CN102158313A - Soft-input soft-out (SISO) minimum mean squared error (MMSE) iteration receiving method based on eigenvalue decomposition - Google Patents

Abstract

The soft input and soft output minimum mean square error iterative receiving method based on eigenvalue decomposition is characterized in that, firstly, the equivalent channel matrix and the equivalent transmission correlation matrix are obtained by using channel estimation and precoding codebook information, and the equivalent transmission correlation matrix is calculated. Carry out eigenvalue decomposition of the matrix to obtain eigenvalues and eigenvectors; and then input the equivalent channel matrix, eigenvalues, eigenvectors, and received signals into a soft-input soft-output minimum mean square error SISO-MMSE detector, and the detector and soft-input The soft output decoder uses the soft information output by each other as prior information to iteratively work, and after reaching a predefined number of iterations, the decoder finally outputs a bit decision. In this method, the matrix inversion operation in each SISO-MMSE iteration process is converted into a division operation by introducing eigenvalue decomposition, thereby effectively reducing the complexity of system implementation.

Description

Soft Input and Soft Output Minimum Mean Square Error Iterative Receiver Method Based on Eigenvalue Decomposition

technical field

The present invention relates to a broadband mobile communication system that achieves high transmission rate through MIMO-OFDM technology, in particular to a wireless signal processing method for a wireless communication receiving end.

Background technique

The combination of multiple antenna (Multiple Input Multiple Output, MIMO) technology and Orthogonal Frequency Division Multiplexing (OFDM) technology can effectively improve system throughput and transmission efficiency, and meet the requirements of future mobile communication systems on system capacity. , Spectrum utilization, data transmission rate and many other requirements. MIMO technology can double the system capacity and spectrum utilization without increasing the bandwidth. OFDM technology converts wideband channels into several parallel narrowband channels, which can effectively combat multipath fading. In the Long Term Evolution (LTE) standard formulated by the Third Generation Partnership (3GPP), the MIMO-OFDM technology is also adopted as the downlink transmission scheme.

In an actual MIMO-OFDM system, the received signal is affected by channel selective fading, channel noise and inter-antenna interference caused by multiple antennas. Error control coding techniques are usually used to combat channel fading and noise, and detection techniques are used to eliminate inter-antenna interference. The iterative reception technology of joint decoding and detection can greatly improve the performance of the receiver by exchanging soft information between the detector and the decoder and gradually approaching the optimal solution through repeated iterations. In the traditional soft input soft output minimum mean square error (Soft Input Soft Output Minimum Mean Square Error, SISO-MMSE) iterative receiving algorithm, since the SISO-MMSE detection process involves complex matrix inversion operations, and matrix inversion operations The number of times increases with the increase of the number of iterations, which has high computational complexity, which limits the wide application of the algorithm. For this reason, the present invention proposes a kind of soft input and soft output minimum mean square error iterative reception algorithm based on eigenvalue decomposition. The inverse operation effectively reduces the computational complexity of the system.

Contents of the invention

Technical problem: the present invention proposes a soft input and soft output minimum mean square error iterative receiving method based on eigenvalue decomposition, and avoids matrix inversion operation in each SISO-MMSE iteration process by introducing eigenvalue decomposition decomposition. While ensuring consistent performance with the traditional SISO-MMSE iterative reception, the calculation complexity is effectively reduced.

Technical solution: The soft input and soft output minimum mean square error iterative receiving method based on eigenvalue decomposition of the present invention includes the following steps: the first step is to obtain the equivalent channel matrix and the equivalent transmission correlation by using channel estimation and precoding codebook information and performing eigenvalue decomposition on the equivalent transmission correlation matrix to obtain eigenvalues and eigenvectors; the second step is to translate the channel estimation, eigenvalues, eigenvectors, frequency domain received signals and soft input and soft output SISO The soft information output by the coder is input into the soft input and soft output minimum mean square error SISO-MMSE detector, and the estimated value and variance of the transmitted signal are obtained through MMSE equalization, and the estimated value and variance of the transmitted signal are input into the soft demodulator to Obtain bit likelihood ratio information; In the third step, the deinterleaver arranges the bit likelihood ratio of the described SISO-MMSE detector output into the order of the decoder, and the interleaver will softly input and softly output the soft soft output of the SISO decoder. The information is rearranged according to the requirements of the SISO-MMSE detector; in the fourth step, the soft-input and soft-output decoder uses the bit likelihood ratio of the SISO-MMSE detection output as prior information, and obtains a new bit likelihood ratio through decoding, It is fed back to the SISO-MMSE detector for reconstructing the mean and variance in the non-final iteration, and the SISO-MMSE detector and the SISO decoder work iteratively using the soft information output by each other as prior information. In the last iteration, The output bit decision is made by the soft-input and soft-output decoder.

