Pattern Synthesis Of Antenna Array Using Intelligent Optimization Algorithm

With improvement of quantities of wireless users and increasing information exchangedata transmission rate, higher requirements are put forward to effective use in limitedresource of wireless system. But deeper researches on corresponding technologies which isnot good enough to improve system performance index such as information codingtechnology and frequency are need. So corresponding system airspace development isgetting hotter in corresponding field, and intelligent antenna has become maincorresponding research objects. Array pattern synthesis of antenna arrays is key technologyof intelligent antenna. Array pattern synthesis of antenna arrays has nonlinear,discontinuous, nondifferentiable characteristics. Especially when there are large quantitiesantenna array elements or synthesis shape is quite complicated, traditional synthesis methodis far away from engineering requirement, thus quick, effective and stable intelligentoptimization algorithm synthesis has been the hotspot in the field.Based on typical intelligent optimization algorithms, such as cultural algorithm,immune algorithm and leapfrog algorithm, this paper improved the algorithm and applied itto array pattern synthesis of antenna arrays. Main research content and innovation points areas follows:Firstly, cultural algorithm is introduced for easy to fall into local optimum problem ofbasic particle swarm algorithm and artificial bee colony algorithm, and culture and culturalcolony algorithm particle swarm optimization algorithm are put forward, and are use foruniform linear array of minimum sidelobe level optimization.Cultural particle swarm algorithm can accumulate and store the experience to solve theproblem and use deeply to guide the process to search optimization of population particle.Cultural algorithm belief space used S,N structure， accept function in a certainpercentage of the choice of the current population spatial local optimal value of belief spaceupdating. Though PSO algorithm combined with CA standard and form knowledge, createdthe algorithm influence function, improve the parent evolutionary properties. And newstructure influence function update cultural particle population space of individualknowledge and statement of form knowledge. Adjust the position change and direction ofthe particle of step, according to form influence function form knowledge and populationparticle speed. Culture bee colony algorithm combined basic bee colony algorithmfood-searching swarm intelligence characteristics and culture algorithm experience knowledge guidance strategy. Culture searching bee judgment food source informationthrough roulette wheel selection form and improve the convergence of the algorithm.Culture bee scout random search make up that cultural colony algorithm easily falling intolocal optimal deficiency to a certain extent. Using the basic swarm evolutionary mechanismand cultural algorithm of situational knowledge and normative knowledge togetherconstruct two different influence functions, and realize the evolution process of theguidance. Applying it to antenna array synthesis of antenna arrays simulation experience,this algorithm has better optimization and convergence performance than basic particleswarm optimization and artificial bee colony algorithm.Secondly, immune algorithm is a higher effective heuristic search of intelligentoptimization algorithm, which through separate spatial search to ensure that the populationdiversity overcome premature phenomenon of the genetic algorithm single search to acertain extent. The use of immune algorithm for sparse array synthetic, rectangular sparsearray pattern function is derived into only two variables of the Kronecker product form, andthen, using matrix calculation instead of iteration, especially suitable for Matlab computing,that algorithm efficiency is improved and the search efficiency, and is validated by thesimulation. Based on combination of importance sampling strategy of cross entropyalgorithm and immune algorithm obtain cross entropy of immune algorithm, usingpolymeric affinity degree concept expressed as antibody comprehensive evaluationparameters. It can not only make the higher affinity individual coming into next generation,also can be used to control the concentration of antibodies, formed the population antibodyto maintain the diversity of good mechanism. For antibodies to replace the use of DNAbinary encoding, it is easier for handling complex optimization problems.Finally, according to the basic leapfrog algorithm is premature, the problem of falling intolocal optimum easily, the chaotic mutation strategy introduces the leapfrog algorithm getthe conclusion based on chaotic mutation strategy of leapfrog algorithm.Hénon and logistic mapping mode got to chaotic sequence that generated initial frogpopulation antibodies and frog antibody complement as eliminated. It could ensure frogpopulations of antibody diversity and expand the search space. Logistic chaos mapping andGauss variation of mixed mutation strategy adjusted the accelerated step, accelerated theconvergence speed of the algorithm, improved the frog antibody population diversity andguaranteed finding out the global optimal solution of maximum probability. After applyingthe algorithm to synthesis of antenna arrays, the simulation result shows a goodoptimization performance.