Hybrid Indoor-Based WLAN-WSN Localization Scheme for Improving Accuracy Based on Artificial Neural Network (original) (raw)

Wlan–Based Indoor Localization Using Neural Networks

Journal of Electrical Engineering, 2016

Wireless indoor localization has generated recent research interest due to its numerous applications. This work investigates Wi-Fi based indoor localization using two variants of the fingerprinting approach. Specifically, we study the application of an artificial neural network (ANN) for implementing the fingerprinting approach and compare its localization performance with a probabilistic fingerprinting method that is based on maximum likelihood estimation (MLE) of the user location. We incorporate spatial correlation of fading into our investigations, which is often neglected in simulation studies and leads to erroneous location estimates. The localization performance is quantified in terms of accuracy, precision, robustness, and complexity. Multiple methods for handling the case of missing APs in online stage are investigated. Our results indicate that ANN-based fingerprinting outperforms the probabilistic approach for all performance metrics considered in this work.

Indoor localization techniques using wireless network and artificial neural network

2017

This research focuses on improving indoor localization using wireless network and artificial neural network (ANN). This involves strategic study on wireless signal behavior and propagation inside buildings, suitable propagation model to simulate indoor propagation and evaluations on different localization methods such as distance based, direction based, time based and signature based. It has been identified that indoor signal propagation impairments are severe, non-linear and custom to a specific indoor location. To accommodate these impairments, an ANN is proposed to provide a viable solution for indoor location prediction as it learns the location specific parameters during training, and then performs positioning based on the trained data, while being robust to severe and non-linear propagation effects. The versatility of ANN allows different setup and optimization possibilities to affect location prediction capabilities. This research identified the best feedforward backpropagati...

Neural Network-based Indoor Localization in WSN Environments

With the advancement of wireless technology even more wireless sensor network (WSN) applications are gaining ground. Their field of application is increasingly widening. This paper examines the WSN application which allows indoor localization based on the Fingerprint (FP) method. The communication between the modules was monitored during the experiment whereby the received radio signal strength indicator (RSSI) values from 5 modules were recorded by a mobile sensor. The received data was used for training of the feed-forward type of neural network. Through use of the trained neural network and the measured RSSI values an indoor localization was realized in a real environment. The neural network-based localization method is analyzed applying the cumulative distribution function (CDF). For the reference model the well-known weighted k-nearest neighbour (WkNN) method was used.

A new method for improving Wi-Fi-based indoor positioning accuracy

Wi-Fi and smartphone based positioning technologies are play-ing a more and more important role in Location Based Service (LBS) industries due to the rapid development of the smartphone market. However, the low positioning accuracy of these technologies is still an issue for indoor positioning. To ad-dress this problem, a new method for improving the indoor posi-tioning accuracy was developed. The new method initially used the Nearest Neighbor (NN) algorithm of the fingerprinting meth-od to identify the initial position estimate of the smartphone us-er. Then two distance correction values in two roughly perpen-dicular directions were calculated by the path loss model based on the two signal strength indicator (RSSI) values observed. The errors from the path loss model were eliminated by differencing two model-derived distances from the same access point. The new method was tested and the results compared and assessed against that of the commercial Ekahau RTLS system and the NN algorithm. The preliminary results showed that the positioning accuracy has been improved consistently after the new method was applied and the root mean square accuracy improved to 3.3 m from 3.8 m compared with the NN algorithm.

Fingerprint-Based Localization Approach for WSN Using Machine Learning Models

Applied Sciences

The area of localization in wireless sensor networks (WSNs) has received considerable attention recently, driven by the need to develop an accurate localization system with the minimum cost and energy consumption possible. On the other hand, machine learning (ML) algorithms have been employed widely in several WSN-based applications (data gathering, clustering, energy-harvesting, and node localization) and showed an enhancement in the obtained results. In this paper, an efficient WSN-based fingerprinting localization system for indoor environments based on a low-cost sensor architecture, through establishing an indoor fingerprinting dataset and adopting four tailored ML models, is presented. The proposed system was validated by real experiments conducted in complex indoor environments with several obstacles and walls and achieves an efficient localization accuracy with an average of 1.4 m. In addition, through real experiments, we analyze and discuss the impact of reference point de...

Development of A Wi-Fi Based Indoor Location System Using Artificial Intelligence Techniques

2020

The main aim of this study is to resolve the problem of indoor positioning in closed areas, which has become a growing need nowadays, by using existing hardwaresolutions. Although the use of the GPS system, which requires satellite communication as an open space location solution, is very common, it cannot provide a solution for indoor. It is a well-known metric to measure signal strengths to determine distances between wireless nodes. However, the signal strength is affected by many external influences and causes erroneous measurements. With the developed approach, the transmission powers of the signals received from more than one transmitter located within a certain closed area are measured and given as an input to an artificial neural network. It has been seen that the outputs produced by the trained neural network are much more successful and reliable than the path-loss calculation.

