ISSN 2278-3091

Volume 10, No.4, July - August 2021

International Journal of Advanced Trends in Computer Science and Engineering
Available Online at http://www.warse.org/IJATCSE/static/pdf/file/ijatcse231042021.pdf
https://doi.org/10.30534/ijatcse/2021/231042021

Efficient Quality of Service Aware Routing In WSN Using

Multi Channel Communication
S.Arockia Babi Reebha1,Dr.T.Suresh2,Dr.D.Saravanan3
1
Research Scholar,Assistant Professor,Department of Computer Science & Engineering,
Pavendar Bharathidasan College of Engg.&Tech, [email protected]
2
Associate Professor, Department of Computer Science & Engineering, FEAT,
Annamalai University, [email protected]
3
Professor, Department of Computer Science & Engineering,
Pavendar Bharathidasan College of Engg.& Tech, [email protected]

ABSTRACT
QoS can be defined as a set of requirements in the service
This paper focuses on ensuring the quality of service to be provisioned in the network during data transmission.
within a Wireless Sensor Network using multi-channel QoS can be considered as a measurable level of service
communication and considering the channel interference provided to the users of the network and it can be
instead of a single channel communication. A interference characterized by the probability of packet loss, bandwidth
aware and probabilistic approach is adopted in this paper to availability, and end to end delay in packet transmission. In
improve the overall efficiency and performance of the conventional wired networks the data transmission between
network. The topology control and QoS provisioning in IEEE the devices are end-to-end in nature without having
802.11 based network with dynamic traffic load is studied knowledge on the type and characteristics of the transmitted
under a simulated environment. Based on the available data. Therefore maintaining end-to-end delay, throughput,
transmission links between nodes established by the channel and packet loss to an acceptable range will ensure the quality
assignment, the set of next highly probable nodes for of service. But the same is not applicable in the case of WSN
transmission can be detected. This paper follows the as the sensor device does not communicate in end-to-end
probability theory and extends the channel interference fashion rather they transmit the sensed data to the
model to give a routing mechanism which implemented using neighboring device and in turn the neighbor nodes forward
a cross-layer (information from MAC layer sent to routing the data packets towards the destination (sink or base
layer for routing decision) method. The simulation results station). The major phenomena to be considered apart from
proved the efficiency of the proposed approach in terms of routing the data packet is the energy cost for transmission.
better packet delivery ratio, throughput, jitter and end to end The various QoS parameters including coverage and network
delay under various network traffic scenarios. life time depend on the energy spent during the transmission.
The coverage problem may rise when the sensors deployed at
Key words: multi-channel communication, probabilistic different geographic locations couldn’t observe the
approach, channel interference, cross-layer atmosphere and communicate the information to the
destination. This can be due to noisy wireless communication
1. INTRODUCTION channel, or improper management of the network
components [10]. The energy cost for transmission depends
A WSN – Wireless Sensor Network consists of multiple on the process of finding an optimal route to the destination
tiny sensor devices scattered at multiple geographic locations from the source.
which has limited energy resource (most probably they are
battery operated). Each sensor device is equipped with a WSN is a scalable network and when the number of
processor, storage unit and a wireless communication device sensor devices increases beyond a threshold level then the
(transceiver). These devices have the ability to observe the QoS tend to deteriorate. For this reason designing a optimal
atmosphere or the surrounding environment. The sensed data routing algorithm and appropriate technique for maintaining
can be processed and communicated to a destination node QoS in the network is desirable. When provisioning QoS for
referred as the base station which receives and analyzes the a WSN network the following factors are to be considered
data collected from multiple sensor devices. The overall
lifetime of a sensor device depends on the remaining energy 1. WSN applications are not end-to-end
in the battery source. The radio frequency data transmission 2. Bandwidth is not a major concern for a single sensor
consumes most of the energy available in the device[9]. device. Because of the burst nature of the sensor
Energy conservation is a main objective in many of the traffic, bandwidth can be an essential consideration
routing protocols design and the main aim in any network is for a sensor group.
to maintain high Quality of Service (QoS) [7]. Many MAC 3. Since multiple sensor devices observe the same
protocols are designed to listen to the channel at a frequent environment there is a higher probability of
interval instead of continuously listening the channel’s redundancy in the data and hence the packet loss of
availability for communication [8].
2798
S.Arockia Babi Reebha et al., International Journal of Advanced Trends in Computer Science and Engineering, 10(4), July – August 2021, 2798 – 2802

