Dynamic Threat Mitigation in Enterprise Networks Using Software-Defined Security and Threat Escalation Consistency Principle
Year: 2026
Type: conference-paper
Abstract: From the past few years, enterprise networks increasingly face sophisticated and continuously evolving cyber threats that bypass conventional security mechanisms and cause operational disruptions. Existing threat detection approaches often depend on static security policies, isolated traffic analysis and predefined response mechanisms by limiting their ability to recognize persistent attacks and dynamically mitigate emerging threats. To address these limitations, this research proposes Software-Defined Threat Escalation and Mitigation Network (SD-TEAM) for intelligent and dynamic enterprise network protection. Initially, network-flow data are preprocessed using robust quantile scaling to reduce influence of extreme traffic values and generate stable feature representations. Next, dualpath temporal convolutional network extracts short-term and long-term behavioral characteristics from sequential network flows. Subsequently, novel Threat Escalation Consistency Principle (TECP) evaluates classification confidence, temporal consistency, threat severity and escalation behavior to determine evolving risk level of detected activities. Furthermore, software-defined security controller dynamically translates threat assessments into appropriate mitigation actions, including monitoring, rate limiting, traffic isolation and blocking. The closed-loop framework continuously updates security policies according to changing threat conditions by enabling rapid and context-aware responses.
Language: English
Subfield: Computer Networks and Communications
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7225074591