Public safety broadband gets sold in two directions at once, either as the thing that finally solves interoperability or as an expensive phone plan with a logo on it. Neither description survives contact with a working incident. What a dedicated public safety broadband effort actually delivers is spectrum, a defined priority and preemption mechanism, a deployable fleet, and a governance structure, and each of those has limits worth knowing before you write it into a plan. This piece covers where the effort came from, what priority and preemption do at the radio level, and the honest answer on whether any of it replaces land mobile radio.

Why a dedicated effort happened at all

The political argument for a nationwide public safety broadband network came out of the same body of post-2001 findings that drove the spectrum fights of that decade. The 9/11 Commission Report, published in 2004, recommended that Congress support legislation providing for what it called the expedited and increased assignment of radio spectrum for public safety purposes, and that recommendation was cited for years afterward by every organization pushing for a public safety allocation at 700 MHz. The Commission was writing mainly about voice interoperability, and the network that eventually got built is a data network, so the lineage is real but it is not a straight line.

The operational argument was different and, in my experience, more persuasive to the people who actually run incidents. Commercial cellular is engineered for the average busy hour in a population center, and it behaves poorly when a hundred thousand people converge on one sector and start uploading video. Responders on scene are competing for the same sector with everyone who is filming the scene. That pattern was reported after the September 11 attacks, after Hurricane Katrina, after the 2013 Boston Marathon bombing, and after most large-attendance events since, and it does not require any conspiracy to explain, because a sector has a finite number of resource blocks and the network hands them out.

One story from Boston is worth labeling as it circulated. On the day of the bombing, reports appeared saying that carriers had been ordered to shut down cellular service in the area to prevent remote detonation. Carriers publicly denied that any shutdown occurred and attributed the difficulty to congestion. That shutdown version was never confirmed, and it should be treated as an unverified claim rather than repeated as history, though it did real work in the policy debate at the time by making the case that responders needed something they controlled.

The D Block, Band 14, and how the spectrum got there

The 700 MHz band became available because television broadcasting moved off those channels in the digital transition. Congress allocated a block of 700 MHz spectrum to public safety in the late 1990s, and part of that allocation became the narrowband channels that many agencies now use for conventional and trunked land mobile radio in the 769 to 775 and 799 to 805 MHz ranges. A separate wideband and then broadband portion sat adjacent to it, and the question of what to do with it consumed most of the decade after 2001.

The FCC’s attempt to solve it in the 700 MHz auction that concluded in 2008 involved a block known as the D Block, which was offered with a condition requiring the winner to build and share a network with a public safety broadband licensee. The bidding did not reach the reserve price and the block went unsold, which left the shared-network concept without a commercial partner and pushed the whole question back to Congress. Public safety organizations then spent several years lobbying to have the D Block reallocated directly to public safety rather than reauctioned.

That is what the Middle Class Tax Relief and Job Creation Act of 2012, signed in February of that year, did. The Act reallocated the D Block to public safety, combined it with the existing public safety broadband allocation to produce a 20 MHz block licensed to a new entity, created the First Responder Network Authority as an independent authority within the National Telecommunications and Information Administration, and directed that auction proceeds fund the buildout along with a state and local implementation grant program. The 20 MHz became 3GPP Band 14. The Authority then ran a competitive procurement and awarded a twenty-five year contract to AT&T in March 2017, and every state and territory elected to participate rather than build an alternative radio access network, with the last decisions landing at the end of December 2017. Anyone writing a plan around the program should read the Authority’s own account of that history rather than a vendor summary, because the statutory structure matters more than the branding.

Priority and preemption: two different mechanisms

These two words get used interchangeably in sales material and they are not the same thing. Priority means that when the network is deciding who gets scheduled and who gets admitted, your session is handled ahead of sessions with a lower priority marking. In LTE this rides on per-bearer quality of service parameters, including a class identifier that describes the traffic type and an allocation and retention priority value that carries a priority level along with flags for whether that bearer may preempt others and whether it may itself be preempted. In 5G the naming changed and the concept did not.

