Europe and the United States both moved public safety radio to digital trunking in the 1990s, and the different ways they did it set up different problems for the next thirty years. Europe wrote TETRA through ETSI and placed it in a harmonized emergency services band just below 400 MHz. The United States wrote P25 through its users and laid it over a licensing system with many bands and thousands of separate licensees. This piece compares those two starting positions and explains how much of each region’s interoperability record follows from them.

Two standards from one decade: what ETSI and the P25 effort set out to build

The European Telecommunications Standards Institute was founded in 1988, and its work on a digital trunked radio standard for professional users began around 1990. The standard was first called Trans-European Trunked Radio and later renamed Terrestrial Trunked Radio, which kept the TETRA acronym while dropping the geographic claim. The core voice-plus-data standards were published in the mid-1990s. Public safety was the most prominent intended user, although TETRA was written for professional mobile radio generally, so transit systems, utilities, airports and industrial sites adopted it as well. Technically, TETRA divides a 25 kHz carrier into four time slots, which gives four voice paths per carrier, and it includes a direct mode so that terminals can talk to each other without infrastructure.

Project 25 began in 1989 as an effort led by APCO with state telecommunications directors and federal agencies, and its standards are published through the Telecommunications Industry Association as the TIA-102 series. The origin story, including the proprietary trunking problem it was meant to fix, has its own article in this section, so I will only summarize the technical shape here. Phase 1 uses FDMA, one voice path in a 12.5 kHz channel, and it works on both conventional and trunked systems. Phase 2 uses two-slot TDMA within that same 12.5 kHz channel for trunked operation, which brings it to the same spectrum efficiency per voice path that TETRA reaches with four slots in 25 kHz.

Vendor comparisons tend to dwell on these choices, but the comparison that explains outcomes sits outside the standards documents. ETSI was writing for a region that was about to harmonize a band for emergency services, so a TETRA terminal could be designed around one frequency range. The P25 committees were writing for a country where an agency might be on low band, VHF, UHF, T-band, 700 or 800 MHz depending on its history, which meant P25 had to work in all of them and could standardize only the air interface, never the frequency. My own work has been on the P25 side, including the reference I wrote on P25 systems, so I am describing TETRA from the published record rather than from time spent running one.

A harmonized band near 400 MHz: what the European “mandate” actually was

A common shorthand holds that Europe mandated TETRA across the continent, and that is a myth as usually stated, because no Europe-wide law required any country to adopt a particular radio standard. What happened was narrower and more useful. Through CEPT, the European conference of postal and telecommunications administrations, the member administrations agreed in the mid-1990s to designate 380 to 385 MHz paired with 390 to 395 MHz for digital land mobile systems serving the emergency services. That range had been largely military, and making part of it available required agreement from the defense side as well. Which standard went into the band was left to each country’s own procurement.

Most countries chose TETRA, though not all of them did. France built its national police and gendarmerie networks on TETRAPOL, a competing digital standard developed in France, and Switzerland’s Polycom network for its security and rescue services is also TETRAPOL-based. Anyone who writes that “all of Europe” runs TETRA has erased those systems, which serve large national populations. The accurate statement is that Europe harmonized a band and most national administrations put TETRA in it.

Harmonization still did a great deal of work. Because the emergency services in most of Europe sit in the same narrow slice of spectrum, a manufacturer can sell essentially one radio-frequency design across many national markets, and a terminal from one country can physically tune the frequencies used next door. Propagation in that band behaves similarly everywhere, so coverage planning methods transfer across borders. The band also arrived early in the digital transition, before national networks were built, so countries were filling an allocation that had been cleared for the purpose rather than squeezing a new system between incumbents.

The United States had nothing comparable. Its public safety spectrum accumulated band by band over most of a century, each band added when the last one filled, as the article in this section on the 1996 PSWAC report describes. When P25 arrived, there was no single band to design for and no prospect of moving every agency into one.

