The Research On Blind Identification Based On Singular Value Iteration For Space Frequency Block Coding

Multiple-Input Multiple-Output and Orthogonal Frequency Division Multiplexing(MIMO-OFDM) is a core technology of the next generation in wireless communication,of which Space-Frequency Block Coding(SFBC) is an important part.It is urgent to study blind identification of SFBC in countermeasure field,which is valuable for reasearch and application.Currently, the blind identification methods are mainly suitable for SpaceTime Block Coding(STBC), which can be divided into maximum likelihood algorithm and time-delay correlation algorithm. Howerver, the maximum likelihood algorithm could not identify STBC with the same number of transmitted time slot and transmitted symbol; Time-delay correlation algorithm has high computation in practice, therefore, it is difficult to be applied in real-time detection.However, the blind identification of SFBC is a new research field and there are no effective methods. At the same time, due to the difference between diversity,blind identification algorithm of STBC can’t be directly applied to the blind identification of SFBC in the scenarios with small sampling numbers and low SNR, which makes it unbale to meet the practical engineering needs.This paper is written to improve the existing blind identification algorithms of SFBC for these problems and a new blind identification algorithm of SFBC based on singular value iteration is proposed.The algorithm achieves the blind identification of SFBC in scenarios with small sampling numbers and low SNR.In this paper, We first introduced the MIMO-OFDM system model, the SFBC technique and two kinds of existing blind identification algorithms of STBC:maximum likelihood and delay-related algorithms with simulation experiments and the performance analysis in terms of simulation results.Then, we showed the model of blind recognition system and transformed the problem of blind identification to the problem of recognizing the number of symbols.We presented a blind identification algorithm of SFBC based on singular value iteration. Througth effective iteration for the singular values of received signal correlation matrix, the difference between signal and noise characteristic feature values are increased, and the systamic performance is also enhanced with small samples and low SNR.The performance of the algorithm is simulated in MATLAB to verify itscorrectness and efficiency. Finally, we analyzed the influence of the residual frequency offset on the blind identification algorithm of SFBC and proposed a blind identification method of SFBC based on residual frequency offset compensation according to analysis. By using searching compensation, we estimated the remaining frequency offset. After compensating for observed signal, the SFBC is identified. The algorithm is simulated in MATLAB and compared with algorithm without frequency offset compensation, verifying its effectiveness and necessity.