Preemption is the part that matters when the sector is already full. Priority alone helps you at the moment you try to establish a session, because the admission control function will take you before it takes a lower-priority request. If the sector is saturated with sessions that are already established, priority by itself leaves you waiting behind them. Preemption means the network will actually tear down or degrade lower-priority bearers to make room for yours. A public safety subscriber on the dedicated network is configured with both, so commercial traffic on shared spectrum can be pushed aside, and Band 14 itself is public safety spectrum that commercial subscribers are only using on a secondary basis in the first place. There is also a related mechanism at the access layer, where certain access classes are exempt from the barring the network applies when it needs to shed load, so a public safety device can still get onto a cell that is turning ordinary consumers away.

What none of this does is create capacity that does not exist. If the sector serving your incident has lost its backhaul, or the site is on generator and has shed its remote radio heads, or the site is simply not there because nobody built one on that ridge, preemption has nothing to work with. I have watched people assume that priority means guaranteed service, and the distinction is worth being blunt about with your command staff, because the same failure that takes out the site takes out your priority along with it.

Local control: who decides who gets uplifted

The dedicated network distinguishes between categories of user. Primary users are the traditional first responder disciplines and get priority and preemption as a standing condition of service. Extended primary covers organizations with a direct role in emergency response, which in practice has included utilities, transportation agencies, hospitals and some public works functions, and the treatment of those users is set by policy rather than being automatic. Beyond that sit users who have service but not the same standing.

The feature that gets underused is the ability of a designated agency official to change priority profiles for specific users during an incident, which is generally called uplift. An agency administrator can raise a group of users for the duration of an event so that, for example, the building department inspectors doing rapid damage assessment after a tornado are not competing at consumer priority. The mechanics are done through an administrative portal, and the practical problem is almost never technical. It is that the person named as the authorized official retired two years ago, or that nobody at the agency has logged in since the account was created, or that the only person with credentials is on vacation when the incident starts.

Treat the uplift authority the way you treat any other delegated authority in an incident. Someone should hold it, someone should back that person up, the credentials should be recoverable outside of one individual’s memory, and the process should get exercised at least once a year on something real like a large planned event. If your county has a communications unit leader or a designated radio administrator, that role is the obvious place to put it, and the decision should be written into the same document that names who approves talkgroup assignments.

Deployables: what they are and what to ask before you request one

A deployable is a portable cell site brought to an incident to create coverage or add capacity where the fixed network cannot. The common forms are a cell on wheels towed behind a truck and fed by terrestrial backhaul, a satellite cell on light truck that carries its own satellite backhaul and therefore needs no local connectivity at all, and smaller compact units that can be flown in or carried in a pickup bed for a single building or a small incident footprint. Aerial platforms have been demonstrated and used, and the fleet composition changes, so ask for the current inventory rather than relying on a count you read somewhere.

The program’s deployable fleet is one of the things a dedicated public safety network buys that a commercial plan generally does not, and it comes with a request process staffed by people whose job is to take that call at three in the morning. The number of units in the fleet has grown steadily since 2017 and has been publicized in the low hundreds in recent years, though I would get the current figure and the current request procedure from the network operator or the FirstNet Authority directly, because both have changed more than once.

The questions to settle before you need one are logistical. Find out who at your agency or in your region is authorized to request a deployable and what information the request requires, because a dispatcher who has never done it will lose an hour figuring it out. Find out the realistic travel time from wherever the nearest unit is staged, which in a regional disaster is competing with every other requester. Establish what the unit needs on arrival, meaning a footprint with clear sky for the satellite dish, a mast keepout distance from power lines, fuel resupply, and physical security if it is sitting in a parking lot for a week. Understand that satellite backhaul on a geostationary link adds meaningful round trip delay, which data traffic tolerates and push to talk audio does not tolerate as gracefully, so if you are running PTT over a satellite-backhauled deployable the feel of it will be different from what your people are used to.

