Triangular Fuzzified Support Vector Regression Based Multivariate Rocchio Robust Boost Classification for Traffic and Congestion Aware Data Delivery in WSN
Wireless sensor network (WSN) is a huge set of sensor nodes distributed geographically with limited power supply. These nodes are used to sense environmental changes, gather information and send the collected data to a remote destination for data analysis and decision making. During continuous data transmission, the network traffic causes congestion. Congestion in WSN is a major cause for packet loss and it degrades the network performance. In order to improve the congestion-aware routing, a novel technique called as the Triangular Fuzzified Support Vector Regression Based Multivariate Rocchio Robust Boost Classification (TFSVR-MRRBC) technique is introduced. The proposed TFSVRMRRBC consists of two processes such as route path identification and congestion aware data delivery. Initially, route paths are constructed by estimating node residual energy. After that, Support Vector Regression is applied to analyze the energy using the fuzzy Triangular membership function. Thus, energy-efficient nodes construct the multiple paths between sources and sink node through control message distributions namely route request with route reply is identified. Then the Multivariate Rocchio adaptive robust boost technique is applied for congestion aware data delivery based on the buffer space and bandwidth capacity. The route path with low buffer space and higher bandwidth capacity is selected as the best route among the multiple paths. Finally, Delay and packet loss is reduced by achieving congestion-aware data delivery. The simulation is achieved by various performance factors namely, energy consumption, packet delivery ratio, packet loss rate, and end-to end delay with versus to amount of data packets and sensor nodes. Experimental results of proposed TFSVR-MRRBC technique improves data packet delivery and energy consumption is minimized, packet loss rate, delay than the conventional routing techniques.