AN IMPROVED ALGORITHM TO FIRE DETECTION IN FOREST BY USING WIRELESS SENSOR NETWORKS

http://www.iaeme.com/IJCIET/index.asp 369 editor@iaeme.com
International Journal of Civil Engineering and Technology (IJCIET)
Volume 9, Issue 11, November 2018, pp. 369–377, Article ID: IJCIET_09_11_036
Available online at http://www.iaeme.com/ijciet/issues.asp?JType=IJCIET&VType=9&IType=10
ISSN Print: 0976-6308 and ISSN Online: 0976-6316
© IAEME Publication Scopus Indexed
AN IMPROVED ALGORITHM TO FIRE
DETECTION IN FOREST BY USING WIRELESS
SENSOR NETWORKS
Ghaida Muttashar Abdulsahib
Department of Computer Engineering, University of Technology, Baghdad, Iraq
Osamah Ibrahim Khalaf
College of Information Engineering, AI-Nahrain University, Baghdad, Iraq
ABSTRACT
Wireless sensor network systems diffuse an intensive, array of small, low-cost
sensors that monitor the environment. The system can be extending anywhere, even in
the place which is inaccessible .This technology can provide observe for forest fires in
the real time. Fire Ignition can be determined quickly, depending on the wake/sleep
table of the system of rules nodes. This research focus on fire detection ability of a
wireless network system. Sub divided system in randomly-spread nodes change the
network from being randomly spread to being arranged, and minimize the period of
working and less energy consumption of each hop. Separate the network into many
sub-networks lead to increases battery lifetime network by 3.7% and increased
performance of power by 69% compared to traditional fire detection networks. The
proposed network requires all nodes to be equipped with temperature sensor. The
analysis of data from sensors can show the fire, also its, behavior, intensity and
direction of deployed, which can assist the firefighting efforts. Traditional fire
detection networks show the fire only so the proposed algorithms show the best result.
Keywords: power consumption, Wireless Sensor Networks, Wireless Sensor Network
Coverage (WSN) and Decision Making.
Cite this Article: Ghaida Muttashar Abdulsahib, Osamah Ibrahim Khalaf, An improved
Algorithm to Fire Detection in Forest by Using Wireless Sensor Networks, International
Journal of Civil Engineering and Technology, 9(10), 2018, pp. 369–377.
http://www.iaeme.com/IJCIET/issues.asp?JType=IJCIET&VType=9&IType=10

An improved Algorithm to Fire Detection in Forest by Using Wireless Sensor Networks
1. INTRODUCTION
Forests important roles in global, environmental, ecological and recreational systems. The
forest greatly effect of the atmospheric carbon absorption and greenhouse gases and decrease
soil erosion. They can regulate rainfall and moderate temperature. Forest fires are the most
dangerous natural incident and occur in all countries. Forest fires are a dangerance of the
plant, environmental and people. Especially if the forest is not inhabited and this means a
great and rapid fire, so we are going to this proposed proposal to control the forest fires.
Unfortunately, forest fires are show only when they have spread over a large area, making its
stoppage and control arduous and impossible. The result is a wasteful loss of lives (people
and of fire-fighter crewmen and others) and ownership (resources from forest such us wood
and resources as well as clusters of buildings and other houses in the outlying areas).
The most problem with forest fires is that forests are distant and full of trees, also dry
wood and leaf litter that act as fuel of fire. These elements are combustible materials and
represent the perfect to ignite the fire. Maybe Fire ignition can be caused by human behavior
like barbeques or smoking, or by natural causes such as very high temperatures on a
summer‟s day. When the ignition starts, then combustible materials may fuel fire. Then the
fire becomes wider and bigger. Initial stage of ignition is referred to as „surface fire‟ stage.
This may then deploy to neighboring trees, and allowing the fire to become higher to higher,
thus becoming a „crown fire. After few minutes fires become uncontrollable. The rule of the
fire response is: wait almost one minute, use one cup of the water; wait almost two minutes,
use hundred liters of water; wait almost ten minutes, use thousand litres of water. The main
aim can present by Figure1:
Figure 1 System Aim
2. COVERAGE
Coverage‟ can be nearly defined as the way that a sensor network monitors a part of
attention, and its identical (QoS). Coverage include providing the connectivity for the hops in
specific way which make each hop capable of connect with another hops in the system until
it reach to the network receiver.
Wireless sensor network coverage can be categories into two kinds: dynamic coverage
and static coverage. Static can be utilized for fixed hops that are used in a shape of the
predefined. It must give a connection between every hop transmission coverage range and
those of neighboring hops. In order to ensure give high performance of transmission coverage
in different applications of environment, so K algorithm of coverage has been proposed. In
the other side the other type which is called as a dynamic, all hops have the capacity to
transfer everywhere.
The algorithm can be utilized in case that the optimal usage cannot be specified because
of a mysterious topology. Random deployment cannot give good coverage and this needs

