Statistical analysis and characterization of signaling and user traffic of a commercial multi-band LTE system

The fourth-generation (4 G) cellular network is currently the dominant mobile technology used worldwide. The analysis of the network behavior can help forecast the traffic and thus improve the network. The characterization of mobile user behavior can be useful in evaluating the emerging concepts. We collected traffic traces on a commercial multi-band 4 G cell to better understand and model the network and user behavior. We evaluate the network utilization, the number of connections, and the uplink and downlink cell achieved throughput. In addition, we analyze and identify statistical models that describe the connection inter-arrival time, connection duration, and connection size. The results show daily and weekly patterns that depend not only on the time of day but also on the frequency band. We find that the frequency band and time of day have a limited impact on user behavior. On the contrary, the connection inter-arrival time strongly depends on the frequency band.