Comparison of Neural Network Architectures for Spectrum Sensing

TL;DR

This paper compares various neural network architectures, including FC, CNN, RNN, and BiRNN, for spectrum sensing, analyzing their detection accuracy, data requirements, and computational efficiency.

Contribution

It provides a comprehensive comparison of NN architectures for spectrum sensing, highlighting their relative strengths and limitations in different resource scenarios.

Findings

CNN, RNN, and BiRNN achieve similar performance with abundant resources.

Fully-connected NN performs worse unless computational resources are limited.

Resource constraints significantly influence the optimal NN architecture choice.

Abstract

Different neural network (NN) architectures have different advantages. Convolutional neural networks (CNNs) achieved enormous success in computer vision, while recurrent neural networks (RNNs) gained popularity in speech recognition. It is not known which type of NN architecture is the best fit for classification of communication signals. In this work, we compare the behavior of fully-connected NN (FC), CNN, RNN, and bi-directional RNN (BiRNN) in a spectrum sensing task. The four NN architectures are compared on their detection performance, requirement of training data, computational complexity, and memory requirement. Given abundant training data and computational and memory resources, CNN, RNN, and BiRNN are shown to achieve similar performance. The performance of FC is worse than that of the other three types, except in the case where computational complexity is stringently limited.