Advanced Position Fixing and Dead Reckoning Techniques: CooperativeGPS Positioning and Context-awarePedestrian Dead Reckoning

Nowadays, always more applications rely on positioning, such as location based services (LBS), intelligent transportation systems (ITS), constructions, Earth science, machine control, surveillance, mapping, and navigation. All navigation and positioning techniques are based on one of two fundamental methods: position xing and dead reckoning. Position xing aims to determine the position using external information such as signals or envi- ronmental features. Alternatively, dead reckoning measures the distance and direction traveled from a previous position to determine the current position. One of the most popular signal-based positioning system using positioning, is the Global Position System (GPS). The latter is one of the global navigation satellite systems (GNSS) available nowadays. The behaviour of these systems is remarkable when the line-of-sight between the satellites and the GNSS receiver is not disturbed. In harsh environments, such as dense urban canyons, light shadowing, or more generally when GNSS signals cannot be received with a suciently high signal-to-noise ratio (SNR) due to obstructions, the performance of GNSS declines. Pedestrian navigation is one of the most challenging applications of navigation technology. Pedestrian dead reckoning is a promising solution for pedestrian navigation. It uses the signal that has been collected by sensors to detect steps and to estimate the step lengths. Step detection and step length estimation algorithms work relatively well when walking in a straight line; however, real paths include turns, sidesteps, stairs, variations in speed, and various actions performed by the subject. These all affect the step detection and the step length estimation, and must be considered to provide an accurate estimate of the distance traveled during daily activities as well as laboratory tests. Chapter 2 gives an overview of the position xing techniques. Chapter 3 focuses on the GPS system and its main characteristics. Chapter 4 presents a peer-to-peer cooperative GPS positioning technique together with its performance assessment. Chapter 5 gives an overview of dead reckoning navigation. Chapter 6 proposes a weighted context-based step length estimation technique as a part of a pedestrian dead reckoning system, which comprises step detection, step mode classication, and context classication. Chapter 7 presents the conclusion of this thesis.