This dissertation discusses models for signal analysis using neural networks. At this timeneural networks have shown significant results in various fields including bio-medicalECGs, automation driving and large language models. It also has potential forimplementation in signal analysis in the RF range. Radio frequency is the most widelyused electromagnetic wave spectrum for communication, for that it has motivated us topursue this research. By going through various review articles and scholarly papers, this dissertation highlightsthe various models of neural networks and the methods implemented in signal analysis.Many of the neural networks modals have been written over the years, each networkhaving some advantages over the other. This study provides a brief description neuralnetwork modals and provides a comparative analysis. Based on this analysis, this studyputs forward hybrid convolutional neural network- long short term memory network alsoknow as CNN-LSTM neural network, that aids in proper signal analysis.The neural network can be implemented in a number of programming languages, keepingthe model’s algorithm indistinguishable. While conducting this research, python languagewas adopted and tensor-flow library by Google was used.