Abstract:

The main focus of this article is Spectrum sensing for efficient data transfer in wireless communications. Currently, there is a need for an effective wireless communication system because the serviceable spectrum is not sufficient due to exponential growth in the number of wireless device users. One approach for effectively using the spectrum is through Radio frequency sensing. Spectrum-aware radio relies on Radio frequency sensing to detect available spectrum for better usage and to reduce harmful interference with licensed users. For spectrum sensing, there are currently established techniques based on energy detection, but the development of machine learning has made spectrum sensing more efficient. A machine learning-based model can perform better than traditional methods. To identify the existence of a licensed user, we suggest a machine learning spectrum sensing technique employing a Convolutional Neural Network model in a Gaussian environment. Such that if the spectrum is not being utilized by the licensed user then the secondary user can benefit from it. This methodology will be very helpful in resolving the spectrum scarcity problem in wireless communications

Description:

Spectrum sensing, Spectrum utilization, Energy detection, CNN