报告详情
Hybrid Deep Network for Network Intrusion Investigation and Packet Classification Using Coordinate Attention-Based TCN with Dilated Res-LSTM Models
编号：28
访问权限：仅限参会人
更新：2026-11-10 16:14:22 浏览：7次
Online
报告开始：2026年11月21日 17:00（Asia/Kolkata）
报告时间：15min
所在会场：[S4] Artificial Intelligence and Intelligent Computing [S4-10] Artificial Intelligence and Intelligent Computing
暂无文件
摘要
Network security is essential due to the increasingly sophisticated cyber threats in recent days. Classical Network Intrusion Detection Schemes (NIDS) have struggled to identify intrusions due to their limited detection capabilities and reliance on preset signatures. This work focuses on developing an innovative model for intrusion detection and packet classification to enhance network security. Initially, the required data are gathered from the benchmark dataset. Then, the most prominent features from the HIKARI-2021 dataset are retrieved using a Modified and Randomized Masterpiece Optimization Algorithm (MR-MOA), which considers multi-objective constraints during the feature selection process. In addition, forensic-based features such as correlation, flow timelines, and header entropy are considered and concatenated with the optimally selected features. The parallel processing capability of the MR-MOA makes it well-suited for detecting intrusions in large-scale network, and network traffic data are also processed rapidly by MR-MOA. Then, the concatenated features are passed to the Hybrid Deep Network (HDeepNet), which consists of a Coordinate Attention-based Temporal Convolution Network (CA-TCN) with Dilated Res-Long Short-Term Memory (DRLSTM) to achieve improved performance in packet classification and intrusion detection. Here, the DRLSTM captures the long-term dependencies of network traffic. The proposed NIDS can facilitate the detection of specific attack signatures or unusual behaviour, aiding forensic specialists in identifying the techniques and tactics employed by malicious actors.
关键词
Network Intrusion Investigation and Packet Classification,Modified and Randomized Masterpiece Optimization Algorithm,Hybrid Deep Network
报告人
Apoorva K A
N/A National Institute of Technology稿件作者
全部评论
重要日期
-
会议日期
11月20日
2026
至11月22日
2026
-
10月31日 2026
初稿截稿日期
-
11月07日 2026
提前注册日期
-
11月22日 2026
注册截止日期
主办单位
United Societies of Science
承办单位
Pondicherry University
协办单位
IEEE Section
IEEE Madras Section
IEEE Madras Section
联系方式
- ICon INDIA 2026
- ic******@usssociety.org
发表评论