Conference PaperPDF Available

Security systems with seamless authentication based on smart phones and surveillance cameras

Authors:

Abstract

A Security system is an essential part of any large infrastructure. It usually comprises of two main parts, an access control system for managing the access to areas and resources within the organisation and a camera surveillance system monitoring unauthorised movement. Both systems are often totally independent despite obvious perform-ance and cost benefits when integrated at the physical level as well as at the logical level. However, before being able to integrate both systems and leverage their complementarities for the access control, a new type of creden-tial has to be introduced, the user's physical position. The user's physical position and his/her past trajectory can support the seamless authentication. This paper claims that the high proliferation of smart phones, Wi-Fi net-works and camera surveillance systems in corporate environments allows the introduction of mobile access cre-dentials, which brings the user's continuous position into the security system as an essential part of the decision process of unobtrusive seamless access authorisation. This will increase the user's working comfort and deliver a higher level of security and system accountability. The paper describes the essential system components which enable this seamless authentication from the smart phone and surveillance camera based localisation subsystems to the architecture for seamless data fusion prosed to work in the cloud in order to achieve high levels of system scalability.
Security systems with seamless authentication based on smart
phones and surveillance cameras
Martin Klepal, Cork Institute of Technology, Ireland
Christian Beder, Cork Institute of Technology, Ireland
Abstract
A Security system is an essential part of any large infrastructure. It usually comprises of two main parts, an access
control system for managing the access to areas and resources within the organisation and a camera surveillance
system monitoring unauthorised movement. Both systems are often totally independent despite obvious perform-
ance and cost benefits when integrated at the physical level as well as at the logical level. However, before being
able to integrate both systems and leverage their complementarities for the access control, a new type of creden-
tial has to be introduced, the user’s physical position. The user’s physical position and his/her past trajectory can
support the seamless authentication. This paper claims that the high proliferation of smart phones, Wi-Fi net-
works and camera surveillance systems in corporate environments allows the introduction of mobile access cre-
dentials, which brings the user’s continuous position into the security system as an essential part of the decision
process of unobtrusive seamless access authorisation. This will increase the user’s working comfort and deliver a
higher level of security and system accountability. The paper describes the essential system components which
enable this seamless authentication from the smart phone and surveillance camera based localisation subsystems
to the architecture for seamless data fusion prosed to work in the cloud in order to achieve high levels of system
scalability.
1 Introduction
A Security system is an essential part of any large in-
frastructure. It usually comprises of two main parts, an
access control system and a camera surveillance sys-
tem. The access control system provides the infra-
structure that enables an organisation to manage ac-
cess to areas and resources within the organisation.
For large infrastructures the access control system
typically includes fixed security devices wired or
wirelessly connected to the central security system and
mobile credentials carried by users to access the areas
(e.g. buildings, offices, manufacturing sites) and re-
sources (e.g. network, data sources). The camera sur-
veillance system monitors the security areas and de-
tects unauthorised movement in those areas.
In any larger organisation many groups of users and
stakeholders exist and have to share the security sys-
tems and collaborate on the access control policy
definition. With the increasing size and complexity of
the premises possibly thousands of devices and users
with various types of credentials exist and are often
required to carry mobile authentication credentials
ranging from ordinary keys to RFID cards and fobs.
To mitigate the problem with the lost and stolen sin-
gle-factor credentials, two-factor authentication by
RFID card with PIN have been gradually enforced.
However, it often results in the requirement to carry
multiple credentials with associated PINs and cumber-
some and lengthy authentication procedures every
time a user needs to access other security areas or re-
sources.
Currently, access control and surveillance systems are
often totally independent despite obvious performance
and cost benefits when integrated at the physical level
(common H/W and S/W infrastructure) as well as at
the logical level (data fusion and business operation).
However, before being able to leverage both systems
complementarities, mobile access credentials have to
become a part of a seamless localisation and tracking
system to be remotely tracked throughout the premise
and the camera surveillance system must be extended
by the target identification and tracking capability.
This integration enables the seamless authentication,
where users provide authentication credentials only at
the entrance to the promise and then their movement
trajectories as obtained by the integrated tracking and
surveillance system are used for subsequent authenti-
cation. For this continuous users’ trajectories are con-
sidered in the follow up automatic seamless authenti-
cation as users require an access to other security ar-
eas and resources without being forced to enter the
credentials again increasing working comfort and
bringing a higher level of security and system ac-
countability.
