SoftBank, in partnership with NVIDIA and Mavenir, announced the successful completion of End-to-End (E2E) communication from user equipment to image processing multi-access edge computing (MEC) applications using Graphic Processing Units (GPUs) in actual machines. This was achieved through virtualized radio access network (vRAN) components. SoftBank had recently launched its research facility, the “AI-on-5G Lab.”
The lab will integrate vRAN and MEC while also demonstrating and applying various AI solutions to the business domain. SoftBank, NVIDIA, and Mavenir also achieved real-time person detection through an AI solution on a server with the same vRAN architecture.
The “AI-on-5G Lab.” is a research facility developed in partnership with NVIDIA that combines virtualized radio signal processing software from network software provider Mavenir, AI processing middleware, AI image processing application software on MEC, and physical Radio Units (RU) from Foxconn Technology Group.
The RAN and MEC AI application work on the same architecture server. SoftBank successfully achieved real-time person detection by wirelessly connecting 5G-compatible camera images through its 5G network to an AI application.
SoftBank plans to use GPUs for both RAN and AI applications, despite many vRAN accelerators being specialized for vRAN software. The company is expected to utilize the GPU platform for AI application as a vRAN platform when adding RU and vRAN software, and enable the vRAN site as a MEC location that uses GPUs. SoftBank aims to achieve dynamic resource allocation depending on RAN and MEC demand while also reducing power consumption of vRAN in the future.