Selecting number of forwarding reports

International Journal of Ambient Systems and Applications (IJASA) Vol.2, No.4, December 2014
DOI:10.5121/ijasa.2014.2401 1
SELECTING NUMBER OF FORWARDING REPORTS
TO IMPROVE ENERGY SAVINGS IN BANDWIDTH
EFFICIENT COOPERATIVE AUTHENTICATIONS FOR
SENSOR NETWORKS
Su Man Nam1
and Tae Ho Cho2
1,2
College of Information Communication Engineering, Sungkyunkwan University,
Suwon, 400-746, Republic of Korea
ABSTRACT
Wireless sensor networks provide ubiquitous computing systems in various open environments. In the
environment, sensor nodes can easily be compromised by adversaries to generate injecting false data
attacks. The injecting false data attack not only consumes unnecessary energy in en-route nodes, but also
causes false alarms at the base station. To detect this type of attack, a bandwidth-efficient cooperative
authentication (BECAN) scheme was proposed to achieve high filtering probability and high reliability
based on random graph characteristics and cooperative bit-compressed authentication techniques. This
scheme may waste energy resources in en-route nodes due to the fixed number of forwarding reports. In
this paper, our proposed method effectively selects a dynamic number of forwarding reports in the source
nodes based on an evaluation function. The experimental results indicate that our proposed method
enhances the energy savings while maintaining security levels as compared to BECAN.
KEYWORDS
wireless sensor network, network security, false data injection attack, bandwidth efficient cooperative
authentication
1. INTRODUCTION
Recent wireless sensor networks (WSNs) have a large number of sensors and a base station in
various applications such as health care monitoring, forest fire detection, natural disaster
prevention, etc. [1]. The sensors enable the development of low-cost, low-power, and multi-
functional sensors [1, 2]. The functions of a sensor node include sensing, computing, and wireless
communication. When a real event occurs, the sensor node senses the event, computes it for a
report, and transmits the report through wireless communication to a base station. The base
station collects the report, analyses the event data of the report, and provides the event
information to users. The technology of the sensor network easily and conveniently provides
diverse information to users. However, the sensor network has the greatest probability of being
captured and compromised because it operates in an open environment [3, 4]. In addition, the
sensor node is readily exposed to diverse attack patterns from malicious attackers. In order to
effectively operate the network, a confident countermeasure is needed against these attacks.
The sensor network is vulnerable to the various attacks because it is easily captured and
compromised. Among those attacks, injecting false data attacks can result in one of two cases that

