Trust-Based Mechanism for the Isolation of Sink Hole Attack in Wireless Sensor Networks

doi:10.21203/rs.3.rs-2960722/v1

Download PDF

Research Article

Trust-Based Mechanism for the Isolation of Sink Hole Attack in Wireless Sensor Networks

https://doi.org/10.21203/rs.3.rs-2960722/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

This paper presents a two-stage technique for detecting malicious nodes in wireless sensor networks (WSNs) where a centralized controller is absent. WSNs are dynamic networks that allow nodes to join or leave, making them vulnerable to sinkhole attacks. In this type of attack, an attacker node pretends to be a sink and intercepts data packets. The proposed technique first determines the presence of the attacker node in the network and then identifies the malicious node. The effectiveness of the technique is evaluated using NS2 simulation, considering metrics such as throughput, packet loss, and delay. The results demonstrate that the developed methodology outperforms traditional methods in detecting malicious nodes.

WSN

Sink Hole

Promiscuous mode

Two Step Scheme

Node Localization

Wireless sensor networks (WSNs) have gained widespread adoption in various applications, such as habitat monitoring, environmental pollution tracking, smart grids, and military operations. However, the decentralized nature and limited resources of WSNs make them vulnerable to malicious attacks [1]. Among these attacks, sinkhole attacks pose a significant threat to the network's integrity and performance [2]. In a sinkhole attack, an attacker node pretends to be a sink and manipulates the hop count information to mislead other nodes into routing their data through it [3]. The consequences of a sinkhole attack can be severe. By strategically disrupting the communication paths between sensors and the actual sink, the attacker can impede the flow of data within the network [3]. This disruption not only wastes valuable energy resources but also compromises the overall functionality of the WSN. Nodes within the attack area of the sinkhole, often referred to as a black hole, face restricted communication and suffer from the inability to establish connections with nodes outside this region [4]. The attacker node intercepts and absorbs any data transmitted by the affected nodes, rendering their communication efforts futile. Detecting and mitigating sinkhole attacks is of utmost importance to ensure the reliability and longevity of WSNs. However, the traditional routing mechanism, which relies on hop-by-hop routing and hop count information, falls short in effectively identifying and isolating sinkhole attacks [5]. Consequently, alternative techniques need to be developed to enhance the security of WSNs and safeguard them against such malicious intrusions.

Given that WSNs are independent networks, the sink's role in communication is quite important. Nodes in a WSN frequently use multi-hop routing to relay data to the sink. Therefore, an attack that can block the communication paths of sensors to the sink may significantly impair WSN operation. An attack on a sinkhole seeks to prevent message exchange between sensors and the sink. In WSNs, data is often moved from sensors to sinks using a hop-by-hop routing strategy. Every node in hop-by-hop routing maintains track of the smallest number of sinks from itself and its neighbouring nodes [3]. When data must be routed to the sink, the sensor node chooses the closest node with the fewest hops to transmit the data to. When the sinkhole attack is inactive, each node keeps track of the precise hop count to the sink. The sinkhole node, however, may choose to launch the assault in the following manner. The sinkhole, or attacker node, initially exposes to the sink its false hop count, which is less than its true hop count to the sink. For instance, an attacker node that claims to have 0 hops to the sink claims that it is the sink. By routing their data to the hostile node, other nodes close to the opponent node give the impression that they have a direct line of communication with the sink [4]. By using a method, the malicious node can access the information of the nodes in its attack area and restrict their communication with the sink, causing severe harm. In worst-case scenarios, although the sinkhole is not obliged to initiate any attacks on other nodes, it causes enough damage to the black hole's surrounding region that it begins to absorb all material that comes within its range, even light. Nodes within this zone are inflowing the blackhole zone and cannot connect to nodes outside of this region. As a result, the region included in the sinkhole attack's coverage area is also referred to as a black hole. Any information the affected node sends won't reach its intended destination because the sinkhole has collected it. As a result, it is highly challenging for the attacked nodes to recognise that they have been attacked. Additionally, the sinkhole attack disrupts coverage and connection by undermining the basic routing architecture [5]. The current routing system affects the attack's detection and isolation. As a result, the sinkhole attack is particularly difficult to detect using the traditional routing mechanism-based security approach. The presence of a fake sink, indicated by the enemy node claiming a hop count of 0, introduces a critical challenge in detecting and mitigating sinkhole attacks within a wireless sensor network [6]. As data convergence regions emerge in the system, centred around a genuine sink and a false sink, nodes located closer to the real sink are expected to broadcast data to the legitimate destination [6]. However, the nodes mistakenly send their data to the fake sink node instead. In Fig. 1, the sinkhole attack is illustrated, where the red line represents the boundary separating the attack region from the defence zone. This line connects nodes with the same hop count to both the genuine and counterfeit sinks. Detecting and defending against a sinkhole attack poses significant difficulties for two main reasons [7]. Firstly, it is challenging to determine whether a sinkhole attack has occurred, as nodes in the network continue to route data normally, making it difficult to discern whether the data reaches a legitimate or bogus sink. Secondly, even if the attack is detected based on the attack range, locating the actual sink becomes problematic, hindering the ability to notify the sink about the attack. Similar characteristics are shared with the hypothetical sink that initiates the attack in a sinkhole scenario, such as the adversary's access to the internal node state and the node broadcasting a fictitious hop count value to neighbouring nodes, lower than its actual hop count to the sink [8].

