DESIGN AND IMPLEMENTATION OF ADVANCED MULTILEVEL PRIORITY PACKET SCHEDULING SCHEME

International Journal of Technical Research and Applications e-ISSN: 2320-8163,
www.ijtra.com Volume 3, Issue 1 (Jan-Feb 2015), PP. 18-22
18 | P a g e
DESIGN AND IMPLEMENTATION OF
ADVANCED MULTILEVEL PRIORITY PACKET
SCHEDULING SCHEME
A.VANI, D. Madhu Babu
Dept of ECE
CBIT
Hyderabad, India
vanialamur@yahoo.in, madhubabud4u@gmail.com
Abstract— Scheduling different types of packets, such as
real-time and non-real-time data packets, at sensor nodes with
resource constraints in Wireless Sensor Networks (WSN) is of
vital importance to reduce sensors’ energy consumptions and end-
to-end data transmission delays. Most of the existing packet-
scheduling mechanisms of WSN use First Come First Served
(FCFS), non pre-emptive priority and pre-emptive priority
scheduling algorithms. These algorithms incur a high processing
overhead and long end-to-end data transmission delay due to the
FCFS concept, starvation of high priority real-time data packets
due to the transmission of a large data packet in non pre-emptive
priority scheduling, starvation of non-real-time data packets due
to the probable continuous arrival of real-time data in pre-
emptive priority scheduling, and improper allocation of data
packets to queues in multilevel queue scheduling algorithms.
Moreover, these algorithms are not dynamic to the changing
requirements of WSN applications since their scheduling policies
are predetermined.
In the Advanced Multilevel Priority packet scheduling
scheme, each node except those at the last level has three levels of
priority queues. According to the priority of the packet and
availability of the queue, node will schedule the packet for
transmission. Due to separated queue availability, packet
transmission delay is reduced. Due to reduction in packet
transmission delay, node can goes into sleep mode as soon as
possible. And Expired packets are deleted at the particular node
at itself before reaching the base station, so that processing
burden on the node is reduced. Thus, energy of the node is saved.
Index Terms—Wireless sensor network, FCFS, CSMA, Q-
MAC
I. INTRODUCTION
The communications in the WSN has the many-to-one
property in that data from a large number of sensor nodes tend
to be concentrated into a few sinks. Since multi-hop routing is
generally needed for distant sensor nodes from the sinks to
save energy, the nodes near a sink can be burdened with
relaying a large amount of traffic from other nodes Sensor
nodes are resource constrained in term of energy, processor and
memory and low range communication and bandwidth.
Limited battery power is used to operate the sensor nodes and
is very difficult to replace or recharge it, when the nodes die.
This will affect the network performance.Energy consumption
and harvesting increase the lifetime of the network. Optimize
the communication range and minimize the energy usage.
Sensor nodes are deployed to gather information and desired
that all the nodes works continuously and transmit information
as long as possible. This addresses the lifetime problem in
wireless sensor networks. Sensor nodes spend their energy
during transmitting the data, receiving and relaying packets.
Hence, designing routing algorithms that maximize the life
time until the first battery expires is an important consideration
designing energy aware algorithms increase the lifetime of
sensor nodes. In some applications the network size is larger
required scalable architectures. Energy conservation in wireless
sensor networks has been the primary objective [2], but
however, this constrain is not the only consideration for
efficient working of wireless sensor networks. There are other
objectives like scalable architecture, routing and latency. In
most of the applications of wireless sensor networks are
envisioned to handled critical scenarios where data retrieval
time is critical, i.e., delivering information of each individual
node as fast as possible to the base station becomes an
important issue.
II. RELATED WORK
To fulfill the objectives of the thesis, understanding the
basics of Wireless Sensor Networks and its characteristics is
essential. Wireless Sensor Networks Technology, Protocols
and Applications [2] (a john Wiley & sons Inc.,
Publication2007) was referred to understand the basics of
wireless sensor networks and its characteristics. This book
refers about the concept of wireless sensor networks,
characteristics, and applications of the wireless sensor
networks. And also refers about the routing protocols for
wireless sensor networks. And also described about the
different MAC protocols useful in wireless sensor network
technology.
In this paper “RAP, new real-time communication
architecture for large-scale sensor networks” [3], Authors
(Chenyang Lu et al, Brian et al) proposed Velocity Monotonic
Scheduling (VMS). VMS assigns the priority of a packet based
on its requested velocity. A packet with a higher requested
velocity is assigned a higher priority. VMS improves the
number of packets that meet their deadlines because it assigns
the “right” priorities to packets based on their urgencies on the
current hop. But there is no detail for packet arrival distance.
And in this he paper [4], “Extending the Lifetime of
Wireless Sensor Networks through Adaptive Sleep”, when the
queue is full, higher priority incoming packets overwrite lower
priority ones. It may problem to be the low priority
information. In the paper [4], author Giuseppe et al and Hong
et al proposed an Adaptive Staggered SLEEP Protocol
(ASLEEP) for efficient power management in wireless sensor
networks targeted to periodic data acquisition. This protocol
dynamically adjusts the sleep schedules of nodes to match the
network demands, even in time-varying operating conditions. It
uses the CSMA scheme for process the data, but it may be not
efficient in fixed WSN network and there is no detail to data
management.
In this paper [5], author presents how to place sensors by
use of a minimal number to maximize the coverage area when
the communication radius of the SN is not less than the sensing
radius, which results in the application of regular topology to

