7COM1084 Specialism Research Report

In order to conduct independent research in computer science you must fully understand the research area. You must have a grasp of the current open questions in this area, as well as any common techniques used to solve problems.

In this assignment you must provide an overview of your MSc Research specialism (AI / Networking / Cyber Security / Software Engineering / Data Science). Students without a specialism may choose any one of these specialisms.

You must discuss the research question presented within the relevant specialist lecture for this specialism, propose some research approaches to investigate this question and identify further work you might undertake which builds on this. You must also explore your personal strengths in this area.




Introduction of research specialism

The networking subject area in computer science is important because of its contribution to installing and maintaining small and large network operations. In recent years, where every business is associated with an international business model, proper networking services and security maintenance are the only solutions to enhance sustainability. Previously, networking was done based on wired connections. However, in recent days, wireless connections, such as cognitive radio connections have created the opportunity to establish the best user experience without network congestion and messy structure (Syed and Safdar, 2015). In this specialism area, spectrum sensing has been a hot topic in research, because it helps to evaluate the frequency bands of networking connections that are coming from licensed and unlicensed users. As a result, it prevents unwanted interference of users and improves network performance. For all these reasons, spectrum sensing techniques in the cognitive radio networks topic was selected as an aspect of future research on the area of networking specialism.

Open research question

From the concerned specialism paper, the following research question has been derived:

What is the energy-efficient and intelligent spectrum sensing technique for improving the efficiency of cognitive radio networks, which can be established as a novel mechanism? (Syed and Safdar, 2015)

Addressing the research problem via the question

Cognitive radio networks are specialised wireless networks that are installed for managing spectrum scarcity in wireless networks. However, Bouabdellah et al. (2018) have stated that these networks are subject to security threats due to the implementation of new functionalities related to TCP/IP protocols of network layers and medium access controls. Such factors will lead to developing safe networking issues for users. Sometimes, spectrum holes are created within these networks, which create false alarms and often miss unlicensed access points. Here, Awin et al. (2019) indicated that continuous spectrum scanning can handle the aforementioned issues in networking. However, Syed and Safdar (2015) highlighted that constant scanning is associated with increased energy consumption, increased operational cost and increased operational time. Thus, addressing for establishing a novel approach towards spectrum sensing will help to manage the operations sustainably.

subject academy

Interesting points about the solution provided based on the aforementioned research question

As per the mentioned problem, the addressed question can produce a possibility of a solution to the loopholes of cognitive radio networks. Syed and Safdar (2015) indicated that the use of Markov two-state chain models will allow the management of networking traffic efficiently for primary and secondary users. The paper will also evaluate the evolutionary trends of spectrum sensing methods in radio networks, which will help to understand the effectiveness of Markov model chains. Such evaluation has confirmed that the “history assisted energy-efficient spectrum scheme” in radio networks helps in sensing information related to networking and database structures. Thus, both primary and secondary users become benefitted from proper energy consumption.

Existing and related work

Previous works have also addressed several issues related to cognitive radio networks and indicated that spectrum sensing is a useful technique in managing traffic and safety of use. For instance, Lu et al. (2016) has outlined that using machine learning models (ML) are worthy to manage the energy vectors of spectrum sensing networks. Here, the authors proposed a low-dimensional probability vector that can improve the performance of the system instead of N-dimensional energy vectors. Thus, primary and secondary users will be benefited from using the concerned network. Despite such effectiveness, the paper has not depicted any significant benefit to energy efficiency approaches on networking, which is the main gap for future research (Syed and Safdar, 2015).

On the other hand, to address the energy efficiency aspects in radio networks, Althunibat et al. (2015) have proposed a cooperative and centralised spectrum sensing process for efficient operations. It was truly efficient in energy savings and increased networking performances. However, the major fault of this model was energy detection accuracy (Kumar et al. 2019). Because of this part, despite having proper energy-efficient features, this model failed due to accuracy issues in detecting spectrum holes.

