Cross Layer- Performance Enhancement Architecture (CL-PEA) for MANET

International Journal of Computer Networks and Communications Security
VOL. 1, NO. 2, JULY 2013, 54–60
C
C
Available online at: www.ijcncs.org
N
ISSN 2308-9830
S
Cross Layer- Performance Enhancement Architecture (CL-PEA)
for MANET
Mehajabeen Fatima1, Roopam Gupta2, T K Bandhopadhyay3
12RGPV, Bhopal, India
3BIST, Bhopal, India
E-mail: 1mehajabeen.fatima@gmail.com, 2roopam_1710@yahoo.co.in, 3tushar_bando1@gmail.com
ABSTRACT
Internet is based on TCP/IP but TCP/IP architecture is not sufficient to cope with the dynamics of MANET.
Number of architectures are already available in literature but cross layer architecture is most popular one.
To enhance performance of protocols a Cross Layer- Performance Enhancement Architecture (CL-PEA) is
proposed. In this architecture one hardware layer is added and all layers can exchange their parameters. CL-PEA
enhances the protocol performances by taking decisions on the basis of information available on
layers.
Keywords: TCP/IP, MANET, CL-PEA, Cross Layer, Protocol.
1 INTRODUCTION
MANET is a special type of wireless network in
which a collection of mobile network interfaces
may form a temporary network without the aid of
any established infrastructure or centralized
administration. Ad Hoc wireless network has
applications in emergency search-and-rescue
operations, decision making in the battlefield, data
acquisition operations in hostile terrain, etc. It is
featured by dynamic topology, multi-hop
communication, limited resources (bandwidth,
CPU, battery, etc.) and limited security [1]. An
accelerating energy crisis in the oil and gas industry
leads the development and investment in MANET
technologies. Actually, the availability of crude oil
in the remote area of seas, it is very difficult to
implement infrastructure oriented network for
communication. Here it is difficult to provide
infrastructure in such places. And deployment of
such network is very costly. Sometimes the
working staff faces disastrous situations in such
remote areas and it is very difficult for them to
communicate with each other without any
infrastructure, hence required an infrastructure less
network (MANET) [2]. MANET was designed to
work in the situations such that whenever certain
number of nodes come together in transmission
area of each other, the nodes should be able to
communicate. The formed network is temporary, so
nodes are equipped with low power and computing
with routing capabilities. Because, in MANET,
topology frequently changes, and nodes have got
limited power so its routing is challenged by these
factors. So conventional routing protocols cannot
be used here. Nodes in an ad hoc environment face
major challenges of mobility, traffic, scalability,
power consumption etc. The mobility in an ad hoc
network can be featured by the speed of nodes in
the systems and the duration of the pause time
during which nodes become stationary. One of the
most widely used mobility models is the random
waypoint model first proposed in [3]. In this model,
nodes choose destination, and move towards it at a
random speed uniformly chosen from (0, Vmax),
where Vmax is the maximal speed of the node. If a
node pauses then it pauses for a constant time and
then it start moving with specified speed [4]. The
average speed of nodes in the system determines
the rate at which route breaks and consequently the
control traffic consumed by the route repairing in
on-demand protocols. The increased routing control
traffic badly affects the packet delivery ratio, power
consumption, delay, throughput etc. As with
increase in traffic, chances of route breakage
increases. The routing protocols of such networks