One network per country: how a common band became national systems

Harmonized spectrum made it practical for a country to build one network for all of its emergency services, and many did. The United Kingdom’s Airwave network carries police, fire and ambulance traffic on TETRA. Germany’s digital radio for its public safety authorities runs on TETRA under a federal agency, BDBOS, created to manage it. The Netherlands built C2000, Belgium built ASTRID, and Finland built VIRVE. Ownership and operation vary, since some networks are government-run and others are delivered by a contractor under a long service agreement, but the pattern of one national infrastructure serving most or all emergency users is common.

That design makes most domestic interoperability a configuration question. When a police patrol and an ambulance crew are registered on the same network in the same band, whether they can talk depends on how talkgroups and permissions are provisioned centrally, and the radio hardware is no obstacle. A national operator can define standard cross-agency talkgroups once and push them to every terminal in the country. The equivalent arrangement in the United States requires agreements among separately licensed agencies, each with its own system owner, budget and programming practices, which is the subject of the article in this section on why the MOU is harder than the hardware.

The cost of that design is concentration. A single national network means that procurement, operation and eventual replacement are each handled as one national project, usually with one operator holding the contract. The United Kingdom’s program to replace Airwave with an LTE-based Emergency Services Network has run years behind its original schedule, and reports from the National Audit Office have documented those delays and the cost of extending Airwave in the meantime. Readers who want the current status should go to those reports rather than to press summaries, because the dates and figures have moved repeatedly. The underlying point is structural: a country that consolidated onto one network also consolidated its exposure to one contract and one program.

Many bands and many licensees: the ground P25 had to stand on

In the United States, the FCC licenses non-federal public safety users one licensee at a time, with applications passing through certified frequency coordinators, and NTIA manages the separate spectrum used by federal agencies. In a single county, the sheriff, the city fire department, the state patrol and a regional transit authority may each be a separate licensee, and each makes its own decisions about band, vendor and replacement timing, constrained mainly by what spectrum is available and what its budget allows. Statewide trunked systems, county systems, city systems and stand-alone conventional repeaters all coexist, sometimes covering the same ground. There is no national operator, and no federal agency has authority to tell a city which system to join.

P25 was built for that environment, which is why it specifies the common air interface and the system interfaces without tying any of them to a frequency. P25 equipment is available across the main land mobile bands, and a Phase 1 radio from one manufacturer can operate on a P25 channel installed by another, provided both are on the same band. That proviso is where most of the US interoperability story lives. Two P25 radios on different bands cannot hear each other any more than two analog radios on different bands could, because the standard governs what is transmitted, not where.

Multiband portables, which cover two or more bands in one radio, have reduced the problem for agencies that can afford them and that program them carefully. They do not remove it, since a multiband radio still needs the right channels, talkgroups and permissions loaded before it reaches the scene. The one place the United States did tie a band to a required standard was the 700 MHz nationwide interoperability channels, where the FCC adopted P25 on the recommendation of the National Coordination Committee after 1999. That decision covered a set of designated interoperability channels within one band, a far smaller footprint than a harmonized national band for all emergency services.

A common air interface: what it fixes and what it leaves alone

The common air interface fixed the problem P25 was written to fix. Before P25, trunked systems in the US were largely proprietary, so an agency that bought one manufacturer’s system was effectively bound to that manufacturer’s radios for the life of the system, and two trunked systems on the same band might not work together at all. P25 made it possible to buy radios from more than one supplier for the same system and gave every P25 radio a common vocoder and a common way to operate on conventional digital channels. DHS later established the Compliance Assessment Program so that agencies spending federal grant money had a way to check vendor claims of P25 compliance against test results, which addressed a real problem with “compliant” being used loosely in sales literature.

What the air interface leaves alone is everything that sits above it. Trunked P25 systems control which radios may register, and an outside radio has to be provisioned with the system’s identifiers before it will affiliate. Encrypted talkgroups require the same algorithm and the same keys, which in turn require someone to agree on how keys are shared and managed. The split between Phase 1 and Phase 2 matters whenever an older fleet meets a newer system. The codeplug in each radio, meaning the programming that decides which channels and talkgroups it can reach, is set by each agency’s radio shop and is often the reason interoperability fails in practice even when every radio on scene meets the standard.

