In the 2.4 GHz ISM band RF interference is becoming an ever-increasing problem. While there have been several attempts to mitigate the impact of RF interference on (body) sensor networks, e.g. via frequency hopping, it is often unclear how these solutions perform in different interference environments and when they are actually useful. This is not least due to a lack of knowledge about the characteristics of environmental 2.4 GHz RF noise as perceived by a BSN in realistic scenarios. Such knowledge would, for example, help to better understand the communication challenges in a BSN and derive design decisions for interference mitigation techniques. Our work targets this under explored area: we present the results from an urban measurement campaign, in which a mobile BSN collected about half a billion RF noise samples in various urban environments (park, campus, residential area, shopping street, urban transportation system). Our setup captured the entire 2.4 GHz band, on five different body positions simultaneously. Among other things, our results indicate that WLAN was the dominating source of 2.4 GHz RF noise, significant spectrum activity was typically detected during about 5% of the time, but there is a large variation among the scenarios, and, to detect the presence of RF interference the body position is of no of major importance, however, the difference in interference power measured at two different body positions is not negligible.