Advanced AI-Based Spectrum Sensing for 5G and Next-Generation Wireless Systems and Implementation of a Real-Time Zone-Based Safety Alert System using C-V2X

dc.contributor.advisorPatel, Dhaval
dc.contributor.authorPatel, Prapti Nimesh
dc.date.accessioned2026-08-21T07:19:02Z
dc.date.issued2026
dc.description.abstractAs 5G and next-generation wireless networks are rapidly evolving, spectrum efficiency has become a key issue. CR technology offers dynamic spectrum sharing, enabling secondary users to access available spectrum bands without interfering with primary users. But efficient spectrum sensing under low Signal-to-Noise Ratio (SNR) and realistic noise uncertainties is still challenging. This thesis investigates new learning-based spectrum sensing strategies to im prove the detection performance and decrease the computational effort. Initially, a Kolmogorov-Arnold Network (KAN) based spectrum sensing approach is explored to enhance the detection performance in low SNR scenarios using features like energy and Zhang statistic. While this approach has improved performance compared to traditional methods, it still fails to take full advantage of the periodic characteris tics of wireless signals. To address these issues, a Fourier Analysis Network (FAN) strategy is proposed, which uses Fourier-based principles in the network to explicitly consider the frequency and periodicity of wireless signals. The proposed method improves detection performance particularly in low SNR regions, and reduces the computational burden allowing for real-time spectrum sensing. This thesis also moves towards practical applications in Intelligent Transportation Systems (ITS) with the development and deployment of a real-time zone-based safety warning system based on C-V2X communication. This system uses pre-programmed Road Side Units (RSUs) which use Cooperative Awareness Messages (CAMs) to detect safety zones based on zone information. Upon entering a zone, Decentralized Environmental Notification Messages (DENMs) are sent to raise the alert. It also supports non-V2X users through a mobile app. Overall, this thesis provides a holistic solution by integrating advanced spectrum sensing techniques with practical design of C-V2X system that addresses theoretical and practical challenges in the next generation of wireless systems.
dc.description.degreeBachelor of Technology
dc.description.idAU2240241
dc.identifier.urihttps://ir.ahduni.edu.in/handle/123456789/8068
dc.language.isoen
dc.publisherAhmedabad University
dc.schoolSchool of Engineering and Applied Science
dc.subjectAI-Based Spectrum Sensing
dc.subjectWireless Systems
dc.subjectSafety Alert System
dc.subjectV2X
dc.titleAdvanced AI-Based Spectrum Sensing for 5G and Next-Generation Wireless Systems and Implementation of a Real-Time Zone-Based Safety Alert System using C-V2X
dc.typeThesis

Files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
AU2240241.pdf
Size:
34 MB
Format:
Adobe Portable Document Format