LOAD AWARE CHANNEL ESTIMATION AND CHANNEL SCHEDULING FOR 2.4 GHZ FREQUENCY BAND WIRELESS NETWORKS FOR SMART GRID APPLICATIONS

摘要

The advanced monitoring and control applications consider Wireless Sensor Networks(WSN) as a promising technology for modern applications like the Internet of Things (IoT), Smart Gridand Wireless Body Area Networks (WBAN). The WSN has important features like less cost, low powerusage, supportable data rates and complexity. There is a need for continuous research on improvingcharacteristics and abilities of WSN. The reliable performance of WSN depends on the latencynecessities depending on the type of application and Quality of Service (QoS) parameters. Thetechnologies like Zigbee, WiFi, and Bluetooth operating in 2.4GHz are mostly considered for deployingthe WSN. Because of coexistence environment, the performance of Zigbee gets affected in terms ofchannel switching and causes the significant amount of delay. Also, the data transmission should beperformed without any collision. In this paper, initially, the pseudorandom – based interferenceevading scheme is introduced for efficient data communication. During this scheme, if node attains achannel it must wait for a network reconfiguration time for moving to next channel. Hence, during thistime other nodes are allowed for moving to the new channel. Secondly, for moving to the new channelload aware channel estimation is proposed to assess the possibility of traffic weight assignment at eachchannel. Finally, the Particle swarm optimization (PSO) based collision avoiding multiple-channelbased superframe scheduling is proposed for IEEE 802.15.4 based wireless networks working underthe influence of IEEE 802.11b network. The channel with best energy function is selected for datatransmission. The work proposed in this paper is evaluated based on the comparison to the existingworks. From the results obtained it is inferred as proposed work shows better performance in terms ofpacket error rate, packet delivery ratio, and energy consumed when compared to the existingalgorithms.