Multiview Features Fusion and AdaBoost Based Indoor Localization on WiFi Platform

With the increasing demand for location-based service (LBS), WiFi-based localization has attracted considerable attentions due to the wide deployment and low cost of WiFi devices. However, most of existing approaches need to work within line-of-sight (LoS) range or require users to take dedicated devices. Thus, a device-free and passive indoor localization scheme is still desired. In this paper, a multi-view features fusion and AdaBoost-based indoor localization scheme by exploiting channel state information (CSI) of WiFi signals is proposed. To this end, we firstly formulate a shared consistent representation by using CSI measurements from all WiFi receivers, modeling the common properties among all views. Based on this, a features extraction is applied to the consistent representation for charactering channel properties. Meanwhile, these features extracted are optimized via a t-distributed stochastic neighbor embedding (t-SNE) algorithm for selecting the effective features, contributing to system performance. AdaBoost then builds a non-linear mapping between the CSI features and locations, realizing the purpose of localizing a user relying solely on WiFi signals with a high level of granularity, without any active engagement from the users. We implement a prototype on commodity WiFi devices and conduct comprehensive experiments in indoor scenarios. Based on real-world CSI measurements, the results confirm that the proposed scheme can achieve average localization errors of 0.8 m and 1.1 m in two indoor scenarios.

I. Introduction

With the wide deployment of wireless networks and rapid development of communication technology and Internet of Things (IoT), localization has attracted increasing interest in industry and research community and been widely used for a wide range of intelligent applications [1], [2], such as smart home, health management, and smart guiding. In practice, accurately determining the locations of a user in a home has the potential to significantly enhance in-home experience. In addition, the home assistant can personalize the responses based on where a user is within sensing area and the temperature, for example, can be automatically controlled, thereby resulting in an energy saving. Future home robots would be more intelligent if they know where a user is in the house, providing high-level analytic service for sensing applications such as responsive smart home and health monitoring. Owing to its importance, many indoor localization systems have been proposed with various advanced techniques in recent years and these approaches can be divided into computer vision-based [3]–[6], dedicated device-based [7]–[10] and received signal strength (RSS)-based [11]–[14]. Specifically, these schemes employing dedicated devices worn or held by the users are inconvenient and cumbersome, especially the elderly often forget to charge them or leave them on tables. In addition, these cameras-based schemes cannot work in non-line-of-sight (NLoS) situation and also incur the users’ privacy concerns. In reality, most people are uncomfortable with being filmed all the time, especially in their own homes. Those RSS-based schemes are vulnerable to the channel dynamics and background noises since it is a superposition of multiple signal propagation paths. Therefore, a device-free, passive and robust indoor localization needs to be developed to overcome these remained limitations of current schemes.