Research On Array Thinning Technology Based On Multi-beam Imaging Sonar

In the multi-beam imaging sonar system,the large-scale transducer array is sampled at the receiving end.After digital beamforming,uniformly and closely distributed receiving beams are formed within a certain angle range in order to obtain a two-dimensional image of the underwater target in real time.Thus the multi-beam imaging sonar is widely used in marine resource development and underwater search and rescue.However,the sonar hardware complexity and cost,volume and power consumption are increased because of the large-scale transducer array.Array thinning technology is one of the effective ways to solve this problem.Most of the existing thinned array designs are directed to single-beam circumstances,which is with few constraints.The thinned array design under the case of multi-beam imaging sonar has not been well resolved yet.Therefore,in this paper,it is of great theoretical and practical value to study the array thinning techniques under multi-beam conditions.The main research contents and achievements are as follows:(1)The array thinnnning method based on Farrow structure multi-beamformer is studied.The method is aimed to obtain a thinned weight vector to achieve the array thinning.The Farrow structure is exploited so that the beam pointing direction can be adjusted by only one parameter.In combination with the multi-beamforming method based on the subaperture rotation,538 beams within the 90° field of view can be formed by using a set of weighting coefficients.A convex optimization model of array thinning based on Farrow structure multi-beamformer is constructed.The simulation results and measured data show that the method can obtain satisfactory solution according to the set threshold,but the thinning rate is not high.(2)The thinning method for multi-beam array based on intelligent optimization algorithms and convex optimization algorithm is studied.The thinning method based on particle swarm optimization for single-beam array proposed by predecessors is extended to multi-beam conditions.The simulation results show that the method has poor optimization performance and its calculation is too large.Then a new hybrid optimization algorithm based on improved binary wind driven optimization and convex optimization is proposed.The algorithm optimizes the element positions and weights at the same time,until a thinnned array of acceptable minimum number of elements is obtained.The simulation results and measured data show that the algorithm can obtain the optimal weights and array layout while satisfying the performance requirement.Both the optimization performance and the computational efficiency are improved compared with the existing method.(3)The thinning method for multi-beam array based on the column space correlation is studied.Considering the large correlation of the received signals,the column space correlation model of the array oriented matrix is studied.The projection error is introduced,and then the relative redundancy of elements is determined based on the projection error of corresponding columns.The algorithm flow is given in detail.The simulation results and measured data show that the thinning method based on the column space correlation has better performance than the previous two methods,and the optimal thinning rate of the array can be achieved under the same performance requirement.In general,three thinning methods for multi-beam array presented in this paper have advantages and disadvantages.The thinning method based on Farrow structure multi-beamformer requires fewer weights and less computation time,but the main beam width is wider and the thinning rate is not high.The latter two thinning methods are superior to the thinning method based on Farrow structure multi-beamformer,but the computation time is increased.In comparison,the thinning method for multi-beam array based on column space correlation has the best overall performance.