Jochen Seitz

Wi-Fi positioning: System considerations and device calibration

Due to an increasing number of public and private access points in indoor and urban environments, Wi-Fi positioning becomes more and more attractive for pedestrian navigation. In the last ten years different approaches and solutions have been developed. But Wi-Fi hardware and network protocols have not been designed for positioning. Therefore, Wi-Fi devices have different hardware characteristics that lead to different positioning accuracies. In this article we analyze and discuss hardware characteristics of Wi-Fi devices with a focus on the so called Wi-Fi fingerprinting technique for positioning. The analysis is based on measurements collected using a static setup in an anechoic chamber to minimize signal reflections and noise. Characteristics like measurement offsets and practical polling intervals of different mobile devices have been examined. Based on this analysis a calibration approach to compensate the measurement offsets of Wi-Fi devices is proposed. Experimental results in a typically office building are presented to evaluate the improvement in localization accuracy using the calibration approach.

A Hidden Markov Model for urban navigation based on fingerprinting and pedestrian dead reckoning

An algorithm for pedestrian navigation in indoor and urban canyon environments is presented. It considers platforms with low processing power and low-cost sensors. A combination of Wi-Fi positioning and dead reckoning, based on a Hidden Markov Model, is used. The positions of the Wi-Fi fingerprints in the database are used as hidden states. Dead reckoning is taken for state transition and a database correlation of the Wi-Fi signal strength measurements is performed in the measurement update. The dead reckoning consists of an accelerometer driven step length estimation and a magnetic field based heading calculation. Simulations and tests demonstrate that in this way ambiguities common in Wi-Fi positioning can be solved and outages can be bridged. Therefore, higher accuracy and robustness can be achieved.

A Hidden Markov Model for pedestrian navigation

We present an algorithm for pedestrian navigation optimized for smart mobile platforms using the present low-cost sensors and the limited processing power. The algorithm is based on a Hidden Markov Model that combines Wi-Fi positioning and dead reckoning. The hidden states are the positions of the Wi-Fi fingerprints in the database. The state transition includes dead reckoning based on step length estimation from acceleration measurements and compass heading calculated from magnetic field measurements. In the measurement update a database correlation of the actual Wi-Fi signal strength measurements with the stored values in the fingerprints has been performed. In simulations and tests we demonstrate that in this way ambiguities common in Wi-Fi positioning can be reduced. Therefor, higher accuracy and robustness can be achieved.

UWB feature localization for imaging

Object recognition, localization and imaging capabilities of ultra-wideband signals are demonstrated in this contribution. Resembling a bat, the experimental setup comprises one transmitter and two receivers. Instead of ultrasound, electromagnetic ultra-wideband (UWB) signals are used to recognize and localize features of the surroundings like e.g. walls, edges or corners. Each feature corresponds to a hypothesis on the dominant reflection and scattering conditions described in terms of simple ray-tracing relationships. The hypothesis matching best the propagation time measurements is picked to identify and localize the feature in front of the moving UWB bat. Henceforth these features could be used as landmarks for bat navigation. Both, position and attitude of the UWB bat could be continuously updated based on a map of landmarks. This kind of navigation is well-known in the literature as simultaneous localization and mapping. Since position and attitude of the UWB bat are tracked, an image of the vicinity in front of the UWB bat can be compiled by superimposing the impulse responses recorded at the various locations. This contribution describes the pivotal elements to this approach: feature recognition and localization as well as bat-type imaging.

Wi-Fi azimuth and position tracking using directional received signal strength measurements

A new approach for estimating and tracking the azimuth angle regarding north and a two-dimensional position of a mobile unit is presented. Outdoors, the azimuth angle of a device can be easily detected using an electronic compass and the position can be calculated using a global navigation satellite system (GNSS). Indoors, magnetic disturbances lead to unreliable compass outputs. Also, indoors there exists no standard positioning system like GNSS outdoors. The presented approach is based on Wi-Fi signal strength measurements collected by four horizontally arranged directional antennas. To proof the concept the well known Wi-Fi fingerprinting based on the normalized Euclidean distance in signal space has been adopted. A test with measurements collected in a laboratory demonstrates the feasibility of the approach. Especially in indoor environments this facilitates the use of electronic guides that offer additional information by means of augmented reality, e.g. on museum exhibits in visual range.

Wi-Fi Attitude and Position Tracking

An approach for pedestrian navigation in indoor environments is presented. It addresses mobile platforms with low processing power and low-cost sensors. Outdoors the horizontal attitude of a device can be easily detected using electronic compasses. Indoors magnetic disturbances lead to unreliable compass outputs. In this paper a novel approach for attitude and position tracking is introduced. Four horizontally arranged directional antennas are used to collect the Wi-Fi signal strengths of transmitters (access points) in range. For attitude estimation an extended Kalman filter is used, and for position tracking Wi-Fi fingerprinting. With this approach the attitude of a mobile device can be estimated and the position can be tracked in indoor environments like e.g. museums. This enables the use of electronic guides that offer additional information by means of augmented reality on exhibits in visual range. Possible accuracies are evaluated in simulations. A test with measurements collected in a museum demonstrates the functionality of the approach.