Optimizing the LoRa network performance for industrial scenario using a machine learning approach

In this article, the performance of the LoRa network for an industrial scenario has been optimized using a machine learning approach. The network performance is analyzed in terms of received power, outage probability, spectral efficiency and bit error rate (BER). A link-level performance of the LoRa network for an indoor industrial area considering both the non-obstructive and obstructive scenarios has been experimentally evaluated in terms of received signal strength indicator (RSSI) and signal-to-noise ratio (SNR). Using the measured values of RSSI and SNR at the LoRa gateway, the received power is mathematically modelled which is further considered as an optimization problem. First, an artificial neural network (ANN) model was built and trained to predict the received power. Particle swarm optimization (PSO) algorithm was further used to find the optimal values of LoRa parameters corresponding to maximum received power. Simulation results reveal that the proposed optimization approach significantly improves the outage probability, spectral efficiency and BER of the LoRa network.