1404 IEEE COMMUNICATIONS LETTERS, VOL. 16, NO. 9, SEPTEMBER 2012
ISAR: Improved Situation-Aware Routing Method for
Wireless Mesh Backbones
Qiang Liu, Jianping Yin, Victor C. M. Leung, Fellow, IEEE, and Zhiping Cai, Member, IEEE
Abstract—Due to the increasing performance requirements of
wireless access services, adaptive interference resistant routing
is crucial for providing reliable access and high bandwidth
services via wireless mesh networks. In this letter, we propose an
improved method for situation-aware routing, which extends the
existing routing protocols via a refined situation-aware routing
metric and a greedy load-balancing routing scheme. Simulation
results demonstrate that the proposed method can achieve a
higher throughput with lower transmission delay compared to
previous routing schemes.
Index Terms—Interference, load balance, routing, wireless
mesh backbones.
I. INTRODUCTION
H
IGHLY efficient routing is crucial for backbone wireless
mesh networks (WMNs) to reliably provide services
for ubiquitous mobile terminals with low delay and good
fairness. Although routing protocols in wireless networks have
been studied extensively, most of them are not suitable for
WMNs. For example, the Ad hoc On-Demand Distance Vector
(AODV) [1] aims to minimize the number of hops in logical
topologies. However, the hop count metric does not work well
in WMNs because the aggregated traffic from a large set of
sources to a few gateways introduces interference amongst
different flows and decreases th e quality of service (QoS).
Many routing protocols and metrics have been proposed for
WMNs [2] [3] to overcome the deficiency of classical proto-
cols. While legacy routing protocols ty pically predeterm ine
one or more paths before delivering the packets, the two
approaches proposed in [2], i.e., opportunistic routing and
network coding, are shown to be promising in exploiting
the intrinsic char acteristics of multi-hop wireless networks.
On the other hand, many routing metrics are found in the
literature, such as the expected transmission count metric
(ETX) [4], the weighted cumulative expected transmission
time (WCETT) [5], the Airtime metric [6], the interference-
aware routing metric (IAR) [7], and the adaptive situation-
Manuscript recei ved March 8, 2012. The associate editor coordinating the
review of this letter and approving it for publication was J. Holliday.
This work was supported by the National Natural Science Foundation
of China (Grant No. 61170287, No. 60970034, No. 61070198, and No.
60903040).
Q. Liu is with the School of Computer, National University of Defense
Technology, Changsha, Hunan, China. He is now a Visiting Scholar in the
Department of Electrical and Computer Engineering, Univ ersity of British
Columbia, Vancouver, BC, Canada (e-mail: libra6032009@gmail.com).
J. Y in and Z. Cai are with the School of Computer, National Uni-
versity of Defense Technology, Changsha, Hunan, China (e-mail: {jpyin,
zpcai}@nudt.edu.cn).
V. C. M. Leung is with the Department of Electrical and Computer
Engineering, University of British Columbia, Vancouver , BC, Canada (e-mail:
vleung@ece.ubc.ca).
Digital Object Identifier 10.1109/LCOMM.2012.072012.120502
aware routing metric (ASA) [8]. However, these measurement-
based methods generally ignore the influence of route length.
While the link states are important metrics for routing in
WMNs, the number of routing hops should also be considered
because long routes will potentially increase transmission
delay and packet loss rate due to inter-flow and intra-flow
interference rooted from the shared nature of the transmission
medium [7].
In this letter, we propose an Improved Situation-Aware
Routing (ISAR) method for WMNs, which extends the Hybrid
Wireless Mesh Protocol (HWMP) that is the default path
selection protocol of IEEE 802.11s [9]. To enhance the over-
all p erformance and fairness of the network, ISAR consists
of two parts: the ISAR metric design and load-balancing
routing. The ISAR metric is designed with the objective
of maximizing routing efficiency. The metric takes several
factors into account, including path length, interference and
transmission delay. Then, a greedy load-balancing routing
scheme is presented to update the optimal path based the
ISAR metric computed with up -to-date network information.
The routing scheme improves previous schemes by selecting
the optimal measuring time to tradeoff between precision
of measurement and node throughput. Our contributions in
this work include: (1) We introduce a novel routing m etric
for improving on the ASA metric [8], an adaptive situation-
aware routing metric. When serving as the basis of path
selection, the proposed metric can achieve a considerable
improvement in routing efficiency; (2) We further propose a
new load-balancing routing scheme to improve existing active
probing based routing schemes by reducing the overhead of the
probing mechanism. Results show that the proposed routing
method outperforms previous routing schemes in terms of total
throughput and average delay.
II. ISAR M
ETRIC DESIGN
Since HWMP in the IEEE 802.11s standard [9] uses some
routing metric in the assessment of a mesh path to the
destination, the first step in our design is to define a suitable
metric. We denote P as the set of candidate paths and define
the cost of a path p ∈ P by
C(p)=
l∈p
ISAR
(p)
(l), (1)
where l is a link of the path p. Furthermore, due to potential
impacts of long paths as mentioned before, a monotone
increasing penalty func tion f
p
(l) is integrated to define link l’s
cost. This penalty function is chosen to be a convex downward
function that is greater than 1, and is applied when link
l’s hop number, denoted by HC
(p)
(l), exceeds a predefined
threshold η. The motivation of such a function is to discourage
1089-7798/12$31.00
c
2012 IEEE