As the organisations are organically growing the ex-
pansion and adaptation of security systems infrastruc-
ture is often patchy and lacks an adequate formal de-
sign methodology, which makes the expansion of se-
curity systems expensive and prone to possible
security flaws in the systems. The software tools for
integrated security system deployment design, com-
missioning and policy access definition is therefore a
critical part towards a security system of the future.
This paper will present the challenges and advances
made in the area of seamless authentication achieved
at the Nimbus Centre for Embedded Systems, Cork,
Ireland, in collaboration with various industry and
academic partners. The paper will cover research on
the enablers which effectively tackle above problems
including the ubiquitous smart phone based tracking
systems, localisation and video data fusion, underlying
embedded middleware and software tools for design-
ing the security systems infrastructure.
The rest of the paper will deal with the system scal-
able architecture suitable for computation in the cloud
(Chapter 2). Then the localisation sub-systems based
on the smartphone and surveillance camera network
are briefly described in Chapters 3 and 4, respectively,
with the references to the detailed implementations.
Finally the paper summarises advances made in the
field of fusion of both complementary localisation
subsystems and its impact on the data fusion middle-
ware implementation in the chapters 5 and 6, respec-
tively.
2 System Architecture
The system architecture (Figure 1) uses the service-
oriented approach to cope with increasing demands on
reusability across various environments and platforms
and to scale up to serve a large number of various
situation-aware security applications.
Figure 1 System Architecture
The architecture can be split into three logical parts,
the Sub-systems being any sensor and actuator towards
the real world, the Data Fusion Middleware process-
ing the sub-system sensory data and providing a trans-
parent service API towards multiple Situation Aware
Security Applications. The next description focuses on
the middleware and subsystem part especially because
this is considered the main contribution of the paper.
Sub-systems:
There are three essential subsystem which have to be
integrated to provide seamless authentication. The
Smart Phone Based Localisation Sub-system, Video
surveillance Sub-system and others security related
systems such as the electronic Door Access System.
Additionally any other sensor being already installed
on site can be integrated and leveraged for improved
system performance such as the widely used Passive
Presence Detectors. More details will be given in the
following chapters.
Data Fusion Middleware:
The Data Fusion Middleware covers all core func-
tionality of the system. It is responsible for the selec-
tion and fusion of suitable location related data as
provided by all connected sub-systems for estimating
the seamless trajectories of all users carrying the mo-
bile credentials/smart phones. In this process it trans-
parently hides all the complexity related to the under-
lining subsystems infrastructure from Situation Aware
Security Applications.
Apart from the seamless and near real time estimation
of users’ location, the critical requirement on the ar-
chitecture is its ability to scale with the number of us-
ers, avoiding any potential processing bottleneck.
Therefore, the proposed architecture utilise the recent
advancement in cloud computing where individual
components are fully decoupled communicating
through a universal cloud messaging service imple-
mented using RabbitMQ [1]. The system is thus able
to replicate its functional components especially its
fusion engines across the corporate cloud without in-
troducing any latency.
All data as proved by the subsystem as well as the
output seamless authentication is logged for the pur-
pose of subsequent Data Mining and security event
analysis as potentially requested by the Context Aware
Security Applications.
Situation Aware Security Application
There may be multiple security applications for differ-
ent stakeholders sharing the same physical space or
building. The Service API and the middleware have to
provide a good data isolation to ensure high levels of
privacy.
Context
Data
Processing
Subsystem Interface
Dispatcher
Data
Logger
Location Fusion
Engine
Manager
Fusion
Engines
Fusion
Engines
Fusion
Engines
Fusion
Engines
Image Fusion
Engine
Manager
Fusion
Engines
Fusion
Engines
Fusion
Engines
Fusion
Engines
Context, Location and Image Data Fusion
Configuration
Platform
Manager
Cloud
Massaging
& Scaling
Subsystem
Manager
Data Mining
Event Broker
Service API
Situation Aware Security Application
Situation Aware Security Application
Situation Aware Security Application
Meta data
Extraction
Video Surveillance
Sub-systems
Video
Compression
Meta Data
Extraction
Smart
Phone Based
Localisation Sub-system
User’s Authen-
tication key
Passive IR
Detectors
Other Security Context Data
Sub-systems
Door Access
System
Data Fusion
Mi
d
dleware
Applications
Sub
-
systems
3 Smart Phone based indoor lo-
cation system
The localisation system is a phone-centric approach
which utilises all location related information and ex-
isting signals readily available in any information rich
device. In contrast to most other indoor localisation
systems, it does not require a fixed dedicated infra-
structure to be installed in the environment making it a
truly ubiquitous localisation service.
