Energy Efficient indoor Localization in Wireless Sensor Networks

swarmtellingMobile - Wireless

Nov 21, 2013 (3 years and 9 months ago)

97 views

Rescom 2007

1


Abstract— Randomly deployed nodes in a wireless sensor
network need to be aware of their location in order to
deliver meaningful information about the physical
environment. We plan to investigate node localization
phenomenon with respect to its energy consumption. In
this regard, we are examining an iterative localization
approach with three non-collinear anchor nodes deployed
close to one another.

Index Terms— Location discovery, localization,
optimization, Wireless Sensor Networks
I. INTRODUCTION
Wireless sensor networks consist of many small, inexpensive
nodes deployed randomly in the area to be monitored. These
nodes are desired to have some way of learning their location
with respect to the environment.

Localization of wireless sensor nodes has been an active area
of research in the past few years. In this regard, various
suggested approaches can be categorized as: infrastructure
based or infrastructure free (ad hoc); centralized or distributed;
range based or range free, iterative or non-iterative [2]. We are
working on an iterative, range-based, distributed localization
approach in 2D that uses three non collinear Anchor nodes
(nodes already aware of their location) deliberately deployed
close to one another.
II. MOTIVATION
GPS is a good option for determining an object's location but
it is not feasible for an in-door sensor network because of cost,
node form factor, line of sight and power consumption. Since
nodes are expected to be of small size they possess very
limited energy. We need to develop specialized energy aware
algorithms for such networks. This leads to the conception of
low power distributed localization approaches that are
optimized both in terms of communication and in terms of
computation yielding accurate position estimates. Also in an
attempt to solve this problem in ad hoc context, we need to



Manuscript received May 15, 2007.
Sadaf Tanvir is a PhD student at INPG/LIG, Saint Martin d'Hères
38402 France. phone: +33 4 76 82 72 28; fax: +33 4 76 82 72 87. E-mail:
sadaf.tanvir@imag.fr.
Benoit Ponsard is a professor at INPG/LIG. E-mail:
benoit.ponsard@imag.fr.


keep in mind reduced dependency on infrastructure like the
number of anchor nodes.

III. STATE OF THE ART
Approach [6] is one of the many initial contributions to the in-
door localization domain which require infrastructure for their
execution. Currently, research efforts are being made to reduce
this infrastructure dependency of localization approaches.

Present literature suggests many range based methods initially
using a small number of anchor nodes. These nodes help
neighboring nodes to get localized and the process continues
in the same manner for the entire network [1],[4],[5]. Some
suggest them to be deployed randomly [1],[4] others propose
them to be located along the perimeter of the network, still
others deploy them close to one another in a non collinear
way[2]. In theory, for a network to get localized in 2D, it
requires only three non collinear anchor nodes. Placing these
nodes close to one another prevents from implementing a
preliminary phase of network topology discovery as in case of
[1],[4] : where localization starts from anchor nodes and
floods throughout the network (accuracy can be improved
afterwards). This advantage has motivated us to work on a
localization approach using the least number (three) of anchor
nodes deployed close to one another.

We see very few localization approaches considering power
efficiency and accuracy as their primary design requirements.
One such example is the LR-DV-Hop [3] which is an
enhancement to the DV-hop localization algorithm but since it
is a range-free method, it is coarse grained. The objective of
our research work is to design and evaluate optimized and
fine-grained localization methods that are distributed, robust
and scalable.
IV. OUR CURRENT WORK
Currently, we are examining an iterative, range-based,
distributed localization approach in a 2D stationary sensor
network called "`Beacon-less two way ranging" approach. In
this approach, a group of three non collinear anchor nodes is
deployed in the corner of the network and localization is done
in a query/response manner. These anchor nodes are aware of
their location either by using GPS or by manuallY configuring
their coordinates. We have assumed same transmission range
for anchor nodes and free nodes and each node is within one
hop connectivity of another node forming a connected
network. A free node (node unaware of its location) near the
Energy Efficient indoor Localization in Wireless
Sensor Networks

Sadaf Tanvir, Benoît Ponsard

Rescom 2007

2
anchor group when gets access to the medium, broadcasts its
query message. This message is received by the three anchor
nodes and each of them sends a unicast reply carrying their
position information to the free node. This “two way ranging"
process is carried out to estimate the distance between the free
node and the anchor nodes. It is done through Radio
Frequency Time of flight (RF TOF) and node position is
calculated using trilateration.

here t0 and t2 are measured at the free node and