Analysis of service oriented network slicing at infrastructure layer in LoRa for IoT applications

Abstract

The Internet of Things (IoT) comprises of innumerable heterogeneous devices. Due to spike in IoT applications, there is a need to enhance flexibility and resource utilization in the network architecture. The modified architecture should acclimate to varying energy limitations, data loads, and power needs. To meet the specific needs of various IoT applications, network slicing has emerged as a solution candidate. This allows the partition of network, fostering the resource optimization, reliability, and flexibility. Long Range (LoRa), one of the Low-Power Wide Area Network (LPWAN) technologies have emerged as reliable solution by providing long-range communication among massive number of connected devices. This paper introduces the concept of strategic service-oriented network slicing framework in the proposed LoRaWAN model, based on communication parameters such as Received Signal Strength Indicator (RSSI), Quality-of-Service (QoS) and Signal-to-Noise Ratio (SNR). Performance evaluation of the proposed LoRa network model highlights the impact of constrained parameter, spreading factor (SF) of LoRa, on different communication parameters. Depending on the changing needs of diverse IoT applications, the derived results from LoRa model have been utilized as the basis for dynamic resource allocation through service-oriented network slicing. Additionally, different machine learning models have been employed to evaluate the effectiveness of the proposed slice classification strategy, confirming the reliability of communication-parameter-based slice allocation.