Iisrt arunkumar b (networks)

International Journal of Networks (IJN)
Vol. 1, Issue. 1, April – 2015 ISSN (Online): 2454-1060
26
QoS - Based Packet Scheduler for Hybrid Wireless
Networks
1
B. Arunkumar, 2
Dr.R. Avudaiammal, 3
Dr.A. Swarnalatha
1
PG Scholar- Department of Applied Electronics,St.Joseph‟s College of Engineering,Chennai, INDIA
2
Professor- Department of ECE, St.Joseph‟s College of Engineering, Chennai, INDIA
3
Professor- Department of ECE, St.Joseph‟s College of Engineering, Chennai, INDIA
Abstract: Packet scheduling is the means by which data
(packet) transmission is governed. During the data transmission,
high quality of service is to be achieved, which leads to the
significant research supporting real-time transmission. The
hybrid network, which combines the Mobile Adhoc Network
(MANETs) and a wireless infrastructure network has been
proven to be a better alternative for next generation wireless
networks. The hybrid networks suffer from transmission delay
and QoS requirements, Resource-Reservation based QOS
routing is adopted for MANETs. In hybrid network, a QoS -
Oriented Distributed Routing Protocol is used to improve the
QoS support capability of hybrid networks. The main two
advantages of hybrid networks are fewer transmission hops and
any cast transmission, by using these advantages, the QOD
transforms the Packet Routing problem to a Resource
Scheduling problem. In this paper a Packet Scheduling
algorithm consisting of Earliest Deadline First algorithm and
Least Slack Time algorithm is proposed for scheduling the
various multimedia applications (Video Conference, Online
Video, Images, etc.). This Scheduling algorithm is used to reduce
the transmission delay and to achieve better QoS requirements.
This is implemented using NS2 simulator.
Keywords: QoS requirements, QOD algorithm, Hybrid
Networks, MANETs,
I. Introduction
The gorgeous development in the field of wireless
networks has triggered numerous wireless applications, used
in many areas such as education, medical, emergency
services and entertainment. The usage of Wi-Fi enabled
devices and handhold devices are increasing exponentially.
For instance, peoples stuck with wireless internet is tripled all
over the world in past 2 years, whereas the smartphone users
in U.S are 144.5 million in 2013, 163.9 million in 2014,
182.6 million in 2015 and will reach around 220 million in
2018 respectively. In this modern world, peoples prefer to
use wireless mobile devices much more than the wired ones.
Users prefer to watch online tv, online videos and online
conferences “In motion”. The recent video streaming
applications such as Netflix, Air Video, Sling Box and Plex
have gained the attention at present. These applications use
infrastructure wireless networks to directly connect the
mobile users for video watching or interaction in real-time.
The ampla use of wireless devices and the rising demand for
multimedia streaming applications in mobile are leading to a
sure shot near future where wireless multimedia services are
widely explicabatur. The emergence and the vorgesetellt
future of multimedia applications and realtime have lead to
the need of high QoS support in mobile networking
environments and wireless. The QoS-support reduces the
end-to-end transmission delay and improves the throughput
to guarantee the nathlos communication between wireless
networks and mobile devices.
On the other side, hybrid wireless networks are proven
to be a better network structure for the next generation
wireless networks (such as A Unified Cell and Adhoc
Network Architecture, Unicast based Multicast
communication in warmhole-routed direct networks,
PARcelS) and help to overcome the precise nend-to-end QoS
requirement of different applications. Hybrid network is a
combination of infrastructure networks and MANETs, hence
hybrid networks are specifically used to enhance the
scalability of MANETs, whereas the MANETs are self-
organizing networks.
In order to improve the QoS-support capability of
hybrid networks and in order to pull down the end-to-end
transmission delay, a scheduling algorithm is proposed here,
which is a combination of Earliest Deadline First algorithm
and least slack time first algorithm.
1. Earliest Deadline Algorithm: The packet with the lowest
deadline value is given the highest priority and it is
forwarded first.
2. Least Slack Time algorithm: The slack time of the packet
during transmission are calculated and the packet with the
least slack time is given first priority and forwarded first
II. Literature Review
The need for a scheduling algorithm arises from the
requirement for most modern systems to perform
multitasking or execute more than one process at a time.
Scheduling is done among large number of queues to provide
certain order in which the packet will be dispatched through
the output queue. The basic need for scheduling is to give
importance to all the types of packets (real-time, non-real-
time, etc) and dispatch them effectively with less amount of
delay and delay variations.
Some of the effective scheduling disciplines in
networking environment are discussed below [1], [2], [3]:
Round-robin is one of the simplest scheduling
algorithms for processes in an operating system, which

