Networks & Cybersecurity
ECE researchers in networks and cybersecurity are investigating fundamentally new techniques for the analysis, design, and optimization of large-scale communications systems; advancing performance limits of internet of things (IoT) systems through optimal coordination of algorithms across multiple layers; developing innovative solutions to complex science and engineering problems arising from wireless IoT; exploring brain-inspired computing; and enabling the vision of smart, connected, and secure cities of the future.
Harpreet S. Dhillon
Scott F. Midkiff
ECE researchers are pioneering a fundamentally new analytical approach to the analysis and design of communications systems that not only captures the current deployment trends accurately, but also facilitates analysis leading to simple and easy-to-use expressions for key performance metrics.
This approach, based on random spatial models, has been applied to the design of different types of wireless communi-cations systems, including heterogeneous cellular networks, vehicular networks, drone-assisted communications networks, energy harvesting communications, and device-to-device communications. Instead of considering network elements as deterministic, we endow them with a probability distribution. The added randomness allows us to use powerful tools from stochastic geometry for performance analysis. This approach has also been useful in areas outside communications, such as smart and connected communities and geolocation.
As the next generation of cellular commu-nication technology, 5G New Radio (NR) aims to cover a wide range of service cases, including broadband human-oriented communications, time-sensitive applications with ultra-low latency, and massive connectivity for the IoT.
NR is likely to operate on a higher frequency range than LTE, with much shorter coherence time. 5G NR is also expected to support applications with ultra-low latency. With such diverse service cases and channel conditions, the air interface design of NR must be much more flexible and scalable than that of LTE.
In our recent work, we invented an NR scheduler that can meet the stringent (~100 µs) time requirement that allows 5G NR to cope with the extremely short coherence times and support ultra-low latency applications (e.g., augmented/virtual reality, autonomous vehicles), many of which require sub-millisecond scale delay or response time.
Joint design of communications, computing, and control is essential to autonomous (ground or air) vehicles in large-scale systems, such as a smart city. Our research provides the fundamental guidelines needed to equip existing communication systems for navigating and controlling autonomous vehicles. We have developed a broad range of machine learning and optimization algorithms that can be used to design fully autonomous wireless and vehicular systems, adapting their network operation to existing users, devices, and data sets.
Cellular network architecture will play a central role in providing connectivity to areas with little or no wireless connectivity, and we have designed a 3-D cellular network that can integrate drone-carried base stations and drone users and enable 5G cellular networks to support drone-based systems. We have also developed one of the first quality-of-experience (QoE) models for deploying virtual and augmented reality (VR/AR) services over wireless networks and designed a broad number of solutions for securing autonomous cyber-physical and IoT systems.
Most existing marine communications technologies are limited and expensive. ECE researchers are working to fill the void of broadband wireless communications at sea by developing self-powered ocean mesh networks. A collection of floating base stations, or nodes, “talk” to each other to create a network connection across a large area. Once they’ve been dropped in the water, the base stations start to harvest energy from ocean waves and automatically form a mesh network. Users can then connect to the internet.