Brain Function Analysis and Brain Controlled Applications based on Communication Theory

In the 21st century, the focus of modern biology has undergone a major shift to neuroscience; that is, to understand how our brain works. Brain research is essential to the diagnosis, prevention and treatment of brain dysfunction, and is also the key for establishing the future interface between the human mind and brain-controlled applications.

Computational analysis for multilevel brain research

To facilitate tractable and systematic analysis, brain research has been divided into different levels based on the size of the unit of study: molecular, cellular, systems, behavioral, and cognitive. The fundamental level is molecular neuroscience, which studies different molecules that play different roles that are crucial to brain functions; the cellular level focuses on studying how molecules work together to give neurons (i.e., brain cells) their special properties; systems neuroscience studies how constellations of neurons form a complex circuit or network that performs a common function, such as vision; behavioral neuroscience focuses on how neural systems work together to produce integrated behaviors; and cognitive neuroscience, the highest level, studies how the activity of brain creates the mind.

Computational modeling and analysis plays a key role in extracting brain signals and revealing brain functions. Innovative computational techniques help us reveal the significant truth or meaningful information hidden in the massive data collected from various experiments. In collaboration with colleagues from Departments of Psychology, Neuroscience, Radiology, and Translational Science & Molecular Medicine, and taking brain as a communication network, we aim to develop modeling and analysis theory and techniques to support different levels of brain analysis.

Currently, we are working on causality analysis, brain network connectivity analysis, neuron population activity level and synchronization analysis, brain information processing capacity and working memory capacity analysis. Brain function analysis provides in-depth information on brain function and dysfunction, and also help us develop new biomarkers for early stage brain disease diagnosis.

Brain controlled applications

Brain-controlled applications have just started to emerge, and are considered to be the user interface of the future. Brain-computer interfaces, though still at infancy with inadequate accuracy, are becoming available for wireless application development. At MSU BAWC Lab, we are currently working to convert brain commands to electrical signals for automatic car control. We can now make the car move and stop just through “thinking”. We aim to develop more accurate brain signal extraction techniques by exploiting high density brain recording system and advanced signal processing techniques. Potentially, our research can help establish the future interface between human mind and brain-controlled sensors, robots, and other Internet of Things (IoT) devices.

Demo Video: Vehicle Control through Mind Power

Secure and Efficient Wireless Communications and Networking

Mainly due to limited total available spectrum and lack of a protective physical boundary, wireless communication is facing much more serious challenges in capacity and security than its wirelined counterpart.

Unlike wirelined networks where new requests on communication services can largely be resolved by adding more transmission lines, in wireless networks, the total available spectrum is mainly limited by the RF technologies, and has to be shared by various users and services at each power defined zone or cell. With the ever-increasing demand on simultaneous multimedia wireless services, increasing the information capacity has been a focal research area in the wireless research community. At the same time, wireless access can easily be shared by anyone within the radio range. Lacking of a protective physical boundary makes wireless communication much more vulnerable than wired communication. As people are relying more and more on wireless networks for critical information transmission, the ubiquitous wireless interconnectivity has also provided a primary conduit for malicious agents to exploit vulnerabilities on a widespread basis. Reliability has become an urgent issue and a great concern in both civilian and military wireless communications.

Reliability includes both security performance (e.g. information confidentiality and jamming resistance) and system performance (e.g. error probability and distortion between the input and output signals). Very often, reliability is achieved at the cost of lower spectral efficiency. A natural but fundamental problem: given the fixed spectrum, how to design wireless systems which are both highly efficient and reliable?

At BAWC lab, we try to address this problem through multilayer methodologies under both benign and hostile environments. Our ultimate goal is to achieve more efficient, secure and timely information exchange.

Secure Monitoring and Control for Smart Home and Smart Grid

Today, long distance information exchange through wide area networks (WANs) has mainly been limited to phone-to-phone or phone-to-computer communications. At the same time, the development of human-to-device interfaces, such as home automation systems and integrated car-driver interfaces, has largely been limited to local area networks (LANs). The separation of WANs and LANs leads to an inconvenient gap in long distance control for general devices, especially moving devices, such as in-car electronics. In this research, we aim to develop cyber-enabled systems that can achieve seamless, secure monitoring and control of localized devices or device networks with a mobile phone. While providing a new class of wireless services, cyber-enabled device monitoring and control also imposes significant capacity and security/reliability demands on wireless networks, we are explore this area along with the emerging 5G wireless systems.