Slide: Towards Fast and Accurate Mobile Fingerprinting for Wi-Fi Indoor Positioning Systems

The deployment of Wi-Fi fingerprint-based indoor positioning systems is severely hindered by the lack of an efficient and low-cost way to establish a signal fingerprint database. In this paper, we present a novel fingerprinting method, slide, that can collect fingerprints in a fast and accurate way. Slide uses a commodity flashlight and a smartphone to achieve linear positioning. This allows automatic mapping from the received signal strength to the position on a line, serving as a building block for fingerprinting in general environments. Slide also features a channel-based scanning method, which acquires fingerprint location after each Wi-Fi channel scanning, to mitigate the fingerprint misalignment problem found in the general mobile fingerprinting. Quantitative analysis and experimental results show that slide is faster than the manual fingerprinting method by up to an order of magnitude with comparable positioning accuracy, and is also more efficient than state-of-the-art mobile fingerprinting methods.