Infragravity waves play an important role in port operations and many nearshore processes, and therefore their characterization is of major interest for oceanographers and coastal engineers. The lack of proper measurement networks and historic databases makes the development of hindcasting techniques essential. This work presents a fully developed infragravity wave hindcast methodology through Artificial Neural Networks (ANNs) and its application to a case study. The characteristic wave-heights of the low frequency band and the swell band inside a port basin are computed for a period of eight years on the basis of the long-term offshore wave conditions, a short record of searlevel oscillations and the historic tidal harmonic constituents. Based on the results, we construct and analyze the single and the joint probability density functions of the two characteristic wave-heights studied. In addition, we study the relationships between the infragravity energy inside the port and the offshore wave parameters and explore the extreme events during which the low frequency band energy exceeds the swell energy. The findings highlight the potential of the methodology to characterize the infragravity wave conditions inside a port basin and its suitability to study other coastal problems in which these waves are involved.