The eigenvalue decomposition includes Jacobi decomposition, a method of obtaining eigenvalues and eigenvectors through numerical calculation of orthogonal triangular QR decomposition.

The matrix inversion operation of the SISO-MMSE detector in each iteration process is converted into a division operation by performing eigenvalue decomposition on the equivalent transmission correlation matrix and saving the eigenvector and eigenvalue.

The soft-input soft-output SISO decoder uses Turbo decoder or LDPC decoding.

Beneficial effects: the present invention proposes a soft input and soft output minimum mean square error iterative receiving algorithm based on eigenvalue decomposition, which avoids matrix inversion operation in each SISO-MMSE iteration process by introducing eigenvalue decomposition decomposition, thereby effectively reducing System implementation complexity.

Description of drawings

Fig. 1 is a block diagram of a specific implementation device of the iterative receiving method for minimum mean square error of soft input and soft output based on eigenvalue decomposition provided by the present invention.

FIG. 2 is an algorithm flow of the soft input and soft output minimum mean square error iterative receiving method based on eigenvalue decomposition provided by the present invention.

Detailed ways

The soft input and soft output minimum mean square error iterative receiving method based on eigenvalue decomposition involves a multiple input multiple output (Multiple Input Multiple Output, MIMO) technology and Orthogonal Frequency Division Multiplexing (OFDM) technology To achieve a high-speed broadband mobile communication system, the soft-input and soft-output minimum mean square error iterative receiving method based on eigenvalue decomposition: The first step is to use channel estimation and precoding codebook information to obtain the equivalent channel matrix and Equivalent transmission correlation matrix, and performing eigenvalue decomposition (Eigen Value Decomposition, EVD) on the equivalent transmission correlation matrix to obtain eigenvalues and eigenvectors; in the second step, the channel estimation, eigenvalues, eigenvectors, Frequency domain received signal and Soft Input Soft Output (Soft Input Soft Output, SISO) decoder output soft information input Soft Input Soft Output Minimum Mean Square Error (Soft Input Soft Output Minimum Mean Square Error, SISO-MMSE) detector, The estimated value and variance of the transmitted signal are obtained through MMSE equalization, and the estimated value and variance of the transmitted signal are input into the soft demodulator to obtain bit likelihood ratio information; the third step, the deinterleaver detects the SISO-MMSE The bit likelihood ratio output by the detector is arranged into the order of the decoder, and the interleaver rearranges the soft information output by the soft-input and soft-output decoder according to the requirements of the SISO-MMSE detector; the fourth step is soft-input and soft-output decoding The detector uses the bit likelihood ratio of the SISO-MMSE detection output as a priori information, and obtains a new bit likelihood ratio through decoding, and feeds it back to the SISO-MMSE detector for reconstructing the mean and variance in the non-last iteration. The SISO-MMSE detector and the SISO decoder use the soft information output by each other as prior information to iteratively work. In the last iteration, the soft input and soft output decoder outputs bit decisions.

The soft input and soft output minimum mean square error iterative receiving method based on eigenvalue decomposition is characterized in that the eigenvalue decomposition includes Jacobi (Jacobi) decomposition, orthogonal triangle (QR) decomposition and other different eigenvalues obtained through numerical calculation and eigenvector methods.

The soft input and soft output minimum mean square error iterative receiving method based on eigenvalue decomposition is characterized in that by performing eigenvalue decomposition on the equivalent transmission correlation matrix and saving eigenvectors and eigenvalues, the SISO-MMSE detector The matrix inversion operation during each iteration is converted to a division operation.

The soft-input soft-output minimum mean square error iterative receiving method based on eigenvalue decomposition is characterized in that the soft-input soft-output (SISO) decoder can use any software including Turbo decoder, LDPC decoder, etc. Input soft output decoder.