a particular nodes data can be tolerated up to a when compared to a single channel. The multi-channel
certain extent. communication mode helps to reduce the impact of channel
4. WSN applications are mission critical which shows interference. When the sensor devices are equipped with a
their importance [1]. multi-channel communicable transceiver then each sense the
availability and send their data through free channel. Two
QoS provisioning must be scalable and must be able to nodes can communicate when they are using the same
support a large number of sensor devices, the QoS should not channel for communication. The increased throughput is due
get degraded when the number of sensor increases. The to the possibility of provisioning permission to parallel
unique characteristics of the WSN poses certain challenges in transmissions among the neighboring nodes. Whereas in case
QoS provisioning when compared to challenges when of a single channel network simultaneous transmissions are
compared to generic network. WSN applications are growing not allowed between two neighboring nodes as the channel is
in large scale and it has become a hot topic for research. QoS shared between the nodes. When multi-channel wireless
of WSN can be characterized by the following factors communication approach is used the problem of collision can
including reliability, timeliness, availability, and security. be avoided as long as the nodes transmit in different
The quality of service provided by the network can be channels. QoS aware routing of data packets in multi-hop
quantified by the following parameters such as throughput, wireless channel poses many challenges due to the
end-to-end delay, packet delivery ratio, and jitter. The QoS interference. Channel interference is a major concern even in
provisioning mechanism adopted in the network must ensure case of multi-channel wireless network when two nodes
the implementation of certain strategies to prioritize flow of attempts to transmit in the same channel [12].
data packets to the destination [11]. The list of strategies that
can be adopted in each of the WSN stack of protocol for QoS An another important challenge to be addressed in case of
provisioning is shown in Table. 1. wireless networks is to reduce the overhead caused due to
transmission of small data packets including the TCP ACKs
Table 1. Summary of strategies for QoS enhancements in WSN in case of large scale IEEE 802.11 wireless networks. A sub-
optimal solution to this issue can be provided by adopting a
frame level or packet level aggregation of data packets.
Layer QoS Strategy Remarks
Packet aggregation can be implemented at the IP layer
without any specific hardware requirements which can be
Application Compression Latency, Available
integrated with any wireless networks easily. In an end-to-
bandwidth
end aggregation category the destination node is capable of
Layer de-aggregating the packets. But in contrast the gateway node
Adjust sensing rate Network Lifetime, can be provisioned with methods to de-aggregate the data
QoS packets before forwarding it to the destination network. End-
to-end aggregation helps to improve the network throughput
Data reduction Latency, Available when constant bit rate is followed.
techniques Bandwidth
In a multi-hop network each node attempts to evenly
Network Routing protocols Reliability, distribute the traffic load to different neighboring nodes thus
Layer Latency, N/W utilizing the spatial and frequency diversity. In contrast
lifetime aggregation attempts to combine multiple packets and
communicates to one of the neighboring node. Packet re-
Packets priority Reliability, ordering has to be executed at the destination node in case of
Latency, N/W multi-path routing. One of the easiest solutions for packet re-
lifetime ordering is to use buffers at the destination node for re-
ordering the packets.
Physical Energy Aware MAC N/W lifetime
Layer This research paper aims at improving the QoSin a
Selection of low Reliability, N/W Wireless multi-hop network by adopting a multi-channel
interference channels lifetime communication and following a multi-path routing along
with packet aggregation. The objective is to balance the load
MAC that avoid Reliability, N/W with in the network by distributing the traffic to multiple
collisions lifetime hops and reduce the probability of bottleneck nodes. The
adopted two approaches have conflicting objectives and may
Link Layer Channel Surfing Reliability, N/W have negative impact on the performance of the network. The
lifetime algorithm works on the local node and do not add any
overhead as no need of transferring extra information
Modifying signal N/W lifetime between the nodes. End-to-End packet delay may increase as
power there is a need for packet re-ordering at the receiving node.