The readily available information depending on the
phone capability is typically a subset or all of the fol-
lowing: GSM/UMTS signal strength, Wi-Fi signal
strength, GPS, reading from embedded accelerome-
ters, compass and Bluetooth proximity information.
The reliability and availability of input information
depends strongly on the actual character of the mobile
client’s physical environment. When the client is out-
doors GPS with GSM/UMTS is a favourable choice
typically combined with pedometry data derived from
the 3D acceleration and compass measurements.
When the client is in an indoor environment the Wi-Fi
signal strength combined with pedometry data per-
form best. If indoor floor plan layouts are available a
map filtering algorithm [5m] can further contribute to
the location estimation reliability.
The fusion engine runs a nonlinear recursive Bayesian
filter to seamlessly combine all incoming location in-
formation. A discrete Bayesian filter is implemented
using the Sequential Monte Carlo Method known as
Particle Filter [5]. A particle filter is based on a set of
random samples with weights to represent the prob-
ability density of the mobile client’s location.
Within the Particle Filter the location related meas-
urements are translated into likelihood observation
functions, which differ for each localisation method
and technology considered. Implemented localisation
methods and technologies include the following:
Wi-Fi is a well-established technology nowadays
and the our smart phone based localisation system
benefits from the omnipresence of Wi-Fi networks
and the advanced RSSI based fusion methods pre-
sented in [4-6]. Depending on the availability of
signal strength fingerprints and the level of detail
of the environment description, either of the fol-
lowing localisation methods are used: fingerprint-
ing (and its variants) described in [5-6], multi-
lateration, or simple proximity
Outdoor GPS and 3G network based localisation is
provided as inherent capability by the majority of
smart phones already and there is no need to dis-
cuss it in any more detail.
The sensory information complementing the pre-
vious technologies are pedometry data estimated
from acceleration and compass measurement. Pe-
dometry data contributes to the seamless transition
between outdoor and indoor localisation. Pedome-
try data consists of number of detected steps and
their length and direction [2-3].
The availability of Bluetooth signal can also be
used to detect the proximity of other Bluetooth en-
abled devices. If such a relation is detected, our
smartphone based system can incorporate this
knowledge in its location estimation process.
4 Surveillance camera based lo-
cation system
One of the major drawbacks of all tag-based localisa-
tion systems, like the smartphone based solution pre-
sented in the previous chapter, is the fact that those
tags need to be carried at all times. Basically not the
people moving in the space are tracked by those sys-
tems but only the devices carried by them. This might
pose a problem especially in the context of security
applications.
For this reason large camera installations are very
popular in the surveillance domain and part of nearly
all large infrastructures. However, apart from very
limited applications for instance in the domain of car
number plate recognition [7] fully automated data
analysis, especially for the task of people tracking, is
not widely used, yet. The reason for this is the fact
that the identification problem, i.e. knowing exactly
who or what is in front of the camera, is still largely
unsolved for generic applications. But because this
particular part of the problem is solved very easily by
the tag-based systems, fusion of both systems is a very
promising approach to overcome the disadvantages of
each of the systems and benefitting from their mutual
advantages.
Because we intend to fuse the information from the
tag-based system and the camera based system we do
not focus on the identification part of the problem but
use the camera network basically for the task of occu-
pancy detection only. Combined with the smartphone
localisation system, which readily provides the identi-
fication information, the cameras help to increase the
localisation accuracy as well as provide a means for
triggering alarms if an unidentified object is detected
within the building. The latter ability will increase the
degree of automation by not requiring any interven-
tion in case of joint detection by the tag-based and the
camera-based system while at the same time allowing
to detect untagged people moving around the envi-
ronment and thereby providing an increased level of
security compared to solely tag-based systems.
Camera based occupancy detection as we understand
it is a two-stage process: first the foreground objects
are extracted from the images [8] and then the camera
calibration and orientation is used to extrude the ex-
tracted silhouette of the foreground object into 3d
space providing an occupancy hypothesis for the sub-
sequent information fusion. In this paper we are not
going to elaborate on the details of either of those to
tasks, but note that for the task of foreground back-
ground discrimination illumination changes posed a
major challenge to us [9] and that camera calibration
and uncertainty propagation [10] is used to forward
project the extracted image foreground object into 3d
space allowing for the geometry based statistical rea-
soning and data fusion outlined later on.