19 | P a g e
WSNS deployment. In this paper [5] author discussed the
details of sensor deployment. Due to optimal coverage sensor
deployment, it reduces the no of sensors usage and also
increases the lifetime of sensors. But till lifetime of sensor need
to increase.
In this paper [6], author proposed a clustering method with
coverage and energy aware TDMA scheduling scheme. And
the cluster formation is done by the base station according to
the current residual energy, and the coverage area of cluster
Member is reduced to avoid the congestion and energy
management. In this paper there is no discussion on the real
time and non-real time packet scheduling.
In this the paper [7] author present a cluster based routing
algorithm. One of author’s main goals is to design the energy
efficient routing protocol. This algorithm makes the best use of
node with low number of cluster head know as super node.
Here author divided the full region in equal zones and the
center area of the region is used to select for super node. Each
zone is considered separately and the zone may be or not
divided further that’s depending upon the density of nodes in
that zone and capability of the super node. In the paper author
considered, cluster head changes when the cluster head is
failed. It may be the problem to sensing in that area.
Author sinem et al [8] proposed the two centralized
heuristic algorithms: one based on direct node-based
scheduling, which is adapted from classical multi-hop
scheduling algorithms for general ad hoc networks, and the
other based on scheduling the levels in the routing tree before
scheduling the nodes, which is a novel scheduling algorithm
for many-to-one communication in sensor networks.
In this paper [9], authors propose Q-MAC scheme that
provides quality of service by differentiating network services
based on priority levels. The priority levels reflect the
criticality of data packets originating from different sensor
nodes. The Q-MAC accomplishes its task through two steps;
intra-node and inter-node scheduling. This paper effectively
handled the priority scheduling, but this paper only considered
the priority packets and it schedules the priority packet as the
first non-priority as second. And in the case of node gives the
first priority to own priority data compare than other nodes
packet.
Author [10] developed scheme by designing the network
with multiple-sized fixed grids while taking into account the
arbitrary-shaped area sensed by the sensor nodes. In this paper
[10], author Tarandeep et al considers the different initial
energy level of sensors, and placed that sensor according to
that energy level. So energy loss was avoided. But calculating
different initial energy level and placing the node according to
that energy level is difficult in real time.
III. Wireless Node Architecture and WSN
A wireless sensor node is capable of gathering information
from surroundings, processing and transmitting required data to
other nodes in network. The sensed signal from the
environment is analog which is then digitized by analog-to-
digital converter which is then sent to microcontroller for
further processing.[5] The block diagram of a sensing node is
shown in figure. While designing the hardware of any sensor
node the main feature in consideration is the reduction of
power consumption by the node. Most of the power
consumption is by the radio subsystem of the sensing node. So
the sending of required data over radio network is
advantageous. An algorithm is required to program a sensing
node so that it knows when to send data after event sensing in
event driven based sensor model. Another important factor is
the reduction of power consumption by the sensor which
should be in consideration as well. During the designing of
hardware of sensing node microprocessor should be allowed to
control the power to different parts such as sensor, sensor
signal conditioner and radio.
Depending on the needs of the applications and on sensors
to be deployed, the block of signal conditioning can be
replaced or re-programmed. Due to this fact a variety of
different sensors with wireless sensing node are allowed for
use. To acquire data from base station remote nodes uses flash
memory.
Fig 1. Sensor Node Architecture
Characteristics of WSN’s:
WSN have the following distinctive characteristics:
 They can be deployed on large scale.
 These networks are scalable; the only limitation is the
bandwidth of gateway node.
 Wireless sensor networks have the ability to deal with
node failures.
 Another unique feature is the mobility of nodes.
 They have the ability to survive in different
environmental surroundings.
1. Advanced Multi level Priority scheduling scheme
In non pre-emptive packet scheduling schemes, real-time
data packets have to wait for completing the transmissions of
other non-real-time data packets. On the other hand, in pre-
emptive priority scheduling, non real time data will be in the
waiting state because of continuous arrival of real time data.
This Advanced Multilevel Priority packet scheduling scheme
ensures a trade off between priority and fairness. And in
existing scheduling schemes, expired packets (Dead Packets)
are removed after reaching the base station. But in this method,
node will check for the expired packets while processing the
data packets and it will delete those packets at the node itself.
Scheduling data packets among several queues of a sensor
node is presented in Figure 2. In the Advanced Multilevel
priority packet scheduling scheme, Nodes are virtually
organized following a hierarchical structure. Nodes that are
same hop distance from the bases station (BS) are considered
to be the same level. Data packets of different levels at nodes
are processed using the TDMA Scheme. Nodes that are at
lowest level and second lowest levels can be allotted time slots
1 & 2 respectively. In advanced multilevel priority packet
scheduling scheme, node has three level of priority queue.
The three priority levels are
Prirority1: Real time packets have high priority. So these
packets will be placed in priority queue level1 (PR1).