27
assigns time slices to each process in equal portions and in
order, handling all processes without. The implementation of
Round Robin technique is easy and all the packets are equally
serviced. It serves as the base for all scheduling algorithms.
Deficit Round Robin algorithm which works well
for Long-term load balancing and Back logged traffic.
Packets from different flows are arranged in separate queues
and assign a quantum size to each flow. If the packet length
is greater than the quantumsize then it value will be stored in
deficit counter and in the next round the amount of packet
can be sent is the value of the sum of the credit in the
previous round and the quantum. In this algorithm, a queue is
deactivated once the last packet is sent and its DC value is
reset to zero. The DRR scheduler services packets in a queue
until its DC is insufficient for forwarding the HoL packet.
Drawback of this algorithm performs very poor for short
term load balancing and non-back logged scenario and also
minimizes the delay of a packet from the queue in service at
the expense of increasing the delay of other queues.
The Quantum Based Earliest Deadline First
Scheduling (QEDF) which works well for Long-term load
balancing and Back logged traffic. It is a non-pre-emptive.
This packet scheduler supports multiservice including time
critical, jitter sensitive, rate based and best effort services. It
uses quantum to control fairness and rate differentiation as
well as to provide protection between flows. It classifies the
flows and services them in a priority order. This algorithm
computes an expected quantum for each queue, thereby can
provide delay guarantees to queues with low delay
requirements in each service round. Expected quantumis still
proportional to bandwidth requirements and therefore
throughput fairness is maintained. The advantage of this
algorithm provides flow protection, minimizes delay, jitter
bound and complexity is O (log N), it provides delay
differentiation that is comparable to EDF scheduler and
almost all the packets are delivered within their deadline as
compared to WDRR. Drawbacks of this algorithm is that
unfortunately, the residual quanta are always carried over to
the next round regardless of the fact that the queues already
run empty, which will degrade the short-termload-balancing.
Efficient Deadline Based QoS Algorithm for High
Performance Networks. This paper describes the Earliest
Deadline First Family of Algorithms to obtain most of the
benefits of the QoS approach. To transmit an MPEG video
sequence, an average bandwidth assignation is not enough
because, during peak-rate periods, it could introduce
intolerable delays. The use of maximum bandwidth to
generate deadlines causes two problems.
(1) If the frame to be transmitted is short, it may end up
unnecessary burst of packets (2) The latency of each frame
will vary a lot since it depend on the size of frames.
III. Proposed System
The hybrid networks suffer from transmission delay
and QoS requirements. In general, QoS-Oriented Distributed
Routing protocol has 5 sub-Algorithms in it, namely: (1).QoS
- Guaranteed neighbour Selection Algorithm (2)Distributed
Packet Scheduling Algorithm (3)Mobility based segment
resizing algorithm (4)Soft-deadline based forwarding
scheduling algorithm (5)Data redundancy elimination based
transmission, in order to improve the QoS support capability
of hybrid network. Algorithm
In this paper, only the scheduling part of the packet
is concentrated, hence, QoS-Gauranteed neighbour Selection
Algorithm is considered for reducing the transmission delay
and Distributed Packet Scheduling Algorithm which selects
the intermediate nodes that can guarantee the QoS of the
packet transmission.
The packets of multimedia applications like online
video, online conference are scheduled using Least Slack
time algorithm and Earliest Deadline first Algorithm
respectively. The block diagram of the proposed system is
shown in the Figure 3.1.
The above block diagram shows the mechanism of
proposed system. The “n” number of packets (for example
say n = 8) are generated using a packet generator. These
packets generated by the packet generator reaches an
intermediate node ni, in which the packets are classified
according to the purpose of multimedia application it has been
generated. After the classification, the packets are placed
over its respective queues as shown in the Figure 3.1. Now
this classified packets are scheduled using a Scheduler. The
packets are scheduled using Least Slack Time Algorithm(for
Online Video) and Earliest Deadline First Algorithm (for
Online Conference) and any Round Robin Scheme.
IV. Proposed System Algorithm
This section describes the algorithms such as Earliest
Deadline First Algorithm and Least Slack Time Algorithm
used in this paper.
4.1. Earliest Deadline First Algorithm
Short delay is the major QoS requirement for traffic
transmission, it is achieved by scheduling the data packets
using Earliest Deadline First algorithm (EDF). It is a
algorithm for data traffic scheduling in intermediate nodes.
Figure 3.1. Block diagram of proposed system
It assigns highest priority to the node with closest deadline
and forwards the packets with highest priority first.
The deadline of the packet is given by,