consume unnecessary energy in the e
as shown in Figure 1.
Figure 1 shows two cases (Figures 1
network. We assume that a source node is compromised and an en
Figure 1-(a), a source node is compromised, and the compromised node injects a false report into
the network. The injected false report passes through the en
station. The en-route nodes consume unnecessary energy due to the reception and transmission of
the false report. If the false report arrives at the base station, the base station may generate false
alarms. In Figure 1-(b), a legitimate report is tr
a compromised node while forwarding it, the compromised node intentionally drops the report.
An alarm cannot be generated at the base station. Thus, if false reports are flooding into the
network, not only will a large amount of energy be wasted in the en
verification burdens will also undoubtedly fall on the base station [5].
Bandwidth-efficient cooperative authentication (BECAN) was proposed in order to prevent false
data injection attacks. BECAN achieves high filtering probability by using a cooperative
neighbors × router-based (CNR) filtering mechanism and high reliability by using a multireport
solution (i.e., multiple paths). When a real event occurs within a cluster, a sourc
message authentication codes (macs) by using the CNR filtering mechanism from its neighbors.
Then the source node produces a MAC by using macs to include in a report. When the node
forwards the report, it uses the multireport solution, whic
paths. Although BECAN simultaneously provides high filtering probability and reliability, the en
route nodes can consume unnecessary energy resources due to the multireport solution.
In this paper, we propose a meth
for the multireport solution based on an evaluation function in the source node. Before forwarding
a report, the source node dynamically selects the number of forwarding reports by using the
evaluation function. Effectively selecting the number of forwarding reports influences the high
reliability and energy savings in the sensor network. Thus, our proposed method more effectively
chooses the dynamic number of forwarding reports in order to enh
against the injecting false data attacks.
The rest of this paper is organized as follows: the background and motivation of this proposal are
described in Section 2. Section 3 introduces our proposed method in detail, and Sectio
consume unnecessary energy in the en-route nodes and generate a false alarm at the base station
Figure 1. Injecting false data attack
Figure 1 shows two cases (Figures 1-(a) and (b)) of injecting false data attacks [5-7] in the sensor
source node is compromised and an en-route node is compromised. In
(a), a source node is compromised, and the compromised node injects a false report into
the network. The injected false report passes through the en-route nodes until it reaches th
route nodes consume unnecessary energy due to the reception and transmission of
(b), a legitimate report is transmitted for a source node. If the report arrives at
nly will a large amount of energy be wasted in the en-route nodes, but heavy
verification burdens will also undoubtedly fall on the base station [5].
efficient cooperative authentication (BECAN) was proposed in order to prevent false
ion attacks. BECAN achieves high filtering probability by using a cooperative
based (CNR) filtering mechanism and high reliability by using a multireport
solution (i.e., multiple paths). When a real event occurs within a cluster, a source node collects
forwards the report, it uses the multireport solution, which is transmitted along multiple routing
paths. Although BECAN simultaneously provides high filtering probability and reliability, the en
In this paper, we propose a method to effectively select a dynamic number of forwarding reports
valuation function. Effectively selecting the number of forwarding reports influences the high
chooses the dynamic number of forwarding reports in order to enhance the energy efficiency
against the injecting false data attacks.
described in Section 2. Section 3 introduces our proposed method in detail, and Sectio
2
route nodes and generate a false alarm at the base station
7] in the sensor
route node is compromised. In
(a), a source node is compromised, and the compromised node injects a false report into
route nodes until it reaches the base
route nodes consume unnecessary energy due to the reception and transmission of
ansmitted for a source node. If the report arrives at
route nodes, but heavy
efficient cooperative authentication (BECAN) was proposed in order to prevent false
ion attacks. BECAN achieves high filtering probability by using a cooperative
based (CNR) filtering mechanism and high reliability by using a multireport
e node collects
h is transmitted along multiple routing
paths. Although BECAN simultaneously provides high filtering probability and reliability, the en-
od to effectively select a dynamic number of forwarding reports
valuation function. Effectively selecting the number of forwarding reports influences the high
ance the energy efficiency
described in Section 2. Section 3 introduces our proposed method in detail, and Section 4

provides the analysis and experimental results. Finally, the conclusions and future works are
discussed in Section 5.
2. BACKGROUND
In this section, we first discuss BECAN among the countermeasures [8, 9] of the sensor network
in Section 2.1, and the motivation of this paper is presented in Section 2.2
2.1. BECAN
BECAN consists of four phases: (1) sensor nodes initialization and deployment, (2) sensed results
reporting protocol, (3) en-routing filtering, and (4) base station (sink) verification
Figure
Figure 2 shows phase (1), which is the initialization of the sensor nodes with the (a) parameters
(params) and (b) keys installation. Before distributing the
elliptic curve E ቀ൫‫ܨܫ‬௣൯, ‫,ܩ‬ ‫ݍ‬ቁ and a hash function h(). All of the sensor nodes set the params with
TinyECC [10] before they are deployed in the sensor field. Each node receives a private key
(where the private key ‫ݔ‬௜ is randomly chosen from
generates the public key ܻ௜ (ܻ௜ ൌ
key and a private key ሺܻ௜, ‫ݔ‬௜ሻ.
Figure 3 illustrates phase (2), which is the sensed results routing phase. A source node
of k neighboring nodes ܰேబ
: ሼܰ
ܴଶ → ⋯ → ܴ௟ → ‫݊݋݅ݐܽݐܵ ݁ݏܽܤ‬ሽ
motivation of this paper is presented in Section 2.2.
routing filtering, and (4) base station (sink) verification.
Figure 2. Sensor nodes initialization
) and (b) keys installation. Before distributing the params, the base station selects an
and a hash function h(). All of the sensor nodes set the params with
before they are deployed in the sensor field. Each node receives a private key
is randomly chosen from ‫ݖ‬௣
∗
) from the base station. It then collectively
ൌ ‫ݔ‬௜‫,ܩ‬ where i is a node’s identifier.) Thus, each node has a public
Figure 3. Sensed results routing
Figure 3 illustrates phase (2), which is the sensed results routing phase. A source node
ܰଵ, ܰଶ, ܰଷ, ܰ௞ିଵ, ܰ௞ሽ, and establishes a routing path
ሽ. When an event occurs in a cluster, the source node senses the
3
, the base station selects an
and a hash function h(). All of the sensor nodes set the params with
before they are deployed in the sensor field. Each node receives a private key ‫ݔ‬௜
) from the base station. It then collectively
is a node’s identifier.) Thus, each node has a public
Figure 3 illustrates phase (2), which is the sensed results routing phase. A source node ܰ଴ consists
, and establishes a routing path ܴேబ
: ሼܴଵ →
occurs in a cluster, the source node senses the

