Research On DOA Estimation Methods Of Underwater Target

After decades of development, the theory of DOA estimation has become more mature,and is widely applied in communication, radar, sonar, medical and other fields. DOAestimation of underwater target plays an important role in battle-field reconnaissance system,and exerts a great influence on the development of military intelligence. In addition, DOAestimation of underwater target also plays an important role in the field of navigation,maritime security operations under water and ocean exploitation. Therefore, the precisedirection finding and positioning of underwater targets is of great important in military andcivil fields.In this paper, DOA estimation methods of the underwater target are discussed. Themain research contents are as follows:1. The feasibility analysis of compressive sensing theory is applied to the DOAestimation and corresponding mathematical models are established. The sparse arraymanifold matrix is taken as the measurement matrix, the correlation was analyzed. Theresults show that the recovery condition based on cumulative mutual coherence possessesthe advantages restricted isometry property (RIP) and mutual incoherence property (MIP),and it is easy to implement.2. Research on the DOA estimation based on the MUSIC algorithm. According to theuniform array element spacing not greater than half the wavelength of the source signal,DOA estimation method that based on non uniform linear array are designed. For DOAestimation method that based on the non uniform linear array (NLA), the underwater targetis sensitive to the noise. When the SNR is low, DOA estimation can cause the pseudo peakswhich leads to the fuzzy estimation. In this paper, a kind of averaged DOA estimationmethod is proposed that based non uniform linear array, and the simulation results areanalyzed. The simulation results show that in the direction, beam width and resolution,non-uniform linear array is far better than the uniform array; and when the SNR is low, theaveraged method can effectively restrain pseudo peaks, while the real spectrum peakposition and amplitude is not changed, thereby the method increases the success probabilityof underwater target DOA estimation. 3. Research on the DOA estimation that based on the compressive sensing theory.Firstly, the DOA estimation model based on compressive sensing theory is established.Secondly, the measurement matrix correlation of the uniform linear array and non uniformlinear array are analyzed. Finally, DOA estimation algorithm that based on the compressedsensing theory, simulation is made under different circumstances of the signal source. Theresults show that: compared with the classical MUSIC algorithm, the proposed DOAestimation algorithm based on compressive sensing overcomes inadvantage that a largenumber of snapshots and multiple array elements are need and failure estimation whensignal sources are coherent.The researches results in this paper provide basic theories and application reference.