Rails and railway cast manganese crossings may develop structural defects while in-service due to the stresses and environmental conditions they are subjected to. Acoustic emission (AE) is a dynamic non-destructive testing technique which is extensively used for structural integrity evaluation. AE has the potential to be employed as a tool to monitor the integrity of railway structures online, thus minimizing delays and maintenance cost on the railway network. In this project, AE signals captured during three-point fatigue crack growth tests performed on different rail and cast manganese samples have been analysed in order to evaluate the applicability of the technique for application in the field. The analysis of the raw data has been based on the application of different AE signal related parameters and signal processing algorithms such as Fast Fourier Transform (FFT) and Spectral Kurtosis (SK). The aim of the analysis is apart from identifying the crack growth events to quantify the severity of damage in the specimens at least in a semi-quantitative way. Field tests have also been carried out to evaluate the effect of rolling stock noise on the AE measurements. One of the key objectives of the project is to evaluate appropriate filtering algorithms in order to minimise the influence of unwanted noise on the AE signals and increase the accuracy of the results acquired.