Spreading Factor Detection for low-cost Adaptive Data Rate in LoRaWAN Gateways

In order to meet the capacity needs of LoRa networks, Adaptive Data Rate (ADR) has been proposed and implemented in LoRaWANs. The network server running ADR determines the optimum data-rate and hence spreading factor setting for each LoRa device in a network. This in turn requires the gateway to be capable of receiving all possible spreading factors. Existing gateways achieve this by using multiple RF front ends, increasing their overall cost and complexity. In this work, we propose a Discrete Wavelet Transform based spreading factor detection algorithm that is agnostic to transmitter settings. This computationally light-weight algorithm can be implemented on any off-the-shelf SDR, bringing down the cost and ease of LoRaWAN gateway implementations. Using experimental, real-world datasets, we show that the proposed algorithm can detect the spreading factor of over 99.5% of the received packets at SNRs down to -10dB.