The ‘memimpedance’ device can control current flow in a circuit, and could make electronics like wearable sensors, flexible medical devices and biodegradable electronics more efficient

A team from Khalifa University has developed a novel electronic ‘memimpedance’ device that can act as a switch and induce tunable resistor and capacitor behavior simultaneously in an electronic circuit.

A resistor is an electrical component that regulates the flow of electrons in a circuit, while a capacitor is an electrical component that collects and stores electrical charge.

The need to control electron flow is what gave rise to transistors, which are at the heart of all electronics today. Transistors are three terminal electronic switches that either permit or prevent electrons from flowing from one terminal to another based on the control provided by the third terminal, which serves as a gate. Other elements, including resistors and capacitors, also play a role in regulating current flow in electronics.

Enter memristors. Memristors are resistors with memory. They were physically realized for the first time in 2008, though they were conceptualized theoretically for decades before that, and have gained popularity for their potential use in computers.�� They are simpler than transistors, smaller, use less energy, can alter their resistance and “remember” the most recent resistance they had. This means they have the potential to replace silicon-based transistors and could be used to create faster, more efficient computer chips that integrate memory with logic.

When memristor and memcapacitor behaviors happen simultaneously in the same device, it is called memimpedance. A memimpedance device, therefore, is designed to control, or tune, the memristor and memcapacitor behavior in an integrated circuit.

Dr. Heba Abunahla, Research Scientist in the Electrical Engineering and Computer Science Department at Khalifa University, and a team from the KU System-on-Chip Lab (SoCL), developed a memimpedance device made out of silver-reduced graphene oxide-silver that can tune the resistance and capacitance behaviors in a circuit.

Dr. Abunahla published her research in the journal, with co-authors Dr. Baker Mohammad, Professor, Dr. Yawar Abbas, Research Scientist, and Dr. Anas Alazzam, Associate Professor.

“Memimpedance has many advantages compared to resistance or capacitance only devices, especially its ability to tune the overall circuit impedance,” Dr. Abunahla said.

Circuit impedance measures how much a circuit impedes the flow of charge. As electrons move through a circuit, they collide with the internal structure of the conductor, which creates friction and slows them down.

The amount of resistance depends on the conductor’s material, shape and size, but conductors generally have low resistance to current. In addition to resistance, circuit impedance also considers capacitance, which is the ability of a component to collect and store electrical charge.

A device that can tune a circuit’s overall impedance would be particularly useful in applications like wearable sensors, flexible medical devices and biodegradable electronics.

Dr. Abunahla and the SoCL team successfully demonstrated that their memimpedance device would, when a suitable voltage was applied, tune the circuit resistance and capacitance concurrently.

They developed the memimpedance device with a unique structure using silver-reduced graphene oxide-silver.

“Using graphene-related materials as a switching material is a great asset due to their low cost and adaptability, and they are environmentally friendly,” Dr. Abunahla said.

The team’s memimpedance device has a planar structure, meaning all the atoms of the molecule sit on a single two-dimensional plane.

“Fabricating the device with a planar design boosts its potential to be deployed in sensing applications, such as wearable electronics. The planar structure allows for a bigger surface area and better interaction with the environment, which increases the efficiency of the sensing unit,” Dr. Abunahla added.

The researchers fabricated the device on a flexible��polymer substrate using a lithography process. They deposited the graphene oxide directly onto the polymer substrate, then immersed it in an acid to create a thin layer of reduced graphene oxide measuring around 60 nanometers thick. They then used a standard lift-off process to pattern a film of silver electrodes onto the substrate.

They intentionally selected a polymer substrate instead of a silicon-based substrate, which is traditionally used to make memristors, because silicon-based devices pose challenges when they are stacked together to create 3D circuits.

The KU memimpedance device, however, is well suited for stacking and to produce 3D integrated circuits, which can achieve better performance than traditional 2D circuits.

Using silver-reduced graphene oxide-silver in a planar structure, fabricated on a flexible substrate using a standard production process, makes the resulting device cost-effective and deployable in flexible electronics and many other potential applications

“This work will be a great asset for tunable emerging applications, especially for communication and AI systems,” Dr. Abunahla said.

