The Binary Beetle Antennae Search Algorithm for Sparse Filter Design

Sparse filter has attracted much attention in the field of signal processing and communication due to its low complexity. However, the sparse optimization problem has non-convex characteristics, which leads to difficulty in solving. A heuristic binary beetle antennae search (BBAS) algorithm is proposed in this paper, which imitates the foraging process of beetle and uses the principle of which beetle uses antennae to detect direction, and finally achieve the purpose of searching the optimal encoding. Encoding vector is introduced to indirectly select filter coefficients, and the sparse solution of filter is obtained by searching the optimal encoding. In each iteration of the algorithm, the synthesized candidate encoding is divided into O-region and 1-region, and the values of encoding elements related to the current iteration are determined by comparing the overall fitness of the two regions. The length of antennae in the algorithm represents the number of elements in the encoding vector that the algorithm can perceive. The larger the antennae length is, the better the global search effect is., but it also means a higher amount of computation. Simulation results show that the proposed algorithm can effectively reduce the number of non-zero coefficients of the filter.