55
Mehajabeen Fatima et al. / International Journal of Computer Networks and Communications Security, 1 (2), JULY 2013
must be able to perform efficiently and effectively,
adapting to any changes. The performance of
network is also badly affected with the increase in
size of network. So scalability is a major concern.
The power consumption increases with increase in
traffic, scalability, mobility. To cope with these
dynamics of MANET different architecture is
required like cross layer approach.
2 CROSS LAYER APPROACH
The internet is based on TCP/IP architecture and
it is a worldwide success. TCP/IP was originally
designed for wired links, characterized by high
bandwidth, low delay, and low probability of
packet loss (high reliability), static routing and no
mobility [5]. But in the wireless links, performance
and resources are limited by the time-varying
channel performance, the scarce availability of
transmission spectrum, the employed modulation
type and the limited transmission power [6, 7, 8].
TCP/IP is a hierarchical model in which only
adjacent layers communicate with each other. But a
strict layered design is insufficient to cope with the
dynamics of ad hoc networks. So instead of using
TCP/IP, cross layer architecture can be a more
promising architecture for ad hoc networks.
Cross-layer design breaks the traditional network
design, where each layer of the protocol stack
operates independently and exchanges information
with adjacent layers only. In the cross-layer
approach information is exchanged among all
layers without following any hierarchy of protocol
stack [9] and performance is optimized by adapting
each layer. Cross-layering breaks the boundaries
between information abstractions to improve
performance of the network. The careful
exploitation of cross-layer protocol interactions can
lead to more efficient performance of the protocol
stack [10].
Physical layer, MAC layer and Network layer
together contend for the network resource in a
wireless network. The transmission power and data
rate decide at the physical layer affects MAC and
routing decisions. Scheduling and allocating the
wireless channel is done at the MAC layer, which
will determine the available bandwidth of the
transmitter and the packet delay. The available
bandwidth and delay in the transmission of packets
also affect the decision at the routing layer to select
the route. The routing protocol in the network layer
chooses the routes to transmit data packets. The
routing decision will change the contention level at
the MAC layer and accordingly the physical layer
parameters [11]. Thus all layers depend on each
other and any change on any layer will propagate to
all layers. Thus cross layer design can be a
promising alternative but unbridled design may
lead to negative consequences. CLD is to combine
the resources available in the different
communities, and create a network which can be
highly adaptive and QoS-efficient by sharing
information between different processes or modules
in the system.
3 DIFFERENT ARCHITECTURES
There are many cross-layer design proposals in
the literature. A survey of different cross layer
architectures are given in Table 1.
A common database that can be accessed by all
layers is discussed in [6]. The common database is
like a new layer, providing the service of
storage/retrieval of information to all the layers.
According to this architecture, all parameters of
layers are stored in database layer and all of the
layers can communicate at any time with any layer.
But all parameters of different layers need not to
store in database layer. If all parameters stores then
it requires large buffer size. Instead of doing this,
an additional hardware layer can be introduced
which can store parameters of hardware, operating
system etc, not the parameters of layers which are
already available at layers. So a cross layer
architecture is proposed.
Table 1: Different Architectures
Author
Name
Issues dicussed Basic principle Pros Cons
Vineet
shrivastava,
Mehul
motani[10]
Survey of
different
proposed cross
layer design in
literature is
given
How TCP/IP architecture
violation can be possible.
Very good
survey is
provided
Only survey is
given,
implementation
is not discussed

57
Vijay T.
Raisingahani
and Sridhar
Iyer[12]
An ÉCLAIR
local-view-based
cross
layer
architecture is
proposed
With legacy protocol stack it
consists of two main
components: an Optimizing
Sub-System, the cross layer
engine which contains many
Protocol Optimizers,
which are the “intelligent”
components of it, and
TuningLayers. Different
tuning layers are included in
TCP/IP architecture. These
tuning layers are new
interfaces. These new
interfaces are included at
different layers
Tuning
Layers.
provide the
necessaryAPI
s to the
protocol
optimizers for
interacting
with various
layers and
manipulating
the protocol
data
structures.
This architecture
requires new
interfaces.
Rolf Winter
et.al.[11]
proposed a
cross layer
architecture
called cross talk
is based on data
dissemination
They propose a data
dissemination process
consists of two data
management entities. One is
responsible for the
organization of locally
available information and the
other entity establishes a
network-wide or global view
of the same type of
information collected in the
local view.
In this
architecture
only data
packets are
transmitted
with different
values
piggybacked
and control
packets are
not
transmitted
Because of
piggybacking
packet size
increases and it
is obvious that
large packet
takes more time
to transmit and
receive,
consumes large
bandwidth
Vikas
Kawadia and
P.R.Kumar[1
3]
Different
principles of
cross layer
design are
discussed.
Principles of cross layer
designs are given like
Interactions and law of
unintended consequences,
dependency graph, timescale
separation and stability, chaos
of unbridled cross layer
design are given
Some
examples are
shown.
improvements
are not given
Enrico
Biermann,
Kathrin
Hoffman and
Julie Pedberg
Layered
architecture is
proposed
A new type of hierarchical
model is proposed. All layers
are different from TCP/IP.
Functions of
all layers is
only for
MANET
This architecture
is made in
consideration of
mobile nodes. It
can’t be used for
stationary nodes.
Marco Conti,
Gaia Maselli,
Giovanni Turi
and Silvia
Giordano[14]
Mobileman
architecture is
proposed
This architecture works on
energy management, security
and cooperation. A network
status is provided which can
communicate with all layers
and all layers can
communicate with network
status. Network status is like
repository.
Whenever a
protocol in
the stack
collects
information,
it will publish
this to the
repository
and thus
MobileMan
reference
architecture is
just a proposal so
far, no per-formance
analysis exists to
back up their
claims.

