Akademik Veri Yönetim Sistemi
SIMULATION-TRANSACTIONS OF THE SOCIETY FOR MODELING AND SIMULATION INTERNATIONAL, cilt.89, sa.6, ss.746-761, 2013 (SCI-Expanded)
In wireless sensor networks, location awareness along with the sensed data is important for most of the applications. Although there are many localization techniques in the literature, some of these techniques involve tiny sensor nodes, which need to be computationally powerful and complex in hardware to estimate their locations. This paper proposes a modified sectoral sweeper-based localization estimation (M-SSLE) method, which is the improved version of our previously studied SSLE method, to achieve position information of sensor nodes. This technique requires only a central node having a smart antenna capability to estimate locations of sensor nodes and simplified message formats. Using the SSLE technique, the computational burden for location estimation is carried out by a central node and no hardware or software modification to tiny sensor devices is required. The M-SSLE is implemented using an OPNET modeler for realistic fading channel conditions. The detailed implementation methodology in OPNET is presented in terms of node and process models. The performance of the M-SSLE method is evaluated through different network scenarios and channel parameters in terms of localization error, average throughput, and node energy consumption. Furthermore, its localization performance is compared with the SSLE and Cramer Rao Bound methods in the two different scenarios for the deployment of the sensor nodes. The M-SSLE average localization errors of 4.3 and 8.9 m are demonstrated under the log-distance and the log-normal shadowing channel conditions in a 250 m x 250 m area, respectively.