Content uploaded by Khaled Elleithy
Author content
All content in this area was uploaded by Khaled Elleithy
Content may be subject to copyright.
Least Distance Smart neighboring Search (LDSNS)
over Wireless Sensor Networks
Abdul Razaque, Member IEEE,Khaled .M. Elleithy, Senior member IEEE
arazaque@bridgeport.eduelleithy@bridgeport.edu
Wireless and Mobile Communication (WMC) Laboratory
Computer Science and Engineering Department, University of Bridgeport, Bridgeport, CT 06604
Abstract: - In this paper, we introduce a novel least distance
smart neighboring search (LDSNS) to determine the
mostefficient path at one-hop distance over WSNs. LDSNS
helps to reduce the energy consumption and speeds up
scheduling for delivery of data. It provides cross layering
support and linking MAC layer with network layer to reduce
the amount of control messages. LDSNS is a robust and
efficient approach that isbased on single-hop communication
mechanism. To validate the strength of LDSNS, we
incorporate LDSN in Boarder Node Medium AccessControl
(BN-MAC) protocol [ 15] to determine the list of neighboring
sensor nodes and choosing best 1-hop efficient search to avoid
collision and reducing energy consumption.
Evaluation of LDSNS is conducted using network simulator-2
(ns2).The performance of LDSNS is compared with minimum
energy accumulative routing problem (MEAR) [12],
asynchronous quorum-based wakeup scheduling scheme
(AQWSS) [14] and Minimum Energy Relay Routing (MERR)
[13]. Simulation results show that LDSNS is highly energy
efficient and faster as compared with MEAR, AQWSS and
MERR. It saves 24% to 62% energy resources and
improves12% to 21% search at 1-hop neighboring nodes.
Keywords: LDSNS, Wireless sensor network, Energy
consumption, MAC layer.
1. INTRODUCTION
WSNs is one of the prominent research areas in recent years.
Unlike traditional networks, WSNs are particularlyusedin
physical environments to develop high degree of perceptibility
[6]. It is one of the rapidly growing fields with attractive
features to use in several application areas[2], [3]. WSNs are
considered as low-cost and easy to set up [4]. The advent of
WSN has improved progress in surveillance and monitoring
systems, home automation devices, earthquake and disaster
applications etc. [1] & [5].
WSNsfaceseveral design and performance issues such as
waste of energy in idle listening, overhearing, extra control
messaging, emitting and congestion. In addition, experiencing
several performance impairing factors such as scalability,
mobility, lack of robustness, uniformity etc. The energy
consumption is considered as one of the major apprehensions
[7] that stimulates challenges for industrial and academic
sectors. Therefore, proper energy handling is one of the key
skills to preserve energy [9].
The radio is the main power consuming section of sensor in
WSNs that can be handled by introducing robust MAC
protocols. Thus, an efficient MAC protocol improves WSN
lifetime. In addition, MAC protocol has capability to handle
the issue of sharing the wireless channel and reduces the
collisions in order to improve throughput.
Several MAC protocols have been proposed to reduce energy
consumption and to providefaster delivery of data but
problemis not fully yet resolved. Two types of mechanisms
are used to support scheduling and routing the data: single hop
and multiple-hop.
According to some protocols, a node takes a part in data union
and consumes more energy. Since, the transmission of power
is quadratically proportional tothe distance, multi-hop
approach consumes less energy than one hop communication,
but it creates more overhead on the network and it also
experiences severe problem when routes are broken.
Furthermore, joining of new node and leaving of working
node reduce the throughput and consumes sufficient amount
of energy [10].In this regard, one-hop communication is
efficient and more reliable.
An energy-efficient minimum transmission energy
consumption (MTEC) protocol is proposedto reduce the
energy consumption and increase throughput during data
transmissions. METC has also discussed that MAC protocol
can dynamicallyadjust the size of contention window on basis
of successful ratios of data transmissions using cross-layer
protocol [8]. MEAR developed heuristic approach to
determine an energy efficient wave-path and compared with
traditional shortest path algorithm.The authors urgedthat the
existing shortest path algorithm has shortcoming of optimal
relaying strategy and channel propagation. MEAR also filsto
decide which node should contribute in transmission schedule
and the order of nodes to transmit and their transmission
power [12].
