Existing licenses for dedicated short range communications equipment in the 5.9 gigahertz band may not be renewed beyond 14 December 2026. The deadline sits in the Federal Register notice of the Federal Communications Commission’s Second Report and Order, published on 13 December 2024 and effective on 11 February 2025, and applies to hardware that state transportation departments installed with federal encouragement over the previous decade. The connected vehicle story in the United States is easiest to understand as the interval between two deadlines nobody met.
It started with fog in Arkansas
In January 1995, nine vehicles collided in fog near milepost 118 on Interstate 40 at Menifee, Arkansas, and five people died. The National Transportation Safety Board investigated, concluded that the collision avoidance systems of the day were limited in exactly those conditions, and recommended that the FCC allocate a dedicated wireless band for collision warning technology. In 1999 the FCC allocated 75 megahertz in the 5.850 to 5.925 gigahertz range for intelligent transportation systems, and the safety board closed its recommendation as acceptable action.
What followed was two decades of research rather than deployment. The board’s February 2024 letter on the federal deployment plan lays out the sequence with the discipline of an organization that has been asking the same question since 1995, and it records that more than a million people have died on American roads since that recommendation was issued.
The strongest field evidence came from the Safety Pilot Model Deployment of 2012 and 2013. The safety board, citing the 2015 independent evaluation of light-vehicle applications, reports that across the eighteen-month deployment there were no missed forward collision warning alerts: of 368 rear-end near-crash scenarios encountered, the vehicle-to-vehicle devices alerted the drivers in every case, an unusually clean result in a field where most claims come from simulation.
The sequence of withdrawals
In January 2017 the National Highway Traffic Safety Administration proposed a rule to mandate vehicle-to-vehicle communication on new passenger vehicles, built on the dedicated short range standard. The industry side was moving in parallel. General Motors had equipped the Cadillac CTS with that technology from the 2017 model year, continuing through model year 2019, and announced in 2018 that it would expand across the Cadillac fleet from 2023. Toyota announced in April 2018 that it would begin equipping part of its fleet by 2021. Ford announced in 2019 that it would begin with cellular technology by 2022.
None of those three deployments happened. General Motors never initiated the announced expansion. Toyota suspended its plans roughly four months after the federal government issued a request for comments, in December 2018, asking about the differences between the two candidate radio technologies, and gave regulatory uncertainty and the absence of other manufacturers’ plans as its reasons. Ford’s plan did not occur and the company made no further announcement. In November 2023 the highway safety administration withdrew the 2017 proposal outright, reasoning that a revision could not reasonably be accomplished given the new communications protocol and the revised regulations governing the band.
The safety board’s summary of that decade is blunt: continued regulatory uncertainty from the department since 2018 led industry and other government agencies to lose faith in the deployment approach. Of its nine open recommendations to the department or its agencies at the date of that letter, four were classified as open with an unacceptable response, including the pair issued after a 2012 school bus collision in Chesterfield, New Jersey that killed one student and injured fifteen.
The spectrum decision, in the regulator’s own words
The FCC’s 2020 First Report and Order kept the upper 30 megahertz of the band, 5.895 to 5.925 gigahertz, for intelligent transportation systems using cellular vehicle-to-everything technology, and opened the lower 45 megahertz to unlicensed use. In 2024 the Commission restated its reasoning, and it amounts to the chicken-and-egg problem described by the party holding the eggs: the services once contemplated across the full band “had not come to fruition in the 20 years since it allocated the spectrum for the ITS service.” The order was affirmed on review by the D.C. Circuit in 2022.
The consequence for infrastructure owners was immediate, because most existing deployments were roadside rather than in-vehicle and were built on the older standard. The safety board reports that in 2021 there were 57 operational vehicle-to-infrastructure projects across 34 states, with more than 15,500 instrumented vehicles and 6,200 intersections and roadside units, a figure that deserves a caveat the board itself supplies in a footnote: it comes from a board member’s recorded conversation with a representative of the state highway officials’ association, which is a thinner provenance than the precision of the numbers suggests. It is the best public count available, and it is not a survey. Deployment counts across transportation technology coverage are mostly of that kind, which makes the evidence gap in this field a counting problem before it is a technology one.
Wyoming’s transportation department had built one of the largest of those deployments along Interstate 80 and described its situation to a legislative committee in December 2020 with striking economy. Three bullets on one slide: the older standard replaced by the cellular one, the majority of the spectrum moved to unlicensed use, and this.
Promise of connected and autonomous vehicles has lessened
The department’s response was engineering rather than complaint. New roadside and on-board units were purchased with dual capability, and the corridor was designed so that infrastructure-to-vehicle information could be delivered either by the short range radio or by satellite. A state that had been asked to bet on one radio standard built for the possibility of losing the bet. The freight case that justified the investment in the first place is set out in commercial truck safety.
The arithmetic of the current plan
The draft plan the safety board reviewed, released in October 2023, sets infrastructure targets the board supports without reservation: deployment on 85 percent of signalized intersections in the top 75 metro areas and full deployment on the National Highway System, in the 2030 to 2034 period. The vehicle side of the same draft calls for 20 vehicle models to carry 5.9 gigahertz capability by 2030 to 2034, which the board calculates at about 6 percent of the passenger vehicle models in the United States, and describes as too slow.
Set the two targets against the physics of the applications and the draft’s logic becomes visible. An infrastructure-to-vehicle warning works when one equipped vehicle passes one equipped signal. A vehicle-to-vehicle warning requires both vehicles to be equipped, which is why the applications with the cleanest evidence from the 2012 pilot are the ones that need fleet penetration the plan does not project. Fully instrumenting the National Highway System while 6 percent of models can hear it produces a network optimised for the weaker half of the safety case.
The interference question compounds it. Unlicensed devices now operate immediately below a 30 megahertz safety band, and the safety board records that a broad range of stakeholders, including vehicle manufacturers and federal transportation agencies, raised interference concerns with the FCC. For comparison it notes that the European Union has allocated 60 megahertz to these applications, with a further 20 megahertz shared with urban rail.
What December tests
The 2024 order gave incumbents two years from Federal Register publication to convert or cease operating, and the FCC’s stated reason for starting the clock at publication rather than at the effective date was that ample notice had already been given. Utah’s transportation department told the Commission that two years was reasonable but that anything shorter would be unacceptable, because of what replacing its own dedicated short range communications system would involve. Other commenters in the docket pointed instead to procurement, engineering, workforce training and installation as reasons a shorter timeframe would not work.
So the equipment installed at those 6,200 intersections faces a federally set expiry while the vehicle fleet that was supposed to justify it is a decade from 6 percent of models. The rest of that roadside estate has a page of its own, what intelligent transportation systems actually do. Whatever survives that date will be roadside units bought by agencies that decided, without a vehicle market to point at, that a corridor should keep talking. Wyoming made that decision in 2020 and hedged the radio.