Scientists can incorporate key computing components into a single flexible fiber and run them through a washing machine. Researchers hope that one day many of these fibers will be united and weaved into a cohesive “fiber computing” network.
Also known as smart fabric or e-textiles, smart textiles are materials that contain electronic components that enhance the functionality of wearable devices and other products. These can be used to create materials using embedded computing components that can be used in clothing, woven displays, and more.
One of the earliest modern applications of this was the creation of Lilypad in 2007. This is a series of sewn electronic components designed for use in interactive clothing, toys and sculptures.
One of the most important limitations of smart textiles is that the computing capabilities of individual fibers are severely limited and individual components are not embedded.
Because these fibers do not have components, basic tasks such as real-time interpretation of biosignals are difficult. And it is difficult to get signals for subsequent data processing.
However, in a new study published in the Nano-Micro Letters Journal on June 6, scientists attached sensing, communication, calculations and storage to a single fiber chain.
Related: New dough can be heated over 50 degrees to keep people warm in ultra cold weather
You might like it
Each elastic fiber has 60% elasticity and can be placed in the washing machine. That is, it can be used to weave practical garments.
The new fiber also allows smart clothing and devices to have better accuracy through detection and real-time interaction between computers and humans, scientists say in their research.
The future of clothing
Each fiber developed in this study incorporated eight devices, including four sensors. Photosector, temperature sensors, accelerometers, photoplethysmugram (PPG) sensors, changes in light absorption by the skin, and microcontrollers, two communication modules, and power management devices, have been mentioned in the study. Together, these components achieved data collection, processing, storage, and outcome transmission.
To monitor how effective the new system is, scientists integrated four smart fibers into the sleeves of clothes and pants legs, asking one person wearing the clothes to do a series of weight exercises, including squats, lunges and boards.
Each fiber performs a separate, trained neural network. This is a collection of machine learning algorithms designed to mimic the way the human brain processes information. This allowed the fiber to recognize a variety of actions in real time, such as squatting, boarding, and arm rotation.
A single fiber achieved 67% accuracy in recognizing a particular movement, but when all four fibers worked together, they achieved 95% accuracy.
“This remarkable improvement highlights the immense potential of multi-fiber co-sensing and decentralized inference, and presents a transformative approach to intelligent wearable systems that seamlessly integrate localized calculations with networked decision-making to provide robust, high-fidelity performance,” the scientist said.
Scientists said these results highlight the potential performance of the collaborative fiber network, but they acknowledged challenges in optimizing communication speeds, reducing energy consumption and increasing bandwidth.
They pointed out that expanding these fiber-based computer networks requires scientists to improve the efficiency of information exchange between individual nodes.
They added that future research could focus on building a higher throughput lowered communications protocol, tailored specifically for these types of fiber computers.
Source link
