Optimization parameter for mpeg-4 data over wireless sensor network

Abstract

Conventional Wireless Sensor Network (WSN) always deals with scalar data such as temperature, humidity, pressure and lights. These types of data are very suitable for low rate and low power networking technology such as IEEE802.15.4. Transmitting a video for IEEE 802.15.4 standard is an interest due to the capability of providing low complexity with low cost but still maintaining the good quality of video in term of Peak Signal to Noise Ratio (PSNR). However transferring video posed new challenges for bandwidth constrained sensor networks like WSN. Furthermore, processing power put a serious limitation on it for any sophisticated processing. Therefore a method needed for reliable video transfer and one way is to compress the video format. A simulation testing has been done to determine the optimal threshold for quantization scale, group of picture and frame per second in order to tailor with the environment of WSN 802.15.4. When these three parameters are changed, the video quality and bandwidth requirement will be affected. Therefore this project aims to create an embedded code in TelG mote according to this concept. Experiment testbeds are conducted for indoor and outdoor to investigate the network performance in term of packet delivery ratio and total packet loss. Results show the packet delivery ratio of optimized MPEG-4 data for Akiyo cif file is 36% better than the non-optimized MPEG-4 data, Foreman cif file is 12% and Mobile cif file is 8%. Thus this project have prooved that optimized parameter of MPEG-4 data has successfully increase the network performance of WSN.