VOICE/DATA COMMUNICATIONS USING VECTOR QUANTIZATION

THE GOAL OF THIS RESEARCH PROJECT IS THE DEVELOPMENT OF A VOICE AND/OR DATA COMMUNICATIONS SYSTEM USING VECTOR QUANTIZATION (VQ). VECTOR QUANTIZATION HAS TRADITIONALLY BEEN USED AS A SOURCE CODING ALGORITH FOR DIGITAL CODING SYSTEMS. BY REVERSING THE ORDER OF THE ENCODER AND DECODER, VQ CAN BE USED TO TRANSMIT DIGITAL VOICE OR DATA IN A LOCALLY OPTIMUM MANNER. VQ ALSO OFFERS THE ADVANTAGE OF PROVIDING LOCALLY OPTIMUM SIGNAL SETS FOR A GIVEN TRANSMISSION SYSTEM, THEREBY COMING CLOSER TO THE SHANNON CHANNEL CAPACITY LIMIT FOR ANALOG CHANNELS. USING THE VQ DESIGN ALGORITHM, SIGNAL SETS CAN BE DESIGNED WHICH YIELD LOWER PROBABILITY OF BIT ERRORS AT A GIVEN BIT RATE THAN TRADITIONAL MODULATION SCHEMES. DUE TO THE UNIQUENESS OF THE CODEBOOK, VQ ALSO OFFERS AN INHERENT COVERTNESS NOT AVAILABLE WITH OTHER MODULATION SCHEMES. THE PHASE I PROPOSED RESEARCH WILL BE COMPOSED OF THE FOLLOWING STAGES: I) COMPUTER MODELING OF THE VQ MODULATION SYSTEM, II) SYSTEM SIMULATION AND SNR/BIT-RATE PERFORMANCE BOUNDS, AND III) REAL-TIME IMPLEMENTATION FOR A ADDITIVE WHITE GAUSSIAN NOISE CHANNEL. ONCE THE BASIC PRINCIPLES OF PHASE I HAVE BEEN ESTABLISHED, PHASE II & III EFFORTS WILL INCORPORATE FULL-SCALE DEVELOPMENTAL EFFORTS TOWARDS AN OPERATIONAL SYSTEM.