Computing Had Moore’s Law. RF Engineering Finally Has an Answer.

For more than half a century, Moore’s Law held the technology world together. The observation that transistor density would double roughly every two years gave the semiconductor industry a shared language, a common benchmark, and a roadmap that drove decades of investment, innovation, and commercial growth. Chips got smaller. Computers got faster. The digital economy was built on the confidence that this curve would hold.

Radio frequency hardware never had that. The engineers building the antennas, waveguides, phased arrays, and RF interconnects that underpin cellular networks, satellite communications, radar systems, and the coming 6G era have worked without a unifying scaling framework. Progress has been real, but it has been measured in scattered benchmarks, vendor roadmaps, and ad hoc comparisons that make it difficult to see the bigger picture. Asif Alam, an RF engineer and researcher at Florida International University, wants to change that.

Alam has developed what he calls a Moore’s Law for RF, a data-driven scaling framework that traces how RF hardware has progressed across more than a century of development, from early kilohertz radio transmitters to modern millimeter-wave and sub-terahertz systems operating at hundreds of gigahertz. The framework, developed with co-author Muhmmad Shah Alam, has been published in IEEE Access and the Educator’s Corner of IEEE Microwave Magazine, two of the most widely read publications in the field.

The core finding is both simple and significant. RF hardware does scale, but not in the clean, linear way that transistors did. Alam’s analysis of RF transistor performance data spanning four decades, from 1985 to 2025, found that record maximum oscillation frequency values increased by approximately 1.6 times per decade, a consistent exponential trend hidden beneath the apparent chaos of frequency-specific engineering challenges. The framework surfaces that trend and maps it across the full spectrum from kilohertz to terahertz.

Why does this matter? Because the wireless industry is currently in the early stages of defining what 6G will look like, and it is doing so largely without a shared benchmark for RF hardware progress. Companies, governments, and standards bodies are making trillion-dollar decisions about spectrum allocation, infrastructure investment, and technology roadmaps based on projections that have no equivalent of Moore’s Law to anchor them. Alam’s framework offers exactly that anchor.

The RF scaling challenge is fundamentally different from the transistor scaling challenge that Moore’s Law described. When chip designers shrink a transistor, the physics cooperates. The same logic gate does the same job in less space and at lower power. When RF engineers push operating frequencies higher, the physics starts pushing back. Signal loss increases. Heat becomes harder to manage. Antennas that worked perfectly at one frequency become finicky at another. The packaging that connects RF chips to antennas, which is already one of the most technically demanding problems in electronics, becomes dramatically harder as wavelengths shrink to millimeter and sub-millimeter scales.

This is the engineering environment in which Alam has built his career. His research at Florida International University focuses on millimeter-wave antennas, air-filled substrate integrated waveguides, RF interconnects, and RF packaging, which are precisely the hardware elements that determine whether a theoretically excellent RF system actually works in the real world. He has presented his work at the IEEE International Microwave Symposium in San Diego, the IEEE Phased Array Symposium in Boston, IEEE VTC in Washington, and the IEEE Texas Symposium in Waco. His papers have appeared in IEEE Access. He holds two granted U.S. patents in millimeter-wave waveguide transition technology, one of which was awarded a Gold Medal at the 50th International Exhibition of Inventions Geneva, among more than 1,000 inventions from over 40 countries.

The Geneva recognition came for work on a systematic design method for connecting rectangular waveguides to air-filled substrate integrated waveguide structures at millimeter-wave frequencies. The problem sounds narrow but its implications are broad. At these frequencies, even a small mismatch at a waveguide connection creates signal reflections and power loss that can degrade an entire system. Most existing designs relied heavily on iterative simulation and trial and error. Alam’s approach introduced analytical design equations that allow engineers to build the connection correctly from the start, reducing the dependence on guesswork and making the technology more accessible to teams that do not have months to spend optimizing a single interface.

That same instinct, bringing systematic rigor to problems that the field had been solving by feel, runs through the Moore’s Law for RF work. The RF engineering community has long known that frequencies are getting higher and systems are getting more capable. What it lacked was a framework that quantified that progress historically and used it to inform expectations about what comes next. Alam’s work provides that framework, drawing on the same kind of empirical analysis that Gordon Moore applied to transistor counts in 1965.

The timing of the Moore’s Law for RF work is deliberate. As 6G standardization accelerates and terahertz systems begin moving from laboratory demonstrations toward commercial prototypes, the need for a shared scaling reference is becoming urgent. The decisions being made now about what frequencies to use, what hardware architectures to pursue, and what performance targets are realistic will shape wireless infrastructure for the next two decades. A framework that tells engineers and investors where the field has been and what the historical rate of progress suggests about where it is going is not just academically interesting. It is practically necessary.

Alam has described his goal simply: to give RF engineering the same shared language that Moore’s Law gave semiconductor engineering. Whether the curve holds as firmly across frequency as it did across transistor counts remains an open question that future measurements will answer. But the framework now exists, which means the question can be asked systematically for the first time.