AQWSS [14] introduced a set of asynchronous quorum-based
wakeup scheduling mechanism to provide a better trade-off
between average delay and energy consumption for neighbor
discovery under variable environments.
MERR is introduced for linear sensor topology to consume
less energy based on optimal transmission distance [13]. All of
the discussed techniques tryto reduce the energy consumption
but from other side, they increased the overhead of network.
Keeping these factors in mind, we introduce LDSNS approach
to support BN-MAC protocol to reduce energy consumption
and speed up the delivery of data without putting any extra
overhead of control packets on network. The reminder of
paper is organized as: section 2, explains the Least Distance
Smart Neighboring Search (LDSNS). Section 3, simulation
and analysis of result are discussed and thepaper is concluded
in section 4.
2. LEAST DISTANCE SMART NEIGHBORING SEARCH
(LDSNS)
In LDSNS, any node monitors the channel after every 500 ms.
ifgain of channel is less than set threshold value, it shows that
there is no activity on medium from its neighbor nodes;
resulting that the node decides to sleep again.When a
transmitter wants to communicate, it first sends short preamble
to alertone hop neighboring nodes for sending the data. When
the targeted receiversenses short preamble, it wakes up and
responds with an acknowledgment (ACK) to transmitter. After
the transmitter gets ACK,it starts to send the data packets.
Pictorial illustration of the protocol is given in Figure 1.
SP SPSP RE-
ACK DATA
TRANSMIT
SHORT PREAMBLE WITH
WITHOUT TARGET ADDRESS
TX
(LDSNS)
RX
(LDSNS) RX
S- W
SE-
ACK DATA
RECEIVE ABP
ENERGY AND TIME
SAVE AT TX & RX
ABP
AUTOMATIC
BUFFER PACKET
SE-
ACK
SENDER EARLY
ACKNOWLEDGEMENT
RE-
ACK
SHORT
WAKE UP
RECEIVER EARLY
ACKNOWLEDGEMENT
RX
S-W
RX WAKE
UP TIME
TIME
TIME
Figure 1: Mechanism of LDSNS to communicate with 1-hop neighbor nodes
Let us prove this idea by using Lemmas and definitions. The
least distance smart neighboring search is based on 1-hop
distance and route discovery. The designed WSN consists of
different regions. The node which communicates within
region that maintains local connectivity, whereas node that
communicates out of region and schedules within region is
called boarder node (BN). Let us assume that directed graph D
= (V, A), consisting of the set of sensor nodes V. The set of
edges are called arcs that are A V2. It helps to differentiate
between 1-hop destination and more than 1-hop destination
nodes. The digraph distance between nodes is simply the
number of shortest path between them [12]. We assign a name
to each sensor node in V. A local route discovery method is
based on relay scheme that works as follows.
For any destination node in „V‟specified by name „v‟, the
scheme targets the 1-hop destination nodes „u‟ on basis of
stored information in routing table regarding the shortest path
1-hop destination node. Each 1-hop destination node delivers
the shortest path to its predecessor during exchange of control
message.Finally destination „v‟ is acquired with efficient path.
We apply method [8] for estimation of global technology of
sensors by dividing nodes into routable boundaries and
extracting adjacency associations between these boundaries.
The objective of creating each boundary is to make the
topology simpler, so that the searching process works
efficiently within the boundaries. For a number of sensor
nodes „V‟ and communication digraph „D‟.
We pretend that D is connected, thus we just consider
connected components autonomously. Therefore,u:((u, v) A)
can be denoted for hop count of neighboring search between
„u‟, „v‟ in communication digraph.
Definition 1. Let P(x, y) denote set of paths from „x‟ to‟ y‟
for 1-hop neighbor nodes in direct graph (Dg). Hence, S (x, y)
is the distance (S) between two neighbor nodes x, y in Dg,
which shows shortest path from node „x‟ to „y‟. It can be
computed as:
=,= ,1
Where
Lmin: Minimum length from one node to other node.
p: path value
If Dg (x, y) = then Dg (x, y) = ∞.
