Adaptive Beamforming And Application In Multiple Signals Recognition

With the increasing complexity of the modern electromagnetic environment, the parameters of multiple signals in space have serious overlap in the time domain and frequency domain. When signals generate overlapping in the time domain, they can't be separated by the the time domain parameters of signals; When signals generate overlapping in the frequency domain, they can't be separated by the the frequency domain parameters of signals. So the extraction of the specific signal from such a complex environment is an urgent problem to be solved. This paper focuses on the research of the adaptive beamforming, and solves the problem of extracting specific signal from multiple signals based on adaptive spatial filtering characteristics of ABF; Then the specific signal is recognized by the method of fingerprint feature extraction.In this paper, ABF is applied to solve the problem of multiple signals recognition.However,In the practical application,the robustness of the ABF algorithm becomes poor,due to the influence of the signal steering vector mismatch or covariance matrix errors. This paper makes a further study on this issue,and analyses the robustness of algorithm under various error, and the algorithm is improved for the lack of existing ABF algorithm.This paper presents two improved diagonal loading algorithm. The improved GLC based on diagonal loading algorithm, which effectively reduce the amount of the computation with the diagonal loading factor. Null broadening and diagonal loading algorithm, the algorithm takes the null broadening and diagonal loading together, and it not only addresses the problem of narrow nulls, but also solve the problem of the worse robustness, as steering vector mismatch existing. In addition, the robust ABF algorithm based on the covariance matrix remodeling is proposed, the algorithm can be used in two-dimensional antenna array,broadening the null and overcoming the situation of interfering signal steering vector mismatch, and the algorithm weight vector calculates process, the desired signal components are not used, so the algorithm also has better robustness for the desired signal steering vector mismatch.Because multiple signals in the complex environment are difficult to be recognized, this paper presents the program based on multiple signals recognition of ABF, the program completes the extraction for the specific signal by the ABF LCMV algorithm based on the covariance of the reconstruction,then the rising edge of the pulse envelope is used to recognize specific signal. Finally, the probability of extracting the specific signal from multiple signals which are overlapping in the time domain and frequency domain by using ABF has been verified by the simulation and the probability of the design program of multiple signals recognition based on ABF has been verified by the experiment on real data.