58
making it
available for
every other
protocol.
Frank Aune -
fa une[15]
A survey is
given on
different cross
layer proposals
LayerTriggers, Event Helix',
MobileMan cross layer
proposals are discussed
Very good
survey is
provided by
this paper
Implementation
is not given
Daniel
G.Sachs et
al[16]
GRACE
(Global
Resource
Adaptation
through Co-opEration)
system)
All system components
(hardware, network, and
operating and applications are
allowed to be adaptive).
These
adaptive
entities co-operate
with
each
other to
achieve a
system-wide
optimal
configuration
its cross-layer
approach
includes no
explicit
consideration of
cross layering
within the
networking
layers or
protocol stack.
Dzmitry
Kliazovich
and Fabrizio
Granelli [17]
WIDENS
(Wireless
deployable
Network
System) cross
layer design has
been proposed
It acquired the
interoperability, cross
layering and re-configurability
at the same
time.
This cross
layer
architecture
seems a
promising
one where
protocol
optimization
is based on
the local state
information
Real
measurement of
efficiency in
terms of
performance
parameter is not
given
Vineet
shrivastava,
Mehul
motani[10]
A common
database that
can be accessed
by all layers is
discussed
All layers are supposed to
store their parameters in
common database layer
Availability
of data in
common
database
which will be
accessed by
all layers and
can utilize
information
easily.
It requires large
buffer size. No
need of keeping
all layer
parameters in
separate
database.
4 PROPOSED CROSS LAYER-PERFORMANCE
ENHANCEMENT
ARCHITECTURE(CL-PEA)
AIM:
1. New architecture should be compatible with
traditional architecture. New architecture can
be used as cross layer architecture otherwise
act as traditional architecture when it is
required.
2. Extra interfaces should not required.
3. Packet sizes should not be increased.
4. Buffer size should not be increased.
5. It is not enough simply sharing the resources
among the different users, it should be done
efficiently and fairly.
Simultaneous optimization of all layers using
cross layer design is very difficult. Unbridled
design of cross layer may lead to spaghetti design

