CAREER Award winner Sanjib Sur is investigating the future sixth generation wireless network technology 

Sixth generation (6G) wireless networks are expected to be available around 2030 to enable the increased demand of data services in an ever-increasing smart society. While 6G implementation is several years away, Computer Science and Engineering Assistant Professor Sanjib Sur is undertaking research to help ensure a smooth transition to a new wireless network. 

Sur was recently awarded a five-year, $560,000 National Science Foundation (NSF) CAREER Award for his research, “Vision and Learning Augmented D-Band Networking and Imaging.” The research started on May 1.

“We expect to have a base version of the 6G standard in 2028. Every generation of cellular communication must have a standard governing the protocols and formats approximately two years before it’s implemented. We hope that some of the research we’re doing here can influence the standardization for 6G in the coming decade,” Sur says. 

Millimeter-Wave (mmWave) is the core wireless technology for enabling new applications in areas such as smart transportation and telemedicine. In 2019, the Federal Communications Commission approved the commercial use of mmWave frequencies exceeding 100 gigahertz (GHz). The D-band, which is between 110-170 GHz, is intended to be the frontier wireless technology for 6G. But the D-band mmWave networks produce new challenges in the optimization of installing a small cellular base station as well as coordinating and adopting mobile links with wide frequency options. Sur’s research aims to address these challenges and improve the mobile D-band network’s performance, reliability and usability. 

“You have a large, contiguous bandwidth with a D-band network, and the wider the bandwidth, the higher data capacity,” Sur says. “Moving devices to the D-band allows 10 to 20 GHz of contiguous bandwidth, which is more than 10 times higher than the current 5G standard. The D-band is unlicensed with no licensing fees, so you just need the right technology to sustain a high data rate.”

The mmWave network was not used prior to the 5G deployment in 2019 and started at a low frequency between 20 to 50 GHz. There are limited devices in today’s market to enable mmWave networking since it is still a new concept. 

“We’re using a prototype device in our lab with a D-band antenna to use with 6G since we don’t have a smartphone with this type of antenna. The antenna will radiate the wireless signal for connectivity, and some of the available prototypes will help us design a better algorithm to create a robust and resilient connectivity,” Sur says. 

Sur’s project includes designing and applying machine learning augmented scalable D-band systems and networks for applications, such as autonomous cars to allow high-quality bandwidth for better utilizing resources. But moving to higher frequencies causes less reach and more signal disruptions and disconnections, which make it challenging to design a robust and resilient network with a long range that sustains many users. 

“The connection is immediately lost if an obstacle appears between a base station and mobile device. For example, an autonomous car driving down the road may require high-speed connectivity with the roadside infrastructure to coordinate and offload data. If a pedestrian walks in front of the car and the connection is lost, it could be fatal,” Sur says. “Designing a highly resilient, robust network with a 10-fold increase in data capacity than existing infrastructure was my motivation to look into the D-band network.”

Sur’s research also intends to investigate new sensing and imaging applications by repurposing those networking devices. 

“The D-band mmWave has a property that when the signal is radiated, it bounces around and returns to wherever the signal was transmitted. All the different parts in the environment will embed a signature inside the reflected signal,” Sur says. 

For example, if a human walks in front of a wireless transmitter, the signature will be unique compared to a static object. These signatures can be used to determine the shape, size, and locations of humans and objects. 

Self-driving cars can be made safer by having the vehicle recognize humans or objects in poor visibility using the D-band mmWave signal. Unlike cameras, the wireless signal works without light and can penetrate through obstacles. 

“There have been instances of Teslas crashing into roadside infrastructure because the camera and vision system are not sophisticated enough and its performance is determined by the ambient light,” Sur says. “But mmWaves will allow us to break away from those limitations. We plan to deploy this application on existing infrastructure in the future when advanced 5G and 6G base stations pop up.” 

Sur admits that one of the project’s challenges will be hardware availability and his ability to implement the research on real world devices beyond prototypes. 

“Hopefully by the end of our five-year project, the hardware will become more mature and start to be deployed at a larger scale,” Sur says. “Hardware innovations cannot be anticipated when you start developing a technology, but after a certain point there’s an avalanche of rapid development. We hope the hardware will catch up with the algorithms we’re developing.” 

The NSF CAREER Award integrates research and teaching for junior faculty members. The combination of helping to cultivate and refine a yet-to-be released technology and integrating the research into his curriculum has Sur excited for the next five years. 

“I'm humbled that the panelists appreciated my research because they receive hundreds of proposals, and they rated mine as highly competitive,” Sur says. “This award allows me to sustain the current direction of my research for at least five years, and hopefully it leads to new directions more proposals.”