Toward System Implementation and Data Analysis for Crowdsensing Based Outdoor RSS Maps

With the explosive usage of smart mobile devices, sustainable access to wireless networks (e.g., Wi-Fi) has become a pervasive demand. Most mobile users expect seamless network connection with low cost. Indeed, this can be achieved by using an accurate received signal strength (RSS) map of wireless access points. While existing methods are either costly or unscalable, the recently emerged mobile crowdsensing (MCS) paradigm is a promising technique for building RSS maps. MCS applications leverage pervasive mobile devices to collaboratively collect data. However, the heterogeneity of devices and the mobility of users could cause inherent noises and blank spots in collected data set. In this paper, we study how to: 1) tame the sensing noises from heterogenous mobile devices and 2) construct accurate and complete RSS maps with random mobility of crowdsensing participants. First, we build a mobile crowdsensing system called iMap to collect RSS measurements with heterogeneous mobile devices. Second, through observing experimental results, we build statistical models of sensing noises and derive different parameters for each kind of mobile device. Third, we present the signal transmission model with measurement error model, and we propose a novel signal recovery scheme to construct accurate and complete RSS maps. The evaluation results show that the proposed method can achieve 90% and 95% recovery rate in geographic coordinate system and polar coordinate system, respectively.