| In this dissertation, we study third party wireless localization and implement the corresponding toolkit, which is infrastructure free and highly portable. We investigate two third party wireless localization techniques. One is rotating a directional antenna to measure the angle of arrival (AOA) of wireless signals transmitted from a target mobile device, and the other is moving an omnidirectional antenna to localize a wireless mobile device. In these two applications, experienced engineers often utilize over sampling to ensure that sufficient samples of received signal strength (RSS) are collected. However, there has been no effort that defines over sampling of RSS theoretically. For example, to identify the AOA of wireless signals transmitted from a target mobile device, an accepted strategy is to rotate a directional antenna slowly, or rotate that antenna for many rounds, and then choose the direction where the maximum RSS is sensed as the targets AOA. Another example is that, to localize a wireless mobile device, engineers move an omnidirectional antenna slowly and choose the position where the maximum RSS is sensed as the targets location. But no research addresses how to guarantee that the maximum RSS can be picked up during the process of RSS sampling.;We study RSS sampling for accurate and efficient third party wireless localization using directional antenna and omnidirectional antenna. 1. In the case of directional antenna, we have proved that the process that a directional antenna measures RSS of wireless packets while rotating is the process that the radiation pattern of the directional antenna is sampled. We propose a set of novel RSS sampling theory to guarantee that a directional antenna picks up packets producing the maximum RSS while rotating. We derive mathematical models to determine the RSS sampling rate given a targets packet transmission rate and the antennas radiation pattern. Our RSS sampling theory is applicable to various types of directional antennas. 2. In the case of omnidirectional antenna, we address the problem of RSS sampling via moving an omnidirectional antenna. We derive a mathematical model to determine the RSS sampling rate given the targets distance and its packet transmission rate. We have applied our RSS sampling theory to the development of BotLoc, which is a programmable and self-coordinated robot armed with a wireless sniffer. We have conducted extensive experiments to demonstrate the effectiveness of the proposed theorems and the system. A video of BotLoc is at http://www.youtube.com/watch?v=tN1U6k3WOH0. |
