Green wireless local area network received signal strength dimensionality reduction and indoor localization based on fingerprint algorithm

Abstract

SUMMARYGreen wireless local area network (WLAN) is an emerging technology to achieve both the purposes of power conservation and high-speed accessing to the Internet because of the working on-demand strategy adoption and high density access points (APs) deployment. Although it is good news to data traffic service, Green WLAN brings severe challenges to the indoor localization service based on fingerprint algorithm. Redundant APs will greatly enlarge the radio map and introduce a much heavier computation burden to the terminal for localization in the online phase. In addition, APs in Green WLAN are powered on and off to make balances between data traffic service demand and energy saving goals so that the received signal strength (RSS) sampled online and recorded in the radio map offline are rarely matched in the same detected AP number, which leads to asymmetric matching problem occurring in the fingerprint algorithm. In this paper, we propose to make a nonlinear dimensionality reduction on the RSS by local discriminant embedding algorithm to realize both the computation burden decreasing and asymmetric matching problem resolving for the fingerprint algorithm in Green WLAN. The simulation results show that our proposed methods could effectively reduce the computation burden in the online phase and make the fingerprint algorithm operate more robustly when the RSS is reduced to the intrinsic dimensionality in Green WLAN.

Random Forest Based Coarse Locating and KPCA Feature Extraction for Indoor Positioning System

“…For the fine positioning performance, according to Figure 8, the accuracy of WKNN fingerprinting method performs roughly the same as in the simulated strong noisy environment (compared with Figure 5). Under the same condition that is set at 4 and is 8, the proposed system based on KPCA still ranks top with 87% confidence probability while the positioning error is within 2 meters, followed by the positioning systems based on PCA, LDE, and LDA algorithms, respectively [8,13,17]. The result verifies the robustness and effectiveness of the proposed system.…”

“…ISOMAP presents capability of representing dataset in global, but it is based on the hypothesis that the high dimensional dataset is isometric to a convex subset of Euclidean space. For the LDE algorithm, it indeed shows a comparable performance in terms of feature extraction and efficiency of dimensional reduction compared with others according to our early works [15,17]. However the shortage of the local method still exists in LDE, in addition to the unsettled optimal number of neighbors and the most appropriate clustering method within LDE.…”

Power boosting factor determination in impulse postfix‐OFDM‐based positioning system

“…However, the performance enhancement is bounded; essentially, the BER performance may be even degraded because of the power insufficient of data signal if p is ever high, occupying too much transmit power and decreasing theγ b as presented in Equation (10).…”

“…Received signal strength-based position scheme attracts an enormous amount of researchers' attention as the solution for indoor position for its superiorities of low cost and practicality. However, existing received signal strength-based position schemes have a capricious and poor performance even though multifarious algorithms are proposed for improving the accuracy [10,11]. Besides, the conventional joint TOA/AOA-based position system [12] has fine accuracy but still remains stuck at the theoretical analysis stage on count of complicated system organization and algorithms.…”

Using Dimensionality Reduction Techniques for Refining Passive Indoor Positioning Systems Based on Radio Fingerprinting

“…Fang et al [4] show that the size of training samples can be reduced in an uncorrelated space, and they make use, for example, of PCA (Principal Component Analysis) [5] to improve the positioning accuracy. Ma et al [6] employ local discriminant embedding [7], a variant of the more classical and widely known LDA technique [8], which is itself an improvement of PCA when treating with supervised data, i.e., when fingerprinting data are labeled by zones, to reduce both the dimensionality of the training radio map and the fingerprints obtained during the on-line phase. These works are focused on the use of dimensionality reduction to improve the accuracy and efficiency of the estimations.…”