The specific implementation device of the soft input and soft output minimum mean square error iterative receiving method based on eigenvalue decomposition of the present invention is shown in Figure 1, which is divided into eigenvalue decomposition, SISO-MMSE detector, interleaving and deinterleaver, SISO translation The four parts of the coder are specifically described as follows: 1. Eigenvalue decomposition Decomposition First, calculate the equivalent transmission correlation matrix according to the channel estimation and precoding matrix information, and perform eigenvalue decomposition and decomposition operations on the equivalent correlation matrix to obtain the eigenvalues and feature vectors.

2. The SISO-MMSE detector uses the eigenvalues and eigenvectors output by eigenvalue decomposition, the channel estimation, the received signal and the soft information output by the SISO decoder as prior information to calculate the bit likelihood ratio. In the calculation process, in order to avoid matrix inversion operation by using eigenvalue decomposition, an assumption is made that the reconstruction variances of each layer are equal, which hardly affects the performance of the receiver.

3. The deinterleaver arranges the bit likelihood ratio output by the SISO-MMSE detector into the order of the decoder, and the interleaver rearranges the soft information output by decoding according to the requirements of the SISO-MMSE detector.

4. The SISO decoder uses the bit likelihood ratio of the SISO-MMSE detection output as a priori information, and obtains a new bit likelihood ratio through decoding, and feeds it back to the SISO-MMSE detector to reconstruct the mean value in the non-last iteration and variance, used to output bit decisions in the last iteration.

In order to enable those skilled in the art to better understand the solution of the present invention, the soft-input and soft-output iterative receiving method based on eigenvalue decomposition provided by the embodiment of the present invention is summarized as a flow chart of the method shown in FIG. 2 .

In the embodiments provided in the present application, it should be understood that the disclosed methods can be implemented in other ways without exceeding the spirit and scope of the present application. The present embodiment is only an exemplary example and should not be taken as a limitation, and the specific content given should not limit the purpose of the present application. For example, several units or components may be combined or may be integrated into another system, or some features may be omitted, or not implemented.

The above is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Anyone skilled in the art can easily think of changes or substitutions within the technical scope disclosed in the present invention. Should be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be determined by the protection scope of the claims.

Claims (3)

1. the soft inputting and soft based on characteristic value decomposition is exported the least mean-square error iteration receiving method, it is characterized in that: the first step, utilize channel estimating and precoding codebook information to try to achieve equivalent channel matrix and equivalence sends relevant battle array, and described equivalence is sent relevant battle array carry out characteristic value decomposition and obtain characteristic value and characteristic vector; Second step, soft information input soft inputting and soft output least mean-square error SISO-MMSE detector with described channel estimating, characteristic value, characteristic vector, frequency domain received signal and the output of soft inputting and soft output SISO decoder, by balanced estimated value and the variance that obtains to send signal of MMSE, the estimated value and the variance of described transmission signal are imported soft demodulator to obtain bit likelihood ratio information; In the 3rd step, the bit likelihood ratio that deinterleaver is exported described SISO-MMSE detector is arranged in the order of decoder, and the soft information that interleaver is exported the output of SISO decoder with soft inputting and soft is arranged according to the requirement of SISO-MMSE detector again; The 4th step, the soft input soft output decode device utilizes SISO-MMSE to detect the bit likelihood ratio of output as prior information, obtain new bit likelihood ratio by decoding, in non-last iteration, it is fed back to the SISO-MMSE detector and be used to rebuild average and variance, the soft information that described SISO-MMSE detector and the utilization of SISO decoder are exported each other is as the work of prior information iteration, in the last iteration, by soft input soft output decode device output bit decision.

2. the soft inputting and soft output least mean-square error iteration receiving method based on characteristic value decomposition according to claim 1, it is characterized in that characteristic value decomposition comprises that Jacobi Jacobi decomposes, the method for passing through numerical computations acquisition characteristic value and characteristic vector that ORTHOGONAL TRIANGULAR QR decomposes.

3. the soft inputting and soft output least mean-square error iteration receiving method based on characteristic value decomposition according to claim 1 is characterized in that described soft inputting and soft output SISO decoder uses Turbo decoder or LDPC decoding.

Images