Association Between Dydog And Coordinator Node

Published Date: 02 Nov 2017

Abstract- The research of Intrusion Detection Systems (IDS) is a mature area wired networks, and has also attracted many attentions in wireless ad hoc networks recently [7] [8] . Nevertheless, there is no previous work reported in the literature about IDS architectures in wireless sensor networks. In this work, we discuss the general guidelines for applying IDS to wireless sensor networks, and introduce a novel technique called Dynamic watchdogs (Dydogs) to optimally watch over the communications of the sensor’s neighbourhood on certain scenarios and detects and prevents the attacks in the network by rerouting the packets safely in another path.

Index terms: IDS, Dynamic Watchdogs, Mudog, Beacon Frame, Rerouting of data

I. INTRODUCTION

Wireless Sensor Networks (WSNs) attract more and more researchers and industrialists because of their potential reliability, accuracy, flexibility, cheapness and easy deployment. In addition, the WSNs application is wide: natural environment monitoring (fire detection, pollution, earthquake, etc.), ecosystem tracking, healthcare, security (video surveillance, objects tracking, etc.) and military (battlefield monitoring, objects localization, etc.). One of the main constraints of these networks is energy limitation due to their small size and wire independence. This constraint must be taken into account in any protocol design and sensor network deployment. The energy limitation creates vulnerabilities that are exploited by attackers. There are two kinds of attacks: passive (like traffic analysis and selfish behaviour) and active (like false routing information injection and impersonation). Propose a solution to counter both passive and active attacks is a real challenge.

In this work, we focus on both passive and active attacks and propose a new analytical model in order to monitor and detect selfish behaviour particularly non-cooperative nodes. The existing solution called Watchdog and Mudog mechanism is proposed for Mobile Ad-hoc Networks (MANETs), but is not adapted to WSNs [3]. Our proposed solution called dyDog is a monitoring mechanism based on MAC IEEE802.15.4 beacon-enabled model and aims at improving the monitoring quality while minimizing the energy cost. The monitoring mechanism is a set of actions allowing specific sensor nodes to monitor the behavior of the other sensor nodes. This mechanism allows to evaluate monitored nodes and update the trust metrics. For example, dyDog mechanism is able to detect the origin of packets loss at the routing nodes which do not cooperate and choose a selfish behavior. This malicious behavior consists in keeping their energy only for their own packets transmission in order to reduce the energy they consume when cooperating. The MAC IEEE802.15.4 beacon-enabled mode standard [1] is used to reduce the energy consumption in WSNs through a sleep/wakeup mechanism [6]. As far as we know, there is no monitoring mechanism adapted to this standard. Therefore, the main goal of this work is to propose a new efficient and optimal analytical model allowing to monitor the network while minimizing the energy consumption. Moreover, the cost of the monitoring mechanism is analyzed in terms of energy consumption and the quality of detection is evaluated by using as metric the probability of monitor correct observation. The impact of nodes density, packet size, network traffic load: saturated/unsaturated cases and distance between monitor and monitored nodes are taken into account in our evaluation. The obtained results illustrate that dyDog is more efficient than Watchdog and mudog whatever the parameter is.This paper is organized as follows: in section 2, we briefly present the MAC IEEE802.15.4 beacon-enabled mode standard and the summary of the existing works related to energy aware and monitoring mechanisms. Section 3 is dedicated to the existing model watchdog and muDog mechanism. The fourth section presents the proposed technique. Finally, section 5 concludes the paper and presents our future works.

II. RELATED WORK

In this section, we briefly present MAC IEEE802.15.4 beacon-enabled mode and the existing energy aware mechanisms.

A. MAC IEEE 802.15.4

The MAC IEEE802.15.4 has two working modes: non-beacon-enabled mode and beacon-enabled mode [1]. The non-beacon-enabled mode is based on non-slotted CSMA/CA and there is no time link between back-off period and beacon. In this mode the coordinator node always stays in active idle listening. However, the beacon-enabled mode is based on slotted CSMA/CA and when the beacon starts each node launches its back-off period. The communication between nodes is controlled by the network coordinator which transmits beacons at regular intervals (Beacon Interval) in order to synchronize the sensors. The nodes use the sleep/wakeup mechanism: they have to wake up in order to receive the coordinator’s beacon. The coordinator is in charge of the data routing in the network. The dyDog receive a beacon from the coordinator node and sends acknowledgement. The advantages of this mechanism are the possibility for the coordinator to communicate with all nodes in activity periods and the reduction of energy consumption when the coordinator and nodes are inactive.

B. Energy aware mechanisms

The MAC layer has an important role to reduce the energy consumption. It is divided into four classes [9]: the consumption related to control packets, collision, idle listening and overhearing. Several mechanisms are proposed to tackle the energy consumption problem in WSNs including the duty cycling [8].

