A new method for emitter source signal fusion identification
Pang Jin-Feng, Lin Yun · 2012
Information fusion is widely used in many fields, such as space remote sensing, environment monitor, emitter recognition, and so on. In this paper, aiming at the drawbacks of DS evidence algorithm, which is difficult to deal with high conflict and mutuality evidence, based on convex optimization theory, a new information fusion algorithm is presented. Firstly, according to the sensor reports, it constructs the cost function of information fusion. By decomposing the cost function, the problem of information fusion is transformed to convexity optimization problem. Finally, it is solved by using interior-point logarithm, which has simple structure, small computation and easy to come true. The theory analysis and simulation result prove that the convex optimization algorithm has better recognition ability and flexibility than traditional DS evidence algorithm, so it can be used in emitter recognition.