EIC Accelerator Winner Radar Reticence and the Car Radar Race

A car radar works by listening for the echo of its own signal, and on a busy road it increasingly hears everyone else's as well. Radar Reticence, a Swedish company built on radar research at Halmstad University, says it has a waveform that keeps those signals from colliding. According to the CORDIS project record, its "patented dual-sweep modulation combined with a signal association method" supports "1,000+ radar units per GHz under realistic automotive conditions", runs as firmware on radar chips already on the market, and needs no coordination between vehicles.

The EIC Accelerator selected the project, SIFFRA, for blended finance in the round it announced on 15 June 2026, and the grant agreement was signed on 28 September 2026. CORDIS records an EU contribution of €2,495,062.50 for 1 November 2026 to 31 October 2028. The project started three days before this article was written, so what follows is an account of what was promised and what can already be checked, not of what was delivered.

Why car radars jam each other

Most automotive radars sweep a frequency ramp, a chirp, across the same band and compare what comes back with what they sent. When another car's radar sweeps the same frequencies at the same moment, its signal can land in the receiver as a false target or bury a real one. The CORDIS objective describes the result as "missed detections, false alarms, and reduced reliability" of driver assistance systems, and argues that the problem grows as cars move to higher automation levels with more radar sensors each.

The two researchers behind the company have worked on this for years. Hans Hellsten and Emil Nilsson are among the authors of a 2019 survey of radar interference mitigation for automated driving, which notes that most automotive radars share one frequency band and proposes several mitigation methods. LEAD, the Linköping incubator that hosts the company, quotes Frida Wygler describing the product as "about efficiently using the frequency bands and eliminating interference."

The project's case against existing fixes is that digital and imaging radar architectures and software post-processing are "costly, compute-intensive, and inherently reactive": they clean up interference after it has arrived. Radar Reticence says its waveform avoids it at transmission. That is the company's characterisation of its competitors, and no independent comparison has been published.

What can be checked so far

The most concrete evidence comes from a Swedish innovation agency project. Vinnova's record of SCARCE (reference 2024-02453) shows SEK 2,176,803 of Vinnova funding for a project coordinated by Radar Reticence with NIRA Dynamics and Halmstad University as partners, from November 2024 to May 2026, and lists it as completed. The project members write that it "achieved its original goals" and that the technology moved from the lab and simulation to working prototypes mounted on vehicles, tested on test tracks and in real traffic, with more testing at Mantorp Park than at AstaZero as first planned. Vinnova notes that this description was written by the project members and not reviewed by its editors. LEAD's March 2025 profile separately reported tests under way at test facilities with NIRA Dynamics and Halmstad University.

No test data from SCARCE has been published, and the 1,000-radar figure appears in the CORDIS objective and on the company website but in no independent measurement found in a search on 4 November 2026. The company says its technology rests on patented research from Halmstad and Linköping universities; its 2022/23 annual report records the decision to file an international PCT application. A patent number could not be confirmed in the searches made for this article.

Industry exposure is documented at events rather than in contracts. EFN filmed the company presenting at MobilityXlab's Tech Day in Gothenburg in September 2025, where Wygler called the method "helt störningsfri", completely interference-free. In April 2026 Jönköping University reported that she had been named to Forbes' European 30 Under 30 list in the manufacturing and industry category. The company has announced no customer, licensee or carmaker partner.

1. Vehicle-mounted prototypesReportedTested on tracks and in traffic in the Vinnova SCARCE project, per the project members; no data published.
2. 1,000+ radars per GHzClaimStated in CORDIS and on the company site; no independent test found.
3. Firmware on existing radar chipsNot shownNo chipmaker or Tier-1 integration announced.
4. First customer pilotPlannedNamed as the use of the 2026 money by LEAD; no pilot partner named.
5. IP licensing to Tier-1 suppliersToo earlyThe business model in CORDIS; no licence announced.

A small company with a large grant

On 1 July 2026 the company announced SEK 37 million of new capital. LEAD's report breaks it into a SEK 10 million share issue and the €2.5 million EIC grant, with Chalmers Ventures, East Sweden Capital and Almi Invest as new investors next to LiU Invest, Halland Invest, Augurium and the chair, Leif Haglund. Dealroom News reported the same round in dollars and described the task as moving from technology readiness level 6 to 8. The money is for the first customer pilot and for doubling the team within a year; Wygler said it "gives us the opportunity to scale our operations, carry out our first customer pilot, and grow our team." MarkLines put the staff at six at the time.

The EIC's own list of the round carries a caveat: selection "does not constitute a formal commitment for funding", which depends on grant preparation. In this case that step is complete, since CORDIS records the agreement as signed. Whether the EIC Fund will take equity under the blended finance award has not been announced, and no amount for an equity component is public.

The filed accounts show how far the EIC grant outweighs what came before. The 2024 annual report records zero net sales, SEK 545,084 of operating income made up of capitalised development work and other income, a net loss of SEK 175,139, equity of SEK 3,075,829 after a share issue, cash of SEK 4,067,154 and an average of one employee. The first, extended financial year ended with a loss of SEK 8,032 and total assets of SEK 35,968. The 2024 report also records that the company bought shares in Comnisens AB and started two Vinnova projects. Accounts for 2025 could not be found in the searches made on 4 November 2026.

Why it matters for European car safety

Radar is one of the sensors behind features that EU law now requires. Under the General Safety Regulation, Regulation (EU) 2019/2144, new cars must carry advanced emergency braking; Germany's transport ministry sets out the stages, with detection of vehicles required for all newly registered cars from July 2024 and of pedestrians and cyclists from July 2026, which the Luxembourg motoring club ACL also reported. More sensors on more cars make interference a reliability question for systems that are no longer optional. The CORDIS objective frames SIFFRA as "a key enabling mobility technology developed and controlled within the EU", which is the project's own strategic claim.

Radar Reticence was not asked for comment. The reporting is based on publicly available sources, listed below.