TETRA users meet most of the same issues once the band question is removed. A TETRA terminal must be provisioned on a network’s subscriber management before it can register, so a terminal from one national network does not simply join another network because the frequencies match. Encryption on TETRA likewise depends on algorithm and key management, and direct mode needs agreed channels and settings. The harmonized band removed one large barrier for Europe, and the remaining barriers are recognizably the same administrative and configuration problems a US radio administrator manages, appearing at a different scale.

ISI and ISSI: connecting systems when one network is not enough

Both standards eventually had to define how separate infrastructures connect, because neither region ever ended up with a single system covering everyone. TETRA has the Inter-System Interface, known as ISI, for connecting one TETRA network to another. P25 has the Inter RF Subsystem Interface, known as ISSI, for connecting P25 trunked systems, along with a console interface that lets dispatch positions from one manufacturer work with infrastructure from another. Both interfaces are meant to carry calls between systems while keeping features such as unit identification and, where configured, encryption.

Where those interfaces get used differs because the network maps differ. In much of Europe, the number of national networks is roughly the number of countries, so the connections that matter are mostly at borders and between national networks. As far as the published material I have read shows, ISI links between national networks have arrived slowly and through specific bilateral or regional projects rather than as a default, and I would point anyone needing current detail to the operators or to TCCA, the industry association that grew out of the original TETRA MoU group. In the United States, the need for system-to-system connection exists at nearly every county line, between counties and the state system, and between local systems and federal users, so the volume of potential links is far larger.

When no standards-based link exists, both regions fall back on the same set of workarounds. A user can carry a second radio, a guest terminal or radio can be provisioned onto the neighboring system, or audio can be patched between systems through a console or a gateway device. Patches are common in the US because they work across bands and across standards, including analog, but they usually carry audio and little else, so unit identification is lost and end-to-end encryption is broken at the patch point. ISSI links preserve more, though they involve feature licensing on both systems, configuration work by both system owners, and a written agreement about who may reach whom, so the paperwork usually takes longer than the engineering.

The border, the contract and the county line: where each region’s problems landed

Putting the pieces together, most of the difference in interoperability outcomes follows from where each standard was deployed rather than from the standards themselves. Europe paired a common standard with a common band and then, in many countries, with a single national operator, so the domestic interoperability problem was largely solved by design before the first terminal shipped. The problems that remained were at the edges of that design, chiefly communication across national borders, dependence on one network and one contract per country, and the very large replacement programs that follow when a national network reaches the end of its life. The migration from TETRA voice toward mission-critical broadband is now one of those national programs in several countries, and its difficulties are the difficulties of a single large project.

The United States paired a common air interface with a fragmented spectrum and licensing system that the standard had no authority to change. P25 made multi-vendor procurement possible and gave neighboring agencies on the same band a common technical language, which was a genuine improvement over proprietary trunking. It could not put a VHF county and an 800 MHz city on the same frequencies, and it could not make five independent agencies agree on shared talkgroups, key management or who pays for a link. The US interoperability problem therefore stayed local, and it is worked out region by region through written agreements and the programming in each agency’s radios.

Neither arrangement is simply better, and I would be wary of any account that ranks them on interoperability alone. The European model concentrates both its strength and its exposure, so a well-run national network delivers domestic interoperability that a US region can only approximate, while a troubled national program affects a whole country at once. The US model spreads its exposure across thousands of licensees, so no single contract can stall the nation, at the price of leaving interoperability to local governance that varies widely in quality. A US agency studying TETRA networks should take from them the value of designing shared talkgroups and permissions centrally across a region, which local governance can do without any change in spectrum policy.

