Various Layers Of Wsn Stack Computer Science Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Wireless sensor network build with large number of small sized, low- cost and computable sensors, which are having limited battery power, memory and computation power. The large numbers of small sensors are deployed to monitor any physical phenomena (i.e. temperature), to collect and process the sensed data and to send the data back to the base station. Some applications of WSN are environmental monitoring, personal healthcare, enemy monitoring, etc. But there are some issues in WSNs i.e. battery backup, security etc. Security is a biggest concern in Wireless Sensor Networks (WSNs) especially those are deployed for military applications and monitoring. They are prone to various attacks and some of them (i.e. passive attacks) are very hazardous as they are difficult to detect and defend. In this paper we have discussed the various issues and challenges in WSN followed by different attacks happen at different layer of wireless sensor network (WSN) stack with available solutions. In last we have done the case study of misdirection attack, with its impact measurement on the performance of WSN.

I. Introduction

Wireless sensor nodes are low power electronic devices, deployed in remote areas, where power resources are limited. The demand of wireless sensor networks (WSN) has extended many real world applications such as environment monitoring, military applications and monitoring etc. There are some issues in WSN like limited battery power, security etc. Sometimes confidential information is being exchanged through insecure medium of WSN. The confidential information can be leaked or altered because many attacks are possible. Therefore, securing the links is important in designing a sensor network. Here we have discussed various issues and challenges in WSN. Then we have explained the various attacks happen at different layer of wireless sensor network with available solutions. A complete case study of misdirection attack is also done in last with its impact measurement on the performance of WSN. Rest of the paper is organized as: section II contains literature survey having the brief discussion of related work done by various authors. The problem definition and novelty of proposed idea is discussed in section III. Various issues and challenges in WSN are discussed in section IV. Applications of WSN are discussed in section V. Different attacks corresponding to different WSN layers are discussed in section VI. Section VII contains the case study of misdirection attack followed by the conclusion in section IX.

