Efforts aim to fill gaps in commercial technology development.
A U.S. Marine from Marine Wing Support Battalion 372 adjusts his radio while on patrol in Iraq. Funding from the National Science Foundation (NSF) aims to build on advances in cognitive radios to integrate communications and networking far beyond current capabilities.
A major thrust of
Basic research funded by the National Science Foundation (NSF) aims at integrating devices or embedding communications in a system. It involves understanding the relationship between different functions and technologies and then building them as a single architecture or system. Even a partial success in this endeavor might empower—and impel—engineers to look at communications in entirely new ways.
The NSF is funding “transformative research” that will change the way experts look at designing communications systems, says the NSF’s Dr. Scott F. Midkiff. He is the program director for integrative, hybrid and complex systems in the Electrical, Communications and Cyber Systems (ECCS) division of the NSF’s Directorate for Engineering. Midkiff does not foresee a major shift in communications research in the coming years. The goal of the NSF’s communications research is to spur development of device and component innovations that will change the way communications is performed 10 years from now, he says.
Most NSF research spending focuses on areas that require specific government attention. The commercial sector conducts the bulk of research and development in the
The hottest NSF communications research covers both optical and wireless communications. Midkiff notes that strong interest prevails both in radio frequency (RF) and optical communications. High-performance networking, collaboration and visualization all will be aided by advances in those disciplines, particularly in optical switching and integrated photonics.
“Putting photonics into the same domain as integrated circuits in the electronics world will allow us to achieve those kinds of scales both in terms of dimensions and manufacturing capabilities—and the cost benefits that come with that,” he says. “There are a lot of really good opportunities in optical communications.”
Midkiff explains that the intersection of RF and optical communications is a key focal point of NSF research efforts. “If you look at traditional cellular systems today, there is a lot of need for backhaul where you have optical networks. So there are opportunities for reducing system cost and improving quality by trying to look at RF and optical communication in a more integrated way,” he declares.
Instead of raising RF communication to a high level for translation back down into the optical domain, the research aims to integrate the two media at the physical level. Modulation would be converted from RF to optical while the signal still is in the digital bit level instead of in the packet level. Just as the optical world is seeing great interest in all-optical networks for potential cost savings and increased throughput, an interface between the RF and optical media would generate similar enhancements.
This builds on efforts to develop hybrid communications, in which the NSF invested a few years back. “The opportunities to improve communications lie more and more at the intersection” of these technologies, Midkiff offers. “We’re not just looking at it as purely a communication link, but [instead] looking at how you leverage other advances for communications systems.”
The ECCS has provided $267 million in funding for efforts that at least broadly relate to telecommunications. Most of this funding, which includes research into devices, systems and applications, has occurred over the past three years. Of that total, $16.7 million focuses on communication systems and subsystems, and $12.9 million involves networking and applications for tightly integrating communications systems. Most of that funding has occurred over the past two years, with some elements accelerating over the past 12 months. These amounts tend to be directed toward transformational basic research that would have significant ramifications on communications, Midkiff points out.
While the NSF is keeping its head above water with regard to finances, some of its funding is heading below the water line. The NSF currently is funding research into underwater communications. That niche area in communications is the target of research aimed largely at acoustic communications for scientific and environmental applications, Midkiff explains.
For example, some scientists at the
Another RF research area involves telemetry in biomedical applications. Implantable or wearable biomedical devices could advance health monitoring, as hospital-based professionals could remotely check on the real-time status of patients in their homes.
Many telemetry issues remain, particularly for implantable devices. Form factor, internal power and transmission power all must be resolved for these remote health sensors to become a reality. But this application could generate substantial benefits in both private health care and military medicine.
Another communications area with ramifications in the commercial, defense and public safety arenas is spectrum. Demand for spectrum ranges across all disciplines, Midkiff notes. A key research area is cognitive wireless communications, and considerable work has been done to develop cognitive radios that would exploit unused bands as needed. In addition to spectrum, these radios would sense channel impairments, waveforms for interoperability and other elements that would affect its performance. That radio also could learn from past situations to select proven configurations when confronted with potentially limiting conditions.
But taking that approach further would allow a network to adapt to its operational environment in the same manner. The cognitive network would coordinate among multiple nodes to adopt an ideal configuration, and it would learn from previous situations.
