Undergraduate Cyber Security Proposal Sample

Here is a sample that showcases why we are one of the world’s leading academic writing firms. This assignment was created by one of our expert academic writers and demonstrated the highest academic quality. Place your order today to achieve academic greatness.

Introduction

Background

Network intrusion detection systems (NIDS) have been quickly developed in industry and academics in response to the escalating cyber-attacks on states and business organisations worldwide. Insider threats, breach-of-system assaults, and web-based attacks are the most damaging types of cybercrime (Haq et al., 2015). And to protect computer systems from unauthorised access, businesses use a firewall, antivirus software, and an intrusion detection system (NIDS) (Liao et al., 2013).

The anomaly detection speed, precision, and reliability are essential success elements for NIDS. Machine learning (ML) approaches are used to develop NIDS to increase recognition performance and minimise false alarms (Halimaa & Sundarakantham, 2019). Deep learning (DL) methodologies have been used in NIDS as an enhanced version of ML (Alrowaily et al., 2019). Therefore, this research compares various crossovers of modern-day technologies, such as ML and DL, with NIDS to show how it can tackle cyber-attacks.

Aim

This research compares computational models such as ML and DL, making NIDS more efficient against cyber-attacks.

Objectives

It has the following objectives:

• Evaluate existing literature in this area to draw insights into the research problem.
• To compare modern-day computational models such as ML and DL to optimise NIDS.
• To identify problems in the existing NIDS to make it more efficient.
• To recommend a suitable model to improve NIDS efficacy.

Product Overview

Scope

It will identify shortcomings in the conventional NIDS. Moreover, it will find modern-day approaches (ML, DL, etc.) to make NIDE more efficient in countering cyber-attacks. It will see how the incorporation of modern computational models can improve NIDS.

Audience

This research targets academics, large corporations, governments, and network security and ML practitioners.

Background Review

Existing Approaches

Currently, the following NIDS are used by large organisations:

• Signature-based intrusion detection systems detect probable threats by skimming network traffic for specified patterns, such as byte sequences or known harmful instruction sequences used by malware. The word “signature” comes from an antivirus program that alludes to these recognised patterns. Although signature-based intrusion detection systems may quickly detect known assaults, they cannot detect novel attempts that no way exists (Kumar & Sangwan, 2012).
• Anomaly-based intrusion detection systems are a relatively new development that perceives and adapts to unidentified threats, mainly due to the explosion of malware. This detection approach uses MLalgorithms to establish a specified prototype of reliable activity, which is then used to compare new behaviour. While this method allows for identifying previously undiscovered assaults, it is vulnerable to false positives, which occur when previously unknown permitted behaviour is erroneously categorised as harmful (Aldweesh et al., 2020).

Related Literature

According to Sultana et al. (2019), because of the advent of customisable capabilities, Software Defined Networking Technology (SDN) provides a chance to better perceive and monitor network sanctuary issues. SDN-based NIDS recently incorporated ML methods to secure computer networks and resolve network security concerns. In the context of SDN, a stream of sophisticated ML methodologies DL– is beginning to emerge. They examined current studies on ML approaches that utilise SDN to achieve NIDS in this survey. They primarily studied DL approaches in the development of SDN-based NIDS. In the interim, in this survey, they explored technologies used to construct NIDS models in an SDN context. This survey concludes with a debate on current issues in executing NIDS using ML/DL and forthcoming work.

Similarly, according to Jiang et al. (2020), Intrusion Detection Systems (IDS) plays a vital role in network security by detecting and stopping hostile activity. The network intrusion observations are drowned in many everyday observations due to the dynamic and time-varying network environment, resulting in inadequate data for model development and detecting results with a high false detection rate. They offer a network intrusion detection technique that combines blended sampling with a deep network model in response to the data imbalance. They use one-side selection to minimise noisy samples in the overwhelming group and then boost subsets of features using the Synthetic Minority Oversampling Technique. This method may create a balanced dataset, allowing the model to thoroughly understand the properties of minority samples while drastically reducing model training time. Second, they create a deep hierarchical network model using a convolutional neural network. Simulations on the NSL-KDD and UNSW-NB15 datasets tested the proposed network intrusion detection system, with classification results of 84.59per cent and 76.76 per cent, respectively.

Lastly, according to Ahmad et al. (2021), a thorough assessment of current NIDS-based publications discusses the merits and drawbacks of the proposed solutions. A discussion of recent trends and developments in ML and DL-based NIDS follows the suggested technique, review criteria, and dataset allocation. They emphasised numerous research obstacles and recommended future research scope in developing ML and DL-based NIDS by using the weaknesses of the presented approaches. According to the study, 61% of the recommended methods were evaluated using the KDD Cup’99 and NSL-KDD datasets, owing to the availability of comprehensive findings utilising these datasets. However, these datasets are too old to address recent network assaults, limiting the performance of the offered approaches in real-time scenarios. For AI-based NIDS approaches, the model should be evaluated using the most recent updated dataset, such as CSE-CIC-IDS2018, for improved detection accuracy for intrusions.

Methodology

Approach

It will be quantitative research based on the secondary data collected through the systematic literature review. Various NIDS approaches based on the literature will be tested through different ML and DL models, such as CNN. Ahmad et al. (2021)’s study will be used as a base to conduct the review.

Technology

The latest hardware with a sound graphics card, such as Nvidia GeForce RTX 3080, will be used to run ML and DL models to test balanced and unbalanced data sets such as KDD Cup’99 and NSL-KDD present for NIDS.

Version Management Plan

It will use git version control as a repository to help all connected users track the progress of the project. All affiliated users can check the source code and test and debug it.

Project Management

The activities of the project are presented in the following Gantt chart:

Bibliography

Ahmad, Z. et al., 2021. Network intrusion detection system: A systematic study of machine learning and deep learning approaches. Transactions on Emerging Telecommunications Technologies, 32(1), p. e4150.

Aldweesh, A., Derhab, A. & Emam, A., 2020. Deep learning approaches for anomaly-based intrusion detection systems: A survey, taxonomy, and open issues. Knowledge-Based Systems, Volume 189, p. 105124.

Alrowaily, M., Alenezi, F. & Lu, Z., 2019. Effectiveness of machine learning-based intrusion detection systems. In. International Conference on Security, Privacy and Anonymity in Computation, Communication and Storage, July.pp. 277-288.

Halimaa, A. & Sundarakantham, K., 2019. Machine learning based intrusion detection system. In. 2019 3rd International conference on trends in electronics and informatics (ICOEI), April.pp. 916-920.

Haq, N. et al., 2015. Application of machine learning approaches in intrusion detection system: a survey. IJARAI-International Journal of Advanced Research in Artificial Intelligence, 4(3), pp. 9-18.

Jiang, K., Wang, W., Wang, A. & Wu, H., 2020. Network intrusion detection combined hybrid sampling with deep hierarchical network. IEEE Access, Volume 8, pp. 32464-32476.

Kumar, V. & Sangwan, O., 2012. Signature based intrusion detection system using SNORT. International Journal of Computer Applications & Information Technology, 1(3), pp. 35-41.

Liao, H., Lin, C., Lin, Y. & Tung, K., 2013. Intrusion detection system: A comprehensive review. Journal of Network and Computer Applications, 36(1), pp. 16-24.

Sultana, N., Chilamkurti, N., Peng, W. & Alhadad, R., 2019. Survey on SDN based network intrusion detection system using machine learning approaches. Peer-to-Peer Networking and Applications, 12(2), pp. 493-501.