Research On Parallel Algorithm For Optimization Of Antenna Arrays In Aperture Synthesis Radiometers

Aperture synthesis radiometer(ASR)is a microwave radiation receiver based on interferometry technology.It forms a large virtual aperture by an array of antennas in a certain structure configuration.Compared to the traditional real aperture radiometer,ASR can reduces the size and weight,instantaneously image without scanning,and improve the spatial resolution of detection.As a crucial technique for ASR,array configuration design determines the final image quality to a large extent.Heuristic algorithms searching for array configuration optimization face the contradiction between accuracy and computational complexity.As the array size is large,the cost of computation time could be extremely high.Therefore,the parallel computing technology is introduced into the array configuration optimization in this thesis.The main contents are as follows:As the high cost of time consumption for searching low redundancy linear array(LRLA),this thesis proposes an effective collaborative parallel strategy based on the classical particle swarm optimization(PSO)algorithm for searching LRLA.And with the use of the LRLA patterns' information,the parallel particle swarm algorithm is applied to search large LRLA.The experimental results show the feasibility of the parallel algorithm by using different number of processor cores.Finding the array configuration in optimal sensitivity is also a very important issue in array configuration optimization,and it always faces the high cost of time consumption as well.Parallel simulated annealing(PSA)algorithm is proposed in this thesis to search the optimal array configuration,and puts forward to the adaptive parallel simulated annealing algorithm.That is using the current calculation result to adjust the cooling rate of the algorithm by the chance of each process communication.The experimental results show the efficiency of the algorithm for searching the array configuration in optimal sensitivity.