LT Codes and the Minimum Spanning Tree Based Distributed Storage in Wireless Sensor Networks

Abstract

In wireless sensor networks, sensor nodes are vulnerable due to the challenging environment and limited energy. Once a sensor node fails, the stored sensed data will be lost. In this paper, we develop a distributed algorithm for solving this problem based on the fountain codes and the minimum spanning tree. Similar to previous works, we adopt the Luby Transform (LT) codes, the first practical scheme of fountain codes, in our algorithm. However, different from the previous schemes, our algorithm deploys the minimum spanning tree instead of the random walk. Due to the fact that all nodes in the wireless sensor networks do not have the location information of other nodes, we use the Gallager-Humble-Spira (GHS) algorithm to generate a distributed minimum spanning tree. Data packets can be transmitted through the tree and easily reach every node in the network. We obtain the performance of the algorithm through simulations, which show that our algorithm consumes less number of the transmissions compared with the existing algorithms using the random walk.