Cross girders of a riveted steel frame bridge were monitored with a wireless sensor network. Since strain sensing is very expensive in terms of power consumption a novel signal conditioning hardware was developed that enabled to significantly reduce the power consumption. An embedded data processing algorithm transformed the recorded raw data into a sequence of maxima and minima representing the strain cycles thus reducing data communication and enabling an additional energy saving. By using a software based event triggering algorithm the embedded data processing was performed only for data acquired during train transits. Strain cycles of more than 900 trains were recorded. The quality of the recorded data was very good and demonstrated that WSNs can be a competitive alternative to conventional tethered monitoring systems for recording operational data for fatigue assessment of steel railway bridges.