Studies On Digital Voice Communication System In Shallow Water Acoustic Channels

Since sound is th,e only effective information carrier for medium or long distance wireless communication in sea water, underwater acoustic (UWA) communication plays an important role in national defense and in the development of marine economy. With the characteristics of large ambient noise, very narrow bandwidth, low carrier frequency, great propagation latency and time-space-frequency variant multipath effect (MPE) , the stochastic ocean channel has demonstrated the greatest complexity and difficulty for underwater acoustic wireless communications. Among them multipath effect (MPE) is the most difficult obstacle that results in signal fading and inter-symbol interference (ISI).The nature of the multipath effect (MPE) has decided that phenomenon that the longer the working distance is adopted, the worse the multipath effect (MPE)plays the role.The new methods should be studied and employed to eliminate or mitigate the negative resulted by multipath effect (MPE).It has been proven that the Spread Spectrum Technologies have good capabilities of overcoming ISI in electromagnetic wireless channel. The feasibilities of their applications to underwater acoustic channel have been analyzed, the Frequency Hopping technology is chosen. Instead of using time-frequency diversity, convolution coding and Viterbi decoding are employed to combat signal fading, consequently the bit-rate and bit-error-rate performance of the underwater acoustic voice communication system can be balanced and controlled.For medium or long distance underwater acoustic communication, the narrow bandwidth and low carrier frequency only supports low bit-rate signal transmission. In order to achieve real time communication, Speech Recognition technology is used to keep the semanticinformation, meanwhile the greatest compression on the information is obtained. Consequently the data rate needed on the channel is reduced dramatically. When the semantic information is received, Speech Synthesis is used to generate highly clear voice in the receiver. In the end hardware electrocircuit have been finished.