Therefore,
Dg(x, Ѐ) between node „x‟ and subset of nodes Ѐ E that is
defined as:
=, É =min,& É (2)
Thus, Ẋ, E, be the distance between two neighbor nodes
that can be computed as:
min , É& É (3)
Thus, we can add random infinitesimal for unique path.
Definition 2. For a digraph D = (V1, V2), be the set of 1-hop
destination nodes for vertex „v‟ that is explained as:
= { . 2}, and beyond of 1-hop
destination nodes are explained as:
+= { . 2},
Where,
λ: Total number of neighboring nodes
D(v): Pair of one hop neighboring nodes
v: Value of link between two neighboring nodes
V1: Vertex of node
V2: Vertex of neighbor node
We describe 1-hop destination nodes of a vertex „V1„as union
with set of 1-hop destination nodesvertex „V2‟. If the distance
exceeds more than 1-hop destination nodes, it can be
expressed as:
= + 4
The range and out of range distance can be found as:
= +5
Above equation (5) shows that node is within range.
= 6
Equation (6) shows that node is out of range.
From equation (5) and (6), we deduce that
We may again exclude subscript if digraph Dg = (V1, V2) is
clear from context. The weighted graphs also get association
of assorted length, cost and strength. We only focus on edge-
weighted graph that is opposite to node-weighted graphs. We
also need to restrict edge weights to 1 that yield an un-
weighted graph.
Consider digraph Dg = (V1, V2) and its subset for boundary of
regions R V1, explain boundary B (v) of a node. Therefore,
v R and whose nearest region is „v‟.
Thus, boundary of all regions can be expressed as follows:
= 1w R, u, v , } (7)
Lemma 1. Let simple path = (,1 , 1,...,1)that
connects tworegion nodes d= a1 and t= with „e‟ edges and
path of length is „p'. The related boundary path „p*‟ has
maximum length in boundary dual graph
such as L(P*) ≤ e
.L(P*).
Proof: The path includes e-1 that is used for more than 1-hop
destination nodes and „e‟ edges that pass through e +1 multi-
hop in the same region. The most of regions e+1 are intrusive
regions, it means that original path does not go directly to
those nodes but shortest path does. The L (P) in original graph
is sum of edge weights that can be defined as:
LP= ds, t= w( , 2
1
=0
+ 1 ) (8)
Equation (8) shows the creation of path from transmitter to
receiver. „e‟ is an edge between two nodes of boundaries on
path P* that is bounded as follows:
= , + 1 ,
+ ,+ 1
+ (+ 1, + 1)] (9)
Nbou : Node in region
d*: Connecting two region nodes
Let us assume that„s‟and„t‟be two nodes of region that could
be source and target nodes, and defined as follows:
( , , ( ,
=( , )
It yields:
,= (, [ (
2
=1
,
+ ( , +1 + + + 1)+1
+ 1,
,+ , (10)
[ (
2
=1
,) + (
+ 1, 1+ 1 )] (
2
=1
,+1)
+ ( ( 1),( 1, )
(,) + (,
2
=1
) + ( , ) (11)
Simplifying the equation (11), we get as:
LP= e. L(P) (12)
From equation (12), we see that Bound is observed to be tight
because constructions exist.
For example, If any choice for m >λ> 0, thus edge graph
weights of graph for two nodes of regions can be described as:
(, (= , w ,+ 1 = (13)
And
,=1, = (14)
Since 2m + (e-2) λ is the length of path and 2m +(e-2) λ 2(e-
1)*(e- λ) is the length of whole region.
Therefore, the worst case for „λ‟ can be written as:
λ → 0, and ratio can be shown as follows:
e 15
If region nodes are available on shortest path, thus maximum
expansion will be shorter than number of edges on shortest
path. We hereby prove that maximum expansion is
proportional to largest gap between region nodes on path.
