International Journal of Scientific and Research Publications, Volume 3, Issue 9, September 2013 1

ISSN 2250-3153

A Review on Cross Layer Routing Techniques

1
Prathibhavani P M, 2Vijayakumar and 3Reshma K
1
Assistant Professor, Grade-3, Department of Information Science and Engineering, Acharya Institute of Technology, Bangalore-90.
2, 3
3rd Semester M. tech Students. Department of Information Science and Engineering, Acharya Institute of Technology, Bangalore-90.

Abstract- Ever since wireless sensor networks (WSNs) have In WSN (wireless sensor networks) the node has small volume
emerged, different optimization techniques have been proposed and uses the battery for energy supply, therefore has created that
to overcome their constraints like energy efficiency, delay, node’s processing, memory property, communication range and
finding the best route, etc. Furthermore, the proposal of new energy is limited. In WSN, reducing the end-to-end delay and the
applications for WSNs has also created new challenges to be energy consumption is very important. The limitation of energy
addressed. The concept of Cross-layer gave a verity of is a fundamental problem in wireless sensor networks (WSNs)
optimization techniques to the world of sensor networks and the because of the limited battery capacity of sensor nodes.
approaches have proven to be the most efficient optimization Communication protocols for WSNs, including routing and
techniques for these problems, since they are able to take the MAC layer protocols should be designed energy-efficiently.
behavior of the protocols at each layer into consideration. Thus, Traditional wireless MAC protocols such as IEEE 802.11 are not
this survey proposes to identify the key problems of WSNs and available for this purpose; since in these protocols, nodes are
gather available crosslayer solutions for them that have been required to keep awake to listen to the medium even when the
proposed so far, in order to provide insights on the identification network becomes idle. This inefficient idle-listening mechanism
of open issues and provide guidelines for future proposals. In this wastes substantial energy.
paper, the need for cross layer design in WSN and the interaction
between the layers is discussed first and later we discussed the The literature survey is discussed in the rest of the paper.
different types of cross layer techniques A. Cross Layer Design
Generally speaking, cross-layer design refers to protocol
Index Terms- Wireless Sensor Networks, Optimization design done by actively exploiting the dependence between
Techniques, Cross Layer design, Constraints. protocol layers to obtain performance gains. This is unlike
layering, where the protocols at the different layers are designed
independently. Cross-layer design states that parameters of two
I. INTRODUCTION or more layers can be retrieved and/or changed in order to
achieve an optimization objective. The concept of cross-layering
D ue to the dynamic topology, limited resources and
unpredictable channel conditions, the strict layered design is
not flexible enough to cope with the wireless sensor network
has been first proposed for TCP/IP networks, when wireless links
were deployed. Since the TCP/IP stack has been proposed for
environment. The widespread use wireless sensor networks be a wired connections, there was a loss of performance when
more and more important and common way to provide Internet wireless technology became part of existing networks.
connectivity. The worldwide success of the Internet, mainly Lately, cross-layering is a field that has been attracting
determined by a layered architecture, has promoted the adoption more attention in WSNs research and it is still in its early
of a similar solution for wireless sensor network. However, a development in this type of networks since it has not been
strict layered design is not flexible enough to cope with the deployed on many test beds or networks yet. Common goals of
dynamics environments, and will thus prevent performance cross-layer optimizations in WSNs are reduction of energy
optimizations. It is because of the unpredictability and consumption, efficient routing, QoS provisioning, and optimal
unreliability of the underlying wireless medium that research on scheduling, as can be verified throughout this work. There are
cross-layer design in wireless ad hoc network and sensor several studies on cross layer protocols for WSNs. To address the
networks has recently attracted a significant interest [10]. receiver based contention, congestion control and duty cycling in
The traditional approach to optimizing performance by WSNs. Some of the most used protocols on cross layer design
separately optimizing different layers of the OSI model may not and new protocol proposals are going to be presented, along with
be optimal. In order to obtain the best results, it might be the optimizations they provide, in order to gather a small
necessary to perform optimization using the information database on what has been proposed so far. Cross-layer design
available across many layers. The concept of cross-design is has been adopted on wireless sensor networks (WSNs) to
based on architecture where the layers can exchanges improve their performance by discussing the state-of-the-art
information in order to improve the overall network literature on the subject. Also, open issues on cross-layering will
performances. Therefore, various cross-layering approaches, be identified to facilitate the work of researchers towards further
where protocol layers actively interact, exchange inherent layer improvements applied to WSNs.
information and fine tune their parameters according to the After presenting the definition, we identify some basic
network status are becoming increasingly popular. types of cross-layer designs and present relevant examples from
the literature. This serves four purposes: first, it clarifies and
illustrates our definition of cross-layer design; second, it creates a

