Optimizing NB-IoT Communication Patterns for Permanently Connected mMTC Devices

The new types of industry-driven applications that need to be supported by low-power wide-area networks (LPWANs), such as remote control or metering of devices within the massive machine-type infrastructures (e.g., Smart Grids), require a permanent connection to the remote server. In addition, there is also a shift in the communication paradigm, as the user equipment (UE) nodes are queried in regular and frequent intervals. Notably, the presence of this type of traffic may drastically deteriorate the performance of LPWAN technologies initially developed to support conventional use-cases characterized by non-synchronized transmissions. Though none of the LPWAN technologies is inherently designed to handle such demanding communication patterns, the narrowband Internet of things (NB-IoT) still stands for the best candidate as it operates within the license frequency spectrum. To optimize the delay performance of both types of traffic coexisting at the NB-IoT air interface, we propose an approach based on spreading the message transmission time instants of regular and stochastic traffic. We show an optimal value of the spreading interval minimizing the message transmission delay of regular traffic and propose a mathematical model to estimate its value. By parameterizing the model using a detailed measurements campaign of NB-IoT, we show that the optimal value of spreading interval and associated mean message delay is a linear function of the number of UEs. We report these values for a wide range of UEs in the coverage area of the NB-IoT base station and show that conventional stochastic traffic does not influence regular traffic performance.