Study On Near Field Beamforming Techniques Based On Spherical Array Measurement

Array signal processing can detect signals and estimate parameters. Recently, it has been widely applied in many areas such as radar, sonar and communications. As an important task of array signal processing, beamforming techniques can be divided into two parts:conventional beamforming techniques and adaptive beamforming techniques. The conventional beamforming techniques are independent of the data received by array, however, the adaptive beamforming techniques rely on the received data. In the thesis, near field sound source model and spherical array are used in conventional beamforming techniques and adaptive beamforming techniques respectively, then the near field beamforming techniques based on spherical array measurement are proposed. Furthermore, theoretical analysis and experimental researches of beamforming techniques are done.Firstly, with theoretical analysis, numerical simulations and experimental verification of some conventional near field beamforming techniques and spherical array measurement, the superiority of spherical array used in conventional near field beamforming techniques is shown. Secondly, the orthogonal projection beamforming technique is deeply studied. In order to reduce the amount of calculation, the propagator method is introduced and the orthogonal projection beamforming technique based on propagator method is then proposed, which provides the possibility for its real-time use. Beamforming performance of the traditional adaptive beamforming techniques will degrade when the received signals include desired signal and the sound field contains coherent sources, the linearly constrained oblique projection beamforming technique is introduced to overcome this problem. The numerical simulations show that it has better beamforming performance in low signal-to-noise ratio, small snapshot number and coherent sources. Under the condition of existing fixed nulls, the transformation-based linearly constrained oblique projection beamforming technique is proposed. It retains all advantages of the linearly constrained oblique projection beamforming technique, furthermore, because of using a transformation matrix, it will keep numerical stability when the adaptive nulls close to or coincident with the fixed nulls. Finally, the conventional and adaptive beamforming techniques based on spherical array measurement are applied to the interior space and some experimental researches are done. The results show that sound sources within a certain frequency range in the interior space can be identified with spherical array measurement. The adaptive beamforming techniques based on spherical array measurement can also be used in the interior space, which expands the applications of beamforming techniques.