Study On DOA Estimation Based On Sparse Signal Reconstruction And Route Planning

In complex battlefield environments,modern radars need to clearly grasp the frequency and location of radiation sources in the battlefield in order to detect targets more accurately.Besides,there is an important development tendency that multiple UAVs collaborate with each other to replace single airplane.Therefore,for airborne radar,it is necessary to carry out the researchs of more accurate DOA estimation algorithms and multiple UAVs cooperating methods to monitor the battlefield.To obtain a more effective DOA estimation algorithm,this thesis studies the theory of compressive sensing and sparse signal reconstruction,applies it to DOA estimation,and proposes two improved algorithms that respectively increase the algorithm's accuracy and improve the algorithm's performance in non-ideal environments such as single-snapshot data and source coherence.For the problem of multi-airborne radars cooperating to monitor the battlefield,this thesis learns from the idea of ant colony optimization algorithm to maximize the coverage of the task area in real time and proposes two routing planning methods for multiple UAVs cooperating to monitor the battlefield monitorin,making airborne radars able to communicate with each other and collaborate to achieve a better coverage of taks area dynamically.The research results of this thesis mainly include:1.Aiming at how to improve the accuracy of DOA estimation algorithm,accuracies of the typical algorithms of sparse signal reconstruction at different SNRs are compared.A new EOLS algorithm is proposed,which embeds the ESPRIT algorithm into the OLS algorithm.In terms of different SNRs,the new EOLS algorithm's performance can exceed both upper limits of the performances of OLS algorithm and the ESPRIT algorithm.The simulation results verify the effectiveness of the new method.2.As to the problem of poor performance or even failure of traditional DOA estimation algorithms in non-ideal environments such as single-snapshot data and source coherence,an improved maximum likelihood sparse parameter estimation algorithm is proposed in this thesis.The algorithm uses the basic principle of minimizing the maximum method and can be solved with second-order cone programming.Simulation experiments show that the algorithm can achieve higher estimation accuracy in the condition of single-snapshot data and source coherence.3.For the coordination problem of multi-airborne radars monitoring the battlefield,this thesis proposes a single-step route planning method for multiple UAVs cooperating to monitor the battlefield.This method uses the idea of ant colony optimization algorithm to achieve a good dynamic coverage of the task area.Then,as to the problem of unstable coverage resulted from occasional aircraft flying out of the boundary and flying around in circles in the single-step method,further optimization is taken on the fitness function in this algorithm including the boundary penalty and angle deflection constraint.At the same time,a multi-step route planning method for multiple UAVs cooperating to monitor the battlefield is also proposed,which solves the problems above and achieves a better coverage curve.