报告详情
AI-Driven Context-Aware Vulnerability Prioritization for Cognitive Radio Networks Using CVSS v4.0 and EPSS
编号：73
访问权限：仅限参会人
更新：2026-11-10 16:38:18 浏览：10次
Online
报告开始：2026年11月21日 14:15（Asia/Kolkata）
报告时间：15min
所在会场：[S7] Networking, Security and Blockchain Systems [S7-2] Networking, Security and Blockchain Systems
暂无文件
摘要
Cognitive Radio Networks (CRNs) dynamically sense and opportunistically access underutilized spectrum through software-defined radios (SDRs) and cognitive engines. While this hardware re-configurability improves spectrum efficiency, it expands the system attack surface across multiple protocol layers, introducing threats such as primary-user emulation (PUE) and spectrum-sensing data falsification (SSDF). Standard vulnerability management frameworks (CVSS and EPSS) evaluate flaws in isolation without representing localized radio environment dynamics, resulting in severe alert fatigue for security operations. This paper presents a context-aware vulnerability prioritization architecture that retains the exact compliance semantics of CVSS v4.0 and EPSS while overlaying a proprietary, CRN-specific evidence-fusion layer. The framework maps dynamic radio assets to public vulnerability intelligence, adjusts contextual severity vectors, and estimates a study-defined 30-day exploitation-observation likelihood using a regularized gradient-boosted classifier. Crucially, the engine treats localized parameters—specifically spectrum-sensing exposure and telemetry-derived anomaly encounter frequencies—as contextual predictors rather than intrinsic vulnerability properties. The signals are mathematically synthesized alongside multi-domain stack criticality to compute a Composite Prioritization Score (CPS), which feeds a theoretical capacity-bounded 0/1 Knapsack remediation queue. Evaluation on a contextually constructed 480-instance vulnerability corpus demonstrates an algorithmic discrimination of 0.941 ROC-AUC and 0.903 PR-AUC. Furthermore, the pipeline contracts the immediate remediation workload from 312 instances under severity-only mandates down to 89 instances, yielding an exact 71.5% reduction in initial triage volume while maintaining high ranking stability (Spearman's ρ = 0.892) across varying operational weight simplexes.
关键词
Cognitive Radio Networks,vulnerability prioritization,,CVSS v4.0,EPSS,exploit likelihood,machine learning,,primary-user emulation,spectrum-sensing data falsification,,SDR security
报告人
Reshma C R
Asst. Professor BMS Institute of Technology & Management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