Research On Advanced Low-Complexity Adaptive Beamforming Technique

Nowadays, people have more and more requirements on the data rate and transmission relia-bility in wireless communication systems, then how to cope with these challenges and increase the system capacity without increasing the bandwidth and transmit power have drawn the most atten-tion in each generation of wireless communication. With the highly speedy development of fourth generation (4G) wireless air interfaces technology, multiple or large scale antenna technique and Orthogonal Frequency Division Multiplexing (OFDM) technique have becoming more and more popular. This dissertation focuses on the research of beamforming, which is the branch of multiple antenna and can be seen as a spatial filter. It can provide flexible beampattern, good processing gain and excellent interference suppression. Moreover, as being a multiple carriers technology, OFDM supports high transmission data rate with high spectrum efficiency and strong immunity to interference, the joint system of these two techniques is worth studying.Firstly, the fundamentals of beamforming and the most simple and common used ULA an-tenna model have been introduced. And then we present the solution of beamforming weighted vectors under different criterion (MMSE, MSINR, CMV, CCM) different process structure (DF-B, GSC) and different iteration algorithm (LMS, RLS). Specially, we give the detailed deviation and analysis under blind beamforming criterion which names CMV and CCM. Simulation results demonstrate that CCM-RLS-GSC algorithm can achieve the best SINR and MSE performance, compared with other beamforming algorithms.And then, we propose a novel low-complexity time-averaged variable forgetting factor (TAVF-F) mechanism to enhance the performance of recursive least squares (RLS) algorithms for adaptive blind beamforming. The beamformer is designed according to the CCM criterion, and the proposed algorithm operates in the GSC structure for implementation. The proposed TAVFF mechanism employs a new component updated by the time average of the CM cost function, to adjust the for-getting factor. A complexity comparison is provided to show its advantages over existing methods. And a detailed study of its operating properties is carried out, including a convexity analysis and a mean squared error (MSE) analysis of its steady-state behavior. The results of numerical exper-iments demonstrate that the proposed VFF mechanism achieves a superior learning and tracking performance compared to other VFF mechanisms in both stationary and nonstationary environ-ment.Furthermore, we propose an adaptive reduced-rank set-membership filtering (SMF) algorithm using the CCM criterion for beamforming. We develop a least mean square (LMS) type algorithm based on the concept of SMF techniques for adaptive beamforming. The filter weights are updated only if the bounded constraint cannot be satisfied. We also propose a scheme of time-varying bound and incorporate parameter dependence to characterize the environment for improving the tracking performance. A detailed analysis of the proposed algorithm in terms of computational complexity and stability is carried out. Simulation results verify the analytical results and show that the proposed adaptive SMF reduced-rank beamforming algorithms with a dynamic bound can achieve superior performance to previously reported methods at a reduced update rate.Finally, we propose two complexity reduction techniques for blind adaptive beamforming in OFDM antenna array systems. For each sub-carrier, a beamformer with generalized sidelobe canceller (GSC) structure is designed according to the constrained constant modulus (CCM) crite-rion with weight vectors adaptation based on the recursive least squares (RLS) method. The two techniques that we propose for complexity reduction rely on frequency domain interpolation and spatial domain clustering, respectively. The former exploits the coherence bandwidth of the radio channel, so that only the weight vectors at selected frequencies need to be adapted, while the re-maining weight vectors are obtained via interpolation. The latter relies on the partitioning of the receiving array into sub-arrays of smaller size. In addition, a combination of these two techniques is proposed to further reduce the complexity. A complexity analysis is presented to quantify the computational savings offered by the proposed techniques. Finally, simulation results are provided to illustrate that these savings can be obtained at the price of only a minor, acceptable loss in perfor-mance when compared to the direct application of the conventional CCM-RLS-GSC beamforming algorithm to the wideband OFDM antenna array systems.