DEFENCE AGAINST INTRUDER IN COGNITIVE RADIO NETWORK OMNET BASED APPROACH. J. Avila, V.Padmapriya, Thenmozhi.K

Transcription

1 Volume 119 No , ISSN: (on-line version) url: DEFENCE AGAINST INTRUDER IN COGNITIVE RADIO NETWORK OMNET BASED APPROACH J. Avila, V.Padmapriya, Thenmozhi.K Department of ECE, School of EEE, SASTRA Deemed University, Thanjavur *Corresponding author: Abstract: Wireless Communication has evolved to fulfill the all-time needs of the current generation. Its widespread usage in almost all the areas has eventually led to spectral inadequacy. Cognitive Radio (CR) has emerged as a successful solution to over the spectral utilization issues. Like any other wireless technique CR also is often subjected to attacks by selfish users. In this manuscript, the security threats in the physical layer of the cognitive radio namely the Primary User Emulation Attack (PUEA) is taken into consideration and the solution to overcome the same has been proposed. Keywords: Cognitive Radio, OFDM, PUEA, Helper Node, OMNeT++, NED, Authentication 1. Introduction Communication plays an imperative role in today s scenario. From the antiquated telegraphy to the modern day cellphones, the evolution of this field has been progressing at a swift pace. One such innovation is the concept of Cognitive Radio. Optimum spectrum utilization is achieved by using the spectrum left by the licensed primary users for unlicensed secondary users whenever it is not in use. Thereby the spectrum management is done effectively. Any wireless system is vulnerable to attacks and the same is applicable to cognitive radio networks. These attacks affect the dynamic spectrum accessing. Hence this paper offer a solution to the threat faced by cognitive and launch a secure connection. Many researchers have proposed various methods to overcome this attack. Some of them are, the aforementioned method makes use of the distance ratio test and the distance difference test in [2]. An algorithm for fast spectrum sensing is given in [1] which deals with the discrete wavelet packet transform based energy detector. In [3], a Primary user emulation attack has been analyzed using dedicated sensor networks. A transmitter location based scheme has been proposed in [4]. Fenton s approximation and markov s inequality based Primary user emulation attack have been performed in [5]. Four dimension continuous Markov chain model to combat PUEA has been proposed in [6]. PU, secondary user, selfish misbehavior secondary user and misbehavior secondary user are the four parameters considered to combat PUEA. Eigen value based PUEA mitigating method has been discussed in [7]. Time Synched link signature scheme to mitigate PUEA has been proposed in [8]. In [9] temporal link signature scheme has been offered to establish link between transmitter and receiver thereby PUEA could be mitigated. Any change in the transmitter location or emulator claiming as transmitter is identified. 513

2 In this paper Helper node based authentication scheme has been proposed. The authentication is achieved with the aid of authentication tag generated by the Helper node. Only the information with the authentication tag is accepted by the cognitive radio [10] and other information are rejected. 2. Materials and Methods The various windows used in this study are outlined here. The signals are chosen in such a way that the spectral leakage occurs and does not occur respectively. Finally, the spectral and SNR analysis is performed with various windows and the results are analyzed for a relative study in the next section. 2.1 Helper Node The tag is generated by the Helper node. Helper node is nothing but the node very close to the primary user and it is trust-worthy. The CR requests the helper node to sense the existence of primary user. On accepting the request the Helper node in turn looks for the availability of the PU. In the absence of the PU it transmits that information to the CR. Absence of PU condition is preferred in order to avoid interference between PU and Helper node [7]. OFDM is chosen as the system model for the Helper node and it is shown in Figure 1. The tag generated is embedded in the parity bits of the error control codes. Error control codes are the extra bits added to the message bit to combat the channel noise. The parity bits which are meant to combat the noise are XORed with tag bits INPUT CONVOLUTION FOURIER TRANSFORM BPSK/QPSK/QAM Figure 1. Transmitter diagram of helper node Spectrum Sensing There are many spectrum sensing schemes available to detect the existence of primary user. In this paper energy detection based spectrum sensing has been considered to find the spectrum holes. The block diagram of energy detector is as shown in Figure 2. The primary user signal is received, filtered and moved on to Fast Fourier Transform (FFT) block. The output of FFT is fed to windowing function block. This is done in order to reduce the side lobs. The out of this block is integrated and the energy of the primary user signal is compared with the predefined threshold. If the signal exceeds the threshold then it is decided that the primary user is present and the spectrum is not vacant. If the signal falls below the threshold then it is concluded that the spectrum is free. The decision Hypothesis is S(t) = u(t) : 514

3 u(t) + n(t) : S(t) is the signal received by the cognitive radio. is null hypothesis which means that only noise is present in the channel. indicates the existence of primary user and noise in the channel. 2.2 Tag Generation Figure 1. Energy detection based spectrum sensing There are many ways to generate the tag. Pseudo Noise sequences, Chaotic sequences etc., can be utilized as authentication tag. Algorithms like AES, DES can also be used to generate the authentication tag. But it is not advisable to utilize them because they are meant to provide security for bulk data transfer. In this case it is only the PU absent information that has to be transmitted from the Helper node to the cognitive radio. Hence any light weight algorithm can be used to generate the authentication tag. 2.3 OMNET Tool The proposed method is explained with the aid of simulated results using OMNET++ tool. OMNET++ is a modular/component based simulation software. In OMNET++ the modules are programmed using C++ and these modules are assembled by the compiler using a high level language called NED. It is the use of this NED language that makes OMNET++ a user friendly software. The user can simply program the modules in a known language (C++) and the compiler assembles them. 3. Simulation Results and Discussion In the graphical view there is one primary transmitter, one primary receiver and three cognitive users. In Figure 3a the cognitive users are sensing the availability of the free holes by receiving the signal from the primary user (PU). Figure 3b shows that PU signal is received by all cognitive users and it is compared with the threshold. Since the received signal exceeds the threshold it is declared that the PU is present. 515

