Title :
Non-linear bandwidth reduction schemes for transmission of multivariate Gaussian sources over noisy channels
Author :
Lu, Bo ; Garcia-Frias, Javier
Author_Institution :
Dept. of Electr. & Comput. Eng., Univ. of Delware, Newark, DE, USA
Abstract :
In this paper we investigate the design of bandwidth reduction non-linear schemes for the transmission of multivariate Gaussian sources over noisy channels. The non-linear mappings are designed utilizing an algorithm called “Power Constrained Channel Optimized Vector Quantization”(PCCOVQ). Practical examples for 2:1, 3:1 and 4:2 systems are provided, and simulation results are compared with the respective Optimal Performance Theoretically Attainable (OPTA).
Keywords :
Gaussian channels; combined source-channel coding; quantisation (signal); OPTA; PCCOVQ; analog joint source-channel coding; multivariate Gaussian source transmission; noisy channels; nonlinear bandwidth reduction schemes; nonlinear mappings; optimal performance theoretically attainable; power constrained channel optimized vector quantization; Communication systems; Computational modeling; Computers; Encoding; Noise measurement; Quantization; Simulation; Analog joint source-channel coding; multivariate Gaussian sources; non-linear bandwidth reduction schemes;
Conference_Titel :
Information Sciences and Systems (CISS), 2012 46th Annual Conference on
Conference_Location :
Princeton, NJ
Print_ISBN :
978-1-4673-3139-5
Electronic_ISBN :
978-1-4673-3138-8
DOI :
10.1109/CISS.2012.6310715
