Standardization aspects of Tactile applications on Multi-access Edge Computing

Abstract

With the advent of 5G communication, it is envisioned to enable the applications which provide real-time immersive experience by offering seamless connectivity, remote interaction, computation and control. Furthermore, such applications require high availability, scalability and reliability with low latency, leading pathway to next generation Tactile Internet (TI). The standard Multi-access Edge Computing (MEC) architecture, proposed by ETSI, is a promising solution offering requirements for such applications. However, to enable TI with MEC, several standardization efforts are required to ensure QoS and QoE standards. An architectural framework leveraging the standardization aspects of MEC for haptics-based telesurgery is presented in this paper. A system consisting of a Robotic Hand (RH) which is operated remotely by a human-agent at the other end through a Haptic Glove (HG) is considered, communicating with each other over the 5G MEC network. An intelligent system at the MEC can proactively predict human intents, without waiting for a feedback, thereby reducing delays. As of now, MEC has very basic bandwidth management APIs for specific bandwidth requirements. Furthermore, the parameters for MEC, considering the aspects related to enablement of edge intelligence involving prediction, caching, training and computation offloading and communication core need to be established. These must be able to fulfill QoS and QoE requirements of the application. Several standardization efforts have been taken up in this direction, though requiring further investigation.