Wifi hybridisation with pseudolites and repeaters for indoor positioning purposes

Indoor positioning is a very important topic, mainly in terms of continuity of services. This leads to many theoretical and experimental works in this ield using a large range of techniques, from purely GNSS approaches to networks of physical sensors or Wireless Local Area Telecommunication Networks (WLAN). Among all these techniques, GNSS based ones present the advantage of making better use of the satellite receiver, which is considered to be the "best" solution for outdoor applications (even with the current limitations in urban canyon environments). Thus, techniques like High Sensitivity GPS (HS-GPS) or Assisted GPS (A-GPS) have been widely investigated within the satellite community: the results are interesting but do not seem to give a definitive answer to indoor positioning. Pseudolites and repeaters are now solutions that could help in a inal system with good accuracy and wide coverage: studies are being carried out and show encouraging results for both approaches. Because of the large deployment of WLANs for communication purposes, a great deal of work is being carried out on location finding with WLANs in order to ind a way to "complement" the outdoor GNSS based systems with indoor WLAN based positioning. This paper describes a irst approach in hybridising the WLAN positioning and the approaches dealing with indoor positioning based on a local GNSS deployment, i.e. pseudolites and repeaters. At irst time, a description of the basic theoretical aspects is given for the standard case of availability of both WLAN and GNSS signals. Then first numerical applications, using real data, are provided. At a inal stage, the main idea is to analyse the possible hybridisation between a WLAN symbolic positioning [1] which is very simple to implement in a real environment, and a repeater and/or pseudolite positioning [2]-[3]. The symbolic the positioning is no longer concentrating on accuracy, but is working in a more symbolic way. That means that it will no longer be possible to say that you are 2 or 4 meters from such an absolute location, but only that you are within a given area, say conference room A, for example. We showed [1] that the global performance of a symbolic WLAN system is as good as that obtained with heavy power level database extraction methods. On the other hand, we also showed [3] that a repeater system required 4 repeaters in order to allow a 3D indoor positioning. The potential gain of a mixed approach can be considered in different ways such as better accuracy, simplified infrastructure to be deployed or better 3D coverage for example. These aspects are dealt with and new directions of investigation proposed.