A New Distributed Topology Control Algorithm for Wireless Environments with Non-Uniform Path Loss and Multipath Propagation

A New Distributed Topology Control Algorithm for Wireless Environments with Non-Uniform Path Loss and Multipath Propagation Harish Sethu and Thomas Gerety Department of Electrical and Computer Engineering Drexel University Philadelphia, PA 19104-2875 Email: {sethu, thomas.gerety}@drexel.edu Abstract Each node in a wireless multi-hop network can adjust the power level at which it transmits and thus change the topology of the network to save energy by choosing the neighbors with which it directly communicates. Many previous algorithms for distributed topology control have assumed an ability at each node to deduce some location-based information such as the direction and the distance of its neighbor nodes with respect to itself. Such a deduction of location-based information, however, cannot be relied upon in real environments where the path loss exponents vary greatly leading to significant errors in distance estimates. Also, multipath effects may result in different signal paths with different loss characteristics, and none of these paths may be line-of-sight, making it difficult to estimate the direction of a neighboring node. In this paper, we present Step Topology Control (STC), a simple distributed topology control algorithm which reduces energy consumption while preserving the connectivity of a heterogeneous sensor network without use of any location-based information. The STC algorithm avoids the use of GPS devices and also makes no assumptions about the distance and direction between neighboring nodes. We show that the STC algorithm achieves the same or better order of communication and computational complexity when compared to other known algorithms that also preserve connectivity without the use of location-based information. We also present a detailed simulation-based comparative analysis of the energy savings and interference reduction achieved by the algorithms. The results show that, in spite of not incurring a higher communication or computational complexity, the STC algorithm performs better than other algorithms in uniform wireless environments and especially better when path loss characteristics are non-uniform. arXiv:0709.0961v2 [cs.NI] 13 Nov 2009 I. INTRODUCTION In a multi-hop wireless sensor network, a node communicates with another node across one or more consecutive wireless links with messages possibly passing through intermediate nodes. The topology of such a network can be viewed as a graph with an edge connecting any pair of nodes that can communicate with each other directly without going through any intermediate nodes. Each node in such a network can choose its own neighbors and thus control the topology by changing the power at which it makes its transmissions or, in the case of nodes capable of directional transmissions, by also changing the set of directions in which it will allow transmissions. The goal of such topology control is to employ algorithms that each node can execute in a distributed manner for the purposes of reducing energy consumption, maintaining connectivity, and increasing network lifetime and/or capacity. In recent years, a large number of topology control algorithms have been proposed and studied for a diverse set of goals [1]. Early work on topology control assumed that accurate location information about its neighbors will be available to the nodes, such as through the use of GPS devices [2]–[6]. This assumption adds to the expense of the nodes and also results in high delays due to the acquiring and tracking of satellite signals. Also, one cannot rely on GPS in many real application environments such as inside buildings or thick forests. Some other topology control protocols that preserve connectivity rely on the more likely ability of a node to estimate the distance and direction to its neighbors. For example, in the cone-based distributed topology control (CBTC) algorithms, a node u transmits with the minimum power pu,α required to ensure that there is some node it can reach within every cone of degree α around u [7]. Assuming a specific loss propagation model, the Euclidean distance to a neighbor can be deduced with knowledge of the power at which a transmission is made by a neighbor and the power at which the signal is received. The direction of a neighbor with respect to itself can be deduced from the angle of arrival of a signal. Wireless communication, however, is often characterized by the phenomenon of multipath propagation wherein a signal reaches the receiving antenna via two or more paths [8]. In addition, there are several other kinds of radio irregularities that have an impact on the topology control algorithms [9]. The different paths, with differences in delay, attenuation, and phase shift, make it difficult for the receiving node to deduce its distance from the sender and the direction of the sender. In this paper, we focus on the design of topology control algorithms that work without the use of any location-based information so that they

…(Full text truncated)…

This content is AI-processed based on ArXiv data.