Using low Reliability, N/W 2.RELATED WORK

interference channel lifetime
Few of the existing researches related QoS provisioning
In general the throughput in a wireless network can be for WSN focus enhancing the life time of the network by
increased by following a multi-channel communications energy harvesting from the network environment. This
2799
S.Arockia Babi Reebha et al., International Journal of Advanced Trends in Computer Science and Engineering, 10(4), July – August 2021, 2798 – 2802

facilitates the network nodes to perform more data intensive them and then will include multiple edges from one node
communication tasks [3-5]. In [14] a multichannel multi-hop to another node.
communication protocol based on TDMA approach suitable
{1,2}
long range transmission is proposed. The throughput in the
network is affected by the interference in the channel B
1 2
communication and it is obvious that energy harvesting {1,2}
procedures introduce channel interference problem. In [6] the 2 {2, 3}
scheduling algorithm for the data routing and charging tasks A D
is proposed to improve the lifetime of the network nodes and
3
prevent data loss due to channel interference caused by the
charging activity. 2
C
The available sensor devices in the market have an
electronic radio chip (CC2420 / CC2520) which enables the {1, 3} 1
multichannel communication in a WSN node. Multichannel E {1, 2}
communication ensures high throughput as it supports
parallel transmission of data packets through multiple
channels. For enhancing the performance of the WSN many Figure 1. Topology of the Network after Channel Assignment
of the literatures attempts to lower the issues caused by
interference, jamming, and congestion in the network. Figure 1 shows the schematic view of the network
Multichannel routing protocols including MMSN [15], topology after channel assignment. In the network only half-
TMMAC [16] and MCMAC [17] show better performance duplex communication are allowed to reduce the self-
when compared to single channel protocols [18]. interference. So that only one NIC out of multiple NIC
carried by a node can be used at a time for transmission/
3.MULTI-CHANNEL BASED PROBABILISTIC reception. With this topology it is obvious that for any pair
ROUTING of nodes denoted as , and if ∈ then ∈ .
Presuming that the nodes , , , of the wireless network
This section presents the polynomial time optimal routing obeys the following constraint ( , ) ≤ and ( , ) ≤
approach followed in this paper which solves the high and there exist a common channel ∈ ( ) ∩ ( ) ∩ ( ) ∩
bandwidth consumption in other routing approaches. Packet ( ). The links = ( , ; ) and ′ = ( , ; ) share the
flow based route is estimated which cannot be a single path common channel and will suffer from collision. Hence it
routing mechanism as it not possible to estimate route in becomes essential to assign channels that define a topology
polynomial time subject to both inter and intra flow with low interference.
contention. There may be a number of single paths exists
between the source node and destination node. The routing A solution is to identify the next hop node based on the
problem can be resolved using a probabilistic approach [8]. probability of the interference and an interference metric that
Let the network topology is represented as =( , ) helps to ensure connectivity among the nodes of the network.
where A is the wireless channel used for communication. For The successful transmission of data packets depends on the
a new packet delivery request from a sender node , and SINR value estimated at the destination node which is
destination node , and required bandwidth . The aim is to defined as follows
find a route that can be represented by the digraph ( ′ =
( ′ , ′ )). The multi access at the MAC layer of each node is
= (1)
handled by the IEEE 802.11 Distributed Coordination ∑ ∈
Function (DCF). The network has a total of wireless
channels operating at different frequencies and each node is Where and represent the signal power of node j
equipped with network interface cards where ≤ . Each received from nodes i and k; Ti denotes the set of nodes that
node is associated with a fixed transmission power and attempts to transmit simultaneously with node j. For
interference range > . Each NIC is tuned to successful transmission of data packets from any node to
communicate in a separate channel to fully utilize the node j, then the SINR at the node j must be less than which
available resources. As described earlier the network can be is considered as the minimum SINR value that the receiver
represented using a directed graph with set of nodes node should have. The signal receiving power estimated at
represented as and set of edges represented as . The node j while receiving a data packet sent from node i can be
wireless link between any two node can be represented by the defined as = where represents the signal
( )
edge ( , )in . A channel assignment strategy is followed to
assign wireless channel for communication to each node attenuation factor (2 ≤ ≤ 6), is the hop distance
which is denoted as ( ) from the set of available wireless between the sender and receiving node. As defined already
channel in the network. The channel assignment defines the the transmission range is fixed as Rand the interfering
topology that is denoted as = ( , ). An edge = neighboring nodes for a receiver lie within a circle of radius
( , ; ) on the wireless channel ( ) = connecting two dLthen ≤ .
nodes and can exist only if the distance between the
nodes ( , ) < and ( ) ∈ ( ) ∩ ( ). Two nodes of the
With this background the probability that a receiving node
network may share two or more common channels between
will successfully receive a data packet can be define

