LAUSR

The classification of error correcting code (ECC) types is an important task for intelligent wireless communication. Previous solutions to this problem usually ignored the time sequences feature and considered the incoming data as separate code words. In this paper, a novel framework for code type classification is proposed based on the Hidden Markov Model (HMM). The structural differences of different types of ECCs are modeled and analyzed using the transition probability and the emission probability. An HMM-based classifier is proposed to recognize the types of ECCs from the intercepting bitstream. To improve the performance in the noisy transmission environment, the basic classifier is enhanced by soft decision message and relative posterior probability. Simulation results show that the proposed scheme outperforms other existing methods in erroneous conditions and the classification accuracy is associated with the sequence length and decision approach.