the cross-correlation of the two vectors. Thus define the spatial correlation coefficient between any two antennas at the transmitting end and between any two antennas at the receiving end. Spatial correlation coefficient, ie

Pll = <U' ₂ ,,"〉 m = 2,---, M is available, assuming that the "line" of the channel matrix #, ^h "' is the mth of H, then, the transmitting end and the receiving end The correlation matrix is

R _M = E[h _m h ] m = l,2,---,M

R _rx =£[/?"/?„] η = \,2,···,ΝΝ

^R H = ^R /ix ® ^R rx where the symbol ® represents the Kronecker product.

1_/；； 其中 ML表示求 R的 p阶范数。 由于矩阵范数的 定义不同， 因而其条件数也不同， 但是由于矩阵范数的等价性， 故在不同范数 下的条件数也是等价的。 在实际估计时， 信道矩阵的条件数就是信道矩阵的最 大奇异值与最小奇异值之比， 即 p=2范数的条件数。 】o Step 42: Estimate the condition number Cd of the channel matrix:

1 _/ ;; where ML represents the p-th order norm of R. Since the definition of the matrix norm is different, the number of conditions is also different, but due to the equivalence of the matrix norm, the condition numbers under different norms are also equivalent. In the actual estimation, the condition number of the channel matrix is the ratio of the largest singular value of the channel matrix to the minimum singular value, that is, the condition number of the p=2 norm.

In order to achieve higher channel capacity, MIMO systems require that each subchannel in the channel matrix be uncorrelated. Therefore, the small-scale fading of the antenna oscillator over a long period of time can be examined by counting the condition numbers of the channel matrix. Step 43: Determine and classify the channel according to the parameters. And discriminating the interval range 44 in which the condition number of the spatial correlation matrix of the channel is; according to the above discriminating result, different channel types 1 44.1, channel type 2 44.2, and channel type 3 44.3 are obtained. Channel type 1 44.1: uncorrelated fading channel, condition number Cd<c0 of spatial correlation matrix, and channel coefficient are independent and identically distributed random complex Gaussian random variables. There are a large number of scatterers in the near field at both ends of the transceiver, and the typical case is the urban microcell channel. Since the angular spread of the transmitting end and the receiving end is large, the correlation between the channel coefficients is weak. Channel type 2 44.2: Semi-correlated fading channel, cO ≤ Cd < cl , at this time there is only a large number of scatterers in the near-end or the near-end near-field and the other end has a small angular spread or a line-of-sight component. This situation may occur in the environment of a city macro cell where the base station antenna is sufficiently high and there are a large number of buildings around the mobile station. Channel type 3 44.3: Fully correlated fading channel, Cd ≥ cl, where the scatterer is located at the far end of the terminating and transmitting ends, and the MIMO channel is the fully correlated channel. For the IEEE802.20 multiple-input multiple-output channel model, the corresponding threshold parameters are c0=10 and cl=40, respectively. The condition number reflects whether the channel matrix is singular. Since the channel is randomly generated, there may be a case where the condition number is large, which may correspond to the deep fading of the channel, resulting in a singularity of the channel matrix. Figure 5) is the relationship between the number of spatial channel correlation matrix conditions corresponding to the wireless communication environment and the number of N runs. It can be seen that when the number of operations is 500, the number of conditions is about 467; when the number of operations is 219, the number of conditions reaches 711, and when the number of operations is 223, the number of conditions reaches 311, and the number of runs is 311. When it is around 275, the number of conditions reaches 1933. It can be seen that except for the case where the number of conditions is small, the other conditions are relatively small. Figure 5 (b) is a spatial channel correlation matrix condition number corresponding to the wireless communication environment and its distribution probability map. By comprehensively analyzing the distribution law of channel condition numbers, it can be seen that about 80% of the condition numbers are less than 10, and about 90% of the condition numbers are less than 40. The probability that the channel response matrix becomes a singular matrix is small. Similarly, the above processing process is also easily applied to the processing of other noisy environments, and the array form is not limited to a uniform linear array, and other forms of arrays such as non-uniform linear arrays, circular arrays, and arc arrays can be easily extended. Of course, the present invention may be embodied in other specific forms and modifications without departing from the spirit and scope of the invention. Modifications are intended to fall within the scope of the claims of the invention.

