Abstract: Based on the statistical characterisitic of speech spectrum, a new algorithm which can distinguish LSB or USB cases of speech from SSB signals without any prior information is presented here. It is simple and very suitable for DSP's real-time implementation, and the adaptive threshold increases the algorithm's robustness. The simulation results show that this algorithm gives good performance even under the conditions of low SNR values.

Key words: single sideband, speech spectrum, Rayleigh channel, adaptive threshold