28
DP = Ai + ( - ) (4.1)
The QoS-gauranted neighbourselection algorithm selects the
qualified neighbour satisfying the QoS requirements. QoS of
the packets going through node, ni satisfied by,
+ + …………… + ≤ 1 (4.2)
Where, Lp (i) is the length of the packet, Ri is the Resource
reservation and ta(i) is the Packet arrival interval time.
In job scheduling model the task consists of number of jobs.
+ + ……… + ≤ 1 (4.3)
Where, tg (j) is the job arrival interval time period, tcp (j) is the
job computing time, “1” refers to CPU Utility when CPU is id
busy all the time.
For a a communication network Tcp is regarded as and
Tg is regarded as Ta and finally the CPU utility is regarded as
Space Utility. The resource reservation is given by,
Ri = 1 – Uc (i) . Cj (4.4)
Therefore by applying the above changes to the equation
(4.3),
Us(i) = + +…………+ 1 (4.5)
Us(i) = + +……………….....+ R (4.6)
Therefore from the equation (4.6) the scheduling feasibility is
affected by, LP – Length of the packet,
Number of packet streams from „m‟ neighbors and the
Resource Reservation „Ri‟.
4.2. Least Slack Time First Algorithm
The Least slack time algorithm is used here to
schedule the packets according the slack time of the each
packet. The slack time of each packet is calculated using,
– – (4.7)
Where DP is the Deadline of the packet, t is the Current time
and C‟ is the Remaining transmission time respectively. For
example if the Packet‟s remaining forwarding time is 7s and
the time interval from the current time to the Deadline of the
packet is 18s, the Slack time is 18s – 7s = 11s.
The Least slack time algorithm assign the packet
with the lowest slack time as highest priority and forwards
first.
VI. Implementation
It is implemented using Network Simulator 2. The
Deadline of the each packet,the distance and finally the type of
packet (i.e. the type ofapplication)are specified in the header of
the packet. The packet interval for the node ni can be
calculated by the source using:
For instance, if the Resource Reservation Ri of the
intermediate node ni, is 100 kb/s, then the threshold of the
workload will be the 80 percent of the overall space utility,
which is 80kb/s. Node ni schedules the packet traffic from
four different source nodes n1, n2, n3, n4, n5 periodically.
2,11,21,31 kb/s are the size of the packet size of the traffic
from n1, n2, n3, n4, n5 with arrival interval 0.2, 0.6, 1 and
1.5s respectively. Then + + + + = 70 kb/s. A
request from the node, n5 is sent to the intermediate node ni,
and if ni checks for the available workload and replies as
10kb/s. if the n5 accepts the reply and send 30kb/s traffic to
the intermediate node ni, now the node ni will reject request
and inform n5 to reduce the traffic to 10kb/s. now the resource
reservation of the intermediate node drops to 90kb/s because
of the interference, ni „s overall space utility is reduced to
68kb/s. then ni informs node n4 under scheduling that the
largest = 21kb/s to change its traffic to 19kb/s.
V. Software and Its Description
This section will discuss about the software used in
this project. NS2 which is known as Network Simulator 2 is
used in this paper. NS2 is widely used tool to simulate the
behavior of wired and wireless networks. The core of NS2 is
also written in C++, but C++ simulation objects are linked to
shadow objects in OTcl and variables can be linked between
both language fields. Simulation scripts are written in Otcl
(Object Tool Command Language) language which is an
extension of Tcl language. It consists of over 300,000 lines of
source code and there is probably a comparable amount of
contributed code. It is two language architecture in which
C++ is used as a compiler (internal mechanism) and tcl is
used as an interpreter (user interface). NS2 is also applied in
traffic generator and simulated applications.
VII. Conclusion
In this paper, the multimedia applications such as
online conference and online video is considered and its
scheduling is carried out in such a way that reduces its end-to-
end transmission delay and improve the QoS requirements by
using Earliest Deadline First algorithm for Online conference,
which is deadline based algorithm. The EDF gives highest
priority to the packet with least deadline value and forwards
first, this makes the packet of online conference to experience
lesser delay and the jitter problem is eliminated. On the other
hand the soft-deadline applications like online video uses
Least Slack Time First algorithm, which assigns highest
priority to the packet with least slack time, thereby reducing
the delay in transmission