III.EXISTING SYSTEM

The monitoring mechanism is defined as the set of actions that are useful to observe the nodes’ behavior. The monitoring mechanism plays a major role in the evaluation of the nodes’ reputation and in the updating of the nodes’ trust level. It deals with some issues that have a negative impact, particularly on the monitoring mechanism, when a collision occurs at the monitor node during the monitoring process. This situation significantly increases the false positive rate. In fact, the presence of non-cooperative nodes can affect the network in a negative way. Many research works were dealing with monitoring mechanisms in IEEE 802.11. Watchdog [3] is a monitoring mechanism based on packets’ forwarding to detect the non-forwarding nodes. It takes into account the routing layer but does not consider the physical and MAC level’s parameters. It consists in listening to the traffic between monitor node’s neighbours and detecting if the monitored nodes forward the packets in routing operations. The monitor node does not check at the routing layer if the monitored node has correctly received the packet. Thus, the ratio of false positives is high. To reduce the ratio of false positives, we proposed in our previous works [2][4] the enhancement of the monitoring mechanism while taking into account the cases of monitor’s misobservation related to monitor/monitored collision. In mudog mechanism there is only one monitoring node available. If that node fails to monitor, the whole network will be lost.Thats why in this paper, we deal with the monitoring mechanism called dyDog in order to detect the non-cooperative nodes in beacon-enabled IEEE 802.15.4..

IV. PROPOSED SYSTEM:

The monitoring mechanism dyDog has as objective the improvement of the monitor’s observation quality with a minimum of energy consumption. The monitoring process in beacon-enabled IEEE 802.15.4 is activated only in CAP (Contention Access Period). That means that the monitor node wakes up to receive the beacon frame and to track the packets transmitted by the monitored node in the CAP duration in order to evaluate the cooperative metric of this node. The main objective of dyDog is to reduce the time of overhearing and then the energy consumption by launching a targeted monitoring.

V. SYSTEM DESIGN

Network Creation

Assumptions on dyDogs i.N thresh < D thresh ii. C present in FP…. iii. D must not be in FP

if above cond satisfy No Yes

D selects coordinator node C

Cannot act as dyDog D

C sends Beacon frame (BF) to D.

If D receives Bf within specified time duration

No Yes

D sends req to C

C repeats the BF

C sends status updates to D

D sends acknowledgement to C

C receives acknowledgement

No Yes

Confirms that the D or the link is not available.

C is activated during data transmission

Selects other node as coordinator node C

C becomes dyDog D

Reroutes the pckts through another path safely.

check and detect attacks in transmission path

Figure 5.1 The global flowchart of dyDog

VI. MODULES DESCRIPTION:

1. Network Creation:

The Wireless sensor Network is established first. The nodes are created based on user needs. Then the topology and movement of nodes is generated. After creating nodes, the link or the connection between the nodes is established. Once the link is established we have to set the queue size and flow of the network. Then the protocol what we going to deploy is set and its agent and services are set.

2. dyDog – The Monitoring Mechanism:

The Monitoring node dyDog is assumed to satisfy the following conditions.

i.Nthresh < Dthresh., the dyDog must have high energy level compared to all other nodes.

ii. D must not be in FP., the dy Dog should not be in a forwarding path.

iii. C must be present in FP., the coordinator node should be present the forwarding path.

The node that satisfies the the above conditions can act as a dyDog.

3. Association between dyDog and coordinator node:

While monitoring the dyDog selects its coordinator with high energy level compared to other nodes in the forwarding path.

Functions of coordinator node:

i.Collects the status information in the forwarding path with two fields, one field is to show that how many packets are there to be sent to the sink node and other field is to show that how many packets are remaining in the source that are to be sent to sink.

ii.Sends beacon frame (BF) to the monitoring node to establish communication with it. The monitoring node dyDog receives that frame and send request to get the status information.

iii.Acts as a intermediate node in the forwarding path and transmits the packets from source to sink.

4. Fault –Tolerant Network:

Once the dyDog D receives the beacon frame (BF) from the coordinator node C, it requests for the status information to C. C sends the data and will be waiting for some specific time duration to get acknowledgement from D. If it does not receive any acknowledgement it will repeat the BF.In this communication, if the dyDog fails to send acknowledgement then the coordinator node will act as Dydog and choose other node as its coordinator node .If the coordinator node does not send BF in a particular interval to D , dyDog will select other coordinator and rejects the current one.The main advantage is that the transmission path can be monitored by more than one monitoring node.Thus the system is fault tolerant and safeguards the wireless sensor networks by providing efficiency and reliability.

5. Detection of attacks and rerouting the packets in safe path:

The misbehaving and non coordinating nodes are monitored and tracked by the dyDogs and the data are sent through another path safely to the sink node.

