A low-complexity DWT module and CRS minimal instruction set computer architecture for wireless visual sensor networks

Abstract

Transmitting cost, processing complexity and data security are three important elements in wireless visual sensor networks (WVSNs). This paper presents a complete low-complexity processing system that performs data compression, data correction and data encryption. In this system, the discrete wavelet transform (DWT) first decomposes the original image into DWT coefficients to ensure low transmitting cost, then the coefficients are encrypted using the Cauchy Reed Solomon CRS(20,16) coding scheme to ensure data security. A CRS minimal instruction set computer architecture with a DWT filtering module is proposed to perform the compression, encryption and error correction encoding in a low-complexity processing system. The proposed system is then implemented in a field programmable gated array (FPGA) to demonstrate the feasibility of the proposed system for implementation in WVSNs. Results on a Xilinx Spartan FPGA show that the proposed system requires a lower implementation complexity of 2536 slices as compared to other existing systems such as Crypto-Processor (4828 slices) and SPIHT CRS MISC (5017 slices).