MIX-MAB: Reinforcement Learning-based Resource Allocation Algorithm for LoRaWAN

Abstract

This paper focuses on improving the resource allocation algorithm in terms of packet delivery ratio (PDR), i.e., the number of successfully received packets sent by end devices (EDs) in a long-range wide-area network (LoRaWAN). Setting the transmission parameters significantly affects the PDR. Employing reinforcement learning (RL), we propose a resource allocation algorithm that enables the EDs to configure their transmission parameters in a distributed manner. We model the resource allocation problem as a multi-armed bandit (MAB) and then address it by proposing a two-phase algorithm named MIX-MAB, which consists of the exponential weights for exploration and exploitation (EXP3) and successive elimination (SE) algorithms. We evaluate the MIX-MAB performance through simulation results and compare it with other existing approaches. Numerical results show that the proposed solution performs better than the existing schemes in terms of convergence time and PDR.

ADR-Lite: A Low-Complexity Adaptive Data Rate Scheme for the LoRa Network

“…Many relevant works have attempted to improve LoRa network performance [3], [6], [7], [9], [10]. Examining the full transmission parameter state space to find the best combination is an exhaustive approach.…”

“…The approaches available for controlling and managing the transmission parameters in the LoRa network are divided into two categories, network-aware [7] and link-based approaches [6], [9]. The transmission parameters between the EDs and the GWs are determined by the NS in a centralized manner (link-based approach), whereas the transmission parameters are determined in a distributed manner by the EDs, in the network-aware approach.…”

“…In the central decision-making Network Server (NS), LoRa executes an Adaptive Data Rate (ADR) mechanism to configure EDs' transmission parameters, resulting in improved performance metrics. To increase the efficiency and scalability of LoRa networks, different techniques such as proposing a new Media Access Control (MAC) [4], message retransmission [5], optimization [6], and machine learning techniques [7] are employed in the relevant literature. This paper aims to review, implement, and analyze a new greedy approach while maintaining minimal space complexity.…”