Ghaida Muttashar Abdulsahib, Osamah Ibrahim Khalaf
difficult methods. To avoid the lack of effectiveness in coverage, different methods
developed. Virtual Force Algorithm it is algorithm to optimize sensor usage to improve
coverage when the random placement specified [1]. Dynamic type used in military
commonly and significantly.
3. PREVIOUS STUDIES
3.1. Broadcasting Communications
Broadcasting of communications is so appropriate in the forest fire applications detection. It
used to assist that the data will be delivered by neighbor hops and they will send the data
again, the data can transfer data to next hop by using various routes in order to send data to
the receiver. In order to avoid nodes from are-broadcasting the repeated messages or same
message, all hops will test if data ID is kept in the buffer or not because each message has i
unique ID in buffer before re-broadcasting. This task is found in AODV (Ad hoc On-Demand
Distance Vector Routing, which makes route only when needs) and can be reduces the
resource consuming like: processing ,power, collision, etc.[2]. Broadcasting communication
will be utilized in this paper in order to keep the energy in networks and to assist the signal
delivery from the sender to the receiver.
3.2. Distribution random hop in network area
A GUI has utilized in the research to make simulation. It depends on node distribution and
the location of the receiver. Because the hop distribution is random, QoS high coverage is not
easy to get initially, but based on localization results method, a GUI can created help identify
gaps coverage. Then, these gaps can be full of by redistributing randomly nodes. GUI can as
well help with network events checking such as battery exhaustion, node failures, forest fires
and coverage gaps [3].
3.3. Studies and Methods to avoid Coverage Problems
Different studies were created in order to avoid shortcomings of coverage algorithm which is
called as K algorithm. This algorithm use K hops to check the domain of the interest in this
algorithm K is a specific number that different based on the application program. Amari and
Dass study [3] showed that the connectivity algorithm of a K-covered net is greater than a 1 -
contained networks and presented various network connection algorithm which calculated
two type of networks which are: heterogeneous and homogeneous WSNs (the term of
homogeneity indicates to similarity of hops types).
3.4. The changing the space from (3D) to a (2D)
Huanng et al. study [4] proposed an algorithm for avoid the problem in (3D) coverage by
converting the data from (3D) to (2D). It created a polynomial period algorithm to test all
locations in an area is K-used or not.
3.5. The Models based Exposure
Srivastava and Adlkha study [5] utilized an "exposure based model" to define sensors
numbers which need to obtain full of the coverage of a specific area.
3.6. Directional coverage
Abozeid and Ai study [6] proposed a pre-directional WSN coverage algorithm for the system
by using the pre-directional sensors, with coverage based on the sensor orientation. Lin and

Du [7] introduced a distinguish coverage system for WSNs, where areas of the network
should have various levels of the sensing algorithm of coverage.
3.7. Arrangement for mobile WSNs
In study [8] suggest that asynchronous algorithm for coverage that can be adapted to WSNs
mobility .In [9] suggest a deployed method which are used to the keep minimum number of
the active sensors without concerning to the links between sensing ranges and
communication.
3.8. Efficiency of power consumption
In [10] suggest dispersed calculations to supply the best inclusion with the less power
utilization.
3.9. Communication Range and Sensing Range
These two terms have different ideas as show in figure 2. Detecting range is the space at
which the gadget can be sense the occasions that it is set up for Communication Range is the
area at which two hubs can connect with one another.
Figure 2 Communication Range and Sensing Range
The hubs can be equipped for imparting more than hundred meters incorporate Libelium-
Waspmote. Hubs can be inclusion to impart from 500 m (XBee 802.15) to the 12 km [11].
The scope of the hubs can from under 1 meter to 80 km in exceptionally costly sensors and
progressed, for example, Firewatcher.
The importance of the correspondence range and detecting area change as per the diverse
application. Sensor extend is vital in the timberland fire recognition since it is assume the real
job in deciding the separation among hubs and hub thickness. Detecting range is lower than
the correspondence run keeping in mind the end goal to the recognize fire in a brief
timeframe after the start, detecting inclusion ought to have just little holes. This thought
prompts the breaking point of the separation between sensors because of the detecting range
space. Then again, prompt builds identification time and lessened the precision. Little range
detecting separation can be deliver less deferral and exact. Little range sensors are sent at
range that are somewhat bigger than their scope of detecting, where it is conceivable to have
the little detecting holes between hubs [12].