What broadband does not replace

The honest answer is that broadband has not replaced land mobile radio for mission critical voice anywhere I am aware of in the United States, and the reasons are structural rather than a matter of the technology needing a few more years. The first is off-network operation. A portable radio with no infrastructure at all still talks to another portable radio a few hundred feet away on simplex, which is the entire basis of fireground communications and of every plan for what happens when the repeater fails. An LTE device without a network is a handheld computer with no path to anyone. The 3GPP standards do define device to device operation, and mission critical push to talk was standardized in the Release 13 timeframe with mission critical video and data following, but the off-network portion has seen limited deployment in commercial public safety devices, so verify what your specific device and service actually support rather than assuming the standard implies the feature.

The second is the behavior of the voice path itself. A trunked radio system grants a channel and delivers audio to every subscriber on the talkgroup with a delay measured in a couple hundred milliseconds, and it does that identically whether five radios or five hundred radios are affiliated. The 3GPP mission critical requirements set access time targets in the same general range, and well-implemented systems can meet them, but the end to end chain over a commercial radio access network includes handovers, congestion, application servers and sometimes a bridge to the radio system, and each of those adds variability. On a fireground, variability in when the audio arrives is a safety problem rather than an annoyance.

The third set of reasons is about how the two kinds of network are engineered and funded. Public safety radio sites are typically built to a coverage design standard tied to a service area boundary, with hardened shelters, substantial battery reserve, generators and redundant paths, because the agency specified them and paid for them. Commercial sites are built where the subscribers are, and site hardening varies. Neither answer is wrong for its purpose. The practical consequence for a rural fire district is that the LMR system covers the territory it was designed to cover, and the broadband coverage map covers the highway and the town. Devices matter too, because a structural firefighting portable is built and tested for an environment that a smartphone is not built for, whatever the case on it claims.

Tying LMR and LTE together without creating a new failure

The realistic architecture for the next decade is both, connected. P25 systems support standardized interfaces for interconnecting systems and consoles, commonly referred to as the ISSI and CSSI, and most of the push to talk applications sold to public safety can be bridged to a talkgroup through one of those interfaces or through a simpler gateway. That gives you a supervisor on a phone in a car hearing the tactical talkgroup, a mutual aid partner on a different system patched in for an event, and a way to get a few people onto the incident without issuing radios you do not have.

The failure modes are worth naming before you build it. A bridge adds audio delay, and a user on the LTE side may key up over someone on the radio side without ever hearing them, which produces exactly the kind of doubling that makes a busy fireground channel unusable. A bridge also creates a talkgroup path that your dispatch console may not display, so the incident commander can have people listening and transmitting who do not appear anywhere in his accountability picture. Recording is the other gap, because your logging recorder captures the radio system, and PTT application traffic may or may not be captured and retained in a form your records custodian can produce on a subpoena.

My rule for this is that a bridge should be built deliberately, documented on the incident radio communications plan, owned by a named person who can tear it down, and understood by the people using it as a secondary path. Where I have seen it go wrong, the bridge was stood up by whoever knew how, nobody wrote it down, and it was still running three days later carrying traffic that nobody was monitoring.

Other carriers, WPS and TSP, and how to check current status

The dedicated network is the statutory program, and it is not the only place to get priority. Other national carriers market public safety priority and preemption offerings, and agencies in areas where a competing carrier simply has better coverage have gone multi-carrier deliberately, carrying devices or routers that can reach more than one network. That is a defensible engineering decision, and the questions to ask are the same ones you would ask about any priority claim: does the offering include preemption or only priority, what categories of user qualify, who at your agency can change it, and what happens to your traffic at a congested sector that belongs to a partner carrier through roaming.