II. literature survey

In paper [1] a topological analysis of WSN in the presence of misdirection attack is done. Authors have proposed an algorithm for the prediction of delay and throughput under the influence of misdirection attack. In paper [2] a comprehensive security model is presented for tailoring the needs of sensor networks. The authors outline the security properties that must be considered when designing a secure sensor networks. The various challenges for sensor networks are also discussed. In paper [3] various types of attacks and countermeasures related to trust schemes in WSNs are categorized. The authors present the development of trust mechanisms along with short summarization of classical trust methodologies emphasizing the challenges of trust scheme in WSNs. In paper [4] a novel approach is proposed for detecting Denial of Service (DoS) attacks in cluster-based sensor networks. This method is based on the election of controller nodes called cNodes which observe and report DoS attack activities. The role of a cNode is to analyze traffic and to send back a warning to the cluster head if any abnormal traffic is detected. In paper [5] an approach for detecting physical layer DoS jamming attack is proposed and analyzed. Authors have proposed a method called physical layer jamming identification based on residual energy where few nodes are taken as monitor nodes. They monitor the jamming attack by checking the receiver signal strength indicator and packet delivery ratio. The system performance improves in the presence of proposed method. In paper [6] an efficient technique that uses multiple base stations deployed in the network to counter the impact of black holes on data transmission is proposed in this paper. The simulation results demonstrate that this technique can achieve more than 99% packet delivery success and can identify 100% of the black hole nodes suffering from very little false positives. In Paper [7] some amendments to the Lightweight Medium Access Control (LMAC) protocol are proposed i.e. data packet separation slot size randomization (DSSSR), round robin (RR) slot size assignment. It also shows that employing RR eliminates the negative impact on the sensor network throughput. In paper [8] a survey on recent advances in WSN research area. It summarizes the special features of sensor data collection in WSNs, by comparing with both wired sensor data collection network and other WSN applications. The issues and prior solutions on the utilizations of WSNs for sensor data collection are also given. In paper [9] authors in lighting the security aspects of sensor network. Wireless sensor network is one of the main future technologies but it has posed many challenges to researchers. Authors have discussed that different set of challenges exist in sensor network. In paper [10] authors discussed various challenges in WSN and then proposed an integrated security mechanism. It will provide security to all services of WSN. In paper [11] a novel power-efficient data fusion assurance scheme has been proposed. It uses silent negative voting mechanism. It is compared with the direct voting based fusion assurance scheme. The proposed scheme shows very good results with better power efficiency and lower network overhead. Sensor security and challenges such as wireless medium, ad hoc network deployment etc are also discussed. In paper [12] it is discussed that the security architectures used in wireless networks are not feasible due to limited resources and wireless nature. Some security threats and challenges faced by WSNs are also explained here. This paper [13] reviews the design and implementation of a novel defense strategy designed to work alongside existing DoS counter measures. The previous approaches were generic and were not capable of filtering out all attack traffic, instead a small amount of attack traffic reached the attackers intended victim. This small level of attack traffic posed a significant threat to the limited resources of WSN. Paper [14] discusses various attacks in WSN with available security mechanism along with various challenges faced. In paper [15] a survey of denial-of-service threats and counter measures is done considering wireless sensor platform resource constraints as well as the denial-of-sleep attack, which targets a battery-powered device's energy supply. The survey of denial-of-service threats is updated with current threats and countermeasures. In paper [16] the problem for some security trends over wireless sensor networks (WSNs) is investigated. A survey of recent trends in general security requirements, key distribution schemes and target localization is presented. In order to facilitate applications that require packet delivery from one or more senders to multiple receivers, provisioning security in group communications is pointed out as a critical and challenging issue. In paper [17] various anomaly detection techniques for wireless sensor networks are discussed. Some of the open issues for research related to WSN are also added. In paper [18] the security related issues and challenges in wireless sensor networks are investigated. They concluded that most of the attacks against security in wireless sensor networks are caused by the insertion of false information by the compromised nodes within the network. For defending the inclusion of false reports by compromised nodes, a means is required for detecting false reports. The development of such a detection mechanism and making it efficient is a great research challenge. In paper [19] the authors have discussed threat models and security issues faced by WSN. On the basis of the survey they motivate the need of a security framework to provide counter measures against attacks in wireless sensor networks. In paper [20] a novel algorithm for detecting a sinkhole attacker is proposed. The proposed algorithm first finds a list of suspected nodes, and then identifies the attackers in the list through a network flow graph method.


The various issues, challenges and application of WSN are discussed. The different attacks happen at various layers of WSN stack with their available solution are discussed in detail. All these aspects of WSN are placed in a single paper and then we have done the complete study of misdirection attack along with its impact measurement on the performance of sensor network.


The ad hoc nature presents significant challenges in the deployment and maintenance of wireless sensor network (WSN). A WSN has many constraint compared to other traditional networks. Some of them as discussed in [14] are:


Wireless medium used in WSN is inherently less secure. In WSN eavesdropping is very simple because of its broadcast nature. Any transmitted data can easily be intercepted, altered, or replayed. The wireless medium allows an adversary to easily interpose valid data packets and easily imbue malicious packets.

B. Ad-Hoc Deployment

Due to the ad hoc nature wireless sensor network does not have defined structure. Due to node failure and high mobility topology changes frequently in WSN. Node deployment can be random because of dropping of node from the air.

C. Hostile Environment

Hostile environment is also a challenging factor in the functioning of sensor nodes. Sensor node can be destructed captured by enemies (attacker). Attackers can easily gain access to a sensor node due to their deployment in hostile environment. Attackers can extract confidential information (e.g. cryptographic keys) from a sensor node, highly hostile environment act as a severe problem for security experts.

D. Inadequate Resources

Each sensor node has limited resources creates a serious challenge to deployed security mechanisms require more battery power and computation power. The implemented security algorithms require more bandwidth, computational power and memory.


Environmental monitoring

Habitat monitoring is an important tool for assessing the threat and conservation status of species and protected areas. This can be used to get the information about the breeding pattern of birds where human cannot go because it can disturb them. This can be done at global and regional scales, where data are available.