4. COVERAGE METHOD
To provide reliable service and accurate sensing in wireless sensor network should separate
by reasonable range. The sensors utilized in the research are just ready to detect temperature
in the earth of the encompassing, so correspondence run is constrained to little scope of
around 25 to 60 meters. Despite the fact that this range will be make holes in detecting range,
in other hand will give detecting of nature. In the 50-hub arrange on NS2 assumed be existent
and the framework was dissected in the present investigation. Hubs areas were at that point
known in gathered system. Nearby data was utilized on the GUI to the speak to each sensor.
As show in Fig 3
Figure 3 The System Supposed for the research
4.1. The Proposal Method and Feature
The main features of this proposed method are as follows:
 Saved of the Resources by a deployment scheme, and less energy consumption
 Distributed randomly for the Nodes, but deployment by using sub-arrange systems
convert the system and area data to composed organization. Network lifespan will be
increased
 The connectivity of the Nodes are guaranteed
 Guaranteed to the full Sensing coverage
 Easily informed to the operator of any gaps, changes, failures or movements these
feature can achieved through dividing system into many sub-systems depending on
the location of the node. In present research, analysis is the limited to division of
network to three sub-networks.
Must turn on the sub-network every ten minutes. The propose algorithm permits every
sub-system, rather than all hops in the system , to be turned on every ten minutes for thirty
seconds, to sense environment after that go back to inactive status. In next ten minutes, next
sub-system will be in active status, and so on.
Rather than traditional method making all nodes in the system will be passed in active
status for three times for ninety seconds in every thirty minutes, sub-systems one, two ,
three work for thirty seconds every thirty minutes. Total, all nodes in the system wake up for
the thirty seconds every thirty minutes.

Through proposed method the nodes deployed are fewer and less traffic in the network.
The proposed method checked in the NS2 and used in the following parts:
1) Proposed method divided system into three of the sub-network depending on the
range from hop which is called (K), where:
 The first Sub-network is consist of some of nodes which located between
(1 + 16n) ≤ K < (6 + 16n)
 The second Sub-network is consist of some of nodes which located between
(6 + 16n) ≤ K< (11 + 16n)
 The third Sub-network is consist of some of nodes which located between
(11 + 16n) ≤ K < (16 + 16n)
Where n = {0, 1, 2 ….. [n + 16n ≤ (range from the receiver to edge of the
transmission space)]}
2) Each hub must check if there accessible one hub in any event in correspondence go,
notwithstanding when the required to obtain from neighbor sub-organize. The scope
of correspondence in the test system is decreased to the twenty meters to the ensure a
large portion of hubs has more one association.
3) To cheek have entire availability between hubs, the check accomplice capacity will be
made to make guarantee each hub in sub-arrange has one association in any event.
Further, an appropriate hub from another sub-system will be select and use as an
accomplice. An arrangement is got hub from neighbor sub-systems and keep control
adjusted in a system. Sub-system will keep rundown of all hubs which can be
borrowed (elective competitors) from neighbor system to the cover its hole. Each time
sub-system will be awakened, and utilize distinctive elective hubs as shown in fig 4
Figure 4 The Method of sub-networks
GUI will be use the location and sub-network information to present the gap and
communication range as show in fig5

Figure 5 Identified of the nodes through Sub-network in the Nam
5. RESULTS
The third alternative of the proposed strategy is the least expensive and appropriate route
because of inclusion hole has checked in the lead position as all hubs are a piece of one major
system. Whenever found the hole doesn't inclusion by any sub-systems, at that point must
make a move to be taken. In another hand, if doesn't hole in mother organizes after that the
holes begin to show up in the sub-arranges, that implies the territory will be secured later by
another sub-arrange. On the off chance that inclusion territory between (- 21, 71) in both of X
and Y so the rates of inclusion add up to the focused zone of inclusion is as appeared in Table
1.
Table 1 The Result of the sub-network Coverage
The proposed technique increment lifetime of the system by 2.8 times and sparing the
vitality will be expanded. It is most vital realize that arbitrarily conveyed hubs can be given
additional inclusion couple of hubs. The fewer holes of the inclusion in the system, the
additional availability that hubs will have. The calendar of wake up/rest can be influence
organize lifetime and deferral in flame recognition, which can be mean what is now caused
by the inclusion of sub-systems. It can change depending on the natural issues. For instance,
in the territory of outdoors in the mid-year might be like clockwork in the day and like
clockwork at the night. In another hand, in winter could be at regular intervals or more. ,
organize perhaps awaken at regular intervals. The timetable of wake up/rest is as per the
climate conditions, region, and history of the flames in the territory. The Evaluation of
proposed strategy utilizes the parameters of inclusion assessment as appeared in Table2.