20 | P a g e
Prirority2: Non Real Time packets which are generated at
lower levels have second highest priority to reduce the average
waiting time. So these packets are placed in second priority
queue level (Pr2).
Priority3: Non Real Time packets which are generated at
the local nodes have third highest priority. So these are placed
in third highest priority queue level (PR3).
Assigning the priority for the packets depending on the type of
the data is shown in the above figure 2.Real packets i.e.
emergency data packets has to reach the base station (BS) as
soon as possible so that these packets are given high priority
compared to the other data packets. Thus, end to end delay for
the real time packets is reduced. In the ready queue, Real time
data packets are processed using FCFS (First come First serve)
method.
Fig 2. Task Scheduling
Real time data packets are placed in pre-emptive priority
queue. Thus these packets on arrival preempt the processing of
non real time data packets. And priority2 packets are placed in
preempt-able queue. The processing of priority1 reduces the
end to end delay for the real time packets. If the low priority
data do not get processed for a long time then it can preempt
the high priority tasks and then starts processing of high
priority tasks. [7] And node also monitors the lifetime of the
non real time packets, if the lifetime of the packet is less than
half of its actual lifetime then node starts processing of those
packets to improve the packet delivery probability. The amount
of real time tasks is very low when compared to other tasks
because real time or emergency data will occur rarely. The
operation at the node is shown in the figure3.
Fig 3. Scheduling Scheme Operation
Each packet has an ID which consists of Level ID and
Node ID, when two equal priority packets arrive the node at
same time, node process the packet which is coming from the
lowest level. Level will be identified by the Level ID. If the
two packets having same priority reach the node at the same
time, then node starts processing of the packet which is having
less size. Node has to process and forward most of the data
sensed at the node itself and also the data which is coming
from below levels in its allocated time slot to reduce the
probability that the ready queue at the node becomes full and
also probability of packet drops. And remaining data will be
sent in further timeslots. Time slot comprised of data sensing
time, data transmission time and CPU Speed. The remaining
time of a timeslot at a particular node will be assigned to other
queues. So the presence of data packets at queue is low. The
quality of service is improved. If processing of data is
completed before expiration of its allocated timeslot, then node
goes into sleep mode. So that energy efficiency is improved.
Dead Packet Removal
Node compares the life time of the packet with the time
packet needs to reach the base station, if it is less than then
those packets are considered as dead packets and automatically
node deletes those packets at node itself so that energy
efficiency is improved. End to End delay for the remaining
packets will be improved by deleting the dead packets.
IV. ALGORITHM (for packet scheduling)
The main aim of this algorithm is how send task (packet)
according to their priority in three types of queue. Considering
level also, if data comes from lowest level then giving more
priority than local data (which is sensed at intermediate level)
then finally consider time slot at particular level.
While (task (k,i))
//task means any packet receiving that node i and K means
level
{
If (that is real time data)
{
Put that is in PR1 queue
}
Else if (node at which data sensed that’s not at lowest
level) //data also non real data
{
If (data is non local) //check whether it came from lowest level
node
{
Put in PR2 Queue
} else {
Put in PR3 Queue // local data sensed at that node itself
}
} else {//data sensed at lowest level
Put in PR2 Queue
}
//Some time slot given to level consider total time slot is Tk at
k level
Consider data sensing time Ts
Remaining time T1(K) = Tk –Ts
Let total real time task at (NODE)i at level at Level k -
nk(pr1)
Proctime(Pr1)k is total time for pr1
If (Proctime (Pr1)k <T1(k))
{
All pr1 tasks of node (i) at l (k) are processed as FCFS
Remaining Time T2 (k)= T1(k)-ProcTime(Pr1)k
Proctime (Pr2)k is total time for pr2
If (Proctime(Pr2) k < T2(K)) {
All pr2 tasks are processed as FCFS pr3 tasks are processed as
FCFS for the remaining time,
T3 (k) = T2 (k) − procTimepr2 (k)
}
else {//this loop for if pr1 task time greater than Remaining
time T2 (k)
pr2 tasks are processed for t2(k) time
no pr3 tasks are processed }//end of inner if loop
} else { //this loop for if pr1 task time greater time than total
time slot of that level
only pr1 tasks are processed for T1(k) time

