PhD Student Mehmetcan Gok Wins Best Paper Award at IEEE DySPAN 2026

As demand for 5G and beyond-5G wireless networks continues to grow, federal radar systems and commercial cell networks increasingly compete for frequencies in mid-band spectrum. Currently, shared federal-commercial access relies on infrastructure that is expensive and difficult to scale.

To facilitate improved use of the limited wireless spectrum, new research from Northwestern offers a way for commercial base stations to monitor radar activity using their existing hardware and signal processing capabilities.

This work, "Radar Detection via RLS Adaptive Filter Residuals for Cellular Coexistence," earned the Best Paper Award in the technical track at the 2026 Institute of Electrical and Electronics Engineers (IEEE) International Symposium on Dynamic Spectrum Access Networks (DySPAN), held May 11–14 in Washington, DC.

The DySPAN symposium showcases pioneering advancements and policy innovations in the domain of radio frequency spectrum, including spectrum management, sensing, access, sharing, coexistence, and utilization across emerging wireless technologies. Celebrating three years in a row of award-winning work, Communications and Networking (Commnet) Laboratory teams earned IEEE DySPAN Best Paper Awards in the policy track in 2024 and 2025.

The winning paper was co-authored by Northwestern Engineering’s Mehmetcan Gok, a PhD student in electrical and computer engineering, and his adviser, Michael Honig, professor of electrical and computer engineering at the McCormick School of Engineering; and collaborator Danijela Čabrić (University of California Los Angeles).

In this work, Gok and the research team developed a spectrum sensing framework that leverages the existing sensing capabilities of commercial cellular base stations to detect radars in the lower 3 GHz range of the mid-band spectrum.

Because the lower 3 GHz range is allocated to military radar systems operating under strict security and covertness constraints, shared federal-commercial access requires a non-cooperative strategy to prevent harmful interference. The team’s approach enables commercial operators to detect incumbent activity without dedicated sensing hardware or prior knowledge of the classified waveforms or noise statistics.

“This award is a real honor,” Gok said. “Seeing our idea recognized by the DySPAN community is deeply gratifying, and credit is also due to my collaborators and the reviewers whose feedback sharpened the work.”

In radar detection, a sudden pulse disrupts the patterns the signal-processing system has learned to expect, causing a sharp, momentary spike in an error signal before the system adapts and normalizes. The team built three lightweight detectors to monitor those spikes—one for wideband interference, one for partially localized pulses, and a third for persistent anomalies—and demonstrated that detector performance depends on how radar energy spreads across users and subcarriers.

“We turned the base station into an opportunistic sensor, reusing the computation it already does to serve mobile users and protect incumbents,” Gok said. “Our results also tell operators which detector to run for different interference patterns.”

As a next step, Gok and the team aim to make radar-cellular spectrum sharing more practical and efficient by tackling the real-world complications of multi-tower interference, developing smarter coexistence strategies that don't default to vacating the spectrum unnecessarily, and understanding the fundamental geometric tradeoffs that link detection capability to interference risk.

This work was supported by the US National Science Foundation (NSF) through SpectrumX, an NSF spectrum innovation center. The center supports several research projects at Northwestern led by Professors Honig, Randall Berry, Dongning Guo, and Igor Kadota.