59
[13]. So Cross Layer-Performance Enhance
Architecture (CL-PEA) is proposed. In this
architecture it is proposed that instead of the
database layer, a system hardware layer should add
in the traditional protocol stack. This hardware
layer can have parameters like speed of a node,
battery status, operating system parameter etc. It
does not have parameters of other layers as already
available at different layers. The hardware layer
stores those parameters which are not parameter of
any four layers. The parameters of all layers can be
called at any other layer and decisions can be taken
accordingly e.g. A routing protocol at network layer
can use the data available at different layers and
decision can be taken on the basis of this. Vertical
coupled layers are sufficient to enhance the
protocol performance. All the layers can
communicate or exchange their information
vertically with each other and can take decision
accordingly as illustrated in Fig.1. Large buffer size
is not required as required in [10], new interfaces
are not required as used in [12] for this architecture.
Here simply all layers can exchange their
parameters to take effective decision. This
architecture is same as TCP/IP, the difference is
that the all layers can exchange their parameters to
take a decision and one more layer is added to
TCP/IP protocol stack named Hardware layer. CL-PEA
is very simple architecture and it is not
complex. Only one layer is introduced and all
layers can exchange their parameters for taking
decision by any layer protocol. With this
architecture performance of network can be
enhanced.
Fig. 1. Cross Layer Performance Enhancement
Architecture (CL-PEA)
The aim of cross-layer design is to preserve the
functions of a layered architecture and allow
performance improvement and adaptability [10,
13].
5 CONCLUSION
QoS in wireless ad hoc networks has become a
crucial issue due to the proliferation of internet
applications, services, emergence and deployment
of wireless ad hoc networks. To cope with
dynamics of adhoc network, a cross-layer
architecture CL_PEA is proposed for QoS. The
novel feature of this is the ability to enhance the
performance of layer protocols. This architecture is
used to reduce power consumption, route breakage,
and delay and to improve QoS parameters. In the
future we plan several improvements to this
architecture.
6 REFERENCES
[1] Energy Issues and Energy aware Routing in
Wireless Ad-hoc Networks. Rango, Marco
Fotino and Floriano De. s.l.: International
Journal of Communication System, Mobile
Ad-Hoc Networks: Protocol Design, pp. 281-
296. 2008.
[2] Energy Efficient Routing Protocol Avoiding
Route Breaks based on DSR, Ashish Kumar,
M. Q. Rafiq, Kamal Bansal International
Journal of Computer Applications (0975 –
8887), Volume 44– No.4, April 2012.
[3] Dynamic Source Routing in AdHoc Wireless
Networks, D.B. Johnson, D.A. Maltz, Kluwer
Academic, 1996
[4] ZAHEDI, KHALID, et al. "Enhancing the
Performance of DSR Routing Protocol Using
Link Breakage Prediction in Vehicular Ad Hoc
Network" International Journal of Computer
Networks and Communications Security 1.1
(2013).
[5] Efficient routing for wireless mesh networks
using a backup path. Amir Esmailpour, Nidal
Nasser and Jehad Al. May 2012.
[6] Integrating machine learning in ad hoc routing:
A wireless adaptive routing protocol. Brian
Russel, Michael L. Littman and Wade Trappe.
2011.
[7] Energy-Efficient Network Selection with
Mobility Pattern Awareness in an Integrated
WiMAX and WiFi Network. Wen-Hsin Yang,
You Chiun Wang, Yu-Chee Tseng and Bao-

60
Shuh P. Lin. International journal of
communication system, 2000.
[8] Cross-layer error recovery in wireless access
networks: The ARQ proxy approach Dzmitry
Kliazovich, Simone Redana, Fabrizio Granelli.
International Journal of Communication
System, 2011.
[9] Survey on Mobile Ad Hoc Network Routing
Protocols and Cross-Layer Design. S.I. : Liang
Qin and Thomas Kunz, Carleton University,
Systems and Computer Engineering,
Technical Report SCE-04-14, August 2004
[10] Cross-Layer Design: A Survey and the Road
Ahead. Vineet Srivastava, Mehul Motani. s.l.:
IEEE Communications Magazine, December
2005.
[11] CrossTalk: Cross-Layer Decision Support
Based on Global Knowledge. Rolf Winter,
Jochen H. Schiller, Navid Nikaein and
Christian Bonnet: IEEE Communications
Magazine, January 2006.
[12] ECLAIR: AnEfficient Cross Layer
Architecture for Wireless Protocol Stacks. , V.
T. Raisinghani and Iyer. Sridhar: WWC, 2004.
[13] A cautionaryperspective on cross-layer design.
Vikas kawadia, p. R. Kumar. IEEE Wireless
Communications, February 2005.
[14] Cross layering in mobile Ad Hoc Network
Design. Marco Conti, Gaia Maselli, Giovanni
Turi, SylviaGiordano. s.l.: IEEE Computer
Society, February 2004, pp. pages 48-51.
[15] Cross-Layer Design T utorial. faune, Frank
Aune –. s.l.: Published under Creative
Commons License, Nov, 2011.
[16] GRACE: A Hierarchical Adaptation
Framework for Saving Energy. al, Daniel
G.Sachs et. ACEED 2005.
[17] A Cross-layer Scheme for TCP Performance
Improvement in Wireless LANs. Granelli,
Dzmitry Kliazovich and Fabrizio. s.l.: IEEE
Communications Society, Globecom 2004, pp.
841-844.

Cross Layer- Performance Enhancement Architecture (CL-PEA) for MANET

More Related Content

What's hot (20)

Viewers also liked (20)

Similar to Cross Layer- Performance Enhancement Architecture (CL-PEA) for MANET (20)

Recently uploaded (20)

Cross Layer- Performance Enhancement Architecture (CL-PEA) for MANET