Addressing the severe implications of sinkhole attacks and the challenges associated with their detection and prevention necessitates further research and exploration. Several existing approaches are insufficient in effectively countering sinkhole attacks [9]. While some studies detect the presence of sinkhole attacks, they fail to pinpoint the exact location of the sinkhole. Moreover, many sinkhole prevention methods lack mechanisms to circumvent such attacks, resulting in delayed recovery of the network's functionality. Additionally, without a bypassing mechanism, nodes within the attack radius are unable to notify the sink about the occurrence of the attack, requiring the integration of additional hardware or alternative methods to provide such notifications. To overcome these limitations and improve the resilience of wireless sensor networks against sinkhole attacks, it is crucial to develop innovative techniques and solutions that address the current challenges posed by these attacks.

Karthigadevi, et.al (2019) suggested an innovative decentralized framework that utilized the NDET algorithm to detect and prevent sinkhole assault [10]. Every node deployed this method to maintain the neighbour table for storing the neighbourhood details. Every node was responsible for gathering the details regarding the neighbour. This method assisted in estimating the network density and recognizing the malevolent node in the area. The distribution of information related to recognized malevolent nodes was done to the neighbouring nodes for avoiding the malicious node at subsequent transmissions. The suggested framework was applicable for mitigating the overhead and maximizing the throughput when the best-effort traffic was maximized.

S. Padmanabhan, et.al (2022) investigated RSR (Reliable Self Reconfiguration) technique for removing the malicious sinkhole assault from the network [11]. The primary stage was executed for detecting the malevolent node. The next stage employed RM (reconfiguration mechanism) to correct it without considering resource loss. A C++-based simulator was applied to simulate RM to fix the sinkhole assault. This technique was computed based on diverse components namely PDR (Packet Delivery Ratio) and energy utilization. The experimental outcomes revealed the superiority of the investigated technique over the traditional methods to discover and remove the sinkhole attack.

N. Al-Maslamani, et.al (2020) focused on constructing and implementing the method to detect sinkhole intrusion with the deployment of the SIO model [12]. In this method, a WE (weight estimation) method was integrated with the ABCO framework to improve the accuracy while detecting the sinkhole assault. MATLAB was executed for quantifying the constructed method concerning accuracy, detection time, convergence speed, overhead, and power usage. The simulation results reported the effectiveness and robustness of the constructed method against the sinkhole assault and offered superior precision.

A. A. Jasim, et.al (2019) designed a protocol known as SEEDA to detect sinkhole attacks [13]. This protocol produced a random value and random timestamp via a secret key for making the network more authentic. The sink aimed to determine the false aggregated data after receiving the packets based on the produced key in advance. An SNA algorithm, DFA, FHE, and AC algorithms were deployed for detecting and preventing the attacks. The initial algorithm helped in preventing the assaults from achieving access to the network. The simulation results reported that this approach offered an accuracy of 98.84% for detecting malevolent nodes, an energy usage of 3.04 joules, a maximum delay of up to 0.038 secs, and a resistance time of up to 0.054.

N. D. L, et.al (2019) introduced a novel secure routing protocol based on trust systems for WSNs [14]. First of all, this algorithm was utilized to generate the SNs (sensor nodes) as clusters. After that, a secure path was developed using a TE (trust evaluation) technique for every SN at CH (Cluster Head) to transmit the data from SN to sink. The trust was computed at CH according to social trust and data trust. Diverse parameters namely duration of the network, and MDR (Malicious Detection Rate) employed to evaluate the introduced algorithm in experimentation. The outcomes validated the applicability of the presented algorithm in contrast to the existing methods during the maximization of the malevolent behaviour of the network.

A. K. Sangaiah, et.al (2022) discussed that the routing assaults such as sinkholes resulted in directing the network data to malevolent users and disrupting the network device [15]. Thus, a novel protocol was established based on CL-MLSP with AODV. The data was encrypted and decrypted using AES (Advanced Encryption Standard) algorithm. A clustering technique was deployed based on power, mobility, and distribution for every node to acquire the shorter route. The established protocol was computed in NS2 (Network Simulator 2) concerning the duration of the network, latency, PLL (packet loss), and security. The outcomes revealed that the established protocol mitigated the energy consumption by up to 6.54%, a drop rate of 12.87%, and delay, and maximized the throughput by 8.12%, and security by up to 9.46%.

D. Kumar, et.al (2022) described that the military areas made the deployment of WSN (Wireless Sensor Network) for monitoring the activities of inconsistent sides [16]. The malicious nodes had the potential for joining the system and activating the security intrusions. Then, SNs (sensor nodes) were employed to initialize the process to transmit the information to the malevolent node rather than BS. A new algorithm was suggested for discovering and segregating malicious nodes from the network. The suggested approach was computed on NS2 concerning diverse parameters. The results demonstrated the adaptability of the suggested algorithm in comparison with the traditional methods for detecting the sinkhole attack.

K. E. Nwankwo, et.al (2019) discussed that the sinkhole assault was launched in WSN when the malicious node pretended as the authentic node nearer to the sink for transmitting the data, and modifying, dropping or delaying the data [17]. Thus, a sinkhole detection system called ACO (Ant Colony Optimization) was presented and employed for detecting the sinkhole more effectively concerning packet drop, PDR, energy exchange and throughput in WSN. An analysis was conducted on the presented approach. The outcomes revealed that the presented approach was capable of enhancing the accuracy to detect the sinkhole attack and mitigating FAR (False Alarm Rate) in WSN. B. M. Devaraju, et.al (2018) analyzed that the sensor nodes had susceptibility to failure due to their implementation in open regions, and tampering by intruders [18]. Malicious assaults such as DoS (Denial of Service), Sinkhole attacks, etc. aimed to modify the complex information which led to degraded the efficacy. Therefore, a CLMPI technique was projected for detecting and preventing the malevolent activities in WSN. The results depicted that the projected method was effective to alleviate the processing delay and communicating delay after avoiding the malevolent activities in networks.