4
event data (m). The source obtains the current timestamp T, selects the neighboring nodes, and
transmits the event (m, T) to the neighboring nodes. Each neighbor transmits a mac (m, T) to the
sources after each neighbor verifies the event data. The source node aggregates all of the macs
(m, T) and produces a report (m, T, MAC) to send to the base station. For transmitting a report,
the multireport solution is used. Once a real event occurs, a source node selects k different
neighbors, produces the reports, and sends them to the base station via different paths.
In phase (3), the en-routing filtering, the en-route nodes (ܴଵ, ܴଶ, ⋯ , ܴ௟) verify the integrity of the
message m and the timestamp T. If T is out of date, the report will be discarded. The en-route
nodes then authenticate the MAC in the report using their public key and the CNR-based MAC
verification algorithm. Once it is verified that the report is legitimate, the report is transmitted to
the next hop node.
In phase (4), the base station verification, when the report reaches the base station, the report is
verified by its keys. When a legitimate report arrives, it is successfully reported.
2.2. Motivation
In a WSN, sensor nodes are easily compromised by attackers due to limited resource hardware.
These attackers can fabricate a bad report through the compromised node and can inject the false
report into the sensor network. BECAN was proposed to detect this attack. This method achieves
not only high filtering probability but also high reliability. Although the existing method performs
well in the sensor network, the sensor nodes may waste energy resources in the en-route nodes
due to the fixed number of forwarding reports by using the multireport solution. In this paper, our
proposed method effectively selects a dynamic number of forwarding reports considering three
input factors before transmitting the reports. Therefore, the proposed method improves the energy
efficiency while maintaining the same security level against the attack compared to BECAN.
3. PROPOSED METHOD
This section describes our proposed method in detail.
3.1. BECAN
The sensor nodes are deployed with fixed positions in a sensor field. The sensor network is
comprised of a base station and a large number of sensor nodes including H-sensors and L-
sensors, e.g., Berkeley MICAz motes [11]. The H-sensor hardware (processors, memory, battery,
storage, etc.) is more powerful than the L-Sensor [12]. An H-sensor consists of some L-sensors in
a cluster. The topology is established by directed diffusion and a minimum cost forwarding
algorithm. Each H-sensor discovers the routing paths toward the base station. Compromised
nodes can inject false reports and interrupt the transmission of legitimate reports in the sensor
network.
3.2. Overview
Our proposed method selects a dynamic number of forwarding reports for the multireport solution
by using an evaluation function.

