The Research On Joint Probabilistic Constrained Robust Beamforming And Antenna Selection

Array signal processing is an important branch in the field of signal processing, in a variety of national economy and military applications of mobile communications, electronic warfare, microphone array speech / audio processing, biomedicine, radar, sonar, and seismology have a wide range of applications prospects.Beamforming technology is an important research direction in the field of array signal processing, interference and noise suppression through picket weighting the purpose to enhance the desired signal. Early beamforming algorithm focus on ideal conditions concern algorithm, but in a real environment, there are various beamformer estimation error, leading to a significant decline in performance, even devastating effect on the desired signal, so robust beamforming technology research is great significant.Under realistic conditions, there is inevitably channel state information CSI error(quantization error) or miscalculation, Classic beamforming algorithm can not guarantee the biggest gain in the desired direction, thereby causing the output signal to noise ratio array of dry decreased significantly, severely damaged the communication performance of the system.Therefore, in the actual project which needs to consider the impact of channel errors beamformer. The current margin of error of two CSI model, determine the boundaries of CSI error(quantization error) model is too conservative, the model based on robust beamforming method is difficult to explore the maximum system performance. Compared to determine the boundaries of CSI error constraint, CSI error probability constraint conditions better adapted to the needs of reality.At present, how to take full use of frequency resources is a hot issue in the scientific research; for scarce spectrum resources, taking into account the various cooperative transmission protocols, technology can not only improve the spectrum utilization of resources, but also the effective control of a particular user interference level; at the bottom of cognitive networks, authorized by the primary user of a secondary spectrum users to use their resources, while secondary user is obliged to control the level of interference to the primary user of the methods to improve the spectrum resource utilization. In the CSI error scenarios, it might appear the phenomenon that the obtained optimal weight vector cannot satisfy the interference level of coexistence of primary user, or the requirements of secondary users; In this case, it is difficult to solve the inequality probability constraints. So this paper will give a certainty probability constraints transform method which is based on Bernstein-type. In the implementation process, through the introduction of Bernstein-type inequality handle complex quadratic Gaussian random variables, and thus the probability of converting constraints, inequality constraints will strengthen the probability matrix inequality, and the use of convex and positive definite sparse regularization method for solving relaxation the original problem, and simulation experiments.Aimed at reducing the transmitter antenna to reduce system cost, we propose to introduce a new weight matrix to improve the iterative algorithm dilution of a steady beam forming the background environment, make a joint antenna selection and probabilistic constrained robust beamforming algorithm. Algorithm iterative algorithm prior information by the binary search method to find its antennas. Then narrow the range to find the best beamforming matrix. And determine the optimal beamforming vectors, so that the transmission power is minimized, and in reducing the transmit antenna case, it can protect the user SINR constraints and meet two primary user interference level constraints. Simulation results show that the algorithm to optimize the efficiency of performing antenna selection and weights than exhaustive search antenna selection algorithms more efficient.