The demand for effective techniques to recognise and categorise various cellular network signals has grown due to the quick development of wireless communication technology. Communication security, spectrum monitoring, and wireless network management all depend on accurate cellular signal identification. The Extreme Learning Machine (ELM) algorithm is used in this study to provide a machine learning-based method for the automatic detection of cellular signal data. Multiple Power Spectral Density (PSD) features representing various wireless communication signals, including 5G, GSM, LTE, and WiFi, make up the dataset used in this work. In order to eliminate inconsistencies and get the features ready for training, the dataset is first preprocessed. The Extreme Learning Machine classifier, which is renowned for its quick learning speed and strong generalisation capacity, is then trained using the extracted PSD bin characteristics as input. Standard performance criteria, such as confusion matrix analysis and classification accuracy, are used to assess the trained model. According to experimental data, the suggested model can accurately and successfully differentiate between various cellular signal kinds. Furthermore, a web-based interface is put in place so that users can train the model and use input metrics to forecast the type of cellular signal. The suggested method can be expanded for real-time wireless signal monitoring applications and offers an effective and useful solution for automatic cellular signal identification.