Figure 4 shows the processes of the proposed method when a source node transmits a report to the
base station. As shown in Figure 4, the source node forwards the report, and it evaluates its
current status in order to select the dynamic number of forwarding report
function. In the evaluation function, there are three input factors (REL, FNR, and HOP) and one
output factor (NFR). The resulting value is compared to a defined value. If the result value is
greater than the defined value, the num
result value is less than the defined value, NFR is decreased. Thus, our proposed method
effectively determines the dynamic number of forwarding reports through the condition of the
sensor network based on the evaluation
3.3. Operation process of the proposed method
In the sensor network, the sensor nodes’ energy resources can be wasted by injecting false data
attacks. In order to effectively detect this attack, our proposed method decides a dyna
of forwarding reports in a source node through an evaluation function. The source node executes
Algorithm 1 before forwarding a report
Algorithm 1 SetNumberForwardingReports(REL, FNR, HOP)
1: state←REL/HOP+FNR
2:
3: ifstate ≥ thresholdthen
4: NFR++;
5: else
6: NFR−−;
7: end if
8: returnNFR;
Algorithm 1 shows the operation process of the proposed method for effectively selecting the
number of forwarding reports in the source node. The source node executes Algorithm 1 and
selects the dynamic number of forwarding reports before forwarding the re
In the algorithm, the input factors are REL, FNR, and HOP, and the output factor is NFR. The
status of the source node is evaluated with these input factors, and a result value
(line 1). If the value state is great
than threshold, NFR is decreased (lines 3
applies the NFR and transmits the report after selecting NFR different neighbors
Figure 4. Overview of proposal
shows the processes of the proposed method when a source node transmits a report to the
current status in order to select the dynamic number of forwarding reports by using an evaluation
greater than the defined value, the number of forwarding reports (NFR) is increased, while if the
based on the evaluation.
Operation process of the proposed method
attacks. In order to effectively detect this attack, our proposed method decides a dyna
Algorithm 1 before forwarding a report.
Algorithm 1 SetNumberForwardingReports(REL, FNR, HOP)
FNR;
then
selects the dynamic number of forwarding reports before forwarding the report to the base station.
status of the source node is evaluated with these input factors, and a result value state
is greater than threshold, NFR is increased, while if the value is less
, NFR is decreased (lines 3-7). The NFR is then returned in line 9. The source node
applies the NFR and transmits the report after selecting NFR different neighbors.
5
shows the processes of the proposed method when a source node transmits a report to the
s by using an evaluation
ber of forwarding reports (NFR) is increased, while if the
attacks. In order to effectively detect this attack, our proposed method decides a dynamic number
port to the base station.
state is stored
, NFR is increased, while if the value is less
7). The NFR is then returned in line 9. The source node

3.4. Evaluation Function
In order to select the dynamic number of forwarding reports, the evaluation function in the
proposed method uses three input factors: REL, FNR, and HOP. The input factors of the
evaluation function are as follows
Remaining energy level: This value is an important factor for saving energy in the sensor
network. If the energy level of the source node is small, the number of forwarding reports
should be decreased to save the network energy. The enhancement of the energy efficiency is
influenced by the number of communications.
Number of Hops: This factor describes the hop count, which is the number of hops from the
base station to a source node. If the source node is far from the base station, reports
transmitted from the base station con
False negative rate: This factor represents the security of the network. This factor is
calculated as the number of true reports that cannot reach the base station over the total
number of true reports. If FNR is small, the sensor network demonstrates high reliability. The
security level is influenced by FNR.
To effectively decide the dynamic number of forwarding reports, we define an evaluation
function by considering these factors. The evaluation fun
In Equation (1), i is an identifier of the source node. The source node calculates the ratio of
energy consumption through energy over the number of hops and evaluates its current status of
the node through FNR. That is, the number of forwarding reports is selected by the eval
function. If the resulting value is greater than a defined threshold, NRF is increased, and if the
result is less than the threshold, NRF is decreased
3.5. Example
Figure 5. Selection of the number of forwarding reports, including a comparison
evaluation function are as follows.
el: This value is an important factor for saving energy in the sensor
luenced by the number of communications.
transmitted from the base station consume a great deal of the energy of the en-route nodes.
. If FNR is small, the sensor network demonstrates high reliability. The
security level is influenced by FNR.
function by considering these factors. The evaluation function is defined as follows:
Fሺiሻ ൌ
‫ܩܰܧ‬
‫ܱܲܪ‬
൅ ‫ܴܰܨ‬
an identifier of the source node. The source node calculates the ratio of
the node through FNR. That is, the number of forwarding reports is selected by the eval
result is less than the threshold, NRF is decreased.
Selection of the number of forwarding reports, including a comparison with a threshold
6
el: This value is an important factor for saving energy in the sensor
route nodes.
. If FNR is small, the sensor network demonstrates high reliability. The
(1)
an identifier of the source node. The source node calculates the ratio of
the node through FNR. That is, the number of forwarding reports is selected by the evaluation
with a threshold

