A Comparative Performance Evaluation of RSS-Based Positioning Algorithms Used in WLAN Network

f X ( x)Biblioteka =1 2π .σ x.e
−
x2
2 2σ x
(2)
Where σx is defined for different models in Table 2.
Model A B C D E F dBP (m) 5 5 5 10 20 30 α1 before dBP 2 2 2 2 2 2 α2 after dBP 3.5 3.5 3.5 3.5 3.5 3.5 Shadow fading (dB) before dBP
A Comparative Performance Evaluation of RSS-Based Positioning Algorithms Used in WLAN Networks
Ahmad Hatami and Kaveh Pahlavan
Center for Wireless Information Network Studies Worcester Polytechnic Institute (WPI) hatami@ , kaveh@
IEEE Communications Society / WCNC 2005
2331
0-7803-8966-2/05/$20.00 © 2005 IEEE
L(d ) = LFS (d ) L(d ) = LFS (d BP ) − 10.α 2 . log10 ( d )+x d BP
d ≤ d BP d > d BP
I. INTRODUCTION
Proliferation of portable computers and hand held devices opened new paradigms in application domain. Location aware and pervasive computing are among these new applications [7]. Traditional methods for wireless positioning are based on time-of-arrival (TOA) or angle-of-arrival (AOA) measurements from several base stations (BS) and application of triangulation [10, 11]. The Global Positioning System (GPS) is the most widely used positioning system in this class. Unfortunately, there are two limitations in GPS based systems. First, these systems do not provide good accuracy inside buildings and other scenarios in which there is no direct Lineof-Site (LOS) between the mobile host (MH) and base station BS [1,13]. The second limitation is the need for an additional hardware and infrastructure that increases overall cost, weight, and power consumption of mobile devices. In recent years IEEE 802.11 established itself as the standard for WLAN. These networks are used in indoor and short distance outdoor areas with a transmission range between 30-200 m. Large number of existing 802.11 networks in addition to new applications with positioning requirements motivated research community for using WLAN as an alternative to other positioning systems. The main purpose of this article is to provide an overview of power based algorithms that have been used for location estimation in WLAN. Table 1 summarizes the acronyms that are common in literature and are used in subsequent sections.
Acronym
AP AOL BS CCF DV GPS LOS MH NLOS QOS RSS TOA VoIP WLAN
Description
Access Point Angle of Arrival Base Station Complementary Cumulative Function Distance Variance Global Positioning System Line of Site Mobile Host Non Line of Site Quality of Service Received Signal Strength [dBm] Time of Arrival Voice over IP Wireless LAN
(1)
LFS (d ) = L0 + 10 .α 1 log 10 (d )
Where L, d, dBP, α1, α2 represent path loss (dB), distance (m), breakpoint distance (m), power-distance gradient; before and after the breakpoint; respectively. x is the shadow fading component with a zero mean Gaussian probability distribution.
(LOS)
Shadow fading (dB) after dBP
(NLOS)
Consider a typical WLAN system in 100 x 100 m2 building as depicted in Figure 3. In this scenario five access points (AP1– AP5) provide coverage within a building. The goal is to locate a MH who is communicating through one AP in this building. We break down the process of positioning system into the following steps as suggested by Figure 4. Step 1 (Training / Profiling): We create a grid of reference points (known locations) within the network coverage area and collect average RSS values from existing AP’s and store them in a reference database. To reduce the effect of noise and interference in profiling, we measure RSS values multiple times in different directions and use the average value as the reference. This step can be done in two ways. First method is to do onsite measurements in all reference points with a user terminal that involves a rather exhaustive and costly measurement campaign [2]. In an alternative approach we can do limited number of onsite measurements and collect physical characteristics of an appropriate channel model. Then generate the reference database by simulation. The accuracy of this system depends on the accuracy of the underlying channel model.
Table 1 WLAN Positioning Systems Channel Model: Existing channel models for WLAN were developed in [14,15]. Different environments are grouped in five different models (A-E) based on various rms delay spread values: • Model A for a typical office environment, non-lineof-sight (NLOS) conditions, and 50 ns rms delay spread. • Model B for a typical large open space and office environments, NLOS conditions, and 100 ns rms delay spread. • Model C for a large open space (indoor and outdoor), NLOS conditions, and 150 ns rms delay spread. • Model D, same as model C, line-of-sight (LOS) conditions, and 140 ns rms delay spread (10 dB Ricean K-factor at the first delay). • Model E for a typical large open space (indoor and outdoor), NLOS conditions, and 250 ns rms delay spread. In our simulations we use a path loss model introduced in [16,17]. This model consists of a free space loss LFS with power distance gradient of 2 up to a breakpoint distance dBP. This corresponds to model B in [17]: