Slicing or Spoofing? Detecting and Containing Resource Hijacking in B5G Networks with Suspicious Traffic

Abstract

Although 5G networks can virtually partition resources, the flexibility provided by slicing in B5G networks raises multi-tenant security concerns, such as coordinated hijacking of resources by spoofed identities. Traditional allocation models focus exclusively on performance and function in a purely reactive mode of post-anomaly detection. In this paper, we propose a proactive resource allocation model with a security-oriented approach, formulated as an integer linear programming (ILP) problem. The model incorporates easy-to-simulate suspicion variables into user-level functionalities, which narrow down multi-slice spoofing, reduce the need to throttle high-risk users, and enable the identification of coordinated attacks in real time, thanks to the conflict-detection variables. Experimental validation demonstrates that the framework proactively detects malicious actors, ensures resources are allocated fairly for legitimate uses, and maintains computational viability at the edge for orchestration.