Radio Mapping for Indoor Environments

The efficient deployment and robust operation of many sensor network applications depend on deploying relays to ensure wireless coverage. Radio mapping aims to predict network coverage based on a small number of link measurements from sampled locations. Radio mapping is particularly challenging in complex indoor environments where walls significantly affect radio signal propagation. This paper makes the following key contributions to indoor radio mapping. First, our empirical study in an office building identifies a wall-classification model as the most effective model for indoor environments due to its balance between model complexity and accuracy. Second, we propose a practical algorithm to predict the Reception Signal Strength (RSS) of links in an indoor environment based on a small number of measurements at sampled locations. A key novelty of our algorithm lies in its capability to automatically classify walls into a small number of classes with different degrees of signal attenuation. Finally, we present a practical Radio Mapping Tool that can predict the coverage areas of relays based on a small number of link quality measurements in the environment. Empirical evaluation in an office building demonstrates that the Radio Mapping Tool reduces the false positive rate by as much as 41% compared to the classical log-normal radio propagation model, with a false negative rate of 9% based on sampling only 20% of the locations of interest. Type of Report: Other Department of Computer Science & Engineering Washington University in St. Louis Campus Box 1045 St. Louis, MO 63130 ph: (314) 935-6160 Radio Mapping for Indoor Environments Octav Chipara, Gregory Hackmann, Chenyang Lu, William D. Smart, Gruia-Catalin Roman Department of Computer Science and Engineering Washington University in St. Louis, USA {ochipara, gwh2, lu, wds, roman}@cse.wustl.edu Abstract The efficient deployment and robust operation of many sensor network applications depend on deploying relays to ensure wireless coverage. Radio mapping aims to predict network coverage based on a small number of link measurements from sampled locations. Radio mapping is particularly challenging in complex indoor environments where walls significantly affect radio signal propagation. This paper makes the following key contributions to indoor radio mapping. First, our empirical study in an office building identifies a wall-classification model as the most effective model for indoor environments due to its balance between model complexity and accuracy. Second, we propose a practical algorithm to predict the Reception Signal Strength (RSS) of links in an indoor environment based on a small number of measurements at sampled locations. A key novelty of our algorithm lies in its capability to automatically classify walls into a small number of classes with different degrees of signal attenuation. Finally, we present a practical Radio Mapping Tool that can predict the coverage areas of relays based on a small number of link quality measurements in the environment. Empirical evaluation in an office building demonstrates that the Radio Mapping Tool reduces the false positive rate by as much as 41% compared to the classical log-normal radio propagation model, with a false negative rate of 9% based on sampling only 20% of the locations of interest.The efficient deployment and robust operation of many sensor network applications depend on deploying relays to ensure wireless coverage. Radio mapping aims to predict network coverage based on a small number of link measurements from sampled locations. Radio mapping is particularly challenging in complex indoor environments where walls significantly affect radio signal propagation. This paper makes the following key contributions to indoor radio mapping. First, our empirical study in an office building identifies a wall-classification model as the most effective model for indoor environments due to its balance between model complexity and accuracy. Second, we propose a practical algorithm to predict the Reception Signal Strength (RSS) of links in an indoor environment based on a small number of measurements at sampled locations. A key novelty of our algorithm lies in its capability to automatically classify walls into a small number of classes with different degrees of signal attenuation. Finally, we present a practical Radio Mapping Tool that can predict the coverage areas of relays based on a small number of link quality measurements in the environment. Empirical evaluation in an office building demonstrates that the Radio Mapping Tool reduces the false positive rate by as much as 41% compared to the classical log-normal radio propagation model, with a false negative rate of 9% based on sampling only 20% of the locations of interest.