With an increasing number of users and demand for higher data transmission rates, the need for bandwidth increases. The solution to this is to introduce spectrum sharing between radar and communication signals, so instead of competing for frequency, the signals can coordinate. This paper mainly focuses on comparing various signal separation and identification algorithms used to classify radar and communication signals in a Dual-functional radar-communication (DFRC) system. The noteworthy applications of DFRC systems are Air Traffic Control, Autonomous Vehicles. Other applications include IoT applications, military applications, drone applications, and civilian applications. The two methods incorporated in this paper are the traditional signal processing method and the deep learning method. In traditional signal processing, we have Independent Component Analysis (ICA) and Principal Component Analysis (PCA). In deep learning, we have Residual Network (Resnet) which uses neural networks for classification. PCA and Fast ICA methods extract independent signals using Blind Source separation.