www.ijsrp.org
International Journal of Scientific and Research Publications, Volume 3, Issue 9, September 2013 2
ISSN 2250-3153

taxonomy for classifying existing cross-layer design proposals; technique similar to our approach to find order statistics,
third, it highlights the different interpretations of cross layer specifically concentrating on the median. The work did not
design in the literature and shows how they can be seen in a more explore extreme value aggregation with more than just a
unified way; and finally, it provides a framework for evaluating straightforward broadcast- converge cast search.
the implementation concerns raised by different kinds of cross- In short, at the beginning of the query, every sensor node in
layer design proposals. As expected, the different kinds of cross- the tree passes the query to all of its children. Once all of the
layer design proposals raise different implementation concerns. children return with the maximum readings of their own
After creating the taxonomy of the cross-layer design proposals, respective subtrees, the parent sensor node then returns only the
we similarly categorize and discuss the initial proposals for maximum reading of its own subtree back up to its parent in the
implementing cross-layer interactions and highlight briefly for network. Building on top of this work, went further by avoiding a
which kind of cross-layer design proposals the different count of distinct data points in the network as well as use
implementation methods are suitable. Differently from the previous query results to improve performance. Both concentrate
conventional networks layers, WSNs consider only the following on an ad hoc tree structure for WSN. This problem can be
layers: application (APP), network (NWK or NET), medium overcome by the proposed cross-layer probabilistic data
access control (MAC), and physical (PHY). Although there is no aggregation method offers superior performance while
transport layer, since it is complex and it would waste sensors demanding much lower energy consumption than existing
energy, some WSN protocols have been designed for congestion algorithms. In comparison to the naive method, which consumes
control and reliable end-to-end communication. power as O (N log (N)), our algorithm offers a marked
improvement at O (1) [2].
B. Cross Layer Interaction
Cross layer interaction means allowing communication of B. Fairness in delay-aware cross layer data transmission scheme
layers with any other possibly non-adjacent layers in the protocol (FDRX)
stack. Traditionally, the network protocols are divided into In paper [3], particularly, delivery of time-critical data
several independent layers. Each layer is designed separately and becomes a significant challenge. For instance, the impact of
the interaction between layers is performed through a well- delay on smart grid applications, considering a WSN that is used
defined interface. for condition monitoring of wind turbines. Timely delivery of
The main advantage of layering is architectural flexibility critical data has been found to be a significant challenge in such
but layering approach is not efficient for wireless networks. applications. This can be solved by designing separately fairness-
Cross layering came into existence because of highly variable aware cross layer scheme and a delay-aware cross layer
nature of links used in the wireless communication systems and technique for WSNs. But here it is proposed a cross layer scheme
due to resource poor nature of the wireless mobile devices there that is both fairness-aware and delay-aware. Our fairness in
has been multiple research efforts to improve the performance of delay-aware cross layer data transmission scheme (FDRX) is
the protocol stack by allowing cross layer interaction by wireless based on delay-estimation and data prioritization steps that are
systems. Because of QoS, energy consumption, poor performed before the data transmission by the application layer.
performance, wireless links, mobility, packet loss, delay [3].
problems observed in the wireless networks much attention is
paid in the cross layer interactions. Typically, sensor nodes avoid C. Link Distance Reliability Cost (LRC) Based Cross Layer
direct communication with distance destination since high Route
transmission power is required to achieve reliable transmission. Shortest path scheme is the most common criteria adopted
Instead in WSNs, sensor nodes communicate by forming a multi by the conventional routing protocols proposed for WSNs. The
hop network to forward messages to the collector nodes, which is problem is that nodes along shortest paths may be used
also called the sink node. frequently and the batteries may exhaust faster. The consequence
is that the network may become disconnected leaving disparity in
the energy and disconnect the networks. Therefore the shortest
II. LITRATURE SURVEY path is not the most suitable metric adopted for routing decision.
A. CLB (Cost Link Based) for routing We have considered the network that is less mobile. MAC and
In paper [1], the physical layer (PHY) plays a very the physical layer information are explored for routing. It can be
important role in wireless communication due to the challenging solved by designing the new routing protocol, concept of the
nature of communication medium. The power consumption of minimum hop scheme is considered.
wireless devices heavily relies on physical layer. The medium We propose Link distance Reliability Cost (LRC) based
access control (MAC) layer manages wireless resources for PHY cross layer route discovery for WSN (Wireless Sensor Network)
layer and directly impact overall performance. This solved by that improve the end-to-end throughput, delivery ratio and node
defining a new cross layer scheme CLB (Cost Link Based) for energy consumption in WSNs. This approach also decreases the
routing [1]. Data aggregation with low energy consumption is chance of loss of data because we have assured that battery
task of finding extreme (maximum, minimum) sensor readings capacity is not below the defined value [4].
within a WSN. Extreme value aggregation has found many
important applications in WSN development. Work by used D. TCLM Model
compression to find approximate solutions to queries to deal with In paper [5] the authors build up a reputation space and
the large amount of data. Used a threshold and binary search trust space in WSNs, and define a transformation from reputation