4 Figure 3a and 3b. Cognitive Radio sensing the Primary User Figure 4a shows the inclusion of malicious user in the wireless environment. Malicious users are selfish users who wish to access the free spectrum on their own and block other cognitive users from accessing the free spectrum. In the absence of the primary user they block the cognitive radio by transmitting a duplicate signal imitating the primary user signal. The cognitive radio on receiving the duplicate signal compares it with the threshold. It exceeds the threshold and the cognitive radio makes a decision that the primary user is present. Actually the primary user is absent and the spectrum is free. Figure 4b shows that the cognitive user 3 is checking the availability of the free spectrum. Figure 4a and 4b. Inclusion of Malicious User in the Cognitive Radio environment Figure 5a shows that the malicious user collects the information about the primary user and in the absence of primary user, mimics it. It conveys the message to the cognitive user 3 that the spectrum is not free which is given in figure 5b. Figure 5a and 5b. Malicious User mimicking the Primary User 516

5 Figure 6a shows the insertion of the Helper node amid the PU and the cognitive radio to overcome the primary user emulation attack. The cognitive requests the Helper node to find the availability of the free spectrum. If the spectrum is free it transmits the information to the cognitive users. Figure 6b shows that CR3 is requesting the Helper node. The Helper node senses that the spectrum is free and sends the information along with the authentication tag (M in this case) to the CR3. Figure 6a and 6b. Helper Node to combat the Primary User Emulation Attack Figure 7 shows that the authentication tag is received by CR3, malicious user and the primary receiver. The malicious user and the primary receiver cannot decode the message without the knowledge of the key. Hence privacy is maintained Figure 7. Data Transfer with Authentication Tag 4. Conclusion The security threats in a cognitive network have been addressed by considering the authentication tag generated and transmitted to the cognitive radio from the Helper node. This scheme is resistance to PUEA which is one the important in the physical layer of the cognitive radio. Eliminating the attacks enables the cognitive radio to make the correct decision about the availability of the free spectrum and occupy it. Hence dynamic spectrum access has been efficiently achieved. References [1] Youngwoo Youn, Hyoungsuk Jeon, Hoiyoon Jung, Hyuckjae Lee. Discrete Wavelet Packet Transform based Energy Detector for Cognitive Radios. IEEE 65th vehicular TecHelper Nodeology conference 2007; Doi: /VETECS

6 [2] Ruiliang Chen,Jung-Min Park, Reed.J.H. Defense against Primary User Emulation Attacks in Cognitive Radio Networks. IEEE Selected Areas in Communications 2008; 26(1):25-37.Doi: /JSAC [3] Jin Z, Anand, S,Subbalakshmi, A.Detecting Primary User Emulation Attacks in Dynamic Spectrum Access Networks. IEEE International Conference on communications 2009;1-5.Doi: /ICC [4] Lu KaiWang, Ke HaiZhou,Yang Jie, Zhang LiangJun. Research of PUE attack based on location.ieee 11th International Conference on Signal Processing (ICSP) 2012; Doi: /ICoSP [5] Anand,S, Jin,Z, K.P.Subbalakshmi.An Analytical Model For Priamry user emulation Attack in cognitive radio networks.ieee symposium on New Frontiers in Dynamic spectrum Access Networks 2008;1-6.Doi: /DYSPAN [6]. WANG Shan-shan, LUO Xing-guo and LI Bai-nan, Primary User Emulation Attacks Analysis for Cognitive Radio Networks Communication, TELKOMNIKA, Vol. 11, No. 7, pp ,2013. [7].Deepa Das and Sumita Das, Eigenvalue Detection Based Method to Mitigate PUEA in Cognitive Radio Networks, IEEE International Conference on Advanced Networks and Telecommunication Systems,pp.1-6, [8].Yao Liu and Peng Ning, Enhanced Wireless Channel Authentication Using Time-Synched Link Signature, Proceedings IEEE INFOCOM,pp ,2012. [9].Neal Patwari and Sneha K. Kasera, Robust Location Distinction using Temporal Link Signatures, IEEE Transactions on Mobile Computing,Vol.10,No.3,pp ,2010. [10].V.P.Charulatha, J.GeethaRamani, Dr.K.Geetha, SPECTRAL PRECODING TECHNIQUES FOR COGNITIVE RADIO SYSTEMS TO IMPROVE International Journal of Innovations in Scientific and Engineering Research (IJISER),vol1no4,pp ,2014 SPECTRUMUTILIZATION [11] Fatty.M.Salem, Maged.H.Ibrahim, Ihab.A.Ali, I.I.Ibrahi.Matched-Filter-based Spectrum Sensing for Secure Cognitive Radio Network Communications. International Journal of Computer Applications 2014;87(18):41-46.Doi /

7 519

8 520