Feature extraction of trajectories for mobility modeling in 5G NB-IoT networks

Industry applications that support massive mobile nodes are an important part of 5G Narrow Band Internet of Things (NB-IoT). The mobile node is abstracted as a terminal scene with many shapes and complex behaviors in the cellular internet of things, which makes the network need a better radio resource management strategy to meet more mobility, larger connection requests and higher access rate services. Therefore, the feature extraction of the node trajectories will greatly facilitate the development of optimal algorithms for radio resource management in the large mobile scene of 5G NB-IoT networks. This paper presents a feature mining of the real trajectories from the urban operating vehicles in the city of Shenzhen, China. Meanwhile, the generated trajectories by four common mobility models are also handled as a comparison. The self-similarity, hot-spots, long-tails and travel time are evaluated due to the widely recognized four features of human traces. Mining results show that the vehicles to serve the daily trip of human in the city always take a short travel and activate in several hot-spots with randomly roaming, while the vehicles to serve the goods are showing the opposite characteristics. Moreover, the trajectory generated by the models does not have the characteristics of short travel and roaming in multiple hot spots, nor does it have different movement trends in various time periods.