Internet of Things (IoT) has created a niche in the last decade. We are in the midst of an unprecedented growth of automation and smart systems, driven by the miniaturization of sensing and computing & communication technologies. Even though battery technology has grown to a large extent, it is still not possible to power these sensors for long, especially in situations where sensors need to provide data at a higher frequency, which drains the batteries fast. Thus, recently researchers are looking into various means of self-powered sensing systems that harvest energy from the ambiance. In this paper, we characterize and optimize a piezoelectric energy harvesting device consisting of a cantilever beam, which is suitable for self-powered Wireless Sensor Network (WSN) nodes in the rail transport networks. The integrated unimorph-piezoelectric sensor harvests energy from ambient vibration. We attune the harvester parameters to the low range of ambient vibration frequencies and demonstrate an experimental model to validate the results of our setup. Vibration frequency and amplitude are measured by performing real experiments inside trains on the routes of the Dutch and German railway network. Each harvester provides 0.72µJ to 0.19mJ per hour depending on the vibration. Multiple of them can be utilized as secondary energy sources inside trains to measure the ambient vibration while harvesting to make it a perpetually powered sensor@en