WiFi Radar: Design and Implementation of an Infrastructure-less Location Tracking System for Pervasive Environment Sheikh I. Ahamed, Nilothpal Talukder and Mehrab Monjur MSCS Department, Marquette University, Milwaukee, WI, USA {iq,ntalukder,mmonjur}@mscs.mu.ed Abstract We design and implement a low cost, easy to deploy and lightweight location tracking solution that operates without the need for a fixed infrastructure. We call it “WiFi Radar,” which uses the signal strength of radio transmissions to determine the position (distance and direction) of devices with near linear approximation. Notwithstanding the challenges that Radio Frequency Signals pose for location determination, the accuracy and precision of our system is relatively high. Our application is user-friendly, customizable and shows graphical as well as list views of the located objects. To the best of our knowledge, our system is the first endeavor that implements a location tracking system without the use of any fixed infrastructure and it may be the cheapest solution built so far. 1 Introduction The prevalent and pervasive use of handheld devices like PDAs, smart-phones and tags like RFID has ushered a new era of location-based services. This has important implications for deployment opportunities in environments, such as, the healthcare sector, manufacturing plants, container terminals, transport facilities, household utilities, and emergency systems. There are various high-tech location-based systems for human, asset, and object tracking. All of them deploy large static infrastructures for accuracy and precision of the information. Though a central infrastructure can address the challenges of precisely detecting an object, it goes against the pervasive environment, which is supposed to be temporally formed, and operated by the nodes themselves. These devices, in reality, have low computational capability and as they need to perform real-time interaction with users, it is hard to fit any complex mathematical model for device tracking in the smart space. Again, the presence of absorbents in the environment makes indoor radio propagation difficult. Therefore, it makes the behavior of the indoor location detection different from that of outdoor. In this paper, we model an infrastructure-less location-detection solution for both indoor and outdoor environment. Location detection systems for both indoor and outdoor environments with the use of high-tech infrastructure are well researched previously. The Global Positioning System (GPS), requires four GPS satellites to calculate positions and synchronize clocks [1].The Cricket [2] developed by MIT uses ultrasound emitters as the infrastructure and embeds receivers in the object being located. Microsoft Research group has developed RADAR, a building-wide tracking system based on the WaveLAN wireless networking technology [3]. For indoor purview, the Active Badge (uses infrared) [4] and the Bat (uses ultrasonic) [4, 6] provide high precision, however, deployment cost of sensors and maintenance are the main disadvantages of these systems. We are motivated by such works but our contribution lies in fulfilling pervasive requirements, such as the follows: • Our solution is deployable without the use of costly sensors and fixed infrastructures. • We use linear approximation to make it lightweight. • We considered the nature of received signal strength indicator (RSSI) for both indoor and outdoor environments and developed two separate equations. • For accuracy, fault tolerance, and to compromise noisy data, we define several error handling techniques. We organize the paper as follows: Section 2 details of our approach, section 3 shows our error handling techniques for accuracy, section 4 evaluates our implementation, and section 5 concludes with future direction. 2 Our Approach We measure the distance and direction of the Wi-Fi Access Points (APs) by walking in two different directions. Our approach has four major sections: a) Samples Collection Phase, b) Distance Determination from received signal strength, c) Direction from Near Linear Approximation of the Samples, and d) Handling Errors for Accuracy. 2.1 Samples Collection and Distance Measurement In our previous work in Smart Tracker [5], we gave details of the samples collection procedure and the empirical model to calibrate distance. The user makes a guided traversal in two directions facilitated by the user-friendly GUI. The choice of the direction in first phase is arbitrary; he can go east west or to north-south direction but second phase is opposite to the direction of the first phase.