Cross-Layer Self Routing: a self-managed routing
approach for MANETs
M.A. Razzaque, Simon Dobson and Paddy Nixon
Systems Research Group
School of Computer Science and Informatics
UCD Dublin IE
Email: abdur.razzaque@ucd.ie
Abstract—Mobile ad hoc networks (MANETs) generally adopt
a peer-to-peer architecture in which the nodes themselves provide
routing and services to the network. Disconnectivity with peer
nodes, induced by mobility, power drains and damage makes
route maintenance difficult and degrade the network’s ability to
offer services reliably to its peer nodes. In this paper, we present a
routing scheme for proactive management of such disconnections,
by fusing and leveraging information derived from multiple levels
of the network protocol stack using cross-layering. In addition
to the disconnectivity information, this routing scheme utilises
node’s service level information and data/service replication
to provide service from an alternate source (if there is one)
even in the absence of the targeted source. Simulation results
demonstrate significant improvements in route maintenance and
service availability over other similar schemes.
I. INTRODUCTION
A key challenge in MANETs is to work out efficient
methods to ensure route availability while incurring minimal
control overhead. Routing techniques may be broadly divided
into proactive and reactive schemes [1]. Proactive protocols
often suffer from excessive control overhead associated with
maintaining routes to destinations even when not required;
reactive protocols experience higher end-to-end packet delays
compared to proactive protocols, since routes must be re-
formed during communication. Hybrid protocols like the Zone
Routing Protocol (ZRP) offer the qualities of both proactive
and reactive protocols. The most desirable routing protocol is
one which offers minimal end-to-end packet delays for real-
time traffic and less control overhead for non-real-time traffic.
This can only be achieved by exploiting link state information
(such as link life time) that is generally ignored in conventional
ad hoc routing protocols: even the existing hybrid protocols
do not utilise link state information, and hence do not offer
enormous performance advantage over existing reactive or
proactive protocols.
Link Life Time (LLT ) is one of the most important link
state information, mainly controlled by disconnectivity due to
mobility and the node failure due to power shortage. If we can
exploit the power- and mobility-related information in routing,
there is an opportunity to provide better route maintenance and
improve the route performances– using link-layer information
to condition transport- and network-layer behaviour. Existing
modular or layered approaches would not allow such cross-
layer interactions, so a cross-layer architecture may provide a
better basis on which to build [2], [3].
The LLT of a wireless link can be predicted exploiting
location and movement pattern information and/or remaining
power information in the node using a cross-layer approach.
Predicted LLT (LLTp) can be utilised for proactive route
maintenance and the corresponding route could maintain ser-
vice only if the providing node is within the coverage area and
not dead. But by exploiting a node’s service level information
and data/service replication we might be able to provide
service from an alternate source (if there is one) even in the
absence of the targeted source. To our knowledge no single
paper has considered all these issues for a MANET routing
protocol. Our present work does not start from scratch rather
builds on an existing reactive ad hoc routing protocol, AODV,
a widely studied routing protocol in MANET environments.
We modify the AODV’s route maintenance and route discov-
ery algorithms to exploit cross-layer information concerning
movement and power to define a new protocol, Cross-Layer
Self Routing (CLSR).
The remainder of this paper is structured as follows. Sec-
tion II briefly describes cross-layer networking. A brief related
study is presented in III . Section IV presents CLSR in
detail, along with the LLTp calculation, and service-level
information, data replication and their corresponding uses in
CLSR. Section V describes the implementation and evaluation.
Section VI concludes the work with some future directions.
II. CROSS-LAYERING
To exploit the link state information like LLT , we need to
capture the interactions between the physical, link and network
layers – but existing strict layering schemes do not support
such interactions. For the exploitation of service/data level
information in routing, the network layer has to interact with
the application layer, but again a strictly layered approach does
not allow interactions amongst non-adjacent layers.
Recent research studies [3], [2] show that the cross-layer
design principle has great importance in wireless ad hoc net-
works, where different layers are more likely to use the same
information in making layer-specific decisions. In particular,
the real-world locations of the nodes and the topology of
the network are commonly used by both the routing and the