Wi-Fi coverage characterization or surveying is crucial at the access points (APs) deployment stage. The technician has to deploy APs according to the building’s floor plan, as the distribution of APs and location of obstacles determine the wireless radio (RF) coverage. Once the APs have been deployed, the technician required to analyze the coverage of the existing wireless network and adjust the location of the APs accordingly.

Site surveying is a task of measuring the received signal strength (RSS) from APs at different locations, using these measurements to predict the coverage of the floor plan. This is time consuming and labor intensive, hence has motivates us to propose using a robot to automate the surveying task. Intuitively, the locations of RSS measurements scatte...[ Read more ]

Wi-Fi coverage characterization or surveying is crucial at the access points (APs) deployment stage. The technician has to deploy APs according to the building’s floor plan, as the distribution of APs and location of obstacles determine the wireless radio (RF) coverage. Once the APs have been deployed, the technician required to analyze the coverage of the existing wireless network and adjust the location of the APs accordingly.

Site surveying is a task of measuring the received signal strength (RSS) from APs at different locations, using these measurements to predict the coverage of the floor plan. This is time consuming and labor intensive, hence has motivates us to propose using a robot to automate the surveying task. Intuitively, the locations of RSS measurements scattered over the site should benefit the prediction accuracy compared to very concentrated measurements, conversely robot consumes less energy to travel for the latter case. Therefore the robot should have a strategic way to measure the RSS at different locations, which is a balance between accuracy and energy. We formulate the problem as Markov decision process (MDP) and solve by an approximation technique to tackle the dimensionality issues. Results show that there is a positive performance gain (in terms of accuracy and energy) compared to a random action or random walk strategy (which humans usually do) to measure the RSS.