Contrasting the results of Althunibat et al. (2015), another study showed that cooperative spectrum sensing is efficient in handling the performance of cognitive radio networks (Anandakumar, and Umamaheswari, 2017). This paper demonstrated that cooperative models of spectrum sensing can detect the probability of false alarms generated by the use of radio networks. Additionally, it is capable of detecting energy and band-limited white noise. In terms of establishing this hypothesis, the authors used some algorithm-based models. These results indicated that the “cooperative CUSUM spectrum sensing algorithm” performed well than existing cooperative radio networks (Anandakumar, and Umamaheswari, 2017). Thus, here also, authors were able to demonstrate the accuracy and effectiveness of the spectrum sensing in a specific condition and self-configuration systems. Such factors were indicating a clear gap in regular networking practices, which need further attention.

Finally, the article of Fu et al. (2018) was an eye-opener in the research of spectrum sensing in cognitive networks. There, the authors proposed a multi-bit data-fusion scheme for radio networks consisting of hard decision rules and soft fusion rules. By application of quantization bits process here, authors stated that the proposed fusion rule was able to manage better network performance and control communication overheads.

Thus, the open research problem lies in: 

  • Specific challenges of the respective spectrum sensing techniques were not known (Fu et al. 2018)
  • The energy efficiency of the spectrum sensing technologies are not properly demonstrated against the standard

Research approach

The concerned paper has undertaken a simulated research design approach to evaluate the energy efficiency of spectrum sensing methods in cognitive radio networks. For that, authors have developed a history-based network analysis technology for the detection of energy efficiency. As seen previously, detecting accuracy of spectrum sensing methods including machine learning models failed to hit the appropriate standards (Lu et al. 2016). However, in this paper, the authors used Energy Detection (ED) techniques as parts of system modelling for ensuring simplicity and widespread usefulness. Based on this, the authors developed a three-state Markov chain model, which was tested against both primary and secondary user conditions. By using this model, authors were able to demonstrate at which stage, what percentage of energy efficiency can be expected (Syed and Safdar, 2015). Based on this decision making of the primary user and secondary user arm was also tested, which gave an appropriate idea about the scenario that can be established after full implementation.

subject academy

Another significant benefit of the spectrum sensing technique was the simplicity of use and optimal spectrum hole detection ability. For example, referring to Pd and Pfa data for developing the prediction model was efficient to ensure proper reliability and accuracy (Syed and Safdar, 2015). For these features, the authors stated that this method is efficient for bandwidth and signal detection while saving energy consumption. On the other hand, the authors of the specialist paper used cognitive radio networks to establish energy efficiency aspects. Here, Namvar and Afghah (2015) stated that cooperative spectrum sensing processes help in the resource allocation of wireless networks. It eases access to networking connections among primary and secondary users (Kalamkar et al. 2016). However, decision-making on the spectrum scanning threshold was not possible through other papers and models, which are possible by this paper. Thus, storing of data, intelligent processing techniques and efficient decision-making on threshold marking are some positive points achieved from this study.

Historical data inclusion in algorithms and decision-making models are always efficient to enhance the accuracy of simulation outputs (Wang et al. 2017). In this paper, the authors used historical data associated models in simulations and decision making. Thus, the authenticity of outcomes is ensured. Despite having such advantages, the following are some of the disadvantages of the paper:

  • No primary data was considered in the paper for analysis and application implementation
  • No consideration was given to centralised and rigid spectrum management policies for developing the Markov chain model (Syed and Safdar, 2015)
  • No comparison was done regarding energy efficiency, to what extent the proposed model is energy efficient compared to standard models, is not known 

Personal investment

Reason of interest

After reviewing the specialist paper and the other papers in the domain, I found that almost all the studies are trying to focus on the effectiveness of the spectrum sensing technique based on the accuracy of the model they propose (Khobragade and Raut, 2017; Zheng et al. 2020). However, no one has reported the efficiency of the prediction model concerning another established model in this domain. Such a gap has motivated me to review the area once before proceeding with the study.


