On the Investigation of Wireless Signal Identification Using Spectral Correlation Function and SVMs

Signal identification is an important notion that leads to significant performance improvements for adaptive wireless spectrum access techniques. Besides identifying the modulation types and other features, standard-based identification has also an important place in signal identification domain. In this paper, a generalized identification method which utilizes the outputs of spectral correlation function as the training inputs for the support vector machines to distinguish wireless signals is introduced. The proposed method eliminates the dependence on the distinct features to identify different signals. The method's performance is tested using the measurements taken in the laboratory environment and various wireless signals are successfully distinguished from each other. The comparative performance of the proposed method is also quantified by the classification confusion matrix.