Research On Distributed Array Optimization Method For Isomorphic/Isomerism Subarrays

With the more complex electromagnetic circumstance and the increasing demand in contemporary electronic warfare,as well as the continuous updating of scientific research conditions and equipment,various new system radars such as distributed radar are constantly emerging.Distributed radar is composed of multiple small subarrays,which are arranged separately according to local conditions,and the subarrays cooperate with each other.It has outstanding advantages in improving radar's anti-reconnaissance,anti-stealth,antidestruction,anti-jamming ability and reducing the complexity of radar signal processing.However,for the distributed array with uniform distribution of subarrays,the subarray spacing is often greater than half wavelength,and there will be many high sidelobes in the pattern,resulting in fuzzy angle and unable to accurately locate the target position.Therefore,how to reduce the high sidelobe of distributed array has become a difficult problem in engineering application.Focus on the problem of high sidelobe in distributed array,this thesis studies the optimization of distributed array structure,in order to reduce the peak sidelobe ratio of distributed array,and seek an optimal distributed array layout.The main contents are as follows:(1)In this thesis,one-dimensional and two-dimensional array models of distributed array are established,and the expressions of distributed array pattern of isomorphic subarray and isomerism subarray are derived respectively.The two-dimensional distributed array patterns with uniform distribution and uneven distribution are calculated by the simulation experiments.By comparing the patterns of these two distributed arrays,the peak side lobe ratio of the distributed array with uneven distributed subarrays is lower,so the high sidelobe of distributed array can be reduced by optimizing the subarray spacing.(2)Distributed array optimization problem is a nonlinear problem with multi variables and multi constraints.Intelligent optimization algorithm is simple in principle and easy to be programmed,which has outstanding advantages in solving such problems.Therefore,this thesis also introduces the principle and process of genetic algorithm(GA),particle swarm optimization(PSO)and quantum particle swarm optimization(QPSO).However,these algorithms have some problems,such as slow iteration speed and easy to fall into local optimization.In order to solve these problems,a distributed array optimization method based on genetic algorithm-quantum particle swarm optimization(GA-QPSO)is proposed.In this algorithm,the method of calculating the median optimal location in QPSO algorithm is introduced into GA to make the mutation move towards the optimal direction,and keep the diversity of the population by updating the individual population in time,so as to speed up the iterative speed of the algorithm and avoid falling into the local optimization.(3)In this thesis,GA-QPSO algorithm is applied to the optimization of distributed arrays with isomorphic and isomerism dense subarray units.The simulation experiments and performance analyses show that this algorithm can effectively reduce the peak side lobe ratio of distributed arrays with dense subarray units,and its performance is better than genetic algorithm and other algorithms.The optimization of distributed array also can be divided into two levels: element level optimization and subarray level optimization.In this thesis,QPSO algorithm is used to optimize the subarray elements to reduce it's peak side lobe ratio,and then GA-QPSO algorithm is used to optimize the subarray spacing to effectively reduce the peak side lobe ratio of the overall pattern,which proves that the algorithm is also effective for the optimization of distributed array with sparse subarray elements.Finally,the optimization results of several different distributed arrays are compared and analyzed.