www.ijsrp.org
International Journal of Scientific and Research Publications, Volume 3, Issue 9, September 2013 3
ISSN 2250-3153

space to trust space. It is proposed a trust-based LEACH protocol Transmission energy for transmitting a k –bit message to a
to provide secure routing, which is an integration of a trust receiver at distance d can be computed as
management module with a trust-based routing module. The trust ETx (k , d ) = ETx − elec (k ) + ETx −amp (k , d )=E elec * k ε * k
management module is responsible for building trust *d2
relationships among sensor nodes with novel methodologies to
provide efficient monitoring, trust exchange and evaluation. There are two common network architectures in WSNs
The trust-based routing module is a modified version of Multi-hop and hierarchical (i.e. clustering). To enable the system
original protocol with the same head-election algorithm and to cope with dynamic load and cover a large area of interest
working phases, while having trust-based decision-making. A without service degradation, cluster-based routing protocols get
multi-angle trust mechanism for nodes in Wireless Sensor more spotlight. We compare our SCL protocol with the
Networks which adding the sensing data and the node's energy in performance obtained using LEACH-C.
the factors of trust assessment in addition to communication , and
new trust models to calculate the trust values of communication F. Simulation Model of SCL
trust, the sensed data and the node's energy. The proposed a new We propose to create a cluster-based MAC protocol that
protocol TTSN (Task-based Trust Framework in Sensor takes advantage of the mechanisms of both CSMA and TDMA
Networks) to construct a trust framework model in wireless protocols. The following assumptions are made in the
sensor networks. The sensor node has different trust rating for architecture of the network:
different task.
The proposed model use watchdog scheme to observe the 1. All nodes are homogenous, i.e. all have equal capacity
behavior in different events of these nodes and broadcast their in terms of power, computation and communication
trust ratings. The authors have proposed an approach called capacity.
BTRM-WSN which is a bio-inspired trust and reputation model 2. The nodes in the network are aware of information
for WSNs aimed to achieve to most trustworthy path leading to about themselves, like the node IDs, locations and
the most reputable node in a WSN offering a certain service. It is energy levels.
based on the bio-inspired algorithm of ant colony system. Most 3. Nodes are stationary.
of existing methods are designed for MANETs and suppose that 4. The Sink node has unlimited resource.
each node have the full route from source to destination (i.e. 5. The number of transmitting nodes is varying, which
designed for source routing), while for the hop by hop routing means the system is event-driven, and traffic load is
they discuss only one hop neighbors not along the full route, that dynamic.
is not suitable for WSNs. In addition to that, no one used the
cross layer concept in computing and updating the trust values. The purpose of our work is to find a balance between
In paper [5] the data available at both DLL and Transport contention- and schedule-based protocols in the real world
layer to make the network-layer decision more accurate, which simulation. As the typical cluster-based architecture, the sensor
improve the performance of the trust model. Wireless Sensor nodes in the network are divided into clusters with one cluster-
Networks (WSNs) are vulnerable to attacks (selfish or malicious head (CH) in each cell, and these cluster-heads form the
i.e. misbehaving nodes) due to the nature of the wireless media, backbone of the sensor network. Therefore, all nodes in one
restricted resource and the natural co-operations of sensors. cluster only send data to their own cluster-head. As described in
Therefore, the security issue is very critical in WSN. The LEACH, to reduce the interference between clusters, we use
decision making in a WSN is essential for carrying out certain CDMA for the communication between CHs and Sink, while the
tasks as it aids sensors establish collaborations. This can be hybrid MAC protocol is used among nodes within each cluster.
solved by new model for trust in WSN, called A Trust-Based Just as in LEACH, the operation of the protocol is broken
Cross-Layer Model, which use cross-layer concept (ACKs from into rounds. For each round, it begins with a set-up phase, when
data link layer and TCP layer) to design trust-based model for the clusters are formed. Then a steady state phase follows, when
sensor networks that guarantee the trust route from source to data is collected by the cluster-heads then transmitted to the sink.
sink and isolate the malicious node. [5]. The time-line of such operation is depicted in Figure 1 below.

