An Optimized ADR Algorithm for LoRa Network Based on EWMA

In the age of rapid technological development, the Internet of Things (IoT) is continuously evolving, giving rise to Low Power Wide Area Network (LPWAN) to meet the needs of IoT devices. In LPWAN, LoRa technology has gained attention due to its simple architecture and the use of unlicensed frequency bands. A typical LoRa network application consists of four layers, with LoRa employing chirp spread spectrum modulation, whose adjustable parameters can affect data transmission performance. The Adaptive Data Rate (ADR) mechanism in the LoRaWAN specification dynamically adjusts the parameters used by the terminal nodes, but it has limitations under unstable channel conditions. This paper introduces the Exponential Weighted Moving Average (EWMA) method to evaluate recent link changes and improves the standard ADR algorithm, proposing a new approach to improve the overall network packet transmission rate and energy efficiency. Simulations based on OMNeT++ demonstrate that the proposed algorithm significantly outperforms the standard ADR algorithm and the ADR+ algorithm.