Soutenance de Thèse de Doctorat en "Telecommunications and Mobile Networks" par Mr. Flavien DONKENG ZEMO

The Euromed University of Fes (UEMF) is pleased to inform the public of

the doctoral thesis defense in ” Telecommunications and Mobile Networks”

The thesis defense will take place on Monday, July 14, 2025, at 09:00 a.m.at l’UEMF

Location: Gallery, Building 1

The thesis will be presented by Mr. Flavien DONKENG ZEMO

under the theme :

“ SECURITY THREATS AND DEFENSE STRATEGIES FOR COOPERATIVE SPECTRUM SENSING IN 6G COGNITIVE RADIO NETWORKS ”

Summary

The exponential growth of connected IoT devices driven by applications like remote surgery, smart cities, and autonomous vehicles demands next-generation wireless capabilities. While 6G networks promise ultra-high performance, they must overcome critical challenges in spectral efficiency and physical-layer security to support ultra-massive IoT deployments. Current static spectrum allocation exacerbates scarcity, with studies revealing utilization rates as low as 15-85%. Cognitive Radio Networks (CRNs) address this by enabling secondary users (SUs) to dynamically access underutilized licensed bands without disrupting primary users (PUs), though their effectiveness hinges on reliable spectrum sensing.

Cooperative Spectrum Sensing (CSS) mitigates limitations like hidden-node problems by leveraging collaborative sensing among multiple SUs. However, CSS systems are vulnerable to physical-layer attacks, particularly Spectrum Sensing Data Falsification (SSDF) attacks, where malicious SUs falsify sensing reports to blind the Fusion Center (FC). This research confronts massive independent SSDF attacks in centralized CSS. We first propose a probabilistic SSDF attack model enabling adaptable, stealthy adversarial strategies. The blind FC problem is then formalized to characterize defense requirements.

Two novel defense mechanisms are developed: The first combines reputation-based Byzantine identification with a sequential weighted fusion scheme, enabling rapid, accurate decisions using minimal samples. The second employs an enhanced reputation method paired with weighted majority fusion for robustness against random SSDF attacks. Extensive MATLAB simulations confirm both mechanisms outperform existing solutions, maintaining detection accuracy even when attackers comprise 90% of SUs across diverse attack strategies.

Keywords: Cognitive Radio Networks, Cooperative spectrum sensing, Spectrum Sensing Data Falsification attack, Primary User Emulation attack.

This thesis will be presented to the jury members:

Full name	Grade	Institution	Quality
Prof. Abdelghafour MARFAK	PES	Euromed University of Fes	Jury Chair
Prof. Farid ABDI	PES	FST, Sidi Mohamed Ben Abdellah University of Fes	Reviewer
Prof. Mohamed EL GHAZI	PES	EST, Sidi Mohamed Ben Abdellah University of Fes	Reviewer
Prof. Mohammed BENBRAHIM	PES	FSDM, Sidi Mohamed Ben Abdellah University of Fes	Reviewer
Prof. Kaouthar CHETIOUI	PH	ENSA, Sidi Mohamed Ben Abdellah University of Fes	Examiner
Prof. Fatiha MRABTI	PES	FST, Sidi Mohamed Ben Abdellah University of Fes	Examiner
Prof. Ahmed El Hilali ALAOUI	PES	Euromed University of Fes	Thesis Director
Prof. Sara BAKKALI	PA	Euromed University of Fes	Thesis co-Director