21 | P a g e
no pr2 and pr3 tasks are processed
}
If (pr1 queue is empty and pr2 task are processing some time
Y since T(K) < Proctimepr2(K))
{
At that time if pr3 task are coming then Pr2 task are pre-
empted
If (any pr1 task coming in this )
{
Pr3 task pre-empted and giving priority to Pr1....
Context are transferred again for processing pr3 tasks
}
If( reach_time > Reamining_time)
{ Drop that packet at node itself
}
} //end of if
}//end of while
V. RESULTS
The below parameters are assumed while simulation:
No. of mobile nodes = 15
Simulation end = 5sec
Packet size = 1000 bytes
Initial energy of the node = 1000 joules
Queue length = 50 packets
Channel Access scheme = MAC-TDMA
AODV protocol is used for creating the nodes. And minimum
distance between the nodes will be calculated to form the route
between the sources to destination via nodes.
Results of the Advanced Multilevel Priority Packet
scheduling scheme are observed by comparing with the FIFO,
multi priority, and hop based multi priority scheduling
schemes.
A. Dead Packet Removal
In this scheduling scheme, node monitors the reach time
and life time of the packets. And if the reach time is less than
the remaining lifetime then node treat them as expired packets
and immediately node deletes those packets because there is no
use of those packets, their lifetime will be completed before
reaching the base station. So that this scheduling scheme
provides good performance in terms of end to end delay and
waiting time of the packets and energy saving when compared
to other schemes. Deletion of expired packets is shown in the
figure 5.5.Reach time is calculated by using the below
equation,
Reach time = remaining hop count *10ms
Fig 4 Dead Packet Removal
B. Non Real Time Packet delay
Scheme Non Real Time Packet
delay
FIFO scheme 300 ms
Priority Based scheme 350 ms
Hop Based Multi Priority scheme 360 ms
Advanced Multilevel Priority
Scheme
260 ms
Table 1. Non Real Time Packet Delay
Fig 5.Non Relay Time Packet Delay Graph
C. Real Time Packet Delay
Scheme Real Time Packet delay
FIFO scheme 300 ms
Priority Based scheme 25 ms
Hop Based Multi Priority
scheme
28 ms
Advanced Multilevel Priority
Scheme
26 ms
Table2. Real time packet delay
Fig 6. Real Time packet delay graph
D. Waiting Time for packets at Node
The comparison of waiting time of the packets at the node
for various scheduling schemes is shown in the figure 7. For
FIFO scheme, waiting time for real time and non real time
packets is equal. In priority scheduling scheme waiting time for
non real time packets is much more when compared with the
real time packet waiting time. In hop based priority scheduling
scheme waiting time for non real time packets is less when
compared to the priority based scheduling scheme. In life based
priority packet scheduling scheme, waiting time for non real
time is reduced when compared to above three methods. And
advanced multilevel priority packet scheduling scheme,
waiting time for non real time and real time packets is reduce
when compared to all the other scheduling schemes