In order to avoid unnecessary heavy network load re-
sulting from the image transmission our embedded
“smart” cameras perform this processing themselves
and transmit only the spatial occupancy data to the
platform. Hence, the platform receives information
from the surveillance camera based system about ar-
eas in the building that are occupied by something. It
is then up to the information fusion within the plat-
form to identify this “something” and trigger actions
accordingly. A compressed video stream is available
to the platform only on demand, so that for instance
security staff is able to have a look at the scene in case
an unidentified occupancy had occurred.
5 Information Fusion
The fusion of the two different sources of information
is crucial for the overall system. This is because the
smartphone based system is well able to identify and
track smartphones but lacks the ability to track any-
thing untagged, while the camera based system is able
to track everything but is unable to identify what it
sees. For this reason the two sub-systems complement
each other very well in terms of the information pro-
vided.
In our system this fusion is performed using a Bayes-
ian geometric reasoning and filtering framework that
allows combining the multiple observations from the
different sub-systems into a single position estimate
for each tracked object. Two problems need to be
solved for this: first the camera derived geometry in-
formation needs to be associated with the currently
tracked objects within the system and second the ob-
servations from the two different sources need to be
combined into a single accurate position estimate.
Because the surveillance camera sub-system is unable
to provide any kind of identification of the object it
sees, the association between the image derived ex-
truded spatial silhouettes and the currently tracked ob-
jects is performed using statistical geometric reason-
ing techniques [10], that allow to identify which tag
corresponds to which tracked image object based
solely on the common spatial occupancy. Once we
have established this correspondence incremental es-
timation and filtering [11, 12] is used to update the
estimated position of the tag obtaining its current loca-
tion with the highest achievable accuracy given the
available input data sources.
For each image derived silhouette that does not corre-
spond to any tag currently tracked within the system
during this process an alarm is triggered requiring
human intervention as this event occurs when some-
body is moving around the building without authorisa-
tion. While the better accuracy obtained through the
information fusion is usually also desirable, especially
the feature of the fusion system being able to detect
untagged objects makes it so useful in security appli-
cations.
5 Conclusion
We have presented an approach for the integration of
an access control system and a surveillance camera
system to leverage their functional complementarities
and introduce seamless authentication for a security
access system. The key aspect is the high proliferation
of smart phones and Wi-Fi networks in corporate en-
vironments allowing the smart phone together with its
seamless localisation to bring the user’s current posi-
tion as an essential part of credential information into
the security system enabling decision processes of un-
obtrusive seamless access authorisation.
The critical requirement on the system is its ability to
combine all location related information for automatic
tracking of both authorised and unauthorised move-
ment of people even without smart phones/mobile
credentials. That was achieved by the seamless inte-
gration with the surveillance camera system.
Another major requirement is the systems scalability
when it has to be able to cope with any sudden in-
crease of the number of people in the premise without
compromising its reliability and responsiveness. That
was achieved by adopting state of the art programing
patterns and software design approaches known from
cloud computing.
6 Acknowledgement
This work has been supported by Enterprise Ireland
through the grant CF/2010/042 and EU FP7 project
LocON (224148).
Figure 2: Fusion between two images. The silhou
ettes
of the two views projected onto the ground plane are
shown in red, while the orange circle is the fused posi-
tion. If there is no smartphone detected within this
area an alarm is triggered.