E. Cross Layer Design for Routing

In [9], there is a growing interest in the use of WSNs in
various applications, such as disaster management, border
protection and security surveillance. In these applications, the Figure 1: Shows an example of a Forwarding Mechanism
sensor nodes are usually expected to be remotely deployed in
large numbers and to operate autonomously in unattended In figure 2 the following algorithm, describes the operation
environments. In WSNs, the transmission medium is shared by of the proposed protocol. A detailed explanation is given bellow,
all nodes within each others’ transmission range, and because of and shows the main protocol scheme of the network.
limited amount of energy and scarcity of the medium spectrum, At the beginning of the procedure, all nodes, which are still
choosing appropriate network architecture for routing is deemed alive, send the information about themselves, such as their
to be an important step for sensor networks. locations and energy levels to the Sink node in one hop at a
specific time, which equals to id*Tinfo-adv, where id is the identity
number of the node, and TInfo −adv is a parameter being set as

www.ijsrp.org
International Journal of Scientific and Research Publications, Volume 3, Issue 9, September 2013 4
ISSN 2250-3153

Lhdr+Iinterval 3. X-MAC is similar to B -MAC but with the advantage

TInfo-adv = *g that it uses short preambles instead of long ones. This
Bw allows the nodes to send acknowledgments to the
preambles to trigger the data transmission which
Step 1: Information advertisements from all nodes to sink reduces idle listening. For this reason and because it
node. provides adaptation to the traffic load, we have selected
Step 2: After Time out occurs, Cluster -Head election X-MAC as the channel access method.
performed by the Sink.
Step 3: After Time out occurs, Clusters setup and Timeslots H. Congestion Control
Assignment for each node in the networks. The congestion control includes forwarding mechanism
(Step 1-3 is said to be SET-UP Phase) (Figure 3) and rate adjustment mechanism.
Step 4: After Time out Occurs, Hybrid MAC Protocol 1. Forwarding mechanics: when a routing node receives a
implementation for Data Collection intra-duster. packet, it doesn’t always forward the packet to a fixed
Step 5: After Time out Occurs, Data Relay from Client node, but selects a downstream neighbor node in its
Hosts to the Sink. routing table as the next hop randomly.
(Step 4-5 is said to be STEADY STATE) 2. Adjustment mechanism: the specific algorithm of
Step 6: After Time out Occurs, goto step 1 and continue the adjustment is as follows.
procedure.
Step 1: When congestion occurs, the adjustment factors of
Figure 2: Shows the algorithm for Information Advertisement probability n will update to n+1. Then the congestion degree C
will reduce greatly. The probability of selecting it to be the
A simple TDMA protocol is used based on our assumption downstream neighbor is also reduced. At the same time, the rate
that all nodes know this information. Furthermore, there is a adjustment will change the rate Rc to Rc/2.
timer for each state for avoiding overlapping [6]. Step 2: When the congestion removes, the adjustment
