Real-Time Locating Systems (RTLS) belong to a class of locating systems capable of remotely determining the location of tagged assets in an environment within a relatively short time frame (real-time). Signals from tags are received wirelessly by readers, which use information about the signal, to continuously determine the location of the tag relative to each reader. Using readers positioned at several locations around a tag offers multiple location readings, which can be compiled together at a central host to estimate the location of a tag in a two or three dimensional space. Location information from several vantage points also reduces the chance of false location estimates of a tag when one or more readers' location estimates contain some error.Indoor asset tracking is a unique specialty of many types of RTLS, as Global Positioning Systems (GPS) have difficulty providing asset locations within buildings. RTLS deployments generally address the problems of determining the location of equipment and products in a warehouse or even tracking employees and files in an office. The most common RTLS uses a Wi-Fi infrastructure to determine the location of tagged assets. These systems employ Time-of-Arrival (ToA) or Received Signal Strength Indication (RSSI) techniques to wirelessly determine asset location. Scaling up a system that uses a Wi-Fi infrastructure to localize large numbers of tags introduces a significant wireless traffic burden, reducing data throughput for other users of the same Wi-Fi network. Both ToA and RSSI methods of localization suffer from significant inaccuracies in reported location of assets and often are unable to determine which room an asset resides.This thesis presents a new method for determining asset location that offers better accuracy than systems using ToA and RSSI techniques, and an RTLS that uses existing Wi-Fi infrastructures and commercial off-the-shelf parts (COTS) without placing excessive need on the Wi-Fi networks.
展开▼
