Pattern Synthesis Of Antenna Array Using Immune Algorithm

The applications of antenna arrays can be found in many areas such as radar, wireless communications and electronic warfare. As the key technologies in antenna array, pattern synthesis have been widely used in interference suppression,probability of intercept and parameter estimation.The optimization of pattern function is non-linear problem. In resent years, As its global search performance and having no restraint of differentiable and continuities, Genetic algorithm has been widely used in pattern synthesis of antenna array. Immune algorithm which is a new global optimal technology based on intelligent search is very similar to genetic algorithm. Some scholars indicate that immune algorithm has the better performance in function optimization.This thesis carried on a research in applying immune algorithm to pattern synthesis of antenna array. Sidelobe suppression and null steering are researched. The control of low sidelob in ultra-low sidelobe antenna array and thinned array are discussed detailedly. Through the study of immune algorithm and chaos optimal algorithm, Multi-map chaos immune algorithm has been presented. The result which gained by applying Multi-map chaos immune algorithm to the synthesis of antenna array is excellent.Pattern synthesis of linear array and rectangular array using immune algorithm is under the ideal condition in the thesis. Conclusions gained are as follows:Firstly, Immune algorithm is the efficient method of pattern synthesis of antenna array.Immune algorithm is a efficient heuristic intelligent bionic algorithm. It uses dispersed and separate searching method to ensure the diversity of the colony, in order to avoid the earliness which can be produced by single searching mode in genetic algorithm. In the thesis, array spacing,amplitude and phase of excitement in linear array and rectangular array are optimized using immune algorithm. The result confirm with the design beacon which is the efficient control of sidelobe,ultra-low sidelobe and depth of the null steering in specified angle.Secondly, The performance of pattern synthesis using immune algorithm is better than which using genetic algorithm, and which is more evidence when problem required optimization is more complexity.The same problem is simulated by both immune algorithm and genetic algorithm for making the research of pattern synthesis applying immune algorithm more objective. Results show that pattern synthesized using immune algorithm is more excellent than which of genetic algorithm, and when the array amount is larger or the pattern function is more complicated, the superiority is more obvious.Thirdly, The pattern function of antenna array is transformed to discrete form, and which of rectangular thinned array is turned to the form of Kronecker product.Through sampling to pattern function of antenna array, a discrete function is gain which is adapted to computer operating and immune algorithm. Circulatory operation is replaced by matrixes' product, which advances efficiency of computing. Pattern function of rectangular thinned array has been transformed to the form of Kronecker product by two matrixes, which not only fits the operation of Matlab ,but also enhances the searching efficiency.Fourthly, Combining respective advantages of chaos optimal algorithm and immune algorithm, Multi-map chaos immune algorithm is presented.There is a farther optimization to the global searching performance of immune algorithm using the ergodicity of chaos sequence and the randomicity with "rules" of chaos disturbance. Two chaos sequences with different rules are applied to control the generation of the antibodies in different periods, which ensure the diversity of antibodies. As a result the searching area is adequacy, earliness phenomenon is conquered. A hybrid mutation method is presented which combining Chaos disturbance and the own mutation method of immune algorithm. As a result the searching efficiency is advanced and the speed of convergence is improved. The synthesizing of antenna array using Multi-map chaos immune algorithm have gain a fine effect. The result validate that the presented algorithm has a strong searching ability and can get the global optimal value while avoiding local convergence.