Improvement of speed in data collection rate in tree based wireless sensor network

Abstract

Convergecast, the process of data collection from set of sensor nodes and forwards to the sink node over a tree based topology is a fundamental traffic pattern in Wireless Sensor Networks(WSN). Time Division Multiple Acess(TDMA) protocol is considered for scheduling, where time is divided into slots. The main aim of this paper is to reduce the number of time slots (scheduling length) using different methods like minimum time slots assignment for Breadth first search (BFS) algorithm, simultaneous assignment of time slots for Local time slot algorithm, balanced tree construction and multiple frequency assignment methods. This proposed work involves construction of a tree structure with random placement of nodes and a BFS algorithm is implemented for aggregated converge cast, which allots minimum time slots to the nodes, a local time slot algorithm is implemented for raw data convergecast, which allots parallel time slots to the nodes to allow simultaneous transmissions. This praposed method attains the less schedule length by using the balanced tree structures. Initially constructed a tree structure with random placement of nodes and applied BFS algorithm for that tree structure. In BFS algorithm, 5 timeslots are required for 10 nodes to send data packets to sink node. The modification of BFS is DCT algorithm, in this algorithm 3 time slots are required for 10 nodes to send data packets to sink node. So from the simulation results we can conclude that, the schedule length is reduced by 40%.