Virtualizing LoRa baseband functionalities to the Edge

Abstract: Traditional network architecture design of Low Power Wide Area Networks (LPWAN) suffer from their inability to dynamically scale resources based on the incoming traffic and involve manual procedures for capacity upgrades. There is a need to develop a cloud-native architecture where the gateway hardware performs signal detection while baseband processing is virtualized in the Edge, provisioning for dynamic scaling of resources. Cloud Radio Access Network (CRAN) as a concept can be leveraged to realize such a functional split architecture. In this thesis work, a LoRa based IoT Radio Access Technology (RAT) system is developed in a cloud-native design where the physical baseband processing is pushed to an Edge computing system. The open source GNURadio software environment is used to prototype the split-architecture design. A preamble detection method is implemented on the gateway, also called the Radiohead (RH), to identify and transmit only the LoRa based packets, which reduces the Fronthaul (FH) link utilization. Our study shows that the proposed split-architecture is feasible with a performance on-par with traditional architecture. The proposed system can operate in all the 3 mandatory channels of EU868 band as specified in the LoRa standard. Measurement results like CPU utilization and network capacity usage, from integration with IEEE 802.15.4 and NB-IoT stacks, further show the feasibility of the multi-RAT scenario. Additionally, implementation comparisons of C++ and Python showed C++ as a better choice for developing modules in the GNURadio environment. Finally, insights were shared into further developing this work like focussing on the drawbacks related to multi-Spreading Factor detection on same channel and interfacing with higher layer open-source architectures like The Things Network.