Communication signal detection technology based on residual neural network and DenseNet

At present, the detection accuracy of communication signals using direct sequence spread spectrum and orthogonal frequency multiplexing technologies is not high. To improve the accuracy and reliability of signal detection, a communication signal detection technology based on residual neural networks and dense convolutional network is proposed in the research. The research takes the autocorrelation sequence of direct sequence spread spectrum communication and the IQ component of orthogonal frequency multiplexing signals as the input of the model. After constructing the detection models of the two signals, the study adopts the method of ensemble learning to further improve the detection accuracy of the model. The experimental results show that the false positive rate and false negative rate of this method are only 0.8% and 0.3% respectively, and it has high computational efficiency, with a calculation time of only 1.32 s. Furthermore, the detection accuracy rate of this method reaches 97.75%. This indicates that the communication signal detection technology based on residual neural network and dense convolutional network proposed in the research can effectively improve the detection accuracy and efficiency. The research designed method can achieve effective communication signal detection, providing a strong guarantee for the stability and security of the communication system.