VII. CONCLUSION:

Here we proposed a dynamic intrusion detection protocol for wireless sensor networks. Each node in the network may have one or more monitoring nodes and identifies the existence of attacks and take appropriate actions accordingly. Thus DIDP ensures high security. The DIDP uses the one hop neighbouring scheme in high data rate conditions. The nodes in the network itself will act as both as forwarding node and monitoring nodes thus there is no need of deploying any special system for intrusion detection mechanism. Thus DIDP reduces the cost substantially to a greater extent. The DIDP is designed in such a way that the consumption of energy is low and to make better utilization of the resources. Thus the proposed system ensures high security, low cost and minimum energy consumption.

VIII. REFERENCES:

[1] Babak Salamat, Todd Jackson, Gregor Wagner, Christian Wimmer, and Michael Franz, "Runtime Defense against Code Injection Attacks Using Replicated Execution", IEEE Transaction on dependable and secure Computing, 2011.

[2] Nam-Uk Kim, Min-Woo Park, Seon-Ho Park, Sung-Min Jung, Jung-Ho Eom and Tai-Myoung Chung, "A study on Effective Hash- based Load Balancing Scheme for Parallel NIDS", ICACT 2011.

[3] Kealan McCusker, Noel E. O’Connor, "Low-Energy Symmetric Key Distribution in Wireless Sensor Networks", IEEE Transactions on Dependable and Secure Computing, 2011.

[4] Ali Modirkhazeni, "Secure Multipath Routing Protocols in Wireless Sensor Networks: A Security Survey Analysis", 2010 Second International Conference on Network Applications, Protocols and Services.

[5] Yun Wang, Xiao dong Wang, Bin Xie," Intrusion Detection in Homogeneous Wireless Sensor Networks", IEEE Transactions on Mobile Computing, June 2008.

[6] Tao Shu, Marwan Krunz, Sisi Liu,"Secure Data Collection in Wireless Sensor Networks Using Randomized Dispersive Routes", IEEE Transactions on mobile Computing, July 2010.

[7] Andrity Stetsko, Vashek Matyas,Masaryk University Brno, Czech Republic "Effectiveness Metrics for Intrusion Detection in Wireless Sensor Networks", Computer Network Defense European Conference On Nov 2009.

[8] T.Claveirole, M.D. de Amorim M.Abdalla and Y.Viniotis," Securing Wireless Sensor Networks Against Aggregator Compromises," IEEE Comm. Magazine, Vol.46,no.4, pp. 134-141,Apr, 2008.

[9] Rodrigo Roman, Jianying Zhou, Javier Lopez, " Applying Intrusion Detection Systems to Wireless Sensor Networks", Institute for Infocomm Research, 2006.

[10] Junbeom Hur, Younho Lee," Trust Management for Resilien Wireless Sensor Networks", LCNS, 2006.

[11] Shi Gaotao, Liao Minghong, " Stochastic Sleeping for Energy- Conserving in Large Wireless Sensor Networks", Journal of Computer Research and Development,2006.

[12] Abderrezak, Rachedi and Hend BakLouti," muDog: Smart Monitoring Mechanism for Wireless Sensor Networks based onIEEE ICC 2011 proceedings.

[13] Rui Tan, Guoliang Xing, Jianping Wang and Benyuan Liu, "Performance Analysis of Real-Time Detection in Fusion Based Sensor Networks", IEEE Transactions on parallel and Distributed Systems, Vol.22, No.9 Sep 2011.

[14] Farzanch Pakzad, Tiran Branch, Islamic Azad University, Tiran, Iran," Intrusion Detection Techniques for Detecting Misbehaving Nodes", Computer and Information Science Vol.4, No.1; Jan 2011.

[15] Michael Krishnan, " Intrusion Detection in Wireless Sensor Networks", Journal of Computer Science and Technology.

[16] Hayoung oh, "A Fault Management and Monitoring Mechanism for Secure Medical Sensor Network, International Journal of Computer Science and Applications, Vol.6, No.3,pp 43-56, 2009.

[17] Ana Paula, Marcelo, bponets," Decentralized Intrusion Detection in Wireless Sensor Networks" Department of Computer Science ,Federal University of Minas Gerais, Brazil.

[18] Issa M.Khalil, " ELMO: Energy Aware Local Monitoring in Sensor Networks", IEEE Transactions on dependable and Secure Computing, Vol.8,No.4, July/Aug 2011.

[19] David Martins and Herve Guyenne, "Wireless Sensor Networks Attacks and Security Mechanisms 2010 13th International Conference on Network-Based Informatioin System.

[20] Xiao Zhenghang, Chen Zhinghang, " A secure Routing Protocol With Intrusion Detection for Clustering Wireless Sensor Networks, 2010 International Forum on Information Technology and Applications.