Secure Localization and Node Placement Strategies for Wireless Networks
Type of DegreeDissertation
Electrical and Computer Engineering
MetadataShow full item record
Localization and placement problems in wireless networks have been widely studied. The adaptation of localization in real world environment can be observed by popular commercial and non-commercial GPS applications. Intelligent placement of base stations in cellular network has led to efficient network deployment. However, with the advent of wireless local area networks (WLAN), ad hoc and sensor networks these problems are again brought into focus. The requirements such as low-cost, nodal resource constrains and multihop characteristics have made difficult problems such as localization and placement very hard to solve for contemporary wireless networks. These problems are considered in this work. A low-cost secure localization scheme is proposed for infrastructure based WLAN. The scheme and its testbed implementation is based on transmission of special messages at different power levels through different access points (APs). Depending on the set of messages received by a user, his/her location is estimated. Measured results have shown this scheme to perform as well as the traditional signal strength based indoor localization schemes. Moreover, the scheme is secure and prevents any intruder from spoofing his location. The problem of node placement is investigated for wireless ad hoc and sensor networks. The placement and interface selection algorithm (PISA) is proposed for placing ΄drones‘ (multi-interface devices) that bridge heterogeneous networks. The objective is to place these ΄drones‘ to maintain a connected network. Furthermore, the problem of placing sophisticated nodes (SNs) in a hierarchical network is considered, given the placement of lower tier nodes called lite nodes (LNs). A HYBRID algorithm comprising of binary integer linear programming (BILP) and genetic algorithm (GA) is proposed to solve this problem. The placement schemes are validated using MATLAB simulations. Moreover, a mobility scheme is proposed to enable auto-configuration of SN placement without the knowledge of LN locations. The scheme is implemented and tested using OPNET. The results of these placement schemes indicate satisfactory placements of nodes for all test cases. These placement schemes demonstrate an increase in network connectivity, capacity and decrease in nodal energy; thus making the network more efficient.