Efficient range free localization scheme for mobile wireless sensor networks / Ammar Moh’d Ammar Abuznaid

Ammar Moh’d, Ammar Abuznaid (2017) Efficient range free localization scheme for mobile wireless sensor networks / Ammar Moh’d Ammar Abuznaid. PhD thesis, University of Malaya.

	PDF (The Candidate's Agreement) Restricted to Repository staff only Download (181Kb)
Preview	PDF (Thesis PhD) Download (1911Kb) | Preview

Abstract

In recent years, wireless sensor networks (WSNs) have been used in several applications to present the reality of operational areas. Owing to the small size and low cost of sensors, they can be efficiently used in mobile objects. However, the location of mobile sensors is a challenging issue because of the need to frequently change location per each time slot. Given this requirement, finding efficient localization method is a significant challenge in mobile objects. This study addresses the impact of localization in mobile WSNs by reviewing the state-of-the-art sequential Monte Carlo method under a range-free scheme. The localization process in range-free schemes is conducted using network connectivity. Thus, movable sensors require the sharing of locations to estimate new locations. The power requirement for communication between sensors is higher than that for computation. Therefore, reducing the communication cost in WSNs can prolong network life. The existing range-free schemes use anchor nodes and normal nodes in a neighborhood to estimate the new location of mobile sensors. Using normal nodes in the neighborhood can increase communication cost without improving localization accuracy. An added challenge in mobile WSN localization is the velocity and number of anchor nodes in the neighborhood. Most localization schemes employ the random waypoint mobility model to transmit the location of mobile sensors. The waypoint model produces a large overlap between anchor nodes and identifies more than three anchor nodes in a neighborhood without improving localization accuracy. In this work, we present a localization framework to solve such problems. The proposed framework solves the first problem by selecting an adjacent normal node in the neighborhood, as in the proposed Low Communication Cost (LCC) scheme. The second problem is solved using the adaptive mobility model (AMM), which selects the anchor node velocity as a function of the overlap degree and the number of anchor nodes in the neighborhood. Results show that the proposed LCC scheme can reduce communication costs (the number of messages sent) by a minimum of 0.02 and a maximum of 0.30 with an average of 0.18 for varying node densities of 6 to 20, while nonetheless able to retain similar MSL* localization accuracy rates. Results to solve the second problem on the other hand, show that the proposed AMM improves localization accuracy with an average of 0.05 and a coverage degree of up to 0.50. We evaluate the proposed LCC scheme and AMM through extensive simulation experiments.

Actions (For repository staff only : Login required)