Implementation and design of underwater duplex speech communication system based on vector hydrophone

This paper realizes 1.2kbps Mixed Excitation Linear Prediction (MELP) speech compression algorithm through multi-frame joint and Vector Quantization technology. In order to adapt to underwater acoustic channel, we apply the Orthogonal Frequency Division Modulation (OFDM) to modulate the source bits, together with synchronization, channel estimation and RS error correcting code technology. The hardware platform of this system relies on TMS320DM642 and AIC23. Moreover, we introduce single vector hydrophone to restrain the local source. The system finished duplex speech signal acquisition and transmission and tank test results prove that the synthesized speech is clear and understandable and it has a stable performance.