Recognition Method of Software Defined Radio Signal Based on Evidence Theory and Interval Grey Relation
Hui Wang, Lili Guo, Yun Lin · 2017
It is known that the features of the radio station vary with the signal to noise ratio (SNR) in a certain range which leads to the uncertainty of the radio station identification system. In this paper, we study the interval evidence recognizer by extending the individual features of the obtained radio from single value to interval and constructing the radio station feature database. Firstly, the range of SNR is divided into several intervals and individual interval entropy features of radio station in every corresponding SNR interval are calculated. Then we transform the interval entropy features to basic probability assignments (BPAs) by comparing with the radio station feature database using the interval grey relational analysis method. Finally we fuse the BPAs by improved weighted evidence theory to obtain the recognition result. Simulation indicates that the interval recognition method proposed in this paper has an excellent recognition rate when dealing with the radio station signals vary in a large range.