Table 1

Comparative Analysis
Author Name	Year	Findings
Karthigadevi, et.al	2019	The suggested model assisted in estimating the network density and recognizing the occurrence of the malicious node in the area. Moreover, this model led to mitigating the overhead to collect the snapshots and routes.
A. A. Jasim, et.al	2019	Designed a protocol known as SEEDA to detect sinkhole attacks.
N. D. L, et.al	2019	Introduced novel secure routing protocols based on a trust system for WSNs. First of all, this algorithm was utilized to generate the SNs (sensor nodes) as clusters.
K. E. Nwankwo, et.al	2019	The primary outcomes as well as further insights were described. The EACO algorithm was deployed further on NS-3.29 in WSN.
N. Al-Maslamani, et.al	2020	The outcomes demonstrated the efficiency and robustness of the projected technique to detect the sinkhole attack at a superior detection accuracy rate.
D. Kumar, et.al	2022	Described the military areas that made the deployment of WSN (Wireless Sensor Network) for monitoring the activities of inconsistent sides.
A. K. Sangaiah, et.al	2022	Discussed that the routing assaults such as sinkholes resulted in directing the network data to the malevolent user and disrupting the network device.
S. Padmanabhan, et.al	2022	Investigated RSR (Reliable Self Reconfiguration) technique for removing the malicious sinkhole assault from the network.

The objective of this research is to address the detection and isolation of sinkhole attacks in wireless sensor networks (WSNs). These attacks aim to divert all network packets through attacker nodes, which disguise themselves as a base station (BS). As a result, the BS misguides the regular nodes, causing them to transmit their information to the attacker node instead of the actual base station. To combat this, the study proposes a two-stage technique for detecting and mitigating these malicious nodes.

In the first stage, the focus is on implementing a method to detect the presence of the sinkhole attack in the network. This involves deploying beacon nodes within the WSN, which flood the network with beacon frames. The available nodes in the network are responsible for responding to these beacon nodes. Based on the responses received, the state of each node is localized. The malevolent node intentionally delays its response, leading to an incorrect localization of its position, thus indicating the presence of an attack in the network.
Moving to the second stage, the emphasis shifts to detecting and isolating the attack from the network. If the localization process identifies an incorrect position for a sensor node, the beacon node floods the network with control messages. Upon receiving the control message, the sensor node transitions into sleep mode. By analyzing the behaviour of the sensor nodes in terms of packet transmission, the malicious node can be identified. Specifically, the node that has a high number of packets and transmits the fewest packets within the network is considered the attacker node.

By employing these two stages, the research aims to effectively detect and isolate sinkhole attacks in WSNs. This contributes to enhancing the security and reliability of the network against such malicious activities.

Proposed Algorithm

1. Execute the network with a restricted number of sensor nodes.

2. Sensor nodes in the network will transmit the ping messages in the network to localize the node.

3. Sensor nodes receiving replies will define the position of sensor nodes.

If (Position node is incorrect)

3.1 Sensor nodes will monitor and maintain a list of the network traffic.

3.2 The sensor node which receives the maximum information and transmits the least packets will be marked as malicious.

4. Else

4.1 Communication continues in the network

5. Repeat Step 3 if the position is incorrect in the network

The purpose of this work is to detect suspicious nodes in the network. The findings are analyzed using a variety of performance indicators, such as energy usage, throughput, and packet loss. Table 2 defines these simulation parameters as follows:

Table 2

Simulation Parameters
Parameter	Value
Number of nodes	100
Area	800*800 meters
Standard	802.11
Queue Type	Priority Queue
Queue Size	50
Range	18-meter

Table 3

Evaluation of energy usage
Time	Node Localization Technique	Proposed Technique
1 second	7 joules	3 joules
3.5 second	8 joules	4 joules
6 second	13 joules	7 joules

Table 4

Throughput Analysis
Time	Node Localization Technique	Proposed Technique
3 second	4 packets	5 packets
4.5 second	15 packets	10 packets
6 second	15 packets	55 packets

Table 5

Packet loss Analysis
Time	Node Localization Technique	Proposed Technique
2 second	2 packets	1 packet
4.5 second	22 packets	3 packets
6 second	5 packets	24 packets

Wireless Sensor Networks (WSNs) consist of a multitude of small-sized sensor nodes that are deployed in an unsupervised open environment. These nodes operate with limited energy resources, which gives rise to challenges such as restricted communication range, limited processing capacity, and constrained storage space at each sensor node. As a consequence, WSNs become susceptible to various security attacks, making the network highly vulnerable. Among these attacks, the sinkhole attack poses a significant threat by targeting the network's efficiency. In order to address this issue, this research work proposes a technique that involves two distinct stages aimed at detecting the presence of malicious nodes in the network. The first stage focuses on determining whether a compromised node exists within the network. This stage employs a specific methodology to identify any abnormal behaviour or compromised nodes. Once the presence of a compromised node is detected, the technique proceeds to the second stage, which is designed to locate the rogue node within the network. By analyzing the network behaviour and employing certain algorithms, the technique is able to isolate and identify the malicious node responsible for the sinkhole attack.

