Study On Speech Coding Technology In Free Space Optical Communication

Free space optical communication has been a hot research area and widely developed. Because of its huge military and commercial value, China also strengthened the research since 1990 s. As one format of the free space optical communication, ultraviolet communication has been an important hotspot, because ultraviolet light with the wavelength of 200 to 300 nm was completely filtered by the atmosphere for its scattering properties, hence ultraviolet communication which has little background noise can be used as NLOS communication. In the speech codec algorithm research, the main research direction is low bit rate speech coding which was leaded by foreign scholars.In the paper, free space optical communication and ultraviolet communication were deeply studied. Plenty of researches have been done in theory, model, system construction, and building experimental system, to lay a solid foundation of the upcoming applications. The main works in the paper are as follows.(1) The basic principle and structure of free space optical communication were deeply studied. Meanwhile, the principle of 200~300 nm wavelength ultraviolet scattering was also introduced, and the simplex and duplex communication systems were designed. What's more, FPGA which is reconfigurable was studied, and used to process digital signal.(2) According to the basic principles of free space optical communication, Cyclone platform, MA24126-P1 voice codec platform and other optoelectronic devices were used to establish a complete continuous work MELP algorithm voice communication system, which enables duplex voice communication for more than 2 h. And its communication rate is 11.25 kbps, and MOS rate is higher than 3.(3) Based on the basic principles of waveform coding, we designed a ?-law algorithm FPGA voice communication system with 637 nm wavelength semiconductor sub-body laser VA-I-DC-637 and acoustic-optical modulators. After related simulation and test, the system can be used to communicate with others for several hours, and MOS rate is higher than 3. According to the simulation and test, the correctness and feasibility of the system was proved.(4) According to the basic principle of linear predictive coding, the program of parameter extraction and encoding was designed, and the core part of the program was simulated. Hence data calculated and coded by the program was send to modulate the MPL-F-266 solar blind UV pulsed laser with the method of pulse position modulation. Finally, voice communication FPGA system of linear predictive coding was set up with the Cyclone platform and other optoelectronic devices. As a prototype system, it was able to accomplish the extraction of voice parameters and NLOS communication, as well as transfer data at a speed of 2.5kbit/s.