Study On Constant Modulus-based Blind Beamforming Algorithm

Blind beamforming, does not need reference signals but utilizes the character of users’ signal to achieve beamforming, is one of the effective methods to overcome co-channel interferences and enforce multi-user beamforming. It is widely used in radar, sonar and wireless communication and turns into a hot research area in recent years. The constant modulus-based blind beamforming algorithm, with the high speed of convergence, and the capability in reducing the effect on constant modulus signals caused by interferences, has gradually become a kind of significant methods to realize the optimum reception of multi-user signals. Therefore, researching on constant modulus-based blind beamforming algorithm has the very important theoretic value and practical significance.Constant modulus-based blind beamforming algorithm is mainly discussed in this paper. Firstly, the background and significance of blind beamforming are stated, the current situation and the developing trend of blind beamforming are concluded. Then, four typical adaptive beamforming criterions are compared and analyzed, and several blind beamforming algorithm are introduced. Finally, the classical blind beamforming algorithms based on constant modulus are deeply researched, especially the least square constant modulus algorithm (LSCMA).Due to the conjugate symmetric structure of the weight vector, the least square constant modulus algorithm with a conjugate symmetric constraint possesses fast speed of convergence and high output signal-to-interference and noise ratio (SINR). However, it can not exactly obtain the desired signal and its convergent speed largely reduces in low signal-to-noise ratio (SNR). To conquer the aforementioned problems, an improved least square constant modulus algorithm is proposed, which increase the constraint of the steering vector of the desired signal to keep the distortionless response of the desired signal. Simulation results show that the proposed algorithm not only has high output signal-to-interference and noise ratio, but also can exactly acquire the desired signal in low SNR. Further more, it can quickly converge to the desired signal.The array beampattern sidelobe of the traditional LSCMA is so high that degrades the performance of its output SINR. Aiming at this problem, an LSCMA with mixed norm constraint is presented in this paper. Making use of the characteristic of sparse distribution for array beampattern of the traditional LSCMA, mainlobe and sidelobe are constrained by l∞norm and l1norm, respectively. This operation can ensure that big array gain concentrates on mainlobe and small array gain focuses on sidelobe, which can achieve the purpose of reducing sidelobe. Simulation results show that compared with traditional LSCMA, the LSCMA with mixed norm constraint possesses lower sidelobe level, faster convergent speed and higher output SINR in low SNR.