Military surveillance

In the battle field some sensor nodes are doing surveillance, monitoring and guiding systems of intelligent missiles. The detection of weapons of mass destruction is also performed.

Smart home/office

Sensors controlling appliances and electrical devices in the house are very popular these days. Better lighting and heating in office buildings.

Radiation Monitoring

Sensor helps authorities and security forces to measure the level of radiation of the affected zones without compromising the life of the workers.

Medical monitoring

Monitoring of physiological data such as cancer detection, glucose, heart rate, diagnosis are performed. Sensor can be extremely useful for medical field.

Traffic Monitoring

Sensors are used for the calculation of average speed of a vehicle which transits over a roadway by considering the time mark at two different points.


The following are the security requirements of a wireless sensor network [5].

A. Data Confidentiality

It is an act to hide messages from an attacker so that message remains confidential. It is the most important aspect of sensor network security. A sensor should not disclose its data.

B. Data Authentication

It assures the reliability of the message by identifying its generator. It is an act verifies the identity of the sender and receiver. It is achieved by symmetric/ asymmetric mechanisms where sender and receiver share secret keys. Due to wireless medium and highly changing topology, it is very difficult to achieve authentication in WSN.

C. Data Integrity

It is an ability to confirm that a message has not been altered. Even a network has achieved confidentiality but there is still a possibility that the integrity of a message has been compromised.

D. Data Availability

Availability signifies that a node is able to access the network resources. If base station does not available to sensor nodes then this can threaten the entire network moreover the performance is also degraded.

E. Data Freshness

It is very important to assure that the sensed data is fresh enough to make correct assumption about any physical phenomena. Freshness signifies that the data is recent and no old messages have been received. To ensure data freshness a time related counter, can be added into the packet.

F. Organization of Network

For network management purpose there is no fixed infrastructure available to WSN, feature further brings a great challenge to WSN security. The absence of self organization may results a damage done by an attacker.


There are five different layers in the layered architecture of wireless sensor network i.e. physical layer, data link layer, network layer, transport layer and application layer. The following attacks are identified at the different layers of WSN stack:

A. Attacks at Physical Layer

Following attacks are identified at physical layer:

1) Jamming: It is Denial of Service (DoS) attack in nature in which an attacker broadcasts a high-energy signal to disrupt the functionality of the network. We use spread spectrum techniques to defend this attack. To handle jamming over the MAC layer Admission Control Mechanisms is used. Network layer can also deals with it, by doing mapping of jammed area in the network.

2) Tampering or destruction: If an attacker accesses a sensor node successfully then he/she can extract sensitive data such as cryptographic keys or other private data from the node. Tamper-proofing the node's physical package is a solution for this problem. If somebody accesses the sensor nodes physically then nodes vaporize their memory contents to prevent any leakage of information. Some Fault Tolerant Protocols are also used to prevent these attacks.

3) Radio interference: For this attack an attacker produces large amounts of interference regularly. Some symmetric key algorithms are used to prevent the attack in which the disclosure of the keys is delayed by some time period.

B. Attacks at Data Link Layer

Following attacks are identified at data link layer:

1) Continuous Channel Access (Exhaustion): An attacker node can disrupt the MAC protocol by continuously requesting over the channel. This leads to starvation for other nodes because they are not able to access the channel during this time.

2) Collision: Collision happens when two sensor nodes attempt to transmit data packets simultaneously on the same frequency. Packets meet and collide and a change will occur in the data portion causes a mismatch in checksum at the receiving end. These packets will then be rejected as they are invalid. We use error-correcting codes to prevent such kind of attacks.

3) Interrogation: When happens exploits the two-way request to send/ clear (RTS/CTS) handshake uses by many MAC protocols to resolve the hidden node problem. A node's resources can be exhausted by repeatedly sending RTS requests to bring CTS responses from a targeted node. For protection purpose a node can limit itself in accepting connections from same identity. Anti replay protection and strong link layer authentication mechanisms can also be used.