factors of probability n will reset to 0. The probability of
G. MAC Layer Protocols selecting it to be the downstream neighbor will also increase.
One of the main techniques for achieving low energy And the rate adjustment will change the rate Rc to Rc+R.
consumption in power restrained WSN is duty cycling. For this
approach, the MAC protocol of each node periodically cycles
between a sleep state and an awake state. To extend the network
life span, each node must undergo a periodic sleep cycle for as
long as it can, since the energy consumed in the sleep state is
much lower than in the awake state.

Some MAC protocols such as S-MAC use fixed duty

cycles. In some sensor network applications such as surveillance
or object-tracking, nodes are inactive most of the time, but when
an event happens, large amount of packets are generated which
yields to network congestion. In such conditions, MAC protocols
Figure 3: Shows an example of a Forwarding Mechanism
with fixed duty cycles experience data loss because they cannot
I. Cross-Layer Design Approaches
adapt themselves to the heavy traffic load. Therefore, in order to
solve this problem, new design approaches that adapt the duty A growing number of recent papers are focusing on the
cycles to the traffic load are being developed. Some load cross-layer development of wireless sensor network protocols
adaptable protocols are X-MAC, T-MAC and B-MAC. As the
and frameworks. Recent works on WSNs demonstrate that cross-
proposed algorithm considers the node’s busyness level, it is
layer design approaches result in a significant performance
necessary to have a MAC protocol which is adaptable to the
improvement. This kind of design is focused on two main aspects
traffic load.
are framework design and protocol design. The previous
1. T -MAC protocol is based on periodic sleep-listen approaches can basically be classified depending on the layer
schedules where nodes belonging to a cluster share the interactions.
same schedule. This protocol has the advantage in that it
Therefore, there are protocol designs based on interactions
reduces idle listening (since each node can only transmit
between MAC and Physical layers, between MAC and Network
during a fixed contention window) but it has the
layers, between Network and Physical layers, etc. This work
disadvantage that it needs synchronization.
focuses on the design of a multi-objective cross-layer algorithm
2. B-MAC is an asynchronous protocol that sends long for application to existing routing protocols based on the
preambles to inform other nodes that they must wait for interactions between MAC, Network layer routing, Physical and
an incoming transmission. This method has the
Application layers. One example of cross-layer routing protocol
advantage that it does not need synchronization but it
design is the Energy Aware Routing Protocol proposed. This
does force nodes into remaining awake during the whole
approach uses the residual energy of the nodes as a cost for the
preamble duration. probabilistic selection of routes with the objective of equalizing

www.ijsrp.org
International Journal of Scientific and Research Publications, Volume 3, Issue 9, September 2013 5
ISSN 2250-3153

