Side-lobe Suppression For Time-Modulated Sparse Array Based On CVX Algorithm

As the electronic technology develops,antenna array has been widely used in fields such as Radar and wireless communication.However,higher requirements of anti-interference ability of radar system is put forward due to the much more complex electromagnetic environment.In order to break through the bottleneck in traditional antenna array design and get better radiation pattern,time modulation technology is applied in traditional sparse arrays,which,named as Time-Modulated Sparse Array(TMSA),can obtain a low/ultra-low side-lobe level(SLL)antenna pattern with a low amplitude dynamic range ratio or even the same amplitude and phase excitation and helps reduce the difficulty of antenna feeding system effectively.Based on convex(CVX)and particle swarm optimization(PSO)algorithm,the side-lobe suppression problem of TMSA is studied to solve the high grating and SLL of time-modulated uniform sparse array(TMUSA).Specific research contents and results are as follows:Firstly,based on the given description of the structure of TMSA,the radiation model in the free space is constructed.Meanwhile,the theoretical derivation and analysis of the switch functions and radiation pattern of five different time-modulation models are shown as the theoretical basis for subsequent chapters.Secondly,the convex optimization model of TMUSA is constructed as the feature that a local optimal solution of the CVX problem is the same as its global optimal solution.Furthermore,a hybrid algorithm based on CVX and PSO is adopted to optimize the high-dimension and non-linear SLL suppression problem of TMSA when the time sequences,positions and static excitations of antenna elements are set as a joint optimization variable.Simulation results show that time-modulated antenna array is prior to traditional antenna array in low/ultra-low SLL array pattern synthesis in convergence speed and SLL.However,the potential weakness of this model is pointed out by the analysis of sideband level(SBL).The last part is about the improvements of the hybrid PSO-CVX algorithm in two aspects.One is the improvement of the math model by adding the peak SLL as a constraint,which means that the peak SBL is chosen as the fitness function instead,and changing the time-modulation model from Variable Aperture Sizes(VAS)into the more flexible Pulse Shifting(PS)model.The other improvement is about the algorithm parameters.The switch time sequences of the array element are designed to be symmetric with respect to the central array element so that the total number of the modified hybrid algorithm optimization variable remains unchanged.The optimization efficiency of SBL suppression is improved with the convergence speed guaranteed.The contrast simulation with the reference shows that the improved hybrid PSO-CVX optimization algorithm is better than the difference algorithm.