5) Sybil Attack: Sybil Attack at data link layer is related to data aggregation. A single malicious node is act as different Sybil Nodes sends many negative reinforcements to make the aggregate message a false one.

C. Attacks at Network Layer

1) Sinkhole: In this attack a sinkhole attacker node tries to lure almost all the traffic toward a compromised node by making a metaphorical sinkhole in the network in which attacker is at the center. Geo-routing protocols are designed to protect against sinkhole attack. A topology is constructed using only localized information and traffic is naturally routed through the physical location of the sink node making it difficult to create a sinkhole.

2) Hello Flood: It uses hello packets that are required in many protocols to find out the neighbors in a network. If a node receives such packets may assume that it is in communication range of that sender node. An attacker can send this kind of packet to all sensor nodes in the network so that they make an assumption that compromised node belongs to their neighbors. So other nodes send packets to this imaginary neighbor node. Authentication is an available solution to such attacks. This can be protected easily by verifying bi directionality of a link.

3) Selective Forwarding/ Black Hole Attack (Neglect and Greed): Multi-hopping is used in sensor network to provide the connectivity inside the network, but the malicious/ compromised node can reside inside the WSN at any place. Due to multi-hopping nature of WSN, the malicious node can drop the packet without forwarding it to the intended recipient. If the malicious node drops all the packets without forwarding it to its neighbor then the resulted attack is known as Blackhole attack. However, if it selectively forwards the packets, then it's pronounced as selective forwarding. To overcome these attacks, random selection of paths in combination with multipath routing can be used, or braided paths ( the paths which have no common link or which do not have two consecutive common nodes) can be used, or use implicit acknowledgments techniques, which ensure that packets are forwarded in a sequence as they were sent.

5) Sybil Attack: A node inside the WSN can show multiple identities to its neighbors, and hence the neighboring node can believe the disjoint routes through the node with muti-identities, it results in Sybil attack. For preventing this attack a shared key can be used between the base station and the nodes present in WSN.

6) Wormhole Attacks: An adversary node inside the network can supply the packet through a tunnel to a node present in the network far from the sender node, basically the node that tunnel the packet and the node that receive the packet from tunnel are involved in the attack, the nodes that are involved in the attack get the estimation the distance between each other through broadcast through a different channel. To overcome this attack, the traffic is routed towards the base station along a path, which is shortest or use very tight time synchronization among the nodes, which is infeasible in WSN environment.

7) Spoofed, Altered, or Replayed Routing Information: In the routing protocol the routing information are exchanged among the network. An attacker may spoof, relay, or alter routing information in order to disrupt traffic in the network. These disruptions can provide the creation of routing loops, attracting or repelling network traffic from selected nodes, shortening and extending source routes, injecting the fake error messages, creating the partitions of the WSN and increasing end-to-end delay. A technique against spoofing and alteration attacks is to append a message authentication code (MAC) after the message. Efficient encryption and authentication techniques can be used to avoid the spoofing attacks.

8) Acknowledgment Spoofing: Deployed Routing algorithms in sensor networks require acknowledgments to achieve reliability. Attacker spoofs the acknowledgments packets destined to neighbor nodes. The available solution to this attack can be achieved by authentication-via-encryption of all sent packets and also in packet headers.

9) Misdirection: In misdirection attack the attacker routes the packet from its children node to other distant nodes, but not necessarily to its legitimate parent node. This produces long delay in packet delivery and further decreases the throughput of the network [1]. To overcome misdirection attack victim node can be scheduled to go into sleep mode for some time.

10) Internet Smurf Attack: In this attack an attacker floods the victim node's network link. The attacker duplicates the victim's address and broadcasts this in the network. Thus attacker floods the network link of the victim node. If this attack is detected then the victim node can be scheduled to go into sleep mode for some time.

D. Attacks at Transport layer

Following attacks are identified at transport layer:

1) Flooding: In flooding attack an attacker make repeated connection requests until the resources of a node are exhausted. For preventing this attack each connecting client determines its duty to the connection by solving a puzzle. A limit can also be put on the number of connections from a specified sensor node.

