Calibration Algorithm For Sensor Position Of Array Antenna Based On Genetic Simulated Annealing Algorithm

Spectral spectrum estimation technology is a new array signal processing technology, and in many areas it has important theoretical significance and practical value. With multiple signal classification algorithm represented, the spectral estimation algorithm has very high resolution and estimation accuracy, but it is based on the accurate array which is known as the prerequisite to achieve precision. However, in the actual system error is prevalent. Sensor position error is caused by the hardware design, production and system disturbance, which deteriorate high-resolution spectral estimation performance of the algorithm. Limited to the current level of technology, to overcome the model error just from the production of hardware design is very difficult, and sensor position error has become urgent for high-resolution spectral estimation technique to solve a problem.In this paper, calibration algorithms for sensor position error of array antenna are studied in depth. A calibration arithmetic is advanced based on multiple signal classification (MUSIC) arithmetic without an exact azimuth. Genetic algorithm which is widely used is analyzed as a randomized search method. It is prone to premature convergence, and less capable of local optimization. Based on in-depth study of simulated annealing algorithm, a genetic-simulated annealing algorithm is advanced combining genetic algorithm with simulated annealing algorithm. Put the simulated annealing algorithm into genetic algorithm to achieve better performance. The introduction of elitist strategy ensures that the algorithm can remember the optimal individual. Finally, genetic-simulated annealing algorithm is applied to sensor position correction of antenna array. The experimental results show that the algorithm can effectively correct sensor position error.