Adapting communication networks for low-latency human-to-machine (H2M) applications
Project Leader: Elaine Wong
Student: Lihua Ruan
Collaborators: Dr Imali Dias (Electrical and Electronic Engineering), Prof Vincent Chan (Massachusetts Institute of Technology (MIT), USA)
Primary Contact: Prof. Elaine Wong (firstname.lastname@example.org)
Keywords: human-to-machine communication networks; low-latency communication networks; Tactile Internet
Disciplines: Electrical & Electronic Engineering
Future real-time and remotely-controlled human-to-machine (H2M) applications such as teleoperation and telemedicine will need to be supported by communication infrastructures that incur very low and stringent end-to-end communication latency of approximately 1 to 10 ms. Addressing this latency challenge necessitates a strategic rethink on how network resource is allocated and managed.
In supporting H2M communications, converged optical fiber and wireless networks have been considered as a promising network infrastructure. In the uplink direction, machines contend for uplink bandwidth. Typically, in order to avoid uplink bandwidth collisions, uplink bandwidth is allocated to each machine by a central office through a process known as dynamic bandwidth allocation.
In remotely-controlled H2M applications, adaptive allocation will promote flexibility and real-time response, thus ensuring that the latency constaint is always satisfied. This project therefore focuses on adaptive allocation schemes based on reinforcement learning. Reinforcement learing is a special branch in machine learning that focuses on the interactions between an agent (the central office) and the environment (machines). The aim here is to reinforce the agent‘s actions for more rewarding outcomes in terms of satisfying the stringent latency constraint.
This project is presently recruiting PhD candidate(s) with a background in Electrical and Electronic Engineering and/or Computer Science. Interested candidates with a keen interest in experimental investigation and analytical modelling, are encouraged to contact Prof. Wong. The successful candidate(s) will be mentored by a team of leading experts with pioneering contributions in broadband access networks, 5G transport networks, converged fibre-wireless networks, human-to-machine communication networks, Internet-of-Things, optical wireless networks, and the Tactile Internet.
Relevant research papers
E. Wong and L. Ruan, “Achieving Low-Latency H2M Communications through Predicting Bandwidth Demand: A Comparative Study of Statistical Prediction and Machine Learning Techniques,” to appear in Proc. of OSA Photonic Networks and Devices, July-August 2019.
L. Ruan, M.P.I. Dias, and E. Wong, "Machine Learning-based Bandwidth Prediction for Low-Latency H2M Applications", to appear in IEEE Internet of Things Journal, 2019. Acceptance date: 25th December 2018.
E. Wong, M.P.I. Dias, and L. Ruan, “Predictive Resource Allocation for Tactile Internet Capable Passive Optical LANs,” J. Lightw. Technol., vol. 35, no. 13, pp. 2629 – 2641, July 2017.
Please contact Prof Wong to access these papers.
Further information: https://findanexpert.unimelb.edu.au/display/person3967