Indoor Positioning System Using Artificial Neural Network

Journal of Computer Science, 2010

Problem statement: Location knowledge in indoor environment using Indoor Positioning Systems (IPS) has become very useful and popular in recent years. A number of Location Based Services (LBS) have been developed, which are based on IPS, these LBS include asset tracking, inventory management and security based applications. Many next-generation LBS applications such as social networking, local search, advertising and geo-tagging are expected to be used in urban and indoor environments where GNSS either underperforms in terms of fix times or accuracy, or fails altogether. To develop an IPS based on Wi-Fi Received Signal Strength (RSS) using Artificial Neural Networks (ANN), which should use already available Wi-Fi infrastructure in a heterogeneous environment. Approach: This study discussed the use of ANN for IPS using RSS in an indoor wireless facility which has varying human activity, material of walls and type of Wireless Access Points (WAP), hence simulating a heterogeneous environment. The proposed system used backpropogation method with 4 input neurons, 2 output neurons and 4 hidden layers. The model was trained with three different types of training data. The accuracy assessment for each training data was performed by computing the distance error and average distance error. Results: The results of the experiments showed that using ANN with the proposed method of collecting training data, maximum accuracy of 0.7 m can be achieved, with 30% of the distance error less than 1 m and 60% of the distance error within the range of 1-2 m. Whereas maximum accuracy of 1.01 can be achieved with the commonly used method of collecting training data. The proposed model also showed 67% more accuracy as compared to a probabilistic model. Conclusion: The results indicated that ANN based IPS can provide accuracy and precision which is quite adequate for the development of indoor LBS while using the already available Wi-Fi infrastructure, also the proposed method for collecting the training data can help in addressing the noise and interference, which are one of the major factors affecting the accuracy of IPS.

Wireless Indoor Localization Using Fingerprinting Technique

Journal of Advanced Research in Dynamical and Control Systems, 2020

Indoor positioning has gained more interest as one of the upcoming applications due to its use in a variety of services. Multiple technologies such as Bluetooth, Wi-Fi, RFID. However, Wi-Fi based localization in indoor environment offers significant advantages utilizing installed wireless infrastructures and good performances with low cost. With this study, we aim to provide a compromise between accurate positioning and feasibility of the system for practical applications. For this purpose, we minimize the fluctuations of Wi-Fi received signal strength (RSS) by filtering and we combine two approaches to locate a mobile user. At first, we implemeted the traditional fingerprinting technique that uses a real time matching of pre-recorded received signal strength (RSS) from the location data of the user transmitted to nearby access points (AP). Secondly, we used distance-based trilateration technique which determines positions using three known access points. The combination of the two methods provides enhancement of accuracy and wide indoor locating coverage. Regardless the locating data number, experiment confirmed a significant and a consistent performance in term of execution time and accuracy.

Location determination of mobile devices for an indoor WLAN application using a neural network

Knowledge and Information Systems, 2009

Due to the popularity of location-based services, determining the location of a device at all times has become a subject of great interests. Although many GPS-based applications have been developed and successfully deployed in various fields, their applicabilities are hindered by the obstruction of the objects in the environment. Essentially, as satellite signals cannot penetrate the walls of buildings, the coverage of GPS systems is limited to outdoor environments. To fully exploit the benefit of location-based services, approaches that determine the location of a device in indoor environments need to be established. This study presents a novel location determination mechanism that uses an indoor WLAN and back-propagation neural network (BPN). A museum is taken as the context of the example indoor environment. Location determination is achieved using the combined strengths of 802.11b wireless access signals. With a significant number of access points (APs) installed in the museum, hand-held devices can sense the strengths of the signals from all APs to which the devices can connect. Using a back-propagation network, device locations can be estimated with sufficient accuracy. A novel adaptive algorithm is implemented for enhancing the accuracy of the estimation.

Indoor 3-Dimensionally Localization of WSNs Using Neural Network

Networks (WSNs) has gained an increasing interest. WSN consist of several sensor nodes that measures data such as temperature, pressure, humidity etc and finally transmits those data to the base station. Data transmitted by sensor nodes are of importance only when the location of the particular sensor node in WSN is known. Idea originated from various researches and papers of localization methods in which all methods were flexible in localizing the sensor nodes when they were placed on the ground only. But as soon as we locate same sensor at some height then this leads to errors and predicts wrong location. This project's technique can be used to get location in both cases when sensor nodes are present on floor or space thus can be referred as improved 3dimenssionally localization with more than 90-96% accuracy. Aim of our project is to make indoor localization system cost effective by limiting use of GPS nodes in WSNs and to increase its accuracy by use of machine learning neural network.