2) De-synchronization Attacks: An attacker repeatedly duplicates messages to the end points request transmission of missing frames. Hence, these messages are transmitted again and again. If exact timing is known to the attacker, it can prevent the end points from exchanging any information causing battery power wastage. The possible solution to this attack is requiring authentication of all packets.

E. Attacks at Application layer

Following attacks are identified at application layer:

1) Overwhelm attack: In this attack an adversary attempts to overwhelm sensor network nodes with sensor stimuli. This causes the network to forward large volumes of traffic towards the base station further consumes network bandwidth and drains the battery of the sensor node. We can prevent this attack by tuning sensor nodes correctly so that only the specifically desired stimulus likes vehicular movement, a counter to any movement. Efficient data aggregation and rate limiting algorithms can also reduce its adverse effect.

2) Path-based DOS attack: It is done by implanting fake packets into the network. Thus traffic from the legitimate sensors can suffer the starvation, because all the bandwidth is consumed by the spurious messages. To prevent this attack we use combining packet authentication and anti replay protection.





Physical Layer

Jamming, Tampering, Sybil Attack, Interceptions

Data Link Layer

Collision, Sybil Attack, Spoofing and Altering Routing Attack, Exhaustion, Unfairness, Replay Attack, Traffic Analysis, Monitoring

Network Layer

Internet Smart Attack, Sybil Attack, Black hole Attack, Spoofing and Altering Routing Attack, Wormhole Attack, Selective Forwarding Attack, Hello flood Attack, Neglect and Greed, Homing, Misdirection Attack, Byzantine

Transport Layer

Flooding Attack, Desynchronization

Application Layer

Spoofing and Altering Routing Attack, False Data Injection


In misdirection attack the attacker routes the packet from its children node to other distant nodes, but not necessarily to its legitimate parent node. This produces long delay in packet delivery and further decreases the throughput of the network [1].

A. Misdirection attack can be performed as

In the presence of misdirection attack packets reach to the destination but not from the original route, from a different route which further produces long delay.

Figure 1. Normal flow of Packets

Figure 1 shows the simulation scenario for normal flow of packets. S1, S2,----- S12 are sensor nodes, sensing any physical phenomenon and send sensed data packets to router R1,R2 and R3. Router R1, R2 and R3 further send this data to the base station (Co).

Figure 2. Flow Packets when R1 becomes Attacker

Figure 2 shows the simulation scenario for flow of packets when router R1 becomes misdirection attacker. Packets are misdirected towards R3 by the malicious node R1. The traffic coming from R3 has the packets of R3 and R1. So packets of R1 are reached to the base station (Co) with some delay. Thus traffic received (bps) at base also reduces.

B. Simulation Design and Results

We have done the simulation of misdirection attack, the both scenario are as shown in figure 1 and 2.

1) Simulation Parameters: We have taken the following simulation parameters:






500x500 met (Fix)

Network Size

Normal Flow

12 Sensor Nodes

03 Routers with normal flow

01 Coordinator

Attacking Scenario

12 Sensor Nodes

02 Routers with normal flow

01 Router (R1: misdirection node)

01 Coordinator



Simulation Time

60 Minutes

Packet Inter- Arrival Time (sec)

Constant (1)

Packet Size (bits)

Constant (1024)

CSMA/CA Parameters


Sensing duration (sec)


2) Results: During the simulation we have compute the effect of misdirection attack on the performance of WSN. The following results are obtained:




Normal Flow


Traffic Received (bps)



Figure 3. Traffic Received (bps) at base station (Co)

Figure 3 shows the traffic received (bps) at base station under normal flow and misdirection attack.


We have discussed the various issues and challenges in WSN. The different attacks corresponding to different layers of WSN stack with available solutions are also discussed. In last we have done the case study of misdirection attack, with its impact measurement the performance of WSN. During the simulations we have observed that the traffic received (bps) at base station reduces to 990.44 bps is a drastic decrement which further degrades the performance of network.