No abstract

In this work we study the performance of Tree-Structured Vector Quantizers (T SV Q) jointly designed with Forward Error Correction Code providing Unequal Error Protection (U EP ) for the transmission of still images over noisy channels. Comparisons have shown that this scheme performs better than Channel Optimized Vector Quantizers (COV Q) scheme and is less affected by the channel distortion. The design of the T SV Q matched to the noisy channel is less complex than the design of COV Q and it has, in addition, a simpler implementation.

No abstract

Abstract-An scheme for the efficient transmission of remote sensing satellite (RSS) images is presented in this paper. A joint source/channel coding solution was used to deal with the conflicting issues of data compression, to reduce the amount of data to be sent, and data protection, to cope with errors introduced by the channel, issues further burdened by the stringent low complexity and low memory requirements brought by the need for on-board compression. Results are presented showing that the proposed scheme attains good PSNR performance while still maintaining low complexity.

No abstract