CM-LSTM Based Spectrum Sensing

Abstract

This paper presents spectrum sensing as a classification problem, and uses a spectrum-sensing algorithm based on a signal covariance matrix and long short-term memory network (CM-LSTM). We jointly exploited the spatial cross-correlation of multiple signals received by the antenna array and the temporal autocorrelation of single signals; we used the long short-term memory network (LSTM), which is good at extracting temporal correlation features, as the classification model; we then input the covariance matrix of the signals received by the array into the LSTM classification model to achieve the fusion learning of spatial correlation features and temporal correlation features of the signals, thus significantly improving the performance of spectrum sensing. Simulation analysis shows that the CM-LSTM-based spectrum-sensing algorithm shows better performance compared with support vector machine (SVM), gradient boosting machine (GBM), random forest (RF), and energy detection (ED) algorithm-based spectrum-sensing algorithms for different signal-to-noise ratios (SNRs) and different numbers of secondary users (SUs). Among them, SVM is a classical machine-learning algorithm, GBM and RF are two integrated learning methods with better generalization capability, and ED is a classical, traditional, and spectrum-sensing algorithm.

A Review of Research on Spectrum Sensing Based on Deep Learning

“…Similar to CNN, LSTM can also learn correlated features through the covariance matrix of the signal. Wantong Chen et al [80] extracted spatially relevant features through the covariance matrix after dimensional changes, then learned temporally relevant features through LSTM and combined the two for joint learning, proposing a CM-LSTM method for spectrum perception, which is able to learn the spatial features of multiple signals received by antenna arrays and the temporal features of individual signals to achieve spectrum perception. The authors conducted experiments to compare the performance of the SVM, GBM, RF, and ED algorithms at the same SNR.…”

“…[ [77][78][79][80][81]83] LSTM Higher accuracy can be achieved at a low SNR through efficient learning of raw signal data or correlation information between signals.…”

1-D Convolutional Neural Network-Based Models for Cooperative Spectrum Sensing

“…RNNs are renowned for their capability to leverage historical data in order to make precise predictions about the future or current state. RNN algorithms are utilized in [15], where the authors use a signal covariance matrix-based SS algorithm and long short-term memory (LSTM) to jointly extract the spatial cross-correlation features of multiple signals received by the antenna array and the temporal autocorrelation features of single signals.…”