7
Figure 5 shows an example of the process for selecting the number of forwarding reports in the
proposed method. The source node outputs its current status by using the evaluation function
before forwarding a report to its neighbors. The result of the function is compared to the defined
threshold (TH). We consider that the source node’s condition has two cases (Figures 5-(a) and 5-
(b)) and the defined TH is 20. If REL is 90, HOP is 10, and FRN is 10, such as the current state
Figure 5-(a), the result of the evaluation function is 19. The result 19 is compared to the TH of 20.
The NFR is decreased because the result is less than the TH. Otherwise, if REL is 60, HOP is 4,
and FRN is 6, such as in the current state in Figure 5-(b), the function has a result of 21. The NFR
is then increased. Therefore, our proposed method can improve the effective performance of the
sensor network because it selects the dynamic number of forwarding reports.
4. EXPERIMENTAL RESULTS
Experimental results were obtained to demonstrate the effectiveness of the proposed method as
compared to BECAN. In the sensor network, the sensor field’s size is 500×500 m2
including a
total of 500 nodes (100 H-sensors and 400 L-sensor). They are uniformly distributed in the field,
and an H-sensor consists of nine L-sensors in a cluster. When a sensor receives and transmits a
report, it consumes 16.25 µJ and 12.5 µJ per byte, respectively. It also consumes 15 µJ and 75 µ
when a MAC is generated and verified, respectively. The report size is 24 bytes. We randomly
generated 2,000 events in the sensor field. To randomly generate injecting false data attacks, we
compromised five sensor nodes after deployment. The compromised nodes inject false reports
with 5% probability into the network.
Figure 6. Average Energy consumption vs. number of hops from source nodes to base station (BS)
Figure 6 shows the average energy consumption as a function of the number of hops in the sensor
network. In the proposed method, the energy used when the en-route nodes are close to the source
nodes is almost the same as for BECAN. Having the en-route nodes between 4 and 10 hops
improves the energy efficiency. That is, the proposed method saves energy resources close to the
base station. The reason for the energy savings in the proposed method is shown in Figure 7 in
detail.
AverageEnergyConsumption(mJ)
0
50,000
100,000
150,000
200,000
250,000
300,000
350,000
Numberof Hops fromSourceNodes fromBS
0 2 4 6 8 10
BECAN
Proposed Method

Figure 7. Number of transmitted reports in source nodes
Figure 7 shows the number of transmitted reports according to the
method reduces the number of transmitted reports as compared to BECAN. The reason is that the
source nodes in the proposed method select the effective number of forwarding reports through
the evaluation function by considering
the energy savings by decreasing the number of forwarding reports as compared to BECAN
Figure 8
NumberofTransmittedReportsinSouceNodes
0
100
200
300
400
500
0
BECAN
Proposed Method
Number of transmitted reports in source nodes vs. number of hops from source node to BS (Base
Station)
Figure 7 shows the number of transmitted reports according to the number of hops. The proposed
the evaluation function by considering three input factors. Thus, the proposed method enhances
the energy savings by decreasing the number of forwarding reports as compared to BECAN
8. Ratio of dropped false reports vs. methods
Numberof Hops fromSource Nodes fromBS
2 4 6 8 10
BECAN
Proposed Method
8
vs. number of hops from source node to BS (Base
number of hops. The proposed
three input factors. Thus, the proposed method enhances
the energy savings by decreasing the number of forwarding reports as compared to BECAN.

