Localization in Wireless Sensor Networks

Abstract

In the past three decades, Wireless sensor networks (WSNs) have gained enormous research across the globe as well as huge industrial interest because they are becoming an integral part of major surveillance and smart projects today, to name a few: hospital surveillance, smart home, and environmental monitoring. For most of these applications, localization is a fundamental issue. The existing localization techniques are broadly divided into following two categories: range-based and range-free localization. Localization algorithms based on range information primarily focus on precisely measuring the range information (the angle or the distance) between the nodes, and then calculating the location based on trilateration, multilateration or triangulation approaches. Contrast to the range-based algorithms, the range-free algorithms use connectivity information between the nodes. In this algorithm, the nodes that already know their locations (through GPS or through localization) are called anchor nodes (LA) while those who do not know their location are called location unaware or normal nodes (LU). LA nodes are mostly fixed while LU nodes are usually mobile. To estimate their locations, LU nodes first of all collect the connectivity information and the location information of the LA nodes and then they calculate their own location. As compared to range-based algorithms, the range-free algorithms are more cost-effective because no additional ranging devices are needed. This work aims to design a new improved range-free localization algorithm that increases the efficiency of the current range-free technique of localization and is more robust.