Wireless Indoor Location Estimation Based on Neural Network RSS Signature Recognition (LENSR) Page: 2 of 7
This article is part of the collection entitled: Office of Scientific & Technical Information Technical Reports and was provided to Digital Library by the UNT Libraries Government Documents Department.
The following text was automatically extracted from the image on this page using optical character recognition software:
Wireless Indoor Location Estimation Based on Neural
Network RSS Signature Recognition (LENSR)
Idaho National Laboratory
2525 Freemont Avenue
Idaho Falls, ID 83415, USA
Abstract- Location Based Services (LBS), context aware
applications, and people and object tracking depend on the ability
to locate mobile devices, also known as localization, in the wireless
landscape. Localization enables a diverse set of applications that
include, but are not limited to, vehicle guidance in an industrial
environment, security monitoring, self-guided tours, personalized
communications services, resource tracking, mobile commerce
services, guiding emergency workers during fire emergencies,
habitat monitoring, environmental surveillance, and receiving
This paper presents a new neural network approach (LENSR)
based on a competitive topological Counter Propagation Network
(CPN) with k-nearest neighborhood vector mapping, for indoor
location estimation based on received signal strength. The
advantage of this approach is both speed and accuracy. The tested
accuracy of the algorithm was 90.6% within 1 meter and 96.4%
within 1.5 meters. Several approaches for location estimation
using WLAN technology were reviewed for comparison of results.
Keywords-RSS, localization, neural network, CPN, k-nearest
neighbor, signature recognition, GPS.
Both indoor and outdoor location estimation is a significant
problem posing serious technical challenges. Location
estimation, also known as localization, concerns the
positioning of mobile devices in some physical space.
Although location estimation represents an active area of
research, proposed solutions are generally cost prohibitive,
inaccurate, or infeasible due to practical issues.
GPS and wireless technologies are useful for determining
location. Although GPS supports LBS, the number of mobile
devices using GPS technology available in the marketplace
today is limited. This is due to high cost, power requirements,
and the inability to work in certain environments, such as
indoors, underground, and in city canyons (an area of a city
with narrow streets and high buildings) [1, 2].
This paper focuses on indoor location estimation from radio
signal strength (RSS) values received by a mobile device with
wireless (WiFi) capabilities as the device moves around an
area of a building; e.g., a smart phone, personal digital
assistant, equipment or packages with WiFi sensors, or a robot
with WiFi capabilities. The ubiquity and low cost of 802.11
technology makes localization based on wireless local area
network (WLAN) technology a viable alternative to GPS,
Department of Computer Science
University of Idaho at Idaho Falls
1776 Science Center Dr., Ste.306
Idaho Falls, ID 83402, USA
enhancing the value of the wireless network.
The building has wireless access points (APs) acting as
anchors deployed at various locations. First, a radio grid map
is obtained offline. This way attenuation and reflection of
signals in an urban environment is recorded as it is, resulting in
both computational time savings and precision of recorded
signals. The mobile device then estimates its position
algorithmically using RSS values received from access points
and the grid map. This is entirely a client-based system in that
the mobile device does not send packets to a server in order to
determine the location of the device. WiFi coverage of the
complete area of interest is necessary for accurate location
In this paper we present a new Counter Propagation neural
Network (CPN) with k-Nearest Neighbor (k-NN) algorithm for
location estimation using WiFi received signal strength. To the
best of our knowledge this algorithmic approach has not been
previously used for location estimation. The paper is organized
as follows. Section 2 discusses related work, section 3 reviews
the principles of mobile device location estimation, section 4
describes the proposed CPN with k-NN algorithm, section 5
presents test results computed in a MATLAB environment, and
section 6 states our conclusions.
Wireless localization schemes are generally categorized as
deterministic or probabilistic . The deterministic techniques
are range or proximity based. The range based approach uses
the characteristics of the channel, such as Received Signal
Strength (RSS), to find the distance from a mobile device to
wireless access points. Alternatively RSS fingerprinting
techniques may used to locate a mobile device in a building.
Neural networks, specifically a generalized regression neural
network, have been used as the pattern matching algorithm in
geo-location systems .
The probabilistic technique [5,6,7,8,9] constructs a
conditional probability distribution over some area of interest
to determine the likelihood of a mobile device being at some
position at a specific point in time. Probabilistic techniques are
computationally more expensive than deterministic techniques
but provide a higher degree of accuracy (90% within 2 meters
Some examples of a wireless localization systems and
Here’s what’s next.
This article can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Article.
Derr, Kurt & Manic, Milos. Wireless Indoor Location Estimation Based on Neural Network RSS Signature Recognition (LENSR), article, June 1, 2008; [Idaho Falls, Idaho]. (https://digital.library.unt.edu/ark:/67531/metadc894963/m1/2/: accessed April 19, 2019), University of North Texas Libraries, Digital Library, https://digital.library.unt.edu; crediting UNT Libraries Government Documents Department.