City-wide V2X for connected passenger vehicles.
The recognisable V2X deployment — equipped cars, roadside units at junctions, a traffic-management centre behind them — is the one everyone pictures and the one that most often stalls. Not for technical reasons. It stalls because seven organisations have to move together, and the party paying for the roadside equipment is rarely the party receiving the benefit.
This page sets out what a city programme actually delivers, in the order the value arrives; which services need infrastructure and which do not; and the three things that make the Indian version of this problem structurally different from the European or North American one.
A city deployment is seven parties with different budgets, procurement cycles and reasons to care.
The technology in a city-wide programme is the easy part. The reason these programmes stall is that no single organisation owns the whole system, and the benefits do not land where the costs do.
| Party | Brings | Needs |
|---|---|---|
| Vehicle OEMs | On-board units in new vehicles, and the vehicle integration around them | A fitment case, a stable message and security profile, and a certification route |
| Road and city authorities | Roadside units, signal controllers, the traffic-management centre | Measurable network outcomes, and a maintenance model for equipment on street furniture |
| Signal and ITS vendors | Controllers, detection, and the interface that turns a signal plan into a broadcast message | A specification stable enough to build a product against |
| Public transport operators | Fleets that are already instrumented and already want priority | Reliable priority, and schedule adherence they can evidence |
| Emergency services | The use case with the clearest public benefit and the fewest objections | Pre-emption that works on the first attempt, at every equipped junction |
| Vulnerable road users | The largest share of urban casualties in Indian traffic | Protection that does not depend on them carrying or charging anything |
| Trust-service and network operators | Credentials, trust lists and the cellular path that delivers them | A published certificate policy, and a defined accountability model |
The asymmetry worth naming: an authority pays for roadside infrastructure, and most of the measured benefit accrues to vehicle occupants and to network efficiency. That is a normal public-infrastructure argument, but it has to be made explicitly, because a programme that assumes the business case is self-evident does not survive its second budget cycle. The measured evidence, and who buys which benefit →
Ordered by how soon it works, not by how impressive it sounds.
| Service | Needs on the street | Needs in the vehicle | Maturity |
|---|---|---|---|
| Signal phase and timing, lane geometry | RSU with a live controller feed | Receive-only is enough | Day 1 |
| Red-light and signal violation warning | RSU plus accurate intersection geometry | Receive, plus reliable own-position | Day 1 |
| Green-light optimal speed advisory | RSU with a predictable signal plan | Receive, plus a driver interface | Day 1 |
| Roadworks, lane closure and event information | RSU or a works-vehicle transmitter | Receive | Day 1 |
| Emergency vehicle approaching | Nothing, for the vehicle-to-vehicle case | Transmit on the emergency vehicle, receive on the rest | Day 1 |
| Signal pre-emption and transit priority | RSU and a controller that accepts a request | Transmit from the priority vehicle | Day 1 |
| Queue, congestion and stationary-vehicle warning | Optional; improves with an RSU | Transmit and receive | Day 1 |
| Intersection collision warning | Optional; better with infrastructure | Transmit and receive, with good positioning | Day 1 to Day 2 |
| Vulnerable-road-user protection | Infrastructure sensing, where user devices cannot be assumed | Receive | Day 2 |
| Cooperative perception at junctions | RSU with sensors and object fusion | Receive, plus the ability to use an object list | Day 2 |
| Network-wide traffic management | RSUs, backhaul and a traffic-management centre | Cellular connectivity | Day 2 |
| Dynamic intersection management | Controller able to act on vehicle requests | Transmit intent | Day 3 |
Two structural observations follow from the middle column. First, the first tranche of value needs receive-only vehicles, which is the cheapest possible fitment and the reason infrastructure-first programmes make sense despite the penetration problem. Second, vulnerable-road-user protection cannot be built on the assumption that pedestrians carry a transmitting device — in a city where most casualties are pedestrians, riders and cyclists, the infrastructure has to do the detecting and the collective-perception message has to do the telling. The V2P mode, and its honest limits →
Two-wheelers first, and infrastructure operated by government.
An Indian city deployment differs from a European or North American one in three ways that change the design rather than the marketing.
