Spectrum sensing using energy detection and comparing with convolution neural network to improve the probability of detection.

Autor:	Reddy, P. Niranjan, Bharathi, P. Shyamala
Předmět:	CONVOLUTIONAL neural networks ENERGY consumption PROBABILITY theory
Zdroj:	AIP Conference Proceedings; 2024, Vol. 2871 Issue 1, p1-5, 5p
Abstrakt:	This study aims to compare energy detection using a unique Convolutional neural network in order to find the ideal channel for spectrum sensing. Two groups of twenty users were selected at random; in our research, the pretest power (G-power) for each group was 80%. A one-of-a-kind convolution neural network outperforms Energy Detection in terms of sensing performance (p=0.016) and detection probability (p=0.05). Based on the results, even under low SNR conditions, the novel convolution neural network detection method significantly increased the sensing probability. [ABSTRACT FROM AUTHOR]
Databáze:	Complementary Index
Externí odkaz:	Zobrazit plný text záznamu