F-TLBO-ID: Fuzzy fed teaching learning based optimisation algorithm to predict the number of k-barriers for intrusion detection

APPLIED SOFT COMPUTING(2024)

引用 0|浏览4
暂无评分
摘要
Ensuring fast and efficient Intrusion Detection and Prevention (IDP) at international borders is crucial for maintaining security and safeguarding nations. In this study, we propose an innovative approach that harnesses the power of machine learning and Wireless Sensor Networks (WSNs) to achieve faster and more accurate IDP. Our novel Fuzzy fed Teaching Learning Based Optimisation regression algorithm (F-TLBO-ID) revolutionises the prediction of the required number of k -barriers for rapid IDP. To develop and validate our approach, we synthetically generated pertinent features using Monte -Carlo simulations. These features encompass essential parameters such as the concerned region's area, effective transmission range, effective sensing range, number of sensor nodes, and the fading parameter. Training the F-TLBO-ID algorithm with these features yielded exceptional results, accurately predicting the required number of k -barriers with an impressive correlation coefficient (R = 0.99), minimal Root Mean Square Error (RMSE = 11.32), and negligible bias (-3.66). To benchmark the performance of our F-TLBO-ID algorithm, we conducted comprehensive comparisons with fine-tuned benchmark algorithms, including AutoML, GPR, GRNN, RF, RNN, SVM, and ANN. Additionally, we evaluated the algorithm against 11 different variants of nature -inspired algorithms. Remarkably, our FTLBO-ID algorithm outperformed all these methods in terms of accuracy, firmly establishing its superiority. Finally, we validated the performance of the F-TLBO-ID algorithm using publicly available datasets. The results were highly satisfactory, exhibiting a strong correlation coefficient (R = 0.84), acceptable RMSE (36.24), and minimal bias (-7.17). This study offers a robust and reliable algorithm to predict the required barriers for fast IDP, surpassing the accuracy of existing benchmark algorithms. By implementing our proposed algorithm, the efficiency of IDP systems at international borders can be significantly improved, ultimately enhancing security and facilitating smooth border operations.
更多
查看译文
关键词
Intrusion detection,Nature-inspired regression,Monte-Carlo simulations,Barrier coverage,Synthetic datasets
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要