Research On Wireless Sensor Networks Based On Small-World Model

The sensor nodes are small, energy limited and cannot be added, so the key of the wireless sensor network research are the limited resources and capability of communication, computing, storage and bandwidth transmission. The small world network presents interesting characteristics such as small average path length and high clustering coefficient and it exist in many real networks.A solution to generate a wireless sensor network with small world features is to add a small number of long range links, called shortcuts,then the performance of the network will be improved effectively. Because the wireless sensor network environment is complicated and the failure of nodes is inevitable, it is very important to research the network running situation after some nodes are failure and the means of improving the network topology. The invulnerability of network describes the network's ability to maintain normal work after some nodes are failure, so it is necessary to determine a standard to measure the invulnerability of wireless sensor network.In this paper, we propose the partition-based uniform energy consumption model(PUECM) to design wireless sensor networks with small world features by adding shortcuts to the network. The proposed model considers the facts that all nodes in the network need to sent information perceived and processed to the Sink node and the energy consumption of nodes closed to the Sink node is faster than that of other nodes, which will cause the collapse of the entire network. Based on this, PUECM model create shortcuts toward the Sink node and the endpoints of the shortcuts are periodically rotated among powerful nodes. In particular, we create the shortcuts based on the H-sensor's region, surplus energy and the directed angulation towards the Sink node. First we determine the optional nodes based on the location of H-sensors. Then select the H-sensor whose surplus energy is the highest to be the endpoint. If there are at least two optional nodes with the highest surplus energy, determine the endpoint based on the directed angulation towards the Sink node. The simulation results show that PUECM model not only presents good small world features, but also leads to a reduction in latency and an increase in lifetime.In view of that the nodes will be failure easily, we propose the invulnerability evaluation measure based on the entropy of nodes important degree, for evaluating the network invulnerability. First of all we assess the importance of nodes in the network taking advantage of the small world theory and the important degree effects of neighbor nodes and adjacent side on them. Comparing with other node importance measure, the proposed evaluation standard is the most comprehensive and accurate. Then we estimate the uniform degree of important degree of each node based on the concept of entropy to measure the network invulnerability. In the simulation, we compare the results of many different invulnerability evaluation measures with failures of different type of nodes. It's clear that the invulnerability evaluation measure based on the entropy of nodes important degree is more accurate and comprehensive in evaluating the network invulnerability, comparing with the invulnerability evaluation measure based on the entropy of nodes degree and the invulnerability evaluation measure based on the number of shortest way.