the energy consumption of all nodes. This cross-layer method has [2] Minglei huang and yu hen hu, “Intelligent Cross-Layer Distributed Data
Aggregation in Wireless Sensor Networks”, Electrical and Computer
the advantage that it considers the health of all nodes from the Engineering, UW-Madison Madison, WI, huang8@wisc.edu , Electrical and
source to the destination and the disadvantage regards one Computer Engineering, UW-Madison Madison, WI, hu@engr.wisc.edu.
parameter (the residual energy) as the performance metric, which [3] Irfan s. Al-anbagi, melike erol-kantarci, Hussein t. Mouftah, “Fairness in
does not provide much flexibility. Another example is the routing Delay-Aware Cross Layer Data Transmission Scheme for Wireless Sensor
protocol that relies on local decisions in an attempt to equalize Networks”, University of Ottawa Ottawa, ON, Canada. Emails: {ialan055,
melike.erolkantarci, mouftah}@uottawa.ca.
the energy consumption of all nodes. This approach uses the
[4] Kanojia sindhuben babulal1, rajiv ranjan tewari, “Cross Layer Design with
level of busyness of the nodes, the number of children of each Link and Reliability Analysis for Wireless Sensor Network” institute of
node and the current node role to choose the paths locally. This technology, Nirma University, Ahmadabad – 382 481, 08-10 December,
methodology has the advantage that the routes can be maintained 2011.
easily, adding very little overhead, and that many parameters can [5] Hosam A. Rahhali, Ihab A. Alp, Samir Shaheen “A Novel Trust-Based
be included to make the routing decisions. Nevertheless, it has Cross-Layer Model for Wireless Sensor Networks” faculty of Engineering,
Cairo University, Giza, Egypt, hrahhal@ieee.org,,2Faculty of Engineering,
the disadvantage that decisions are made locally without Helwan University, Helwan, Egypt, ihabaly@ieee.org, faculty of
considering the entire path from the origin to the destination. Engineering, Cairo University, Giza, Egypt, shaheen@eng.cu.edu.eg.
These cross layer designs and other similar approaches are [6] Zilong Li, Weixia Zou, Tao Qi, “A cross-layer congestion control strategy
limited in that they attempt to only reduce or equalize energy in wireless sensor network”, Proceedings of IEEE IC-BNMT 2011.
consumption and this single performance optimization severely [7] Berta Carballido Villaverde, Susan Rea & Dirk Pesch, “Multi-objective
cross-layer algorithm for routing over wireless sensor networks”, 2009
restricts their flexibility. The objective of the tunable multi- Third International Conference on Sensor Technologies and Applications.
objective algorithm presented here is to balance multiple [8] Fei Tong, Wan Tang, Rong Xie, Lei Shu and Young-Chon Kim. “P-MAC:
performance requirements, as demanded by the overlaying a cross-layer duty cycle mac protocol towards pipelining for wireless sensor
applications while also considering the limited available node networks”. IEEE Communications Society subject matter Experts for
resources so that the system response to application demands is publication in the IEEE ICC 2011 proceedings.
dynamic and flexible. To achieve this goal, our cross-layer [9] Min Zhang, Alireza Babaei and Prathima Agrawal, “SCL: A Cross-Layer
Protocol for Wireless Sensor Networks”, 44th IEEE Southeastern
approach looks to exploit the advantages of previous cross layer Symposium on System Theory University of North Florida, Jacksonville,
methodologies, as highlighted earlier, firstly by considering the FL March 11-13, 2012.
complete source-destination routing path for decision making, [10] Xiongwei ren and jianqiang zhang, “Review of the Cross-Layer Design in
which gives a more global view of the network and secondly by Wireless Ad hoc and Sensor Networks”, Natural Science Foundation of
Hubei Province under Grant No. 2007ABA187.
using several parameters in selecting the path with an aim of
providing flexibility. In addition, parameters can be weighted
depending on the application running on the nodes and these
weighting factors can be dynamically reset post network AUTHORS
initialization if the application needs to change its priorities. First Author – Prathibhavani P M, BE ME (Phd), Acharya
Institute of Technology and prathibhavani@acharya.ac.in .
Second Author – Vijayakumar, BE (Mtech), Acharya Institute
REFERENCES of Technology and vijayakumar.mtcn.12@acharya.ac.in .
[1] Kanojia sindhuben babulal1, rajiv ranjan tewari , “Cross Layer Energy Third Author – Reshma K, BE (Mtech), Acharya Institute of
Efficient Cost Link Based Routing for Wireless Sensor Network”, Technology and reshma.mtcn.12@acharya.ac.in .
Department of Electronics and Communication. J.K Institute of Applied
Physics and Technology University of Allahabad, Allahabad, India.

www.ijsrp.org