Lemma 2. For any node u B (v), the shortest path from
node u to v is completely included in B (v).
Proof: If lemma were incorrect, there would exist
()on the shortest path from node ‟u‟ to destination node
„v‟.
Therefore,
,< ,
And such that:
, ,+ ,< ,+ ,
=,16
This statement contradicts with hypothesis, such as ;
thus lemma must be correct. One inference of this lemma is
connection of boundary cells on spanning graph. Region cells
are dirichlet, connecting all points of sensor field. Region has
simple topology in all dimensions that is stronger point of
connectivity. The simpler topology helps to make subsets of
sensor fields, when sensor filed experiences large holes. Thus,
edges u1, u2 є B (v).
Lemma 3: For node in each region of WSN calculates
maximum cost for all one-hop neighboring nodes for selection
of lowest cost path.
Proof: Let „x1‟ be node, which calculates the cost for each 1-
hop neighboring nodes „x2‟
Here, „Tcost‟ is total cost for all 1-hop neighbor nodes and
„Scost‟ is the cost for one neighbor node, which can be
calculated as follows.
= 1+ 2+1,217
We set value zero to Tcost(x1) because „x1‟ is initiating node
that calculates the path cost that will be starting point. Energy
Level is used to calculate transmitting and receiving cost of
node with remaining energy of nodes. Nodes with value of
high cost are discarded and the cost of each 1- hop neighbor
node is saved into routing forwarding table (RFT).
Thus, „x2‟ calculates the minimum cost distance „D‟ for
reaching at 1-hop destination node with RFT using following
formula.
1= 1
,2 18
It is proved that minimum cost for establishing path from „x1‟
to „x2‟ is set in RFT of „x1‟.
3. SIMULATION AND ANALYSIS OF RESULT
The Realistic environment of WSNs use low power radios
with stochastic link and high asymmetrical communication
range. The simulation results could be different from expected
realistic results.We simulate LDSNS, MEAR, AQWSS and
MERR using NS 2.35-RC7. For simulation, we have designed
WSN that consists of different regions. Each region has
boarder node (BN) that forwards the collected information of
its region to BN of next region. We have simulated different
realistic scenarios:mobileand static. The main goal of
contribution is to reduce energy consumption and
supportingfaster search at one hop neighbor nodes.
The simulation scenario consists of 140 nodes with
transmission radius of 30 meters. The Bluetooth enabled
(BT)sensor nodes are uniformly and randomly placed in
geographical area of 300 * 300 square meters. Area is divided
into 75m x 75m different regions. The initial energy of nodes
is set 40 Joules. The bandwidth of node is 50 Kb/Sec and
maximum power consumption for each sensor is set 16 Mw.
Sensing and idle modes 12 mW and 0.5 mW respectively but
in our case, there is no idle mode. Sensors either go to active
or sleep mode. Each sensor is capable of broadcasting the data
at 10 power intensity ranging from -20 dBm to 12 dBm.
Total simulation time is 35 minutes and set 30 seconds
pause time for initialization of phase at start of simulation.
During this phase, only BN remains active and remaining
sensors of all regions go into power saving mode
automatically. The results demonstrate an average of 10
simulation runs. The energy consumption pertaining with
different radio modes and simulation parameters are summed
up in Table 1.
Table 1: Summarized simulation parameters for WSN
Name of parameters
Description
Transmission Range
30 meters
Type of sensors
BT node sensors
Sensing Range of node
10 meters
Initial energy of node
40 Joules
Bandwidth of node
50 Kb/Sec
Number of sensors
105 BT node rev-3
Size of network
300 * 300 square meters
Size of each region
75 * 75 square meters
Packet transmission rate
30 Packets/Sec
Mobility model
Random way- point
Simulation time
35 minutes
Initial pause time
30 Seconds
Tx energy
16 mW,
Rx energy
12 mW,
Power intensity
-20 dBm to 12 dBm.