Jade Sterling
Science Writer
6 October 2021

The post Khalifa University Researchers Develop Next-Generation Electronic Tuning Device as New Building Block for Modern Computers appeared first on Khalifa University.

The original computers were designed around a human brain model. Since then, developments in artificial intelligence and computer science continue to take inspiration from the brain.

Read Arabic story .

The human brain has always been under study for inspiration of computing systems. Although there’s a very long way to go until we can achieve a computing system that matches the efficiency of the human brain for cognitive tasks, several brain-inspired computing paradigms are being researched. Convolutional neural networks are a widely used machine learning approach for AI-related applications due to their significant performance relative to rules-based or symbolic approaches. Nonetheless, for many tasks machine learning requires vast amounts of data and training to converge to an acceptable level of performance.

A PhD student from Khalifa University, Eman Hasan, is investigating another AI computation methodology called ‘hyperdimensional computing’, which can possibly take AI systems a step closer toward human-like cognition. The work is supervised by Dr. Baker Mohammad, Associate Professor and Director of the System-on-Chip Lab (SOCL), and Dr. Yasmin Halawani, Postdoctoral Fellow.

Hasan’s work, which was published recently in the journal, analyses different models of hyperdimensional computing and ��ݮ��Ƶ the advantages of this computing paradigm. Hyperdimensional computing, or HDC, is a relatively new paradigm for computing using large vectors (like 10000 bits each) and is inspired by patterns of neural activity in the human brain. The means by which can allow AI-based computing systems to retain memory can reduce their computing and power demands.

HDC vectors, by nature, are also extremely robust against noise, much like the human’s central nervous system. Intelligence requires detecting, storing, binding and unbinding noisy patterns, and HDC is well-suited to handling noisy patterns. Inspired by an abstract representation of neuronal circuits in the human brain, developing an HDC architecture involves encoding, training, and comparison stages.

The human brain is excellent at recognizing patterns and using those patterns to infer information about other things. For example, humans generally understand that just because a chair is missing a leg, that doesn’t mean it’s no longer a chair. An AI system may look at this three-legged chair and decide it is a completely new object that needs a new classification. HDC vectors, however, offer some margin for error. With HDC, recognizing certain features will generate a vector that is similar enough to a chair that the computer can infer the object is a chair from its memory of what a chair looks like. Hence, the three-legged chair will remain a chair in hyperdimensional computing while in traditional object recognition this is a difficult task.

“In a HD vector, we can represent data holistically, meaning that the value of an object is distributed among many data points,” explained Hasan. “Therefore, we can reconstruct the vector’s meaning as long as we have 60% of its content.”��

The structure of the vectors leads to one of the strongest advantages of the HDC approach, which is that it can tolerate errors and therefore is a great option for approximate computing applications. This arises from the representation of the hyper vectors, where a bit value is independent of its location in the bit sequence.

HDC is also powerful in that it is memory-centric, which makes it capable of performing complex calculations while requiring less computing power. This type of computing is particularly useful for ‘edge’ computing, which refers to computing that’s done at or near the source of data. In a growing number of devices, including in autonomous vehicles, computations must be carried out immediately and at the point of the data collection, instead of relying on computing done in the cloud at a data center.

��

“Hyperdimensional computing is a promising model for edge devices as it does not include the computationally demanding training step found in the widely used convolutional neural network,” explained Hasan. “However, hyperdimensional computing comes with its own challenges as encoding alone takes about 80 percent of the execution time of its training and some encoding algorithms result in the encoded data growing to twenty times its original size.”

Hasan studied the HDC paradigm and its main algorithms in one-dimensional and two-dimensional applications. Research has shown that HDC outperforms digital neural networks in one dimensional data set applications, such as speech recognition, but the complexity increases once it is expanded to 2D applications.

“HDC has shown promising results for one dimensional applications, using less power, and with lower latency than state-of-the-art simple deep neural networks,” explained Hasan. “But in 2D applications, convolutional neural networks still achieve higher classification accuracy, but at the expense of more computations.”

Hasan concluded that HDC is still considered a new paradigm and faces challenges requiring further analysis.��

Jade Sterling
Science Writer
29 April 2021

The post A Step Closer to Brain Like AI with Hyperdimensional Computing appeared first on Khalifa University.