29
References
[1] Avudaiammal, R. and Seethalakshmi, P.,“Design and
Implementation of a Hybrid Packet Scheduling Algorithm
on NetworkProcessorBased Router for Enhancing QoS of
Multimedia Applications”, European Journal of Scientific
Research,Vol.72, No.2, pp. 245-262,2012.
[2] Avudaiammal, R. and Seethalakshmi, P., “Packet
Scheduling Disciplines- A Survey”, International Journal
of Engineering Research and Industrial Applications,
Vol. 3, No. 1, pp. 375-399, 2010.
[3] Avudaiammal. R and Seethalakshmi, P., “Bandwidth-
Adaptive Scheduling for QoS Enhancement of Real-Time
Multimedia Applications in Network Processor based
Router”, Journal of Computer Science, Vol. 5, No. 12, pp.
1068-1074,2009.
[4] A QoS-Oriented Distributed Routing protocol for Hybrid
Wireless Networks, Ze Li, Student Member, IEEE, and
Haiying Shen, Member, IEEE. IEEE TRANSACTIONS
ON MOBILE COMPUTING, VOL. 13, NO. 3, MARCH
2014.
[5] “A Majority of U.S. Mobile Users Are Now Smartphone
Users,”
[6] http://adage.com/article/digital/a-majority-u-s-
mobile-userssmartphone-users/241717, 2013.
www.datastax.com/wp-content/uploads/2011/09/CS-
Netflix.pdf
[7] www.onairvideo.com/pdf-downloads.html
[8] forums.macrumors.com › iPhone, iPod and iPad › iPad
[9] X. Du, “QoS Routing Based on Multi-Class Nodes for
Mobile Ad Hoc Networks,” Ad Hoc Networks, vol. 2, pp.
241-254, 2004.
[10] S. Jiang, Y. Liu, Y. Jiang, and Q. Yin, “Provisioning of
Adaptability to Variable Topologies for Routing
Schemes in MANETs,” IEEE J. Selected Areas in
Comm., vol. 22, no. 7, pp. 1347-1356, Sept. 2004.
[11] M. Conti, E. Gregori, and G. Maselli, “Reliable and
Efficient Forwarding in Ad Hoc Networks,” Ad Hoc
Networks, vol. 4, pp. 398-415, 2006.
[12] G. Chakrabarti and S. Kulkarni, “Load Balancing and
Resource Reservation in Mobile Ad Hoc Networks,” Ad
Hoc Networks, vol. 4, pp. 186-203, 2006.
[13] A. Argyriou and V. Madisetti, “Using a New Protocol to
Enhance Path Reliability and Realize Load Balancing in
Mobile Ad Hoc Networks,” Ad Hoc Networks, vol. 4,
pp. 60-74, 2006.
[14] C. Shen and S. Rajagopalan, “Protocol-Independent
Multicast Packet Delivery Improvement Service for
Mobile Ad Hoc Networks,” Ad Hoc Networks, vol. 5,
pp. 210-227, 2007.
[15] C.E. Perkins, E.M. Royer, and S.R. Das, Quality of
Service in Ad Hoc On-Demand Distance Vector Routing,
IETF Internet draft, 2001.
[16] Z. Shen and J.P. Thomas, “Security and QoS Self-
Optimization in Mobile Ad Hoc Networks,” IEEE Trans.
Mobile Computing, vol. 7, pp. 1138-1151, Sept. 2008.
[17] Y. Li and A. Ephremides, “A Joint Scheduling Power
Control and Routing Algorithm for Ad Hoc Networks,”
Ad Hoc Networks, 2008.
[18] ec1.imagesamazon.com/media/i3d/01/A/.../MANUAL00
0075454
[19] H. Luo, R. Ramjeey, P. Sinhaz, L. Liy, and S. Lu,
“UCAN: A Unified Cell and Ad-Hoc Network
Architecture,” Proc. ACM MobiCom, 2003.
[20] P.K. Mckinley, H. Xu, A. Esfahanian, and L.M. Ni,
“Unicast-Based Multicast Communication in Wormhole-
Routed Direct Networks,” IEEE Trans. Parallel Data and
Distributed Systems, vol. 5, no. 12, pp. 1252-1265, Dec.
1992.
[21] H. Wu, C. Qiao, S. De, and O. Tonguz, “Integrated Cell
and Ad Hoc Relaying Systems: iCAR,” IEEE J. Selected
Areas in Comm.vol. 19, no. 10, pp. 2105-2115, Oct. 2001.
[22] J. Zhou and Y.R. Yang, “PAR CelS: Pervasive Ad-Hoc
Relaying for Cell Systems,” Proc. IFIP Mediterranean Ad
Hoc Networking Workshop (Med-Hoc-Net), 2002
[23] C. Liu and J. Layland, “Scheduling Algorithms for
Multiprogramming in a Hard Real-Time Environment,”
J. ACM, vol. 20, pp. 46-61, 1973.
[24] Cheng, Real-Time Systems: Scheduling Analysis, and
verification, first edition, Wiley-Interscience 2002

Iisrt arunkumar b (networks)

More Related Content

What's hot (19)

Viewers also liked (20)

Similar to Iisrt arunkumar b (networks) (20)

More from IISRT (20)

Recently uploaded (20)

Iisrt arunkumar b (networks)