Practical assessment of RSS-based localization in indoor environments

Transmitter localization is a widely and extensively addressed topic in the area of wireless networks. Its broad usage in various applications, such as sensor networks, cognitive radio, military communications etc., continuously drives the need for more robust, more flexible and more efficient localization solutions. The Received Signal Strength (RSS) based methods are most popular in the area due to their simplicity and applicability on off-the-shelf radio hardware. Most of the current work in the area targets only theoretical performance assessment with negligible efforts made to practically evaluate the RSS localization algorithms. The practical evaluation is crucial for gaining an insight into the practical feasibility and applicability of a certain localization technique since it considers all possible factors affecting the accuracy of the localization. Therefore, this paper presents a practical test-bed based performance evaluation of Maximum Likelihood (ML) and Least Squares (LS) RSS based single localization methods. It also compares the obtained test-bed results with extensive simulations. The paper proposes an LS-grid search localization technique that proves to be a sensible solution for indoor environments, offering the best trade-off between complexity and reliability among all inspected techniques.