To evaluate the effectiveness of the proposed technique, simulations are conducted using NS2 (Network Simulator 2). Several metrics, including throughput, packet loss, and energy utilization, are considered during the simulation process. The results indicate that the projected technique yields significant improvements, achieving a detection rate of approximately 5–6% higher than traditional methods for identifying malicious nodes. By employing this two-stage technique, the research work aims to enhance the overall security and performance of WSNs by effectively detecting and mitigating sinkhole attacks. This contributes to ensuring the reliability and integrity of the network in the face of malicious activities.

Compliance with Ethical Standards:

This research paper adheres to the highest standards of ethical conduct in accordance with the principles outlined by relevant regulatory bodies and ethical guidelines. The following aspects were carefully considered and addressed throughout the research process:

-Informed Consent: Prior to data collection, informed consent was obtained from all participants, clearly explaining the purpose of the study, procedures involved, potential risks and benefits, confidentiality measures, and their right to withdraw at any time without consequences.

-Confidentiality and Privacy: Measures were implemented to ensure the confidentiality and privacy of participants' personal information. Data were anonymised and securely stored to prevent unauthorized access or disclosure.

-Conflict of Interest: Any potential conflicts of interest that could influence the objectivity or impartiality of the research findings were disclosed. Steps were taken to minimize and manage such conflicts throughout the study.

Competing Interests:

The authors declare that they have no competing interests that could potentially influence the research, data interpretation, or presentation of findings in this research paper.

Research Data Policy and Data Availability Statements

The datasets generated during and/or analysed during the current study are available from the corresponding author upon reasonable request.