Table 2 The saving power of proposed method
6. CONCLUSION
There are two contributions in the proposal method to fire detection, overcome coverage and
alarm faulty problems:
1. The sub-arrange inclusion technique can change the systems with arbitrary
dissemination to a sorted out sending by separation the system to three of the sub-
systems to the lessening number of hubs, increment systems lifetime by 2.8% and
expands execution of the vitality by 64% whenever contrasted and a typical fire
location organize.
2. Fire detection systems do not require gas board detection or another device,
connection with databases in network. Only simple, temperature sensors are required
at every node. Temperature sensors help in the decision-making by distinguishing
between fires, false alarms. In addition, provide fire behavior can improve
performance of firefighting.
REFERENCES
[1] Z. Chuan, Z. Chunlin, S. Lei, H. Guangjie, "A survey on coverage and connectivity issues
in wireless sensor networks", Journal of Network and Computer Applications, vol. 35, no.
2, (2011), pp. 619-632.
[2] V. Rao, "The Simulative Investigation of ZibBee/IEEE 802.15.4", Department of
Electrical Engineering and Information Technology Dresden University of Technology.
Master of Science, (2005).
[3] H.M. Ammari and S.K. Das, “Coverage, Connectivity, and Fault Tolerance Measures of
Wireless Sensor Networks,” Proc. Eighth Int‟l Symp. Stabilization, Safety, and Security
of Distributed Systems, (2006), pp. 35-49.
[4] CM. R. C. Qazani, H. Asadi, S. Pedrammehr and S. Nahavandi, "Performance analysis
and dexterity monitoring of hexapod-based simulator," 2018 4th International Conference
on Control, Automation and Robotics (ICCAR), Auckland, 2018, pp. 226-231.
[5] S. Adlakha and M. Srivastava, “Critical Density Threshold for Coverage in Wireless
Sensor Networks,” Proc. IEEE Wireless Comm. and Networking Conf, (2003), pp. 1615-
1620.
[6] J. Ai and A. Abouzeid, “Coverage by Directional Sensors in Randomly Deployed
Wireless Sensor Networks,” J. Combinatorial Optimization, vol. 11, no. 1, (2006), pp. 21-
41.

[7] X. Du and F. Lin, “Maintaining Differentiated Coverage in Heterogeneous Sensor
Networks,” EURASIP J. Wireless Comm. and Networking, vol. 5, no. 4, (2005), pp. 565-
572.
[8] J. Cortes, S. Martinez, T. Karatas, and F. Bullo, “Coverage Control for Mobile Sensing
Networks,” IEEE Trans. Robotics and Automation, vol. 20, no. 2, (2004), pp. 243-255.
[9] H. Zhang and J. Hou, “Maintaining Sensing Coverage and Connectivity in Large Sensor
Networks,” Ad Hoc & Sensor Wireless Networks, vol. 1, nos. 1-2, (2005), pp. 89-124.
[10] X. Li, P. Wan, and O. Frieder, “Coverage in Wireless Ad-Hoc Sensor Networks,” IEEE
Trans. Computers, vol. 52, no. 6, (2003), pp. 753-763.
[11] M.R.C., Qazani, S. Pedrammehr, & M.J. Nategh, “An Investigation on the Motion Error
of Machine Tools‟ Hexapod Table” Int. J. Precis. Eng. Manuf. (2018) 19: 463.
[12] Chamkouri, N., Niazi, A., &Zare-Shahabadi, V. (2016). Development of a novel pH
sensor based upon Janus Green B immobilized on triacetyl cellulose membrane:
Experimental design and optimization. SpectrochimicaActa Part A: Molecular and
Biomolecular Spectroscopy, 156, 105-111.
[13] S. Syed Abdul Syed, Senthil Kumaran, A. Sarfaraz Ahmed, E. Syed Mohamed, A
Manikandan B and S. Rajesh, A Novel Broadcasting Method For Code Dissemination In
Wireless Sensor Networks. International Journal of Electrical Engineering & Technology,
6(7), 2015, pp. 61-70.
[14] Manan Mehta. ESP 8266: A Breakthrough in Wireless Sensor Networks and Internet of
things. International Journal of Electronics and Communication Engineering &
Technology, 6(8), 2015, pp. 07-11
[15] Dr. Syeda Gauhar Fatima, Dr. Syed Abdul Sattar and Mohammed Sadiq, Efficient
Defense System for Jamming Attacks in Wireless Sensor Networks. International Journal
of Electronics and Communication Engineering and Technology, 9(4), 2018, pp.22–35.

AN IMPROVED ALGORITHM TO FIRE DETECTION IN FOREST BY USING WIRELESS SENSOR NETWORKS

Related slideshows

More Related Content

AN IMPROVED ALGORITHM TO FIRE DETECTION IN FOREST BY USING WIRELESS SENSOR NETWORKS