Just one eNodeB (the "Radio Access Network," i.e. part that connects to the antennas, sourced from Star Solutions), two antennas (13 dBi sectorial antennas with

Just one eNodeB (the "Radio Access Network," i.e. part that connects to the antennas, sourced from Star Solutions), two antennas (13 dBi sectorial antennas with spatial diversity), and an EPC (the "backend," also from Star Solutions) as a VM on a single server. The goal is serve a detector array of 35 indepedent detectors up to 10 km away from the base station (which is mounted on the roof of the tallest building in the research station, some pictures of both a station and the base station that I happened to have taken are available here: https://www.desy.de/news/news_search/index_eng.html?openDirectAnchor=2107&two_columns=0 https://www.desy.de/news/news_search/index_eng.html?openDire... ). Each station has an omni 8 dBi LTE antenna mounted at several meters height. I also did put a sim card in my phone and it worked fine though that was not the primary purpose (but it allowed me to video call from the middle of the ice sheet far away from the station WiFi, which was kind of a novelty). The only feasible alternative was 900 MHz WiFi which would have been similar, but lower throughput and much less available technology, whereas LTE modems are cheap and abundant. It was deployed about a month ago and so far so good. It's data only, and the server the EPC is on just acts as a gateway. Due to the conditions, the eNB is indoors. Installation of antennas was done by station carpenters as I was not allowed on the roof without training :). I am a physicist, not a telecom engineer but it was not too complicated.