2800
S.Arockia Babi Reebha et al., International Journal of Advanced Trends in Computer Science and Engineering, 10(4), July – August 2021, 2798 – 2802

( )
mathematically as = Pr > =( ) .
100
( )
=( ) is the probability a receiving node Proposed

Packet Delivery Ratio (%)

successfully receives the packets when interfering nodes lie 80
at a farthest circle of node j. Multi-path
AODV
60
The Network Allocation Vector (NAV) estimated at the
MAC layer can be used to approximate the channel
40
occupancy period and PNAV can be defined as the sum of
the NAV/ unit time.
20
∑
= ∈∆∆ (2)
0
0 20 40 60 80
The probability of transmitting a data packet at node i can
be estimated as follows Transmission Rate (packets/s)
Figure 2. Comparison of Packet Delivery Ratio
∑ ∈∆
=1− 1− (3) The variation in the packet delivery ratio when the
∆
transmission rate is varied during the simulation of the
network is presented. The higher packet delivery ratio is
Here NAV can be read from the MAC layer and sent to
achieved in case of proposed probabilistic based routing
the routing layer and the maximum communication range
approach as it considers the interference in the route
Rmax can be defined as
discovery process.

= (4) 100
90
Considering the interference at the sending and receiving Throughput (kbps) 80
node the routing metric can be defined as 70
60
50
= . (5) 40
30 Proposed
20
where denotes the distance between the node j
10 Multi-path
(neighboring node of sender) and the destination node. The AODV
0
routing decision is taken by estimating the probability of
transmitting a data packet successfully from the sender node 0 20 40 60 80
to a destination node. The sender and every intermediate Transmission Rate (packets/s)
node on the routing path select the next hop node based on
the estimated probability. The neighboring node with the Figure 3. Throughput Analysis
highest probability is selected as the next hop node for
transmission. Figure 3 shows Throughput analysis is presented. The
throughput is plotted against the transmission rate in the x-
4.SIMULATION AND RESULTS axis. The proposed approach is capable of sending more
packets in a period of time as it considers the interference
For evaluating the performance of the probabilistic during the routing process.
routing approach the simulations are done in network
simulator. The density of nodes in the network is varied 0.12
between 300 and 600 with in a network region of 300 x 300 Proposed
m. The maximum communication range of node is fixed as 0.1
Multi-path AODV
50m. Multiple nodes are allowed to transmit at the same
Average Jitter

0.08
instance as each node is configured with multi-channel
wireless radio. During the simulation the sending rate is 0.06
varied and the performance of the proposed model is
0.04
evaluated. The total duration of the simulation is conducted
for a period of 500 seconds. The simulation includes an 0.02
analysis of multiple sources and a single sink. The
performance of the network is analyzed using the following 0
metrics including jitter, average delay, and average -0.02 0 20 40 60 80
transmission route length under various network conditions.
Sending Rate (packet/s)
Figure 2 shows the performance of the probabilistic routing
is compared with the performance of the multi-path AODV.
Figure 4. Plot of Average Jitter
2801
S.Arockia Babi Reebha et al., International Journal of Advanced Trends in Computer Science and Engineering, 10(4), July – August 2021, 2798 – 2802