The fleet is dominated by powered two-wheelers. MoRTH's draft mandate covers categories L, M and N — which, if finalised, would make it the first national V2V fitment requirement anywhere to include two-wheelers. That changes the antenna problem, the power budget, the human-machine interface and the cost ceiling, and it makes the motorcycle-specific use cases in the ETSI catalogue — filtering, overtaking, blind-spot interaction with a turning vehicle — central rather than peripheral. The draft mandate and its timetable →
Roadside operation is proposed to be restricted. The TRAI consultation, following MoRTH correspondence, indicates authorisation limited to central and state governments and their authorised agencies, with private entities considered only for non-safety applications. A private organisation planning to operate roadside units in India should read that carefully; supplying equipment to an authorised operator is a different proposition from operating it. Why an RSU is a trusted insider →
The trust framework is not published yet. India's existing digital-trust regime recognises X.509 certificates, and V2X uses IEEE 1609.2 and ETSI ITS formats. Until a certificate policy exists, a city programme cannot specify credential lifecycle, provisioning or revocation for the equipment it is buying — which is precisely why the policy should precede the first tender rather than accompany it. The trust-system position in full →
Also worth building into a programme from the start: spectrum and equipment approval are separate things at different widths, so an on-board unit that is licence-exempt is still subject to equipment type approval, and roadside units are not covered by the on-board exemption at all. The allocation and the exemption, side by side →
Corridors before coverage, and a test domain before either.
- A test trust domain. Separate credentials, separate roots, certificates that cannot cross into production. Every mature programme has one, and the ones that skipped it regretted it.
- A pilot corridor. A defined route with instrumented junctions, a known fleet and measurable outcomes — rather than a thin layer of equipment spread across a whole city.
- Signal services first. Phase, timing and geometry deliver value to receive-only vehicles and force the hardest integration — getting live controller state out of the traffic system — to be solved early.
- Priority and pre-emption next. Transit and emergency vehicles are a small, professionally maintained transmitting fleet with a clear benefit case.
- Infrastructure perception where the casualties are. Junctions and crossings with a vulnerable-road-user record, rather than uniform coverage.
- System-level interoperability testing. Roadside and on-board units tested together, not only certified apart. Component conformity does not prove that two vendors' equipment cooperates. The laboratory gap →
- Then extend. Coverage is the last step, not the first.
Worth stating because the counter-example is well documented: China built more than eleven thousand roadside units and ran twenty pilot cities, and vehicle penetration remained around four percent. Infrastructure alone does not create a connected fleet. What the Chinese programme actually shows →
Equipment and integration for the parties who deploy, not the deployment.
Ambimat develops and customises the endpoints and the engineering around them for vehicle OEMs, tier suppliers, ITS and infrastructure vendors, research programmes and system integrators: on-board and roadside development platforms, embedded and hardware engineering, device-side security and credential integration, and application and integration work scoped per programme. V2X development services →
Ambimat does not operate city networks or traffic-management centres, does not bid as a turnkey public-infrastructure operator, and is not a certification, homologation or regulatory-affairs consultancy — type approval sits with ARAI or the appropriate authorised body. The customer in a city programme is the party that holds the authorisation; the work here is the equipment they deploy and the engineering that makes it fit.
Where this fits.
V2X use cases
The full catalogue, and the two other operating environments.
V2I
What a roadside unit broadcasts, and the integration surface behind it.
India regulation
TRAI, the OBU licence exemption, the draft mandate and AIS-230.
Spectrum
The designation, the exemption, and what still needs equipment approval.
AmbiRSU
The roadside development platform and the authorisation question around it.
Test and certification
The accredited-laboratory capacity a city programme depends on.
Questions this page answers.
What does a city-wide V2X deployment actually deliver first?
Signal phase and timing with lane geometry, in-vehicle information, roadworks and lane-closure warning, and signal priority for transit and emergency vehicles. These are the services that deliver value to receive-only vehicles, which is the cheapest possible fitment, and they force the hardest integration — getting live signal-controller state out of the traffic system — to be solved early.
Do pedestrians need a device for V2X to protect them?
Not if the system is designed properly. Vulnerable-road-user awareness messages assume a transmitting device, which cannot be relied on in a city where most casualties are pedestrians, riders and cyclists. The alternative is infrastructure sensing at junctions with a collective perception message telling approaching vehicles what was detected, which requires nothing of the person being protected.
Who can deploy roadside units in India?
Not finally settled. The TRAI consultation, following MoRTH correspondence, indicates authorisation restricted to central and state governments and their authorised agencies, with private entities considered only for non-safety applications. TRAI's recommendations were still pending as of mid-August 2026. Supplying equipment to an authorised operator is a different proposition from operating it.
Is building roadside infrastructure enough to create a connected fleet?
The available evidence says no. China allocated spectrum in 2018, built more than eleven thousand roadside units and ran twenty pilot cities; as of 2025, C-V2X penetration in Chinese vehicles was around four percent across fewer than thirty mass-production models. Infrastructure and vehicle fitment are separate problems and need separate instruments.
Last updated 2026-09-07 · Technical reference maintained by Ambimat Electronics, Ahmedabad, India. Corrections: neel.shah@ambimat.com