Inter­na­tional Success: Austrian Re­search Part­ner­ship Sets New Stan­dards in Neuro­mor­phic Computing

Neuro­mor­phic Computing is an emerging tech­nology designed to make arti­fi­cial intel­li­gence (AI) faster and more energy-effi­cient. For the past three years, the Insti­tute for Signal Process­ing at Johannes Kepler Univer­sity Linz (JKU), Silicon Austria Labs (SAL), and the Soft­ware Compe­tence Center Hagen­berg (SCCH) have been working closely together, pooling their exper­tise to advance this exciting field.

Neuro­mor­phic Computing – Inspired by the Brain

Unlike clas­sical computers, which process infor­ma­tion in bits, the human brain uses short elec­trical impulses—known as spikes. What matters is the precise timing of these spikes, which carries the actual infor­ma­tion.

This timing-based repre­sen­ta­tion allows the brain to transmit more complex infor­ma­tion per signal compared to a digital bit, which can only be “0” or “1.” The inter­vals between spikes and their patterns carry addi­tional meaning, making this highly effi­cient for tasks such as detecting patterns or sequences in real time. In tech­nology, these prin­ci­ples are repli­cated by so-called Spiking Neural Networks (SNNs).

“Our goal is to develop systems that mimic biolog­ical struc­tures and processes—using minimal energy while achieving maximum respon­sive­ness,” explains Priv.-Doz. Dr. Bern­hard A. Moser, Tech­nology and Inno­va­tion Manager at SCCH. “Poten­tial appli­ca­tions include robotics, medical tech­nology, and envi­ron­mental moni­toring.”

Dispelling Common Myths

Despite growing interest, several miscon­cep­tions about Neuro­mor­phic Computing (NC) persist—such as the assump­tion that it requires expen­sive, special­ized hard­ware, is slower than tradi­tional methods, or lacks indus­trial matu­rity. “We show that many of these assump­tions are outdated,” says Dr. Thomas Buchegger, Head of the SAL site in Linz. “NC offers tremen­dous poten­tial not only in energy effi­ciency but also in speed. More­over, it can be effec­tively imple­mented on stan­dard hard­ware plat­forms—not just special­ized analog chips.”

A recent joint study by JKU, SAL, SCCH, and TU Graz has demon­strated NC’s versa­tility in the medical field. Researchers found a way to collect and process elec­tro­car­dio­gram (ECG) data far more effi­ciently—reducing data volume by more than 80% without sacri­ficing analysis quality. This means lower storage require­ments while main­taining the same accu­racy, opening up signif­i­cant oppor­tu­ni­ties for the future of medical diag­nos­tics and beyond.

Milestones of the Research Partnership

Since the beginning of their collaboration, the partners have achieved a number of pioneering successes that impressively underscore the international potential of neuromorphic computing „Made in Austria“:

World Record for SNN Inference on Standard FPGAs

Using a newly devel­oped demon­strator, the team has shown that Spiking Neural Networks can run at excep­tional speeds on widely avail­able FPGA (Field Program­mable Gate Array) systems.

“Our method processes over 2.5 million images per second—more than 100 times faster than previous systems using the same hard­ware and data,” says Assoc. Prof. Dr. Michael Lunglmayr from JKU’s Insti­tute for Signal Process­ing. “It also oper­ates at low power, achieving more than 3 million images per second per watt.”

This break­through illus­trates how modern AI can be both highly powerful and energy-effi­cient—on off-the-shelf hard­ware.

Patent Filed

The team has filed a patent for a novel, ultra-energy-effi­cient data acqui­si­tion method based on neuro­mor­phic prin­ci­ples inspired by the human brain.

Austrian Workshop Series and Publications

Through the SNNSys work­shop series on Neuro­mor­phic Computing and Spiking Neural Networks, orga­nized in collab­o­ra­tion with TU Graz as part of the AIROV confer­ence, the part­ners are driving knowl­edge exchange within Austria’s re­search commu­nity. Find­ings have been published in leading jour­nals such as the Journal of Neuro­mor­phic Computing and Engi­neering and presented at top-tier confer­ences, including the IEEE Inter­na­tional Sympo­sium on Circuits and Systems.

Inter­na­tional Grand Chal­lenge

In September 2025, at the pres­ti­gious IEEE Inter­na­tional Confer­ence on Image Process­ing (ICIP) in Alaska, the re­search team will host an inter­na­tional chal­lenge under its motto “Low Energy, High Speed.” The compe­ti­tion aims to iden­tify the most effi­cient and fastest AI solu­tions.