Security claims and the 2023 TETRA findings: what this comparison does not settle

This comparison says nothing about which standard is more secure, and I would not accept a blanket claim in either direction from a vendor, a consultant or a conference speaker. The security of any digital radio system depends on the encryption algorithm in use, how keys are generated and distributed, whether encryption is actually enabled on the traffic that needs it, and how the system’s management interfaces are protected. Two systems built on the same standard can differ enormously on all of those points. The article in this section on checking whether your encryption is actually on covers the most common practical failure, which is sensitive traffic going out in the clear on a system fully capable of encrypting it.

As a dated footnote, in July 2023 researchers at the Dutch firm Midnight Blue published a set of vulnerabilities they called TETRA:BURST, concerning TETRA’s air-interface encryption algorithms and related functions. ETSI and TCCA issued public responses, and the findings remain recent enough that their practical consequences for particular networks depend on which algorithms and configurations those networks use. Operators of TETRA systems should follow guidance from their national authority and their supplier, which will be more current than anything I can write. These findings concern TETRA’s own algorithms and tell us nothing about how P25 compares.

P25 has had its own published scrutiny. A 2011 study by researchers at the University of Pennsylvania examined P25 as used by federal agencies and focused largely on how often traffic users believed was protected went out in the clear, and on the protocol’s exposure to interference, which I will leave at that awareness level. Spectrum efficiency is another claim that does not separate the two standards much, since TETRA’s four slots in 25 kHz and P25 Phase 2’s two slots in 12.5 kHz both work out to one voice path per 6.25 kHz. A procurement decision between digital standards in the US is in practice a P25 decision for most public safety agencies, driven by what neighbors run and by the 700 MHz interoperability rules, and an honest comparison with TETRA is useful mainly for what it shows about spectrum and governance.

What to do at your agency

• The radio administrator should write a one-page table this month listing every neighboring agency the department works with routinely, the band and P25 phase of that agency’s primary system, and whether the connection between the two systems is an ISSI link, a console patch, a shared programmed talkgroup, or nothing, so the agency’s actual position on the band map is on paper.
• The communications director should ask the system owner or vendor for a written statement of which P25 system interfaces, including ISSI and the console interface, are licensed and active on the system the agency uses, and file that statement with the agency’s interoperability documentation.
• The radio shop should pull the codeplug from one radio of each model in the fleet and confirm whether that model can operate on each neighbor’s system as recorded in the administrator’s table, then report any radio that is Phase 1-only on a system where neighbors run Phase 2 talkgroups.
• The communications director should add one item to the next meeting of the regional radio or interoperability committee that already exists: for each system-to-system link in the region, name the agency that owns the written agreement and the date it was last reviewed.
• The grants coordinator or purchasing officer should check the current radio procurement specification and confirm it asks vendors for Compliance Assessment Program documentation for each radio model offered, rather than accepting the phrase “P25 compliant” on its own.
• The training officer should add a ten-minute segment to the next scheduled radio refresher explaining that a shared standard does not make two systems interoperable, using the agency’s own neighbor table as the example.

Takeaways

• TETRA was written through ETSI beginning around 1990 and P25 through an APCO-led effort beginning in 1989, so the two standards come from the same decade but were written for very different spectrum environments.
• Europe did not legally mandate TETRA; CEPT harmonized a band near 380 to 395 MHz for emergency services, most countries chose TETRA for it, and France and Switzerland built national networks on TETRAPOL instead.
• A harmonized band let many European countries build one national network for most emergency users, which turned domestic interoperability into a matter of central talkgroup and permission design.
• P25 had to work across many US bands and thousands of separate licensees, so it standardized the air interface and system interfaces while leaving frequency, system access and agreements to local decision.
• Two P25 radios on different bands cannot talk to each other, and band, phase, provisioning, keys and codeplugs explain most US interoperability failures among radios that all meet the standard.
• Europe’s remaining problems sit at national borders and in the concentration of each country’s radio service in one network and one contract, as the delays in the UK’s Airwave replacement program illustrate.
• Neither standard can be called more secure as a blanket statement, and the TETRA:BURST findings published in July 2023 concern TETRA’s own encryption algorithms and should be followed through national authorities and suppliers.