References
[1] www.rabbitmq.com, RabbiMQ - Open source
enterprise messaging
[2] Maarten Weyn and Martin Klepal, OLS – Oppor-
tunistic Localization System for Smart Phone
Devices, Ref: Mobile Phones: Technology, Net-
works and User Issues, Nova 2011
[3] Widyawan, G. Pirkl, D. Munaretto, C. Fischer,
Chunlei A., P. Lukowicz, M. Klepal, A. Timm-
Giel, J. Widmer, D. Pesch, H. Gellersen, “Virtual
Lifeline: Multimodal Fusion for Robust Naviga-
tion in Unstructured Environments”, submitted
to Elsevier Journal on Pervasive and Mobile
Computing, January 2011
[4] M. Klepal et al, “OLS Opportunistic localisa-
tion system for smart phones devices”, Mobi-
Held, ACM SIGCOMM 2009 Int. Workshop,
Barcelona, Spain, August 2009
[5] Widyawan, Klepal, M., Beauregard, S. 2008. A
Backtracking Particle Filter for Fusing Building
Plans with PDR Displacement Esti-
mates, WPNC, Hannover, Germany
[6] Widyawan, Klepal, M., Beauregard, S., Pesch,
D. 2008. A Novel Backtracking Particle Filter
for Pattern Matching Indoor Localisation, The
First ACM MELT, San Fransisco,
[7] Anagnostopoulos, C.-N.E.; Anagnostopoulos,
I.E.; Psoroulas, I.D.; Loumos, V.; Kayafas, E.:
License Plate Recognition From Still Images and
Video Sequences: A Survey. In: IEEE Transac-
tions on Intelligent Transportation Systems,
vol.9, no.3, pp.377-391, Sept. 2008
[8] Piccardi, M.: Background subtraction tech-
niques: a review. In: IEEE International Confer-
ence on Systems, Man and Cybernetics, vol. 4,
pp. 3099-3104, 2004
[9] Beder, C., Breuels, A., Klepal, M.: Extracting
foreground objects from a video stream while
compensating for global brightness and contrast
changes. 2011, to appear
[10] Beder, C.: Grouping uncertain oriented projec-
tive geometric entities with application to auto-
matic building reconstruction. PhD thesis, Pho-
togrammetry Department, Institute for Geodesy
and Geoinformation, Bonn University, Germany,
2007
[11] Beder, C. and Steffen, R.: Incremental estima-
tion without specifying a- priori covariance ma-
trices for the novel parameters. Proceedings of
VLMP Workshop 2008, Anchorage, AL, June
2008
[12] Steffen, R. and Beder, C.: Recursive Estimation
with Implicit Constraints, DAGM Symposium
2007, Heidelberg, Germany, pages 194-203,
2007
ResearchGate has not been able to resolve any citations for this publication.
Full-text available
Chapter
People are eager to locate their peers and stay connected with them. The Opportunistic Localization System (OLS) makes that simple. OLS’s core technology is an opportunistic location system based on smart phone devices. One main advantage is that it seamlessly works throughout heterogeneous environments including indoors as opposed to GPS based systems often available only outdoors under the unobstructed sky. OLS is a phone-centric localization system which grasps any location related information that is readily available in the mobile phone. In contrast to most of the competing indoor localization systems, OLS does not require a fixed dedicated infrastructure to be installed in the environment, making OLS a truly ubiquitous localization service. The latest version of OLS strives to reduce the system ownership cost by adopting a patent covered self-calibration mechanism minimizing the system installation and maintenance cost even further. OLS’s architecture migrated from the original client-server to the current service oriented architecture to cope with the increasing demands on reusability across various environments and platforms. Thus, to scale up service for a large variety of clients. The location related information used for the estimation of mobile device location are used by existing signals in the environment and can be sensed by a smart phone. The readily available information, depending on the phone capability, is typically a subset or all of the following: the GSM/UMTS signal strength, WiFi signal strength, GPS, reading from embedded accelerometers and Bluetooth proximity information. The reliability and availability of input information depends strongly on the actual character of the mobile client physical environment. When the client is outdoors under the unobstructed sky, GPS is a favorable choice typically combined with pedometer data derived from the 3D acceleration and compass measurement. When the client is in a dense urban and indoor environment the GSM/UMTS and WiFi signal strength combined with pedometer data usually performs best. If indoor floor plan layouts are available, the map filtering algorithm can further contribute to the location estimation. However, the information about the actual type of environment is not available and proper importance weighting of all input information in the fusion engine is paramount. This chapter will describe how OLS developed and particularly give insight in OLS core technology, which is the adaptive fusion of location related information that is readily available in smart phones. The experiments carried out, shows when considering all location related information for the object location estimation namely the GSM and WiFi signal strength, GPS, PDR and map filtering, the presented opportunistic localization system achieves a mean error of 2.73 m and a correct floor detection of 93%in common environments. The achieved accuracy and robustness of the system should be sufficient for most of location aware services and application, therefore having a potential of enabling truly ubiquitous location aware computing.