Claims

An adaptive multi-antenna communication method is characterized in that channel classification is performed according to a correlation matrix of a received signal of an array and a correlation matrix characteristic of a transmitted signal, and different modulation modes and space-time processing modes are respectively selected for different types of channels, and packet 4 The following steps of the tongue -  Step 3: The measured value information of the channel quality generated by the channel estimation of the receiving end is given in the form of the total signal to noise ratio of the link; and the channel matrix RH reflecting the channel characteristic value is given; the channel matrix contains each pair of transmission and weight Channel response relationship between receiving antennas;  Step 2: feedback the total signal to noise ratio of the link obtained in the above step 1 and the channel matrix reflecting the channel characteristic value; Step 3: Estimating the condition number Cd of the channel matrix according to the channel matrix R _H reflecting the channel characteristic value, and comparing Cd with the threshold value cO, cl; according to the interval range in which the condition number of the spatial correlation matrix of the channel is Further determining the channel classification;  Step 4: According to the result of the channel classification determined in step 3, the different modulation modes and space-time processing modes are selected;  Step 5: Implement a modulation mode and a space-time processing mode selected according to the channel classification result. The adaptive multi-antenna communication method according to claim 1, wherein: In the above step 1, the information of the channel matrix R _H is as follows; if the multi-antenna input/output system is configured with M transmitting antennas and N receiving antennas, the channel impulse response is: h , k

 The spatial correlation matrix is R _H = E[H ^H ] wherein H represents a conjugate transpose; The spatial correlation matrix of the channel is completed according to the following formula The symbol ® represents the Kronecker product, RRX represents the spatial correlation matrix at the transmitting end, and RTX represents the spatial correlation matrix of the transmitting end. The adaptive multi-antenna communication method according to claim 2, wherein:  In the above step 3, the condition number Cd of the channel matrix is estimated, and is completed according to the following formula: Cd _p (R _H )

MI _P represents the p-th order norm of R, and the condition number of the channel matrix is the ratio of the maximum singular value to the minimum singular value of the channel matrix, that is, the condition number of the p=2 norm;  The method for judging channel classification is:  For example, the condition number Cd<c0 of the spatial correlation matrix belongs to the uncorrelated fading channel, and the channel coefficient is an independent and identically distributed random complex Gaussian random variable, and the correlation between the channel coefficients is weak;  For example, the condition number cO ≤ Cd < cl of the spatial correlation matrix belongs to the semi-correlated fading channel. At this time, there is only a scatterer in the near-end or the near-end near-field and the other end has a small angular spread or a line-of-sight component exists;  For example, the condition number Cd ≥ cl of the spatial correlation matrix belongs to the fully correlated fading channel. At this time, the scatterer is located at the far end of the receiving end and the originating end, and the multi-input and output channels are fully correlated channels. For the IEEE802.20 multiple input multiple output channel model, The corresponding threshold parameters are c0=10 and cl=40, respectively. The adaptive multi-antenna communication method according to claim 3, wherein:  In the above step 4, according to the channel classification result determined in the above step 3, the method of selecting different modulation modes and space-time processing modes is;  For the uncorrelated fading channel, the corresponding space-time coding mode is selected as a vertical Bell real cell layered space-time code unit; For a semi-correlated fading channel, when only one of the transmit channel correlation matrix and the receive channel correlation matrix is a unit matrix, it is also necessary to simultaneously consider the value of the total signal to noise ratio reflecting the channel quality information: For a higher total signal to noise ratio, If the normalized total signal-to-noise ratio is 1 to 10, the channel quality is always When the signal-to-noise ratio is increased, the space-time grid coding mode is selected when the total signal-to-noise ratio is >5. When the total signal-to-noise ratio is less than or equal to 5, the multi-antenna vertical Bell-dimensional layered space-time code unit is selected. ;  For the fully correlated fading channel, when the transmit channel correlation matrix and the receive channel correlation matrix are both non-unit arrays, the adaptive beamforming mode is selected. An adaptive multi-antenna communication device, comprising: a transmitting end and a receiving end:  Send "" end packet ^^ space-time processing mode selection unit, vertical Bell-real-layer layered space-time code unit, space-time trellis coding unit, adaptive beamforming unit, transmit antenna array unit; channel estimation based on receiver The channel state information fed back, after the transmission data is selected by the space-time processing mode selection unit, is adaptively selected and sent to the vertical Bell Lab layered space-time code unit or the space-time trellis coding unit or the adaptive beam forming unit, and then passes through Channel coding, modulation, sent by the antenna antenna unit;  The receiving end includes: a receiving antenna array unit, a space time receiving unit, and a channel state information unit;  The array information received by the receiving antenna array unit is transmitted to the space-time receiving unit, demodulated, the estimated output data is decoded, and the channel state information of the channel estimation of the receiving end is fed back to the transmitting end.