Gateway Planning for the Long-Range Wireless Networks in Smart Cities

Smart cities use internet to make the day-to-day living of city inhabitants more comfortable and secure. In this regard, long-range wide area networks (LoRaWAN) are ideally suited to provide IoT wireless access networks in smart city applications. LoRaWAN uses the long-range (LoRa) technology at the physical layer to provide long-range wireless connectivity. Considering the shadow fading of the large and closely located buildings in the cities, one of the key challenges in the LoRaWAN deployment is finding the optimal location of gateways while guaranteeing network coverage. In particular, the LoRa gateway deployment becomes more complicated due to the stochastic shadowing of the city buildings due to the random locations of IoT devices. In this paper, a novel analytical framework is proposed that enables the joint gateway placement and spread factor (SF) assignment in LoRaWAN while being cognizant of random building shadow fading. Inspired by the chance-constrained method, a joint stochastic gateway placement and SF assignment problem subject to network coverage constraint is formulated for LoRaWANs. Then, a new unsupervised learning-aware greedy algorithm based on the ''size-constrained weighted set cover' concept is proposed to find an approximate solution to the formulated LoRa gateway planning problem in the city environments. The proposed algorithm is simulated over the campus of Shahid Chamran University of Ahvaz at IRAN. Simulation results demonstrate the effectiveness of the proposed approach compared to the benchmarks. For example, on average, the network coverage probability under our proposed unsupervised learning-aware greedy algorithm is 15 % and 10 % more than the greedy and clustering ones.