Some of my strengths, for which this research is possible for me to carry out:

  • The ability of critical thinking
  • Knowledge of spectrum sensing, Markov model and cognitive radio networking
  • Efficiency in historical data-assisted analysis

Some of my experiences that will help to proceed in this study:

  • Class-experience of teamwork
  • Research work experience in assignments

Altogether, it is clear that there are some gaps to be filled in this specialised area to implement flawless solutions into real-life situations for better networking.

Reference list

Althunibat, S., Di Renzo, M. and Granelli, F., 2015. Towards energy-efficient cooperative spectrum sensing for cognitive radio networks: An overview. Telecommunication Systems, 59(1), pp.77-91. https://link.springer.com/article/10.1007/s11235-014-9887-2

Anandakumar, H. and Umamaheswari, K., 2017. An efficient optimized handover in cognitive radio networks using cooperative spectrum sensing. Intelligent Automation & Soft Computing, pp.1-8. https://www.tandfonline.com/doi/abs/10.1080/10798587.2017.1364931

Awin, F.A., Alginahi, Y.M., Abdel-Raheem, E. and Tepe, K., 2019. Technical issues on cognitive radio-based Internet of Things systems: A survey. IEEE access, 7, pp.97887-97908. https://ieeexplore.ieee.org/abstract/document/8766798

Bouabdellah, M., Kaabouch, N., El Bouanani, F. and Ben-Azza, H., 2018. Network layer attacks and countermeasures in cognitive radio networks: A survey. Journal of information security and applications, 38, pp.40-49. https://www.sciencedirect.com/science/article/abs/pii/S2214212617301515

Fu, Y., Yang, F. and He, Z., 2018. A quantization-based multibit data fusion scheme for cooperative spectrum sensing in cognitive radio networks. Sensors, 18(2), p.473. https://www.mdpi.com/1424-8220/18/2/473/pdf

Kalamkar, S.S., Jeyaraj, J.P., Banerjee, A. and Rajawat, K., 2016. Resource allocation and fairness in wireless powered cooperative cognitive radio networks. IEEE Transactions on Communications, 64(8), pp.3246-3261. https://ieeexplore.ieee.org/abstract/document/7086843

Kumar, A., Thakur, P., Pandit, S. and Singh, G., 2019. Analysis of optimal threshold selection for spectrum sensing in a cognitive radio network: an energy detection approach. Wireless Networks25(7), pp.3917-3931. https://link.springer.com/article/10.1007/s11276-018-01927-y

Lu, Y., Zhu, P., Wang, D. and Fattouche, M., 2016, April. Machine learning techniques with probability vector for cooperative spectrum sensing in cognitive radio networks. In 2016 IEEE wireless communications and networking conference (pp. 1-6). IEEE. https://ieeexplore.ieee.org/abstract/document/7564840

Namvar, N. and Afghah, F., 2015, March. Spectrum sharing in cooperative cognitive radio networks: A matching game framework. In 2015 49th Annual Conference on Information Sciences and Systems (CISS) (pp. 1-5). IEEE. https://ieeexplore.ieee.org/abstract/document/7086843

Syed, T.S. and Safdar, G.A., 2015. On the usage of history for energy efficient spectrum sensing. IEEE Communications Letters, 19(3), pp.407-410. https://ieeexplore.ieee.org/abstract/document/7004859/

Wang, M., Cui, Y., Wang, X., Xiao, S. and Jiang, J., 2017. Machine learning for networking: Workflow, advances and opportunities. Ieee Network, 32(2), pp.92-99. https://arxiv.org/pdf/1709.08339

Zheng, S., Chen, S., Qi, P., Zhou, H. and Yang, X., 2020. Spectrum sensing based on deep learning classification for cognitive radios. China Communications17(2), pp.138-148. https://arxiv.org/pdf/1909.06020

Scroll to Top