22 | P a g e
Fig 7. Waiting timing of packets
E. Energy Consumption
Scheme Energy Saving (%)
FIFO scheme 88
Priority Based scheme 88
Hop Based Multi Priority
scheme
88
Life Time Priority scheme 88
Advanced Multilevel
Priority Scheme
90
Table 3. Energy Saving in %
Fig 8. Energy Saving Graph
VI. CONCLUSION AND FUTURE SCOPE
Advanced Multilevel Priority (DMP) packet scheduling
scheme for Wireless Sensor Networks (WSN’s). The scheme
uses three-level of priority Queues to schedule data packets
based on their types and priorities. It ensures minimum end-to-
end data transmission End-to-End Delay of all Tasks
(microSec) for the highest priority data while exhibiting
acceptable fairness towards lowest-priority data. Experimental
results show that the proposed DMP packet scheduling scheme
has better performance than the existing FCFS and Multilevel
Queue Scheduler in terms of the average task waiting time and
end to-end. It assigns the priority based on task deadline
instead of the shortest task processing time.
As a Future enhancement, by forming the nodes into zones
(consisting node levels) based topology, end to end delay
further can be reduced and also by using other channel access
techniques, deadlock situation can be avoided further.
REFERENCES
[1] Nidal Nasser, Lutful Karim, and Tarik Taleb, “Dynamic
Multilevel Priority Packet Scheduling Scheme for Wireless
Sensor Network,” IEEE Transaction wireless communications,
vol. 12, no. 4, April 2013
[2] Wireless Sensor Networks Technology, Protocols and
Applications (a john Wiley & sons Inc., Publication2007)
[3] Chenyang Lu, Brian M. Blum, Tarek F. Abdelzaher, John A.
Stankovic, Tian He, “RAP: A Real-Time Communication
Architecture for Large-Scale Wireless Sensor Networks”, 2002.
[4] Giuseppe Anastasi, Marco Conti, Mario Di Francesco,
“Extending the Lifetime of Wireless Sensor Networks through
Adaptive Sleep”, 2007
[5] “Maximizing Networking Lifetime in Wireless Sensor Networks
with Regular Topologies”, Hui Tian, Hong Shen and Matthew
Roughan.
[6] Ms.Namrata S. Baghele, “An Adaptive Energy Aware Node
Scheduling Scheme In Clustering for Wireless Sensor
Networks”, 2012
[7] Ashim Kumar Ghosh1, Anupam Kumar Bairagi2, Dr. M. Abul
Kashem3, Md. Rezwan-ul-Islam1, A J M Asraf Uddin1 “Energy
Efficient Zone Division Multihop Hierarchical Clustering
Algorithm for Load Balancing in Wireless Sensor Network. ”,
2011.
[8] Sinem Coleri Ergen and Pravin Varaiya, “TDMA Scheduling
Algorithms for Wireless Sensor Networks”, 2009
[9] Y. Liu, I. Elhanany, and H. Qi, “An energy-efficient QoS-aware
media access control protocol for wireless sensor networks,”
Nov. 2005.
[10] “A Strategic Deployment and Cluster-Header Selection for
Wireless Sensor Networks”, Tarandeep Kaur, and Jinsuk Baek,
Member, IEEE.
[11] A book on “Introduction to Network Simulator NS2” by
Teerawat Issariyakul, Ekram Hossain; Springer Publications.

DESIGN AND IMPLEMENTATION OF ADVANCED MULTILEVEL PRIORITY PACKET SCHEDULING SCHEME

Recommended

More Related Content

What's hot (16)

Viewers also liked (15)

Similar to DESIGN AND IMPLEMENTATION OF ADVANCED MULTILEVEL PRIORITY PACKET SCHEDULING SCHEME (20)

More from International Journal of Technical Research & Application (20)

Recently uploaded (20)

DESIGN AND IMPLEMENTATION OF ADVANCED MULTILEVEL PRIORITY PACKET SCHEDULING SCHEME