127
to neighbours in the direction of the sink, and subsequently
128
prolonging the MWSN lifetime. ERTLD reports high perfor-
129
mance in terms of delivery ratio, end-to-end delay, and
130
power consumption. It has been successfully studied and
131
verified through simulation experiment using Network
132
Simulator-2 (NS-2) [8]. Secondly, it proposes a mobility
133
detection mechanism that used corona architecture based
134
on the position of a mobile sink. Corona architecture di-
135
vides MWSN area into a dynamic corona based on a mobile
136
sink which is assumed to be in the centre of coronas as it
137
will be explained in Section 3. An acronym table for fre-
138
quently used terms is shown as Table 1.
139
The rest of this paper is organized as follows: Section 2
140
will present related work on real-time communication for
141
MWSN. The design of ERTLD will be described in Sections 3
142
and 4 will describe the simulation study of ERTLD. Finally,
143
Section 5 will conclude the paper.
144
2. Related work
145
While most existing wireless sensor network deploy-
146
ments are still terrestrial networks with static sensor
147
nodes, mobile wireless sensor networks have received
148
increasing attention. During the past few years, several
149
mobile wireless sensor networks have been successfully
150
deployed in which sensor nodes are either equipped with
151
motors for active mobility or attached to mobile objects
152
for passive mobility. For example, researchers have at-
153
tached wireless sensor devices to Micro Air Vehicles [9],
154
bikes [10], vehicles [11,12], and animals [13,14]. In addi-
155
tion, wireless sensors are equipped with motors to move
156
underwater to collect data from static sensor devices
157
[15]. The related research for this paper can be classified
158
into two categories as follows:
159
2.1. Real-time routing protocol for static WSN
160
RAP is a real-time architecture and protocols based on
161
velocity [16]. It provides service differentiation in the time-
162
liness domain by velocity-monotonic classification of pack-
163
ets. Based on packet deadline and destination, its required
164
velocity is calculated and its priority is determined in the
165
velocity-monotonic order so that a high velocity packet
166
can be delivered earlier than a low velocity one. Similarly,
167
SPEED [17] is a stateless protocol for real-time communica-
168
tion in WSN. It bounds the end-to-end communication
169
delay by enforcing a uniform communication speed in
170
every hop in the network through a novel combination of
171
feedback control and non-deterministic QoS aware geo-
172
graphic-forwarding. MM-SPEED [18] is an extension to
173
SPEED protocol. It was designed to support multiple com-
174
munication speeds and provides differentiated reliability.
175
Scheduling messages with deadlines focuses on the prob-
176
lem of providing timeliness guarantees for multi-hop trans-
177
missions in a real-time robotic sensor application [19].In
178
such application, each message is associated with a dead-
179
line and may need to traverse multiple hops from the
180
source to the destination. Message’s deadlines are derived
181
from the validity of the accompanying sensor data and
182
the start time of the consuming task at the destination.
183
The authors propose heuristics for online scheduling of
184
messages with deadline constraints as follow: schedules
185
messages based on their per-hop timeliness constraints,
186
carefully exploit spatial reuse of the wireless channel and
187
explicitly avoid collisions to reduce deadline misses.
188
A routing protocol called real-time power control
189
(RTPC) uses velocity with the most energy-efficient for-
190
warding choice as the metrics for selecting a forwarding
191
node [20]. A key feature of RTPC is the ability to send the
192
data while adapting to the power of transmission.
193
RTLD is a real-time with load distribution for WSN. It
194
computes the optimal forwarding node based on the packet
195
reception rate (PRR), remaining power of sensor nodes and
196
packet velocity over one-hop. It consists of four functional
197
modules that include location management, routing man-
198
agement, power management and neighbourhood manage-
199
ment. The location management calculates the sensor node
200
location based on the distance to three pre-determined
201
neighbour nodes. RTLD reports high performance in terms
202
of delivery ratio, control packet overhead and power con-
203
sumption. However, RTPC, MM-SPEED, and RTLD are de-
204
signed for static WSN and unsuitable for MWSN.
205
2.2. Real-time routing protocol for MWSN
206
EAR2 is an expected area-based real-time routing pro-
207
tocol in Wireless Sensor Networks [21,22]. It depends on
208
an Expect Area (EA) of the mobile sink and exploit flooding
209
of real-time data within EA. EAR2 exploits multicasting
210
and one-hop forwarding time. To support a real-time data
211
with a desired time deadline, EAR2 guarantees that the
Table 1
An acronym table for frequently used terms.
Acronym Full spelling Acronym Full spelling
ERTLD Enhanced Real-time with Load Distribution CCP Corona Control Packet
CCP_ID Corona Control Packet Identity CD Corona Discovery
C_ID Corana Identity LM Location Management
LN Local Neighbour MWSN Mobile Wireless Sensor Network
MS Mobile Sink NC Neighbour Discovery
MN Mobile Node NS-2 Network Simulator-2
NM Neighbour Management NT Neighbour Table
OF Optimal Forwarding PE Performance Evaluation
PM Power Management PRR Packet Reception Rate
RM Routing Management RPH Route Problem Handler
RSSI Received Signal Strength Indicator RTR Request to Route
A. Ali Ahmed / Computer Networks xxx (2013) xxx–xxx
3
COMPNW 4944 No. of Pages 15, Model 3G
25 February 2013
Please cite this article in press as: A. Ali Ahmed, An enhanced real-time routing protocol with load distribution for mobile wirele ss sensor
networks, Comput. Netw. (2013), http://dx.doi.org/10.1016/j.comnet.2013.02.003