Full-text available
Conference Paper
Recursive estimation or Kalman filtering usually relies on explicit model functions, that directly and explicitly describe the effect of the parameters on the observations. However, many problems in computer vision, including all those resulting in homogeneous equation systems, are easier described using implicit constraints between the observations and the parameters. By implicit we mean, that the constraints are given by equations, that are not easily solvable for the observation vector. We present a framework, that allows to incorporate such implicit constraints as measurement equations into a Kalman filter. The algorithm may be used as a black-box, simplifying the process of specifying suitable measurement equations for many problems. As a byproduct, the possibility of specifying model equations non-explicitly, some non-linearities may be avoided and better results can be achieved for certain problems.
Full-text available
Conference Paper
Particle Filter (PF) techniques has been widely used in in- door localization systems. They are often used in conjunc- tion with pattern matching based on Received Signal Strength Indication (RSSI) fingerprinting. Several variants of the particle filter within a generic framework of the Sequential Importance Sampling (SIS) algorithm have been described. The purpose of this paper is to show how a variant of PF, the so-called Backtracking Particle Filter (BPF), can be used to improve indoor localization performance. The BPF is a technique for refining state estimates based on exclusion of invalid particle trajectories. Categorization of invalid trajectory determined during importance sampling step of the PF. The BPF can also take advantage of available building plan information using the so-called Map Filtering (MF) technique. The incorporation of MF allows the BPF to exploit long-range geometrical constraints. This paper evaluates BPF with indoor localization based on WLAN RSSI fingerprinting. The filtering schema is evalu- ated using the propagation simulation in an office building, a typical environment for fingerprinting technique. Favor- able result are obtained, showing positioning performance (1.34 m mean 2D error) superior to the PF-only no MF case (1.82 m mean 2D error), or up to 25% improvement. It is also shown that the performance is far better than the po- sition estimates from conventional Nearest-Neighbour (NN) and Kalman Filter (KF) approaches using the same RSSI measurements.
Full-text available
Conference Paper
It is known that Particle Filter and Map Filtering techniques can be used to improve the performance of positioning systems, such as Pedestrian Dead Reckoning (PDR). In previous research on indoor navigation, it was generally assumed that detailed building plans were available. However, in many emer gency / rescue scenarios, there may be only limited building plan information on hand. The purpose of this paper is to show how a novel Backtracking Particle Filter (BPF) can be combined with different levels of building plan detail to improve PDR performance. We use real PDR stride length and blunder-prone stride azimuth data which were collected from multiple walks along paths in and out of a small office building. The PDR displacement data is input to the BPF estimator that in turn uses the building plan information to constrain particle motions. The BPF can take advantage of long-range (geometrical) constraint information and yields excellent positioning performance (1.32 m mean 2D error) with detailed building plan information. More significantly, this same filter using only external wall information produces dramatically improved positioning performance (1.89 m mean 2D error) relative to a PDR-only, no map base case (8.04 m mean 2D error). This effect may very well occur for many other realistic wall layouts and path geometries. Moreover, this result has a substantial practical significance since this level of building plan detail could be quickly and easily generated in many emergency instances.
Full-text available
Article
License plate recognition (LPR) algorithms in images or videos are generally composed of the following three processing steps: 1) extraction of a license plate region; 2) segmentation of the plate characters; and 3) recognition of each character. This task is quite challenging due to the diversity of plate formats and the nonuniform outdoor illumination conditions during image acquisition. Therefore, most approaches work only under restricted conditions such as fixed illumination, limited vehicle speed, designated routes, and stationary backgrounds. Numerous techniques have been developed for LPR in still images or video sequences, and the purpose of this paper is to categorize and assess them. Issues such as processing time, computational power, and recognition rate are also addressed, when available. Finally, this paper offers to researchers a link to a public image database to define a common reference point for LPR algorithmic assessment.
Conference Paper
Background subtraction is a widely used approach for detecting moving objects from static cameras. Many different methods have been proposed over the recent years and both the novice and the expert can be confused about their benefits and limitations. In order to overcome this problem, this paper provides a review of the main methods and an original categorisation based on speed, memory requirements and accuracy. Such a review can effectively guide the designer to select the most suitable method for a given application in a principled way. Methods reviewed include parametric and non-parametric background density estimates and spatial correlation approaches.
OLS – Opportunistic Localization System for Smart Phone Devices, Ref: Mobile Phones: Technology, Networks and User Issues
  • Maarten Weyn
  • Martin Klepal
Maarten Weyn and Martin Klepal, OLS – Opportunistic Localization System for Smart Phone Devices, Ref: Mobile Phones: Technology, Networks and User Issues, Nova 2011