Novel Multi-strategy Optimization Techniques For Sparse Array And Subarray Synthesis

Antenna arrays have the advantages of high gain,narrow beam(high resolution),shaped beam and beam(orientation)steering abilities,which enables a huge number of applications in various areas,such as radar,communication,remote sensing,radio astronomy,autopilot,biomedical imaging.The degree of freedom of a large number of elements makes the antenna array have high flexibility,however,it brings great challenge to array synthesis technology.Furthermore,elements are usually integrated with active devices and phase shifters,which causes a considerable fabricating cost and systematic complexity As a consequence,there is automatically the expectation of finding a suitable trade-off between array performance and cost/complexity.This dissertation starts with the above issues,and some key techniques of the array synthesis theme are studied,including the joint optimization of array excitation and layout,the sparse array synthesis and the subarray synthesis.Progress has been made in the following aspects.1.The joint optimization technique of array excitation and layout based on hybrid optimization methodologies.Aiming at the disadvantage that the global optimization algorithm is difficult to deal with the constraints,a genetic algorithm adaptive to the linear constraints is proposed in this dissertation,which is called as linear constraint genetic algorithm(LCGA).Subsequently,LCGA algorithm is combined with the convex optimization techniques to form a hybrid optimization algorithm,which is suitable for the joint optimization of the array excitation and array layout.Thanks to the adaptation of linear constraints of LCGA,the element spacings can be perfectly constrained.Subsequently,a series of extensions are carried out on the above basic hybrid algorithm to form a set of hybrid optimization methodologies,which can be applied to array synthesis problems in a wide range of scenarios.Furthermore,corresponding results significantly superior to those of existing technologies can be obtained in many cases.2.The introducing of spatial taper in subarray synthesis.The idea of spatial taper is in introduced by the author to compensate for the constraints of amplitude taper caused by subarraying,and results significantly superior to the uniformly spaced arrays can be obtained.Usually,several subarrays can achieve good beam shaping and low side lobe level performance.Furthermore,a mixed-integer programming model is proposed for subarray synthesis problem in this dissertation,and a no relaxation solving of subarrays partition is achieved by the developed algorithm.3.The equal-power subarray and its synthesis method.In this dissertation a novel subarray form with actual or potential use or value,which is named equal-power subarray,is proposed.Specifically,it means to partition elements into groups wherein elements are of different magnitudes,while all of the groups are of equal power.The advantages of such subarrays are manifold: Firstly,since the total power of each subarray is the same,all power amplifiers can operate with full capacity if constructing them at the subarray level,which can effectively improve the efficiency of active arrays; Secondly,equalpower subarrays can form a series of equal-power nodes in arrays,which can significantly simplifying the design of array feeding networks; Finally,comparing to traditional subarrays,equal-power subarrays allow amplitude taper between different elements(as long as the total power of all subarrays are equal).Therefore,equal-power subarrays can often achieve better beamforming capability.Aiming at the synthesis of equal-power subarray,the author also developed an algorithm to solve it,and excellent results are obtained.