A. Rehman, S. Rehman and H. Raheem, Sinkhole Attacks in Wireless Sensor Networks: A Survey (Springer, 2018).
R. Kumar, Sundararajan and U. K. Arumugam, Intrusion Detection Algorithm for Mitigating Sinkhole Attack on LEACH Protocol in Wireless Sensor Networks (Journal of Sensors, 2015).
K. Karthigadevi, S. Balamurali and M. Venkatesulu, Based on Neighbor Density Estimation Technique to Improve the Quality of Service and to Detect and Prevent the Sinkhole Attack in Wireless Sensor Network (IEEE International Conference on Intelligent Techniques in Control, Optimization and Signal Processing (INCOS), Tamilnadu, India, 2019), pp. 1-4.
S. Ali, M. A. Khan, J. Ahmad, A. W. Malik and A. ur Rehman, Detection and prevention of Black Hole Attacks in IOT & WSN (Third International Conference on Fog and Mobile Edge Computing (FMEC), Barcelona, Spain, 2018), pp. 217-226.
J. Li, Z. Yang, X. Yi, T. Hong, X. Wang, A Secure Routing Mechanism for Industrial Wireless Networks Based on SDN (14th International Conference on Mobile Ad-Hoc and Sensor Networks (MSN)).
G. G. Gebremariam, J. Panda and S. Indu, Secure Intrusion Detection System for Hierarchically Distributed Wireless Sensor Networks (International Conference on Industrial Electronics Research and Applications (ICIERA), New Delhi, India, 2021), pp. 1-6.
S. Rizvi, J. Gualdoni and A. Razaque, Securing Wireless Networks from Sinkhole and Sybil Attacks Using Secure Data Aggregation Protocol (17th IEEE International Conference On Trust, Security and Privacy In Computing And Communications/ 12th IEEE International Conference On Big Data Science And Engineering).
T. X. Zheng, X. Chen, C. Wang, K. K. Wong and J. Yuan, Physical Layer Security in Large-Scale Random Multiple Access Wireless Sensor Networks: A Stochastic Geometry Approach, (IEEE Transactions on Communications, Volume 70, Issue 6, 2022), pp. 4038-4051.
M. Kumar, P. Mukherjee, K. Verma, S. Verma and D. B. Rawat, Improved Deep Convolutional Neural Network Based Malicious Node Detection and Energy-Efficient Data Transmission in Wireless Sensor Networks (IEEE Transactions on Network Science and Engineering, Volume 9, Issue 5, 2022), pp. 3272-3281.
L. K. Ramasamy, F. Khan, A. L. Imoize, J. O. Ogbebor, S. Kadry and S. Rho, Blockchain-Based Wireless Sensor Networks for Malicious Node Detection: A Survey (IEEE Access, Volume 9, 2021), pp. 128765-128785.
N. Xie, Y. Chen, Z. Li and D. O. Wu, Lightweight Secure Localization Approach in Wireless Sensor Networks, (IEEE Transactions on Communications, Volume 69, Issue 10, 2021), pp. 6879-6893.
K. Karthigadevi, S. Balamurali and M. Venkatesulu, Based on Neighbor Density Estimation Technique to Improve the Quality of Service and to Detect and Prevent the Sinkhole Attack in Wireless Sensor Network (IEEE International Conference on Intelligent Techniques in Control, Optimization and Signal Processing (INCOS), Tamilnadu, India, 2019), pp. 1-4.
S. Padmanabhan, R. M. and R. Anitha, An experimental study to recognize and mitigate the malevolent attack in wireless sensors networks (Global Transitions Proceedings, 2022).
N. Al-Maslamani and M. Abdallah, Malicious Node Detection in Wireless Sensor Network using Swarm Intelligence Optimization (IEEE International Conference on Informatics, IoT, and Enabling Technologies (ICIoT), Doha, Qatar, 2020), pp. 219-224.
A. A. Jasim, Secure and Energy-Efficient Data Aggregation Method Based on an Access Control Model (IEEE Access, Volume 7, 2019), pp. 164327-164343.
N. D. Labana and V. S. Katoch, Secure and Composite Routing Strategy through Clustering In WSN," 2019 2nd International Conference on Innovations in Electronics (Signal Processing and Communication (IESC), Shillong, India, 2019), pp. 119-123.
A. K. Sangaiah, A. Javadpour and W. Zhang, CL-MLSP: The design of a detection mechanism for sinkhole attacks in smart cities (Microprocessors and Microsystems, Volume 25, Issue 9, 2022), pp. 3416-3427.
D. Kumar and E. N. Kapoor, Novel Scheme for Mutual Authentication to Isolate Sinkhole Attack in Wireless Sensor Networks (International Conference on Engineering and Emerging Technologies (ICEET), Kuala Lumpur, Malaysia, 2022), pp. 1-5.
K. E. Nwankwo and S. M. Abdulhamid, Sinkhole Attack Detection in A Wireless Sensor Networks using Enhanced Ant Colony Optimization to Improve Detection Rate (2nd International Conference of the IEEE Nigeria Computer Chapter (NigeriaComputConf), Zaria, Nigeria, 2019), pp. 1-6.
B. M. Devaraju and G. T. Raju, Cross Layer and Management Plane Integration Approach for Detection and Prevention of Malicious Activities in WSN (International Conference on Electrical, Electronics, Communication, Computer, and Optimization Techniques (ICEECCOT), Msyuru, India, 2018), pp. 1831-1838

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Trust-Based Mechanism for the Isolation of Sink Hole Attack in Wireless Sensor Networks

Status:

Version 1

Abstract

Figures

INTRODUCTION

LITERATURE REVIEW

RESEARCH METHODOLOGY

RESULT AND DISCUSSION

CONCLUSION

Declarations

Research Data Policy and Data Availability Statements

References

Additional Declarations

Status:

Version 1