Research Interests

I am interested in communication theory, signal processing, and machine learning. The emphasis of my research is on coding, wireless communications, image processing, and wireless networks.

My recent work in machine learning and artificial intelligence includes the theoretical and practical challenges of designing distributed and quantized learning algorithms, e.g. federated learning algorithms, that do not need to transmit the collected data over networks. In many applications, multiple access nodes collect and/or disseminate local information over a certain geographical area of interest. In such distributed networks, both data and computational power are distributed. The main challenge is to design robust distributed learning algorithms that only need to communicate a limited number of bits with neighboring nodes instead of sharing the entire collected data.

I have been working on the theoretical and practical challenges of designing communication systems and networks that use multiple antennas and/or relays, especially in the presence of interference. There are many challenges in designing space-time coding and beamforming schemes for multiple-input multiple-output (MIMO) systems and designing distributed methods for cooperative communications especially in the presence of interference. One of these challenges is to study the theoretical tradeoffs and limits from an information theory perspective. Another is to design practical coding and beamforming schemes that satisfy those limits using signal processing methods.

On a related topic, I have been looking at optimizing resources across different layers of a wireless network. This is in particular important for mobile ad-hoc networks and wireless sensor networks where the traditional layer hierarchies may not exist. I am interested in UAV networks, cross-layer adaptation, network connectivity, power-aware wireless network protocols, optimal node deployment especially for heterogeneous sensor networks and IoT, cooperative communications, and interference cancellation especially for networks with multiple-antenna nodes and relays. Recent challenges involve the design of quantized and distributed machine learning algorithms and trajectory optimization for UAVs.

Also, I have been involved in developing data compression algorithms especially for image and video coding. My work in this field includes the transmission of multimedia information over wireless networks and the Internet. A related topic of interest is the image/video quality estimation especially for a heterogeneous network that supports different resolutions and bandwidth. I am interested in developing new coding schemes and network protocols that improve end-to-end recovery and enhance the quality of service.

Multiscale Image Quality Estimator (MIQE) Software is available here

Multiscale Video Quality Estimator (MVQE) Software is available here

In addition, I have worked on segmenting cardiac images, especially MRIs, using traditional image processing algorithms and machine learning methods. This includes the design of deep learning algorithms for automatic segmentation of heart chambers for both cardiac MRIs and Echocardiograms. In addition, we have designed generative adversarial networks (GANs) to generate synthetic cardiac MRIs and improve the performance of the convolutional neural networks (CNNs) by augmenting the training dataset.

Cardiac MRI LV segmentation software is available here

Cardiac MRI RV segmentation software is available here


H. Jafarkhani, "Taking to the Air to Help on the Ground: How UAVs Can Help Fight Wildfires ," IEEE ComSoc Technology News, Oct. 2022.

SMAC-FIRE: Closed-Loop Sensing, Modeling and Communications for WildFIRE

Increases in temperatures and drought duration and intensity due to climate change, together with the expansion of wildlife-urban interfaces, has dramatically increased the frequency and intensity of forest fires, and has had devastating effects on lives, property, and the environment. To address this challenge, this project's goal is to design a network of airborne drones and wireless sensors that can aid in initial wildfire localization and mapping, near-term prediction of fire progression, and providing communications support for firefighting personnel on the ground. Two key aspects differentiate the system from prior work: (1) It leverages and subsequently updates detailed three-dimensional models of the environment, including the effects of fuel type and moisture state, terrain, and atmospheric/wind conditions, in order to provide the most timely and accurate predictions of fire behavior possible, and (2) It adapts to hazardous and rapidly changing conditions, optimally balancing the need for wide-area coverage and maintaining communication links with personnel in remote locations. The science and engineering developed under this project can be adapted to many applications beyond wildfires including structural fires in urban and suburban settings, natural or man-made emergencies involving radiation or airborne chemical leaks, "dirty bombs" that release chemical or biological agents, or tracking highly localized atmospheric conditions surrounding imminent or on-going extreme weather events.

The system developed under this project will enable more rapid localization and situational awareness of wildfires at their earliest stages, better predictions of both local, near-term and event-scale behavior, better situational awareness and coordination of personnel and resources, and increased safety for fire fighters on the ground. Models ranging from simple algebraic relationships based on wind velocity to more complex time-dependent coupled fluid dynamics-fire physics models will be used to anticipate fire behavior. These models are hampered by stochastic processes such as the lofting of burning embers to ignite new fires, that cause errors to grow rapidly with time. This project is focused on closing the loop using sensor data provided by airborne drones and ground-based sensors (GBS). The models inform the sensing by anticipating rapid growth of problematic phenomena, and the subsequent sensing updates the models, providing local wind and spot fire locations. Closing this loop as quickly as possible is critical to mitigating the fire's impact. The system we propose integrates advanced fire modeling tools with mobile drones, wireless GBS, and high-level human interaction for both the initial attack of a wildfire event and subsequent on-going support.

Every Important Idea Is Simple !