Separate from cellular data, the Cybersecurity and Infrastructure Security Agency administers a set of priority telecommunications services worth enrolling in, which most small agencies have never heard of. Wireless Priority Service provides priority call setup for cellular voice on participating carriers for authorized users, the Government Emergency Telecommunications Service does the equivalent for calls placed through the wireline network, and Telecommunications Service Priority provides priority for the provisioning and restoration of critical circuits, which is the one that matters when your 911 trunks or your microwave backhaul circuit are in the repair queue behind everybody else. Enrollment takes paperwork and lead time and is worthless if you start it after the storm, so check current eligibility and application procedures with CISA.

One item of genuine uncertainty deserves a plain statement. The 2012 Act as enacted contained a sunset provision for the FirstNet Authority fifteen years after enactment, which places that date in February 2027, and legislation to reauthorize or extend the Authority has been introduced in Congress in recent sessions. I am not going to tell you what the law says as of today, because this is precisely the kind of thing that changes between when an article is written and when it is read. Check the current status with the FirstNet Authority and with the congressional record before you build a long-term plan or a budget assumption on it, and get subscriber counts, deployable fleet size, coverage claims and contract terms from the Authority’s and the carrier’s own published material with a date attached.

What to do at your agency

• Have your communications officer identify by name the person who currently holds administrative access to your broadband carrier’s priority portal, confirm that person can log in this week, and name a backup in writing so the uplift capability does not live with one individual.
• Ask your carrier representative in writing whether your agency’s devices are configured with preemption or with priority only, which user category each group of your devices falls into, and get the answer as a document you can keep in the communications file.
• Put one item on the agenda of the next regional interoperability or 911 board meeting that already occurs: document the deployable request process, including who may request, what information is required, and the realistic response time to your county, and distribute that one page to dispatch.
• Have a supervisor drive your two worst known coverage areas with an LTE device and a portable radio, record where each one fails, and add the result to the coverage documentation you already maintain rather than filing it separately.
• Have your records custodian determine whether push to talk application audio used by your agency is captured by your logging system and how long it is retained, and write the answer into your existing records retention policy.
• Check with your 911 center manager and your IT vendor whether your agency is enrolled in Telecommunications Service Priority for its critical circuits, and if not, start the application through CISA this quarter rather than after the next outage.
• Add one paragraph to your existing communications SOP stating that fireground and tactical voice remains on the radio system, that broadband push to talk is a secondary path, and that any bridge between the two must be documented on the ICS-205 and owned by a named person.

Takeaways

• The Middle Class Tax Relief and Job Creation Act of 2012 reallocated the D Block to public safety, created the First Responder Network Authority within NTIA, and produced the 20 MHz of 700 MHz spectrum known as Band 14, after the 2008 auction attempt at a shared commercial network failed to reach its reserve price.
• Priority determines who gets admitted and scheduled first, while preemption allows the network to tear down or degrade lower-priority sessions already in progress, and an offering that includes only the first is materially weaker on a saturated sector.
• Neither priority nor preemption creates coverage, backhaul or generator runtime, so a site that is down takes your priority down with it.
• The ability to raise priority for specific users during an incident depends entirely on someone at your agency holding working portal credentials, which is an administrative failure point rather than a technical one.
• Deployable cell sites are a real capability of the dedicated network, and their value on the day depends on knowing the request path, the travel time, the site requirements and the backhaul delay before you need them.
• Broadband has not replaced land mobile radio for mission critical voice, principally because off-network device to device operation is limited in deployed equipment, because voice delay over a commercial radio access network varies more than a trunked system does, and because radio sites are engineered and hardened to a coverage standard the agency specified.
• Bridging LTE push to talk to a P25 talkgroup is useful and introduces audio delay, users invisible to the dispatch console, and recording gaps, so it should be documented and owned by a named person.
• Wireless Priority Service, GETS and Telecommunications Service Priority are separate programs administered through CISA that many small agencies have never enrolled in, and enrollment takes lead time.
• The FirstNet Authority carried a statutory sunset date in February 2027 under the 2012 Act as enacted, and the current legal status should be verified with the Authority and the congressional record rather than assumed from any article, including this one.