Research And Application Of Smart Antenna Multi-target Fast Angle Estimation Technology

Smart antenna technique with simultaneous multi-beamfomming and nulling generation can effectively improve the channel capacity of the communication system and improve the system's anti-interference capability in space domain.The fast angle estimation of interference and target is the fundamental importance of the smart antenna technique,which has very important research significance and clear application background..This paper focuses on the research of multi-target fast 2D angle estimation algorithm for the to-ground communication smart antenna system.Firstly,three algorithms for arbitrary array of 2D angle estimation are reviewed and perfected,which are 2D MUSIC algorithm,ESPRIT based fourth-order cumulant and space-time DOA matrix method.The specific derivation process of each algorithm is given,and perfonnance of the three algorithms are analyzed through the simulation experiments and comparations.Then,an improved algorithm,combining the space-time DOA matrix method and correlation method,is proposed to deal with the problem of the cyclical ambiguity under the circumstance of element spacing larger than half wavelength.The proposed algorithm obtains the unambiguous array phase by performing the phase unwrapping of each estimated steering vector,the least square solution is calculated by combining all elements' phase,which improves the algorithm's robustness and accuracy.The improved algorithm directly uses the estimation of steering vector to get the closed form solution of 2-D DOA,without parameter alignment and multi dimension search,the computation load of the algorithm is small and suitable for practical application.Finally,realization and real-time testing of the above mentioned improved algorithm and MUSIC algorithm are implemented in the C6678 DSP processor under the platform of a smart antenna system with 19-element equilateral triangular grid anoay.In compact field microwave anechoic chamber environment,the interference and target estimation performance of the algorithms are tested and verified.According to the problems encountered in the actual test,the algorithm implementation process of the angle estimation module in system is improved.The experimental results verify the effectiveness and practicability of the algorithms proposed in this paper.