Research

About ICNS

Advances in electronic devices have had a massive impact on society. Increasing transistor density has so far enabled this growth but is now running into physical limitations.

Neuromorphic Engineering offers an alternative approach to increased computational performance by changing the structure of electronic circuits. Current electronic processors need increasing transistor density for increased performance, but Neuromorphic Engineering instead takes inspiration from the architecture and signal processing in biological neural systems to enhance performance. Western Sydney University is home to world leading researchers in neuromorphic engineering who undertake world changing research.

Spearheading this globally vital research is Western Sydney University’s new International Centre for Neuromorphic Systems (ICNS) - Western's next step in staking our claim in international excellence in this space.

Our Research

The International Centre for Neuromorphic Systems (ICNS) is a leading research group focussed on the development of neuromorphic sensors, processors, and algorithms.

We focus primarily on real world applications of neuro inspired perception and processing, where biological systems have natural advantages over conventional solutions: where robust, low power, high speed processors must respond autonomously to noisy, unpredictable environments.

Our Vision

To perform world-leading research to develop neuromorphic sensors, algorithms, and processing hardware, and apply these to solve existing problems in modern society.

Our mission statement reflects the three levels of technology ICNS will develop, and our focus on the applications of this technology as our developmental driver and potential source of funding. Our specific combination of neuromorphic sensor, algorithm, and processing hardware (platform) development gives us our niche and enables applications.

For the past thirty years, neuromorphic sensors have been developed and improved for technology’s sake. The rapid progress in electronics fuelled by Moore’s law meant that there was little interest outside our small research community in these sensors. Now, with the end of Moore’s law near, Industry and Defence have become keenly interested in Neuromorphic Engineering as an alternative technology. It is now up to us to show that we can deliver enhanced solutions to their problems using neuromorphic technology. We need to do this within the next decade, or the current interest in Neuromorphic Engineering will just have been hype. At ICNS, we are perfectly positioned to do this.


Western Sydney University has established the International Centre for Neuromorphic Systems (ICNS) – the only dedicated neuromorphic laboratory in Australia ­– as a home and global hub for leading researchers and students in this increasingly important field. The work of ICNS encompasses all three essential components of data-based decision-making systems, as our vision is to perform world-leading research to develop neuromorphic sensors, algorithms, and processors, and apply them to solve problems in modern society.

ICNS aims to become a world leader in neuromorphic engineering education, research, and technology transfer, developing innovative solutions for agriculture, industry, and government, growing jobs and training the future workforce to meet burgeoning demand. We will also find answers to big scientific questions, like How do brains work?

We welcome your interest in our research and impact. To learn more, please explore our website or get in touch.

What is neuromorphic engineering and why do we need it?

  1. We are drowning in data
  2. As the world’s population grows and associated challenges mount, we need computing power to keep pace. Agriculture, industry, education, medicine, transport, communication, energy, water, defence and emergency systems now depend on big data. The Internet connects billions of people and things. Increasing complexity demands ever faster, precise data collection, processing, and decision-making to avoid disaster.

  3. Data processing is draining our energy resources
  4. The data centres of large technology companies already consume around 3% of the world's electricity, with exponential growth anticipated over the next decade. It’s estimated that if we continue to use classical computing with centralised data processing, then by 2040, computing will require more energy than our entire existing energy grid. Clearly, the current approach is not scalable or sustainable.

  5. The end of Moore’s Law is nigh
  6. In 1965, Gordon Moore (co-founder of Intel) predicted the exponential growth of computing power that has shaped and supported our civilisation, when he proposedthat the number of transistors on a silicon chip would double every year (Moore’s Law). As we approach the end of Moore’s Law due to physical limits on the miniaturisation of transistors, the need for alternative solutions to enhance computing power is becoming critical.

What about quantum computing?

The promise of quantum computing espoused in the 1980s remains unfulfilled. Existing quantum computers are physically large with high power consumption and low functionality. Their functionality might increase in the next decade, but they will remain large and power hungry for the foreseeable future. One of several barriers to progress is that the slightest temperature change or vibration triggers quantum decoherence, generating computing errors. When might a quantum computer outperform a classical supercomputer in solving real-world problems? Optimists say it will take a decade, while others estimate twenty years or more. We can’t afford to wait.

To solve real-world problems, we look to nature.

To survive, living organisms must conserve precious energy while quickly and correctly perceiving any threat or opportunity in their surroundings, processing the information from their sensors to determine the appropriate response. Natural selection drives the evolution of highly optimised solutions to these computational challenges.The compact human brain uses less power than a lightbulb to outperform computers on many tasks.

Neuromorphic engineering draws inspiration from nature’s solutions to develop high-performance electronic systems for sensing and decision-making. Neurons are the cells from which biological nervous systems (including brains) are made, while -morphic refers to form, so neuromorphic means ‘in the form of neurons (or brains)’. Neuromorphic engineering is lower risk than quantum computing, because it repurposes existing infrastructure and tried and trusted electronics to develop novel sensing and processing solutions. Neuromorphic computers will solve problems that current classical computers and future quantum computers, by their nature, will not crack. Guided by biology, neuromorphic solutions are typically fast, compact, robust and energy-efficient, and therefore especially ideal for mobile devices and edge (decentralised, distributed) computing. Globally, governments and corporations are investing millions in neuromorphic R&D.