A new approach to realizing wireless computing in parallel memory

This article was reviewed based on Science X’s editorial process and policies. The editors have highlighted the following attributes ensuring the credibility of the content:


peer-reviewed publication

trusted source


The use of in-memory computing enables extremely energy-efficient wireless communications, operating on radio, acoustic and light waves. Credit: Wang et al. (Nature’s Electronics, 2023)

Advanced communication technologies, such as the fifth generation mobile network (5G) and the Internet of Things (IoT), can greatly benefit from devices that can support wireless communications while consuming a minimal amount of energy. Since most existing devices have separate components to perform calculations and transmit data, reducing their power consumption can be challenging.

Researchers at Nanjing University, Southeast University, and Purple Mountain Laboratories in China recently devised a parallel in-memory wireless computing scheme that simultaneously performs computation and wireless transmission on the same hardware. This design, introduced in Electronics of natureit is based on the use of memristive crossbar arrays, grid structures containing memristors, electrical components capable of both processing and storing data.

“In one of our previous works published in Nature Nanotechnologywe proposed the realization of massively parallel in-memory computing using continuous-time data representation in a nanoscale crossbar array,” Shi-Jun Liang, one of the researchers who led the recent study, told Tech Xplore.

“In this past paper, we demonstrated that processed analog signals can be transmitted via RF module. Inspired by our findings, we started thinking about whether it is possible to leverage in-memory computation based on a memristive crossbar array to realize parallel digital signal transmission of data in an ultra-low-power manner, which is desirable for intelligent IoT or other edge computing devices.”

The wireless in-memory computing approach pioneered by Liang, Feng Miao, and their colleagues uses the analog in-memory computing capabilities of memristors to also process analog wireless signals that are transmitted or received by a device. This could significantly improve the energy efficiency of devices, smoothing out the sharp separation between the digital and analog domains that characterizes most current technology projects.

“The unique features of this computing paradigm is that information-carrying wireless signals can be processed (for example, modulated or demodulated) in the data transmission path in real time and consuming a low amount of power,” Liang explained.

“First, this approach can reduce the demand for ADC (analog-to-digital converter) used in wireless communication systems and dramatically increase the energy efficiency of real-time wireless digital communication. Our study also identifies a promising application scenario for memristive devices, whose practical application has been limited by the noise associated with the change in conduction state.”

Researchers created a prototype system based on the proposed design and then evaluated its performance in a series of tests. They found that it transmits a 480-bit binary stream with a very promising bit error rate of 0/480. In addition, their device consumes much less energy (that is, about two orders of magnitude) than conventional devices based on digital-to-analog and analog-to-digital converters.

In the future, the parallel in-memory wireless computing approach pioneered by Liang, Miao and their colleagues could enable the development of electronic devices that consume less power and are better equipped to meet high computational demands. While the team has so far tested their design to create electronics, it could also be applied to acoustic and optical wireless communication devices.

“Using memristive devices in wireless communication, where precise computation is not required, provides a great