Multiband Massive IoT: A Learning Approach to Infrastructure Deployment

We consider a novel ultra-narrowband (UNB) low-power wide-area network (LPWAN) architecture design for uplink transmission of massive number Internet Things (IoT) devices over multiple multiplexing bands. An IoT device can randomly choose any the bands to transmit its packet. Due hardware constraints, base station (BS) is able listen only one band. Our main objective maximize packet decoding probability (PDP) by optimizing placement BSs and frequency assignment develop two online approaches that adapt environment based on statistics (un)successful packets at BSs. The first approach predefined model environment, while second measurement-based model-free approach, which applicable environment. benefit model-based lower training complexity, risk poor fit in model-incompatible simulation results show our proposed band BS offer significant improvement PDP baseline random perform closely theoretical upper bound.