Extended Range Wide Area Network (LoRaWAN) has recently gained a lot of attention from the industrial and research community for dynamic Internet of Things (IoT) applications. IoT devices broadcast messages for neighbor gateways that deliver the message to the application server through an IP network. Hence, it is required to deploy LoRaWAN gateways, i.e., network planning, and optimization, in an environment while considering Operational Expenditure (OPEX) and Capital Expenditure (CAPEX) along with Quality of Service (QoS) requirements. In this article, we introduced a LoRaWAN gateway placement model for dynamic IoT applications called DPLACE. It divides the IoT devices into groups with some degree of similarity between them to allow for the placement of LoRaWAN gateways that can serve these devices in the best possible way. Specifically, DPLACE computes the number of LoRaWAN gateways based on the Gap statistics method. Afterward, DPLACE uses K-Means and Fuzzy C-means algorithms to calculate the LoRaWAN gateway placement. The simulations’ results proved the benefits of DPLACE compared to state-of-the-art LoRaWAN gateway placement models in terms of OPEX, CAPEX, and QoS.

Low Power Wide Area Network (LPWAN) technologies recently gained interest from the research and industrial community. Internet of Things (IoT) devices communicate directly with gateways, which act as bridges towards a central network server and the Internet. In this context, it is important to study how to place multiple gateways in an area considering Quality of Service, Capital expenditure (CAPEX), and operational expenditure (OPEX) requirements. This is because network planning and optimization are considered to be significant issues that impact on the application performance, CAPEX, and OPEX. In this paper, we propose an optimal LoRa gateway placement (PLACE). It considers the Gap statistics method to find the number of LoRa gateway, which is used to compute the gateway placement using the Fuzzy C-Means algorithm. Simulation results show that PLACE reduced in 36% the CAPEX and OPEX compared to the grid and random gateway placement, while keeps a similar Packet Delivery Ratio.