9
Figure 8 shows the ratio of dropped false reports between the two methods. As shown in Figure 8,
the two methods are approximately equal in terms of the ratio of dropped false reports. Therefore,
our proposed method maintains the same security levels as BECAN.
5. CONCLUSIONS AND FUTURE WORKS
In WSNs, injecting false data attacks are easily generated in compromised nodes because sensor
nodes are vulnerable to diverse attacks. These attacks not only cause unnecessary energy
consumption in the en-route nodes, but they also generate false alarms at the base station. To
detect the attack, BECAN was proposed to achieve both high filtering probability and high
reliability. In BECAN, the sensors may consume needless energy resources in the en-route nodes
because of the fixed number of forwarding reports. In this paper, our proposed method effectively
selects a dynamic number of forwarding reports in the source nodes through an evaluation
function. The experimental results demonstrated improved energy savings while maintaining the
same security levels as compared to BECAN. Therefore, the proposed method improves the
energy efficiency up to 10% with the same security level against attacks. In the future, we
propose to optimize the evaluation function in the proposed method to increase the lifetime of the
sensor network.
ACKNOWLEDGEMENTS
This research was supported by the Basic Science Research Program through the National
Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and
Technology (No. 2013R1A2A2A01013971).
REFERENCES
[1] I. F. Akyildiz, W. Su, Y. Sankarasubramaniam and E. Cayirci, "A survey on sensor networks,"
Communications Magazine, IEEE, vol. 40, pp. 102-114, Aug., 2002.
[2] K. Akkaya and M. Younis, "A survey on routing protocols for wireless sensor networks," Ad Hoc
Networks, vol. 3, pp. 325-349, 2005.
[3] H. Y. Lee and T. H. Cho, "A Scheme for Adaptively Countering Application Layer Security Attacks
in Wireless Sensor Networks," IEICE Transactions on Communications, vol. E93.B, pp. 1881-1889,
2010.
[4] A. Ferreira, M. VilaÃ§a, L. Oliveira, E. Habib, H. Wong and A. Loureiro, "On the Security of
Cluster-Based Communication Protocols for Wireless Sensor Networks," vol. 3420, pp. 449-458,
2005.
[5] Rongxing Lu, Xiaodong Lin, Haojin Zhu, Xiaohui Liang and Xuemin Shen, "BECAN: A Bandwidth-
Efficient Cooperative Authentication Scheme for Filtering Injected False Data in Wireless Sensor
Networks," Parallel and Distributed Systems, IEEE Transactions On, vol. 23, pp. 32-43, 2012.
[6] F. Ye, H. Luo, S. Lu and L. Zhang, "Statistical en-route filtering of injected false data in sensor
networks," Selected Areas in Communications, IEEE Journal On, vol. 23, pp. 839-850, 2005.
[7] F. Li, A. Srinivasan and J. Wu, "PVFS: A Probabilistic Voting-based Filtering Scheme in Wireless
Sensor Networks," International Journal of Security and Network, vol. 3, pp. 173-182, 2008.
[8] H. Chan and A. Perrig, "Security and privacy in sensor networks," Computer, vol. 36, pp. 103-105,
2003.
[9] A. Perrig, J. Stankovic and D. Wagner, "Security in wireless sensor network,"Communication of the
ACM, vol. 47, pp. 53-57, 2004, 2004.
[10] A. Liu and P. Ning, "TinyECC: A configurable library for elliptic curve cryptography in wireless
sensor networks," in Proceedings of the 7th International Conference on Information Processing in
Sensor Networks, 2008, pp. 245-256.

10
[11] Crossbow, MICAz.
http://bullseye.xbow.com:81/Products/Product_pdf_files/Wireless_pdf/MICA2_Datasheet.pdf.
[12] J. Yu, Y. Qi, G. Wang and X. Gu, "A cluster-based routing protocol for wireless sensor networks with
nonuniform node distribution," AEU - International Journal of Electronics and Communications, vol.
66, pp. 54-61, 1, 2012.
Authors
Su Man Nam received his B.S. degrees in computer information from Hanseo
University, Korea, in February 2009 and M.S degrees in in Electrical and Computer
Engineering from Sungkyunkwan University in 2013, respectively. He is currently a
doctoral student in the College of Information and Communication Engineering at
Sungkyunkwan University, Korea. His research interests include wireless sensor
network, security in wireless sensor networks, and modelling & simulation.
Tae Ho Cho received the Ph.D. degree in Electrical and Computer Engineering from the
University of Arizona, USA, in 1993, and the B.S. and M.S. degrees in Electrical
Engineering from Sungkyunkwan University, Republic of Korea, and the University of
Alabama, USA, respectively. He is currently a Professor in the College of Information and
Communication Engineering, Sungkyunkwan University, Korea. His research interests
are in the areas of wireless sensor network, intelligent systems, modeling & simulation, and enterprise
resource planning.

Selecting number of forwarding reports

More Related Content

What's hot (20)

Similar to Selecting number of forwarding reports (20)

More from ijasa (20)

Recently uploaded (20)

Selecting number of forwarding reports