Hybrid long range wide area network-5G-artificial intelligencearchitecture for enhanced reliable internet of things

Abstract

Internet of things (IoT) has made long range wide area network (LoRaWAN) a key component of low-power communication; nevertheless, its adoption in mission-critical domains is limited by its latency, throughput, and quality of service (QoS) restrictions. To bridge these gaps, this paper proposes a hybrid three-layer structure that combines LoRaWAN, fifth generation (5G), and artificial intelligence (AI) to achieve a dynamic trade-off between scalability, reliability, and energy efficiency. According to the findings, the proposed hybrid model significantly improves network performance. It is based on a random forest regressor that integrates LoRaWAN and 5G technologies. The model increases throughput to 40-80 Mbps while maintaining moderate energy consumption, and reduces latency from an average of over 300 ms to less than 150 ms. Additionally, the random forest algorithm improved the overall network performance stability and the packet delivery ratio to roughly 97%. This approach is more effective than earlier studies that relied on descriptive methods or a single technology. This work establishes the foundation for the applications of IoT in smart agriculture, healthcare, and fourth industrial revolution (Industry 4.0).