# Robust Deep Sensing Through Transfer Learning in Cognitive Radio

08/01/2019

∙

by [Qihang Peng, et al.](/content/profile/qihang-peng/index.html)

We propose a robust spectrum sensing framework based on [deep learning](/content/machine-learning-glossary-and-terms/deep-learning/index.html). The received signals at the secondary user's receiver are filtered, sampled and then directly fed into a [convolutional neural network](/content/machine-learning-glossary-and-terms/convolutional-neural-network/index.html). Although this deep sensing is effective when operating in the same scenario as the collected training data, the sensing performance is degraded when it is applied in a different scenario with different wireless signals and propagation. We incorporate [transfer learning](/content/machine-learning-glossary-and-terms/transfer-learning/index.html) into the framework to improve the robustness. Results validate the effectiveness as well as the robustness of the proposed deep spectrum sensing framework.

[READ FULL TEXT](http://arxiv.org/pdf/1908.00658v1)