A Spike Neural Controller for Traffic Load Parameter with Priority-Based Rate in Wireless Multimedia Sensor Networks

Abstract

Wireless Multimedia Sensor Networks (WMSNs) are a type of sensor network that contains sensor nodes equipped with cameras, microphones; therefore the WMSNS are able to   produce multimedia data such as video and audio streams, still images, and scalar data from the surrounding environment. Most multimedia applications typically produce huge volumes of data, this leads to congestion. To address this challenge, This paper proposes Modify Spike Neural Network control for Traffic Load Parameter with Exponential Weight of Priority Based Rate Control algorithm (MSNTLP with EWBPRC). The Modify Spike Neural Network controller (MSNC) can calculate the appropriate traffic load parameter μ for each parent node and then use in the EWPBRC algorithm to estimate the transmission rate of parent nodes and then assign a suitable transmission rate for each child node. A comparative study between (MSNTLP with EWBPRC) and fuzzy logic controller for traffic load parameter with Exponential Weight of Priority Based Rate Control algorithm (FTLP with EWBPRC) algorithm shows that the (MSNTLP with EWBPRC) is more efficient than (FTLP with EWBPRC) algorithm in terms of packet loss, queue delay and throughput. Another comparative study between (MSNTLP with EWBPRC) and EWBPRC with fixed traffic load parameter (µ) shows that the MSNTLP with EWBPRC is more efficient than EWBPRC with fixed traffic load parameter (µ) in terms of packet loss ratio and queue delay. A simulation process is developed and tested using the network simulator _2 (NS2) in a computer having the following properties: windows 7 (64-bit), core i7, RAM 8GB, hard 1TB.