Study reveals how analog computing delivers faster solutions to complex signal processing challenges

Rochester, New York – A new study is offering a clearer picture of how analog computing could reshape the way modern communication and radar systems process information. Researchers at the Rochester Institute of Technology (RIT), working alongside international partners, have demonstrated that analog-driven approaches can carry out complex signal operations faster and at lower power than many current digital systems. Their findings outline how a shift toward analog circuits could help solve long-standing bottlenecks in signal processing, particularly in applications that rely on electromagnetic radio waves.

The research team, led in part by Mohammad-Ali Miri, an associate professor of electrical engineering in RIT’s Kate Gleason College of Engineering, focused on one of the most demanding areas in computational work: matrix operations. These operations are at the heart of nearly every advanced computation, powering technologies such as artificial intelligence, radar systems, wireless communications, and image processing. Because matrix calculations require the system to process large grids of values all at once, performance slowdowns can appear quickly when digital platforms reach the limits of their speed or energy efficiency.

The team’s work, featured in a fall issue of Nature Communication under the title “Programmable circuits for analog matrix computations,” brings attention to these challenges while introducing a path around them. Miri collaborated with researchers led by Rasool Keshavarz, a professor and senior research fellow at the University of Sydney; Kevin Zelaya, a post-doctoral fellow in RIT’s Department of Electrical and Microelectronic Engineering; and Negin Shariati, an associate professor at the University of Sydney. Together, they explored how the physics used to manipulate electromagnetic waves could directly support matrix computations—without converting signals into digital form first.

Read also: Bull’s Head prepares for a transformative year as new streets a workforce center and newly constructed homes move closer to reality

Their approach builds on the fact that photonic systems, which rely on the behavior of light, follow mathematical principles that closely resemble those in conventional computing. Previous work in this area often focused on optical frequencies, but the team extended those ideas into microwave and radio-frequency ranges—areas important for radar, satellite systems, and wireless devices. As Miri explained, “We know the physics may be the same, but the technology is completely different when you go to a different frequency range and it is a whole different design and fabrication process and even characterization.”

The motivation behind the research goes beyond theory. Industries worldwide are pushing for faster, more efficient chips that can manage heavy computational loads without draining energy resources. Artificial intelligence, in particular, depends on rapid matrix operations, and digital processors are beginning to run into physical limits on how much more speed and power efficiency they can deliver. “We are in a race for advanced chips, and the artificial intelligence that makes certain industries or companies very popular is that they can do matrix operations very fast and at low power. But the thing we know with digital computing, we are reaching the limits with those approaches,” Miri said.

To test alternatives, the researchers examined how analog signals travel through radar systems. Radar antennas collect electromagnetic waves reflected from surrounding objects, and these waves must be converted from analog to digital before computers can analyze them. This conversion step, while essential in digital systems, creates energy demands and processing slowdowns. The team questioned whether some of that conversion could be avoided entirely.

Their answer emerged in the form of a new type of microwave circuit designed to process radio signals directly. Instead of converting waves into digital data, the circuit manipulates the waves themselves, carefully shaping interactions in a way that performs matrix computations as the signals pass through. Because the processing happens at the speed of the waves—and not at the speed of a digital processor—the operations occur in real time.

The result, according to the study, is a device that bypasses the “digital bottleneck” and completes complex tasks at the speed of light. Early tests show that the analog-based system achieves performance levels comparable to traditional processors but uses less power, making it well-suited for energy-sensitive technologies such as radar arrays, remote sensing, and portable communication platforms.

Read also: Ahead of Cyber Monday the New York State Division of Consumer Protection urges shoppers to stay alert as AI-generated scams spread across online marketplaces

“That is huge pre-processing that can significantly reduce the load in the next stages of the whole system,” Miri said. “This is how we begin to encode information in this system. The way the device is doing the operation is that physics of the system. That is the new and interesting thing with this device. Its architecture is what we have developed.”

The work represents a substantial step toward computing models that rely less on digital conversions and more on the natural physical behavior of electromagnetic waves. While digital systems remain essential across nearly all areas of modern technology, the study shows that analog hardware—when carefully designed—can handle certain operations far more efficiently. For fields that depend on rapid pattern detection and large-scale signal interpretation, analog computing could offer a powerful new toolset.

As research continues, the team hopes their developments will spark broader interest in analog-based solutions, paving the way for next-generation devices that combine speed, efficiency, and adaptability in ways not currently possible with digital-only designs.