A significant narrative shift is starting in spectrum management and will likely continue through 2026 and beyond: “spectrum sharing” is slowly pivoting to “spectrum coexistence,” with recent US policy moves around wireless accelerating this.
This isn’t just semantics. It reflects a subtle shift in the underlying mental, technical, and regulatory models for accommodating multiple spectrum users in a band, both cross-technology (military radar vs commercial mobile networks) and intratechnology (coordination between neighbouring private wireless networks, or adjacent 5G systems).
Rather than simple binary sharing with rigid boundaries in time and space, coexistence implies a more dynamic optimisation of spectrum resources.
To make it work, regulators will generally need a blend of sensing, databases, and, increasingly, an analytics/AI layer—with one set of techniques during “regime transition” and another for in-operation assurance.
Good spectrum options are limited and contested. Over the past thirty years, a huge range of wireless technologies has emerged, demanding access to spectrum that often overlaps with other users. Until recently, the default approach was to clear or “repackage” bands, then auction or allocate the now “clean” spectrum to new, usually exclusive, high-power uses. Incumbents have had (often inefficiently large) allocations reduced or been moved to alternative bands.
Mobile is the prime example: each new “G” generation has driven a hunt for fresh spectrum, pushed by smartphones and by governments’ desire to raise large sums through licensing.
But the “easy” bands have already been cleared and assigned. The remaining incumbents—especially defence, aviation, and meteorological—are increasingly difficult to move because of cost, timing, and criticality.
And at the same time as mobile’s growth, WiFi, advanced satellite communications, and aviation/drone connectivity have all increased their incremental demands alongside the cellular industry’s.
The net outcome is that clearing spectrum bands entirely, for new exclusive high-power uses, is becoming close to impossible—time-consuming, expensive, and often legally and politically fraught.
The past decade has seen a surge in spectrum-sharing schemes and enabling technologies. Many countries have enabled local low- or medium-power networks in bands with incumbents, often with geographic protections:
In many cases, sharing has implied a deliberate two-party arrangement: a primary user lets a secondary one “borrow” the band under strict conditions. In practice, the newcomer carries most of the risk, while the incumbent keeps broad protection. Even advanced versions, such as CBRS, still rely on rigid hierarchies.
But those simpler models are also running out of scope. Keeping fixed carve-outs or strict priorities is hard when missions are critical, applications evolve, and users move quickly. The edge cases are becoming normal.
What’s left is harder, but increasingly essential: actively monitoring and understanding the spectrum landscape, and developing coexistence methods that manage trade-offs rather than rigid boundaries. It shifts the focus towards in-operation modification and enforcement, rather than pre-judging everything.
This acknowledges a messier reality: a dynamic ecosystem where multiple users—old and new—must harness the airwaves without a permanent fixed hierarchy or unassailable rules.
In a sense, spectrum is becoming a crowded party: public 5G and emerging 6G, private 5G, WiFi and FWA, military radars, public safety, satellites, aviation, and maritime systems. Less-visible niches such as audiovisual systems and scientific observation are also important.
Some users are static, others are slow-moving, but the trickiest ones are fast and disruptive—airborne radars or low-altitude drones that force other users to vacate a band in a split-second. Some behaviour is predictable (satellites); other usage is intermittent. The challenge is optimising this heterogeneity without descending into interference chaos.
Today, the most common coexistence cases involve different technologies, such as radars and mobile networks (CBRS), or fixed links and WiFi (6GHz AFC). That will continue, but the mechanisms will need to get finer-grained: predicting and mitigating interference at a tighter level, and handling three or more coexisting systems, not just two.
US pressure and politics are making all this more salient. Under the One Big Beautiful Bill Act (OBBBA) mandate for the FCC to find 800MHz of new commercial spectrum, candidate ranges such as the 3.98-4.2GHz C-Band, 4.4–4.9GHz, and 2.7–2.9GHz are being studied.
These have multiple incumbents, including hard-to-clear tenants like military communications and weather radar. C-Band has both incumbents and “sensitive neighbours—aviation altimeters. The 3.1–3.45GHz band is tougher still: it includes existing military operations and is likely tied to future defence developments such as the proposed “Golden Dome” missile defence system. That makes it an extremely challenging coexistence target, yet also an obvious proving ground for leading-edge trials.
A newer class of coexistence is where multiple users exploit the same underlying technology. 5G and future 6G networks (3GPP / OFDM) could be operated by:
Sometimes coexistence can be managed “above” the radio layer—roaming, shared RAN, or one radio network connecting to multiple cores. But in many cases, that coordination is not practical: separate systems will still need to coexist in the same frequencies (or close neighbours) via coordination, constraints, or sometimes pre-emption.
To these scenarios, we must also add receive-only users (radioastronomy) and occasional but very powerful users (meteorological radar). Spectrum-usage awareness and adaptability become unavoidable.
Static databases and fixed protection areas are not sufficient when incumbents are mobile and fast-moving. Three or more coexisting user groups mean rigid rules need to become dynamic optimisation engines.
The core toolkit is:
In the US, the NTIA’s National Spectrum Strategy has pushed advanced dynamic sharing demonstrations (notably in lower 3GHz), and the National Spectrum Consortium is funding large-scale demonstrations under its ASC (Advanced Spectrum Coexistence).
It is also important to distinguish a coexistence transition from steady-state operation:
Sensing can be basic (energy detection for occupancy) or sophisticated (feature/waveform detection and geolocation tied to databases). Some functions can be embedded in mainstream infrastructure like cell towers. Others need dedicated installations with higher dynamic range, stronger security/isolation for sensitive sources, and tight synchronisation across sensor grids.
Regulators need occupancy data to inform allocations and detect unauthorised use. Military sites need continuous monitoring to identify threats or compromises. Critical infrastructure operators need to detect subtle interference that could degrade essential services—especially when there are more legitimate emitters in the mix, plus more opportunities for anomalies.
For the most advanced coexistence models, specialised hardware can be deployed at hotspots like airports or nationwide in monitoring stations and mobile units. There are also complexities for protecting passive incumbents, such as receive-only satellite earth stations or scientific observation sites.
The shift from sharing to coexistence is not wordplay. It’s a recognition that spectrum management must accommodate complexity, rather than forcing simplicity. That drives a parallel evolution in management tools: sensing, analytics, and AI become central; databases and workflows become more dynamic and programmable; and policy needs to move from fixed priorities to pragmatic adaptability.
Ultimately, coexistence demands we rethink spectrum not as a scarce resource to hoard, but as a commons to optimise across multiple stakeholders. This won’t yield “perfect harmony”, but it can deliver workable, evolving mechanisms that let multiple stakeholders use crowded spectrum bands without pretending the old clear-and-auction model can keep running forever.
Dean Bubley, founder of Disruptive Analysis, writes guest posts for CRFS. He is an independent analyst and advisor to the wireless and telecoms industry and has covered the evolution of private cellular networks since 2001.
Join thousands of professionals who subscribe for exclusive insights and early access to our reports and analysis.
