MUMBAI, India, May 1 -- Intellectual Property India has published a patent application (202641049568 A) filed by Srinivasa Ramanujan Institute Of Technology; Dr. G. Rasheed; V. Madhu Shalini; M. Ashok; and P. R. Raghu Vamshi, Ananthapuramu, Andhra Pradesh, on April 18, for 'cnn-based enhanced spectrum prediction model for cognitive radio networks.'
Inventor(s) include Srinivasa Ramanujan Institute Technology; Dr. G. Rasheed; V. Madhu Shalini; M. Ashok; and P. R. Raghu Vamshi.
The application for the patent was published on May 1, under issue no. 18/2026.
According to the abstract released by the Intellectual Property India: "The primary aim of this project is to design and develop a CNN-based spectrum occupancy prediction system that addresses and overcomes the limitations of traditional spectrum sensing techniques in cognitive radio networks. The project focuses on creating an efficient, intelligent, and low-overhead prediction framework that enables secondary users to access unused licensed spectrum without causing interference to primary users. This initiative provides an effective solution for improving spectrum utilization and supporting dynamic wireless communication environments. This project analyzes the shortcomings of continuous spectrum sensing methods and conventional machine learning models such as Support Vector Machines. By addressing these limitations, the proposed system reduces sensing delay, minimizes energy consumption, and enhances prediction accuracy. The developed CNN-based model enables automatic feature extraction from measurable spectrum parameters such as frequency, received signal power, and duty cycle. Given the rapid growth of wireless communication technologies and the increasing demand for spectrum resources, improving spectrum efficiency is essential. The proposed model operates by collecting historical spectrum data, normalizing input parameters, and processing them through convolutional layers that reduce dimensionality while increasing discriminative feature representation. The learned features are then classified into idle or occupied channel states. The effectiveness of the CNN-based system directly impacts spectrum utilization efficiency and interference reduction. Therefore, improving prediction reliability and robustness is vital for dynamic spectrum access. This project explores these aspects to ensure optimal performance and enhanced wireless communication efficiency."
Disclaimer: Curated by HT Syndication.
