Abstract: in This Paper Indoor Localization Problem Is Addressed From Theoretical and

Abstract: In this paper indoor localization problem is addressed from theoretical and
application stand points. In order to overcome this problem, an algorithm has been developed.
The execution of the algorithm on each nodes results in the localization of all the blind nodes
in the network. A decentralized algorithm is presented in this paper, which proceeds locally
on each node based on the sole knowledge of distance measured from and estimated positions
of nearby nodes. The repeated execution of algorithm on each node guarantees the
convergence to positioning of whole network. Simulation of simple network has been
performed to quantify the proposed scheme in realistic conditions, and the results are
presented. This paper can be implemented on a campus to track the location of staff or/and
students.
1 INTRODUCTION
The localization and positioning of generic wireless node with in a space domain,
with time synchronization can be considered as a prerequisite for the deployment of large
context aware services [1]. Here the emphasis is given to indoor case, where GPS-like
solutions are unreliable or expensive. Here already available low cost dedicated technologies
or standard protocols are encouraged so as to take advantage of wireless access points and
peer-to-peer connections.
The indoor situation is particularly prone to coverage problems due to the massive
presence of obstacles in the line-of-sight condition with respect to anchor nodes. So based on
this mixed convex/nonconvex algorithm we can localize all the blind nodes in network, with
limited number of anchor nodes. The positioning calculations are made on each node based
on the distance measured from the nearby nodes and the position of that nearer nodes. The
execution of the algorithm after some particular time may converge and result in tracking the
exact location of all blind nodes in the network.
The application of this paper is, it can be implemented on any network to track the
location of users, provide the users must have some device were the algorithm has to be
executed. It can be used on a campus to track the location of students and/or staff. For
example, at each time we can find out the location of each student, whether they are present
inside the campus etc.
2 PROBLEM STATEMENTS
Consider a network composed of n nodes, each capable of wireless communication,
and connected to an arbitrary number of neighbouring nodes as specified by the connectivity
of the wireless network. If (i, j) L, the two nodes are referred to as adjacent only if the two
nodes i and j can communicate.
As far as the localization problem is concerned, the nodes of the network can be
classified as anchor nodes and blind nodes. A node who knows its exact position is referred
to as anchor nodes. A node who does not know its exact position is referred to as blind nodes.
The localization problem is concerned with the determination of exact position of all blind
nodes. In this paper, we denote with
 xi the vector of exact coordinates of node i.

 xi the vector of estimated coordinates of node i.
 dij the actual distance between two nodes i and j.
Obviously, xi = xi for all set of anchor nodes, ,i.e., the estimated position and the exact
position converges for anchor nodes.
According to the local character of solution for localization, the only information
being available at node i is following:
 the distance dij and
 the estimated coordinate xj.
It is easy to verify that to determine the blind node positions; the following optimization
problem must be solved:
(nc)
= min rij (1)
s.t, | ||xi – xj || - dij | = rij
Where rij are the optimization residuals, and ||. || indicates the Euclidean norm of a vector.
The major concern with formulation (1) is that the optimization problem is not
convex, in view of the equality sign in the constraint | ||xi – xj || - dij| = rij..As a consequence, it
(nc)
admits local minima. This implies that the condition = 0 may not be reached in a finite
computing time. To avoid the presence of local minima, we transformed the problem into a
convex one, replacing the equality constraint of (1) by an inequality, and removing the
absolute value. The result is the following convex formulation:
(c)
= min rij (2)
s.t, ||xi – xj || - d ij rij ,

rij 0 ,
In Fig 1.a, x1, x2, and x3 are the exact coordinates of three anchor nodes. In nonconvex
formulation, a generic node i obtaining its distance dij from the RSS (based on received signal
strength) coming from its anchors j, represented in Fig 1.a by the dashed lines (graphically,
the radius of the corresponding circle), can find in xi the only admissible solution to the
(nc)
positioning problem. In fact, xi is the only point returning = 0. When formulation (2) is
used, instead, the inequality constraint allows as admissible (i.e., corresponding to zero cost)
any solution belonging to the shaded area in Fig 1.a that we called Zero Cost Region (ZCR),
that is to say to the intersection of the three transmission coverage circles centered in xj and
having radius dij , j=1,…, 3.
Zero cost region (ZCR)
X2
Xi
X3
X1
Another relevant concept can be extracted from the analysis of Fig. 1.b, representing a
network topology the difference with Fig 1.a is that x i lies inside the triangle of points x 1, x2,
and x3. Here it can be easily observed that, under the exact knowledge of the distances d ij ,
the inequality sign in the constraint (i.e., the convex formulation) can be adopted without
losing the uniqueness of the solution. In this case, the intersection among the coverage circles
collapses into a single point.
X1
X3
X2

Abstract: in This Paper Indoor Localization Problem Is Addressed From Theoretical and

Uploaded by

Copyright:

Available Formats

Abstract: in This Paper Indoor Localization Problem Is Addressed From Theoretical and

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Abstract: in This Paper Indoor Localization Problem Is Addressed From Theoretical and

Uploaded by

Copyright:

Available Formats

Abstract: In this paper indoor localization problem is addressed from theoretical and

 xi the vector of exact coordinates of node i.

s.t, | ||xi – xj || - dij | = rij

s.t, ||xi – xj || - d ij rij ,

having radius dij , j=1,…, 3.

Zero cost region (ZCR)

You might also like