IEEE Wireless Communications - April 2017 - page 28

IEEE Wireless Communications • April 2017
26
1536-1284/17/$25.00 © 2017 IEEE
A
bstract
This article presents a scalable route map for
the least cost deployment of wireless hetero-
geneous networks that support traffic from the
advance metering infrastructure (AMI). We first
explore the performance of a common scenar-
io in which a single technology is employed to
connect smart meters sending traffic to the util-
ity. Based on simulations with actual city maps,
we study the coverage provided to smart meters
by an LTE cellular network. In order to improve
the coverage, an optimization model that con-
siders network capacity and range is proposed to
determine the optimal location of base stations
to achieve a target coverage of smart meters.
According to these preliminary results with a sin-
gle access technology, we propose an evolved
network architecture that considers several alter-
natives of wireless heterogeneous networks to
guarantee the coverage to smart meters with the
least use of resources. We introduce a heuristic
model that involves elements from base stations,
universal data aggregation points, number of
smart meters, and an optimal routing to achieve
the desired connectivity from the group of smart
meters. We employ geo-referenced models to
consider actual characteristics of cities as well as
geographical conditions. Results from the evolved
model demonstrate that by combining technolo-
gies and employing data aggregation points with
optimized localizations, the network is able to
achieve a target coverage of smart meters with a
reduced cost in terms of technological resources.
I
ntroduction
The advanced metering infrastructure (AMI) in
smart grid requires communication technolo-
gies able to support the large number of smart
meters (SMs) deployed in both urban and rural
scenarios. Accordingly, wireless heterogeneous
networks (WHN) have emerged as a candidate to
provide the desired connectivity to the SMs [1].
Different from a traditional AMI network where
only a single wireless technology (e.g., WiFi, IEEE
802.15.4g, GPRS/LTE) is employed for SMs’ con-
nectivity [2, 3], there are a plethora of technol-
ogies available in a heterogeneous architecture;
hence, there should be an objective criteria or
route map that allows the definition of the opti-
mal use of resources in this new approach.
This work focuses on scenarios with different
population densities where it is assumed each
house has one SM installed. We introduce the
wireless heterogeneous network architecture,
which involves universal data aggregation points
(UDAP) with multi-radio capabilities [4]. In this
way, each UDAP may employ different wireless
technologies and variable clustering capabilities
to collect data from SMs [5, 6]. Therefore, thanks
to the UDAPs, it is possible to reduce resource
costs according to the number of houses with
SMs to be covered. In addition, it also makes it
possible to achieve a scalable deployment in the
long run, according to the policies of the utility
companies. Previous works propose clustering
mechanisms based, for example, in k-means [7]
to create conglomerates. In this work we present
an optimization process based on capacity and
coverage, which is contrasted with simulations of
a real scenario.
As illustrated in Fig. 1, the proposed architecture
also considers multi-hop capabilities for routing traf-
fic from the SMs to the UDAP. In this way, traffic
from the SMs is concentrated in the closest UDAP
with no need to incorporate cellular connectivity in
each SM. Once the UDAP aggregates data from
different SMs, a single aggregated packet will be
sent to the closest base station (BS), hence reduc-
ing traffic overload in the cellular network, especial-
ly for dense deployments scenarios [8].
We present a heuristic model that solves the
routing problem between SMs, UDAPs, and cel-
lular BS, in order to obtain the optimal number
and locations of UDAPs in the proposed WHN
for AMI. In this way, we minimize the costs asso-
ciated with the number of deployed UDAPs, but
at the same time guaranteeing the coverage of
the SMs. Different from other networks, such as
wireless sensor networks, in AMI it is prevalent to
guarantee coverage to be able to collect the read-
ings from each SM, as well as to provide other
services such as remote management (i.e., discon-
nection and reconnection), electric demand mon-
itoring, and failure detection in the distribution
system, among others [9, 10].
The remainder of this article is organized as
follows. The next section elaborates on the con-
E
steban
I
nga
, S
andra
C
éspedes
, R
oberto
H
incapié
,
and
C
esar
A
ndy
C
árdenas
S
calable
R
oute
M
ap
for
A
dvanced
M
etering
I
nfrastructure
B
ased
on
O
ptimal
R
outing
of
W
ireless
H
eterogeneous
N
etworks
S
mart
G
rids
Esteban Inga is with the
the Universidad Politécnica
Salesiana.
Sandra Céspedes is with the
Universidad de Chile and
with the Universidad Icesi.
Roberto Hincapié is with
the Universidad Pontificia
Bolivariana.
Cesar Andy Cárdenas is with
the Universidad de Chile.
Digital Object Identifier:
10.1109/MWC.2017.1600255
1...,18,19,20,21,22,23,24,25,26,27 29,30,31,32,33,34,35,36,37,38,...132
Powered by FlippingBook