Considerations for the implementation of the radio interferometric positioning system on a single wireless node

Abstract

The ability to localise objects and persons is a useful ability, that is currently used in everyday life in the form of Global Positioning System (GPS) navigation. Localisation is also useful in data networks. The ability to localise nodes in a network paves the way for applications such as location based services, beamforming and geographic routing. The Radio Interferometric Positioning System (RIPS), is a method originally designed for localisation in wireless sensor networks. RIPS is a promising method due to the fact that it is capable of localisation with high accuracy over long ranges. This is something which other existing methods are not capable of. RIPS makes localisation measurements in a different manner from conventional methods. Instead of making pairwise measurements between a transmitter and receiver, RIPS uses sets of four nodes in each of its measurements. Furthermore, RIPS requires multiple measurements to obtain the correct RIPS measurement value. This value is referred to as a q-range. Multiple q-ranges are required in order to localise a node. This creates overhead in terms of co-operation between the nodes participating in a RIPS measurement. The focus of this research is to provide a possible solution to this problem of overhead. In this dissertation an investigation is launched into the considerations and benefits of implementing RIPS on a single node. This is done by creating a conceptual design for a single wireless node capable of implementing RIPS through the use of multiple antennas. In order to test this conceptual device, a simulation model is created. This simulation model is then validated, verified and used in experiments designed to test the effects of certain design considerations and variables on the conceptual device's localisation accuracy. The analysis of the results from these experiments shows that the conceptual device's use of multiple antennas makes RIPS sensitive to errors. Increasing the distances separating the conceptual device's antennas is found to decrease this sensitivity to errors. This is shown to be caused by the distances separating the antennas imposing limits on the range of q-ranges values that are possible, with smaller distances resulting in smaller ranges of possible q-range values. It is also found that the use of higher frequencies in RIPS measurements results in greater accuracy. This is with the assumption that these frequencies can be accurately transmitted.