Figure 4 shows the average jitter of the proposed approach [10] Chen, D., and Varshney, P. K. QoS support in wireless
is smaller when compared to that of the multipath distance sensor networks: A survey. In Proc. of the 2004
vector routing. International Conference on Wireless Networks (ICWN
2004) (Las Vegas, Nevada, USA, June 2004), vol. 1, pp.
5.CONCLUSION 227–233.
[11] D Petrović, R Shah, K Ramchandran and J Rabaey,
A multichannel interference aware topology control and “Data Funneling: Routing with Aggregation and
QoS enhanced routing in IEEE 802.11 based multichannel Compression for Wireless Sensor Networks” In
WSN is studied. A probability based routing mechanism is Proceedings of First IEEE International Workshop on
derived which enabled the sending network nodes to discover Sensor Network Protocols and Applications, Page(s):
the routing path one hop at a time. The proposed approach is 156- 162, May 2003.
implemented using cross layer design as the network [12] Tang, Jian, GuoliangXue, and Weiyi Zhang.
allocation vector read from the MAC layer is communicated "Interference-aware topology control and QoS routing
to the routing layer for estimating the sending probability. in multi-channel wireless mesh networks."
This approach considers both the sender and receiver In Proceedings of the 6th ACM international
interference levels and estimates a routing metric which is symposium on Mobile ad hoc networking and
used deciding the next hop node in the routing path. The computing, pp. 68-77. 2005.
simulation results exhibited the superior performance of the [13] Tang, J., Xue, G., & Zhang, W. (2005, May).
proposed approach when compared to a conventional multi- Interference-aware topology control and QoS routing in
path routing protocol. multi-channel wireless mesh networks. In Proceedings
of the 6th ACM international symposium on Mobile ad
REFERENCES hoc networking and computing (pp. 68-77).
[14] Toldov, Viktor, Laurent Clavier, and Nathalie Mitton.
[1] S. Kumar, Z. Feng, F. Hu, Y. Xiao, “E2SRT: enhanced "Multi-channel Distributed MAC protocol for WSN-
event-to-sink reliable transport for wireless sensor based wildlife monitoring." In 2018 14th International
networks,” Wireless Communications and Mobile Conference on Wireless and Mobile Computing,
Computing, Volume 9, Issue 10, pages 1301–1311, Networking and Communications (WiMob), pp. 1-8.
October 2009. IEEE, 2018.
[2] Lu, J. and Wang, X., 2012. Interference-aware [15] G. Zhou, C. Huang, T. Yan, T. He, J. A. Stankovic, and
probabilistic routing for wireless sensor T. F. Abdelzaher, “Mmsn: Multi-frequency media
networks. Tsinghua Science and Technology, 17(5), access control for wireless sensor networks,” in 25th
pp.575-585. IEEE International Conference on Computer
[3] Tang, Q.; Yang, K.; Wang, J.; Luo, Y.; Li, K.; Yu, F. Communications, ser. INFOCOM ’06, vol. 6, April
Wireless Sensor Network MCDS Construction 2006, pp. 1–13.
Algorithms with Energy Consideration for Extreme [16] J. Zhang, G. Zhou, C. Huang, S. H. Son, and J. A.
Environments Healthcare. IEEE Access 2019, 7, 33130– Stankovic, “Tmmac: An energy efficient multi-channel
33144. mac protocol for ad hoc networks,” in IEEE
[4] Cao, D.; Zheng, B.; Ji, B.; Lei, Z.; Feng, C. A robust International Conference on Communications, ser. ICC
distance-based relay selection for message ’07, June 2007, pp. 3554–3561.
dissemination in vehicular network. Wirel. Netw. 2018, [17] X. Chen, P. Han, Q.-S. He, S.-L. Tu, and Z.-L. Chen,
1–17. “A multichannel mac protocol for wireless sensor
[5] He, S.; Zhang, D.; Xie, K.; Qiao, H.; Zhang, J. networks,” in The Sixth IEEE International Conference
Distributed low-complexity channel assignment for on Computer and Information Technology, ser. CIT ’06,
opportunistic routing. China Commun. 2012, 9, 9–22. September 2006, pp. 224–224.

[6] Tan, Q., Qu, Y., Xu, K., Wang, H., Wang, D. and Shen, [18] Y. Wu, J. Stankovic, T. He, and S. Lin, “Realistic and
M., 2019. Interference-aware lifetime maximization efficient multichannel communications in wireless
with joint routing and charging in wireless sensor sensor networks,” in The 27th IEEE Conference on
networks. CCF Transactions on Networking, 2(3), Computer Communications, ser. INFOCOM ’08, April
pp.188-206. 2008.
[7] Akyildiz, I. F., Su, W., Sankarasubramaniam, Y., and
Cayirci, E. Wireless sensor networks: a survey. IEEE
Communications Magazine 40, 8 (August 2002), 102–
114.
[8] Willig, A., and Karl, H. Data transport reliability in
wireless sensor networks – a survey of issues and
solutions. In Praxis der Informationsverarbeitung und
Kommunikation (April 2005), vol. 28, pp. 86–92.
[9] Estrin, D., Girod, L., Pottie, G., and Srivastava, M.
Instrumenting the world with wireless sensor networks.
In International Conference on Acoustics, Speech, and
Signal Processing (ICASSP 2001), Salt Lake City, Utah,
May 2001. (2001).

2802