PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
May 9, 2026InformationOpen Access

Edge-Prioritize IDS: Zero-Retraining Class Prioritization for Real-Time Edge Intrusion Detection

View Full Paper
Ask AI
Bookmark
Share

Authors

PPPruthviraj PawarGEGregory Epiphaniou

Discussion

Loading...

Member takes

Overview

Randomized trial evaluates class-prioritized early-exit framework for real-time intrusion detection, indicating effective performance under variable conditions.

Key Points

  • The aim is to enhance real-time intrusion detection on edge devices by prioritizing high-risk attack classes without retraining.
  • Developed Edge-Prioritize IDS framework for early exit based on class prioritization.
  • Utilized a K-dimensional control vector for encoding runtime priorities.
  • Conducted evaluations on benchmarks including NSL-KDD and CIC-IDS2017 using NVIDIA Jetson TX2.
  • Achieved baseline accuracy up to 99.6% while reducing latency by 55% and energy consumption by 50% for prioritized classes.
  • Isolated contributions of each component through ablation studies.
  • Demonstrated recovery of near-baseline latency within 500 samples under class-frequency drift.

Cite This Study

Pawar et al. (2026) studied this question.

synapsesocial.com/papers/69fed19ab9154b0b82879031https://doi.org/10.3390/info17050451
View Full Paper
Ask AI
Bookmark
Share