These chips could be used in applications with limited energy resources, such as in space, environmental sensors or biomedical implants
Researchers at Boise State University in the US are developing a new computer chip that will mimic the human brain than based on a traditional digital computer.
The chip is being developed under a project titled "CIF: Small: Realizing Chip-scale Bio-inspired Spiking Neural Networks with Monolithically Integrated Nano-scale Memristors", is being carried out by the varsity’s electrical and computer engineering faculty Elisa Barney Smith, Kris Campbell and Vishal Saxena.
According to the researchers, the currently available complex computing chips incorporate billions of nano-scale transistors that allow developing rapid, high-performance computers, pocket-sized smartphones that outpace early desktop PCS and lead to an explosion in handheld tablets.
Barney Smith said: "By mimicking the brain’s billions of interconnections and pattern recognition capabilities, we may ultimately introduce a new paradigm in speed and power, and potentially enable systems that include the ability to learn, adapt and respond to their environment."
The project completely depends on a memristor, which is a resistor that can be programmed to a new resistance by applying electrical pulses and it can memorise its new resistance value upon the removal of power.
The team’s research will further work on developing the already derived mathematical algorithms that describe the electrical interaction between brain synapses and neurons.
"By employing these models in combination with a new device technology that exhibits similar electrical response to the neuralsynapses, we will design entirely new computing chips that mimic how the brain processes information," Smith said.
The new chips are also expected to consume power at an order of magnitude which is less compared to the already existing computing processors, even while being similar to existing chips in physical dimensions.
Further, the project would result in development of ultra low-power electronics meant for applications with limited energy resources, such as in space, environmental sensors or biomedical implants.
Upon the development of an artificial neural network, the researchers are planning to engage neurobiologists into their ongoing project.