Midkiff describes this effort as a system architecture approach. The devices would have a broader range of operating points and would interact with each other. The design and verification and the algorithm techniques to work in that domain are challenging, he offers.
Quoting Professor P.R. Kumar of the
“We can build things that are far more complex than we can design or can analyze and predict how they will behave with a high degree of assurance,” he continues. “So now, if you are putting cognitive radios or lots of other very adaptive, highly cyber-enabled systems out into the world—especially in safety-critical or life-critical applications—how can you really certify that their operation is going to be correct?”
Midkiff adds that the NSF is funding related research into what he describes as a cyberphysical system. This entails coupling physical engineered systems tightly with communication, computation and control. In effect, the physical environment is the RF environment. Radios and network nodes would be able to sense their environment, make decisions and respond appropriately through reconfiguration. Midkiff explains that the NSF funds this area broadly as well as specifically for projects that apply this approach to communications systems.
What makes cognitive radios possible is the ability to build into a handheld terminal the computing power that was not possible on desktops just a few years ago, Midkiff says. Similarly, integrated photonics will generate different ways of looking at system architectures and capabilities.
Leveraging these new devices to improve communications will change the fundamental rules of communications, he continues. Engineers no longer will be able to take off-the-shelf communications solutions to use in a system. Instead, system design will require multidisciplinary interaction between people who understand an application and those who understand how to enable communications.
“It’s no longer incremental improvements such as slightly better modulation schemes or slightly better ways of extracting information from signals to make them more resistant to noise,” Midkiff emphasizes. “[It’s] really fundamentally different ways of realizing systems that really require fundamentally different architectures and algorithms. That’s what we try to fund.”
Midkiff admits that the NSF may not be doing enough to promote the synergy among different capabilities and enablers. It is funding a project that is examining how to establish short-distance networks using ultrawideband links for high multimedia traffic. Other efforts look at integrating field programmable gate array (FPGA)-based architectures with communication algorithms. The FPGA would serve as a more generic platform for computation to support communications, and it could support other areas as well.
Many related research areas affect communications. The NSF has invested heavily in nanotechnology, which should lead to nanoscale devices with many applications. NSF nanotechnology research ranges from basic materials to algorithms for analysis at the electron level, which will be essential for understanding nanotechnology devices.
Another research effort includes magnetic devices and storage. One related area involves spintronics, in which a storage system uses electron spin instead of charge to represent information. Midkiff allows that this technology has the potential to be transformative, as it could change the scale of memory devices.
The ECCS is funding other related research through its electronics, photonics and device technologies program. These technologies include bioelectronics, flexible electronics, micromagnetics, microwave photonics, molecular electronics and opto-electronics.
NSF ECCS: www.nsf.gov/div/index.jsp?div=ECCS
Research Aims to Beget More Research
Some of the NSF’s efforts touch on providing the infrastructure to further research. Dr. Scott F. Midkiff explains that a key presence within NSF is the concept of cyberinfrastructure. This concept entails using state-of-the-art computing, communications, networking and storage for performing science and engineering research. It includes supercomputing for large-scale problems, visualization for extracting knowledge out of large datasets, virtual organizations and collaboration for teaming in search of solutions. All of this is enabled by today’s cyberinfrastructure, he notes.
The next-generation cyberinfrastructure will be more than just high-performance computing and networking, he adds. It will feature pervasive computing in which devices communicate and share information much more often than currently. Some of that interaction will not even be visible to the user.
Another effort within Midkiff’s division focuses on using computation and communication to develop entirely new types of engineered systems. For example, today’s radio represents a continuous push for analog and digital interfaces toward the antenna. This has increased the digitization of the radio; and software-defined and cognitive radios are becoming computational-based. Add high-performance computing and multi-core architectures, and “you have a lot of capability there to use,” Midkiff points out.
One new NSF initiative is cyber-enabled discovery and innovation, or CDI. This new program is a five-year initiative to revolutionize research through advances in computational thinking. According to NSF officials, the goal is to change understanding in a wide range of science and engineering aspects along with socio-technical innovations. The NSF wants change that ultimately would lead to broad improvements in quality of life. CDI focuses on three thematic areas: data to knowledge; complexity in natural, built and social systems; and virtual organizations.