Start time of BN-MAC
(0,30) Seconds
Sink location in each
region
(60, 40)
MAC protocol
BN-MAC
Type of protocols
MEAR, LADSNS, AQWSS
and MERR
Mobility
0.5 m/sec to 3.5 m/sec
Size of packets
8, 16,32,64,128,256 and
512 bytes
Routing Protocol
Pheromone termite
To demonstrate the validity of this novel environment of WSN
for handling several devices, we conduct several tests from
different perspectives. Having proved the mathematical
model, let us now evaluate the e efficiency of WSN.
We target preserving amount of energy of the network after
deploying from sensors as shown infigures 2 to 6. Our
simulated network shows that our proposed paradigm achieves
almost 100% efficiency and saves 62% energy using
maximum 140 sensors. We establish 15 sessions
simultaneously in order to determine the actual behavior of the
network in a highly congested environment. If we have less
number of sensors, it is hard to establish many sessions at the
same time. Figure 2 shows 14 hops in WSN and the used duty
cycle. We notethat LDSNS has consumed minimum duty
cycle as compared with MERR, MEAR and AQWSS. During
1 hop destination, all of the techniques use 1.5 % to 2.0% duty
cycles as the hops increase then their consumption ratio of
duty cycles decrease. The reason is that there is no overhead
on single hop-node and work is distrusted on other nodes that
are part of other hops. LDSNS uses less duty cycle because it
is integrated with BN-MAC protocol that has fully support of
semi synchronous mechanism as well as robust nature of
LDSNS makes it capable to use low duty cycle.
2
0.5 1 1.5 2.0 2.5
0
NUMBER OF HOPS
DUTY CYCLE (%)
LDSNS
MEAR
MERR
04 6 8 10 12 14
AQWSS
Figure.2. Consumed duty cycles versus number of hops
InF 3, we show consumption of energy for each scheme
against variable size of packets. LDSNS consumes less energy
with variable size of packets. In case of broken link, LDSNS
uses alternate link to send data packets but other schemes do
not have sound alternate link to forward the data. Further,
LDSNS uses both proactive and reactive approaches, when
nodes are static and there is possibility of leaving and joining
the new node.If the node decides to leave or other node wants
to join then LDSNS uses reactive approach to get run time
informationaccording to change of topology.
8
0.5 1 1.5 2.0 2.5
0
DATA PACKET SIZE (BYTES)
ENERGY CONSUMPTION (JOULE)
LDSNS
MEAR
MERR
016 32 64 128 256 512
AQWSS
Figure 3: Consumption of energy VS variable size of packets
In Figure 4, we show consumption of energy versus number of
hops. LDSNS consumes less energy than MERR, MEAR and
AQWSS. The increase in the number of hops affects all
protocols but LDSNS has better energy consumption. LDSN is
based on an efficient approach whichis integrated mobility and
scalability aware BN-MAC protocol. Meanwhile,the
otherapproaches experience energy increase due to the limited
support at MAC layer.
Figure 4. Energy consumption at variable number of hops of WSN
4
0.02 0.04 0.06 0.08 0.1
0
NUMBER OF NEIGHBORING NODES
EFFECTIVE DUTY CYCLES (%)
LDSNS
MEAR
MERR
08 12 16 20 24 28
AQWSS
Figure.5. Effective duty cycles due to neighboring nodes
In Figure 5, we show the effective duty cycle versus the
number of neighboring nodes. Due to the increase in the
number of neighbor nodes, LDSNS, MERR, MEAR and
AQWSS are affected Meanwhile LDSNS is not highly
affected. In addition, LDSNS is supported with semi
synchronous approach inherited by BN-MAC protocol. In all
of cases, LDSNS performs better than other protocols and
consumes less energy and gets higher data packet rates.
4. CONCLUSION
In this paper, we have introduced the Least Distance Smart
Neighboring Search (LDSNS) to reduce energy consumption
and provide faster scheduling for data delivery.LDSNS
performs better than other techniques even under low duty
cycle and different packet sizes. LDSNS is integrated with
BN-MAC protocol to support several applications such as
disaster, home automation, hospital and monitoring
applications of WSNs. To test the strength of LDSNS, our
designed WSNis divided into N-regions. Each region consists
of one boarder node that communicates with a neighboring
region. The boundary node israndomly elected on the basis of
high energy available inthe node. To validate the proposed
LDSNS, we have simulated LDSNS and compared its
performance with other known protocols: MEAR, AQWSS
and MERR by using ns2.35-RC7. On the basis of simulation
results, we demonstrate thatLDSNSperformsbetter than other
mechanisms (protocols) in terms of low duty cycle, energy
consumption, increasing number of neighbor nodes and size of
packets. LDSNS saves 24% to 62% energy resources and
improves by 12% to 21% search at 1-hop neighboring
nodes.In the future, we plan tocombine LDSNS with different
analytical models to test its strength in different scenarios.
REFERENCES
[1] F. Restuccia et al., “Performance analysis of a hierarchical discovery
protocol for WSNs with Mobile Elements”, In proceedings of IEEE
conference on World of Wireless, Mobile and Multimedia
Networks(WOWMOM), pp. 1-9, 2012.
[2] J. Chin et al., “Soft-appliances: A vision for user created networked
appliances in digital homes. Journal on Ambient Intelligence and Smart
Environments (JAISE), 1(1):69–75, 2009.
[3] A. Razaque and K.M.Elleithy,” Ambient Intelligence Paradigm to
support pedagogical activities over wireless sensor networks,”
Ubiquitous Computing and Ambient Intelligence (UCAmI), ACM/
Springer, December 3-5, 2012.
[4] T.Zhu et al., “Energy-synchronized computing for sustainable sensor
networks” International Journal for Elsevier on Ad Hoc Networks, vol.
11, no.4, pp. 1392-1404,2013.
[5] M.D.Francesco et al., “Reliability and Energy-Efficiency inIEEE
802.15.4/ZigBee Sensor Networks: An Adaptive and Cross-Layer”,
IEEE Trans. On selected areas in communications, vol.29, no.8, pp.
1508-1524, 2011.
[6] Debraj et al., “EAR: An energy and activity-aware routing protocol for
wireless sensor networks in smart environments”, Comput. J. 55(12),pp.
1492-1506, 2012.
[7] G Anastasi et al., “Energy conservation in wireless sensor networks: A
survey. International journal of Elsevier on Ad Hoc Networks, vol 7, no.
3,pp. 537–568, May 2009.
[8] C-C. Weng et al., “A Minimum Transmission Energy Consumption
Routing Protocol for User-Centric Wireless Networks” In proceedings
of IEEE International SI conference, pp.1143-1148, 2011.
[9] P Tarrio et al., “A Transmission Rate and Energy Design for Power
Aware Localization in Ad Hoc and Sensor Networks. PIMRC”, In
proceedings of IEEE,pp. 1-5, 2007.
[10] D. Bhattacharyya et al., “Review: A Comparative Study of Wireless
Sensor Networks and Their Routing Protocols”, International Journal of
sensors, vol.10, pp.10506-10523, 2010.
[11] Y. Zhang, K. Partridge and J. Reich, ”Localizing tags using mobile
infrastructure”, In Proceedings of the 3rd International Symposium on
location and context awareness (LoCA). Lecture notes in computer
Science (LNCS 4718). Springer, (Oberpfaffenhofen, Germany,
September 20-21, 2007, 279-296.
[12] J.Chen et al., “Minimum Energy Accumulative Routing in Wireless
Networks”, In proceedings of IEEE for 24th Annual INFOCOM,
pp.1875-1886, 2005.
[13] M. Zimmerling et al., Energy-Efficient Routing in Linear Wireless
Sensor Networks”, In proceedings of IEEE 2009.
[14] Shouwen Lai, “Duty-Cycled Wireless Sensor Networks: Wakeup
Scheduling, Routing, and Broadcasting” thesis, 2010.
[15] A. Razaque and K.M.Elleithy, “Automatic energy saving (AES) modelto
boost ubiquitous wireless sensor networks (WSNs)“ International
Journal of computers and technology 2013.