“We’re excited to see the creative and inno­v­a­tive approaches this will inspire,” says Moser. “The strong interest from the scien­tific commu­nity shows the incred­ible momentum in this field.”

About the research partners

Software Competence Center Hagenberg (SCCH)

Soft­ware Compe­tence Center Hagen­berg (SCCH) is a leading non-univer­sity re­search center in Austria, special­izing for over 20 years in applied re­search in the fields of Data Science and Soft­ware Science. SCCH excels in project imple­men­ta­tion for digi­tal­iza­tion, Industry 4.0, and Arti­fi­cial Intel­li­gence, serving as an inter­face between inter­na­tional re­search and domestic industry. The center employs around 130 staff from 25 nations, conducting world-class re­search. As a COMET Center (Compe­tence Centers for Excel­lent Tech­nolo­gies), SCCH is funded by the Austrian Federal Ministry for Trans­port, Inno­va­tion and Tech­nology, the Federal Ministry for Digi­tal­iza­tion and Busi­ness Loca­tion, and the State of Upper Austria.

Press contacts:

• Mag. Martina Höller, Science Communication, Software Competence Center Hagenberg, (+43) 50 343 882, martina.hoeller@scch.at
• Priv.-Doz. Dr. Bernhard A. Moser, Technology and Innovation Manager at SCCH and co-initiator of this cooperation, also works with a double affiliation at the Institute for Signal Processing at Johannes Kepler University Linz (JKU). bernhard.moser@scch.at

Johannes Kepler Univer­sity Linz (JKU) – Insti­tute for Signal Process­ing (ISP)

The Insti­tute for Signal Process­ing (ISP) at Johannes Kepler Univer­sity Linz is a leading acad­emic re­search unit special­izing in the algo­rithmic, archi­tec­tural, and hard­ware-oriented aspects of signal process­ing systems. Its re­search spans infor­ma­tion and commu­ni­ca­tion systems, high-frequency and base­band inte­grated circuits (ICs), bio- and sensor signal process­ing, and auto­mo­tive appli­ca­tions.

For nearly 60 years, the Johannes Kepler Univer­sity has been rooted in regional tradi­tion but endowed with a strong inter­na­tional perspec­tive. As Upper Austria’s largest insti­tu­tion of educa­tion and re­search, the univer­sity offers degree programs and conducts re­search in the fields of law, engi­neering, natural sciences, medi­cine, educa­tion, social sciences, economics and busi­ness. The JKU campus is home to approx­i­mately 25,000 enrolled students and close to 4,000 employees, boasting a modern infra­struc­ture with ample space to study, learn, conduct re­search, and enjoy univer­sity life.

Press contacts:

• Mag. Christian Savoy, University Communications, +43 732 2468 3012, christian.savoy@jku.at
• Associate Professor Dr. Michael Lunglmayr, co-initiator of the cooperation,
• Michael.lungl­mayr@jku.at

Silicon Austria Labs (SAL)

Silicon Austria Labs (SAL) is Austria's top re­search center for elec­tronics and soft­ware-based systems (ESBS), which are the tech­no­log­ical back­bone of digi­ti­za­tion. At three loca­tions in Graz, Villach and Linz, SAL conducts cutting-edge re­search in Microsys­tems, Sensor Systems, Power Elec­tronics, Intel­li­gent Wire­less Systems, and Embedded Systems to develop future-oriented solu­tions for indus­trial produc­tion, health, energy, mobility, and safety.

The Intel­li­gent Wire­less Systems divi­sion focuses on devel­oping reli­able and secure commu­ni­ca­tion systems for indus­trial envi­ron­ments. This enables broad­band connec­tions for mobile devices (e.g., smart­phones, tablets, laptops), wire­less machine-to-machine commu­ni­ca­tion, and increas­ingly wire­less sensor networks in indus­trial appli­ca­tions.

One of SAL’s flag­ship initia­tives is Digineuron, a pioneering project aimed at devel­oping ultra-effi­cient inte­grated circuits that imple­ment AI in minia­ture formats. These chips mimic the struc­ture and func­tion of the human brain, using layered neural networks that commu­ni­cate locally to mini­mize energy consump­tion — opening new possi­bil­i­ties for smart, low-power appli­ca­tions.

SAL brings together key players from industry, science, and re­search, combining valu­able exper­tise and know-how to conduct coop­er­a­tive, appli­ca­tion-oriented re­search along the value chain. Coop­er­a­tive projects are co-financed by SAL and enable a fast and unbu­reau­cratic project start.

Press contacts:

• DI Dr. Thomas Buchegger, thomas.buchegger@silicon-austria.com