Optimization Algorithm for Distributed Target Detection Integration Decision

Existing distributed object detection systems generally use the enumeration method, SFFO method or SOFF method. These methods are computationally complex, and their optimization results strongly depend on initial value. A climbing variation ant colony algorithm is proposed in this paper. The algorithm is used to simultaneous optimization of sensor decision threshold and the fusion center decision rules. The numerical results show that the fusion system reduces Bayes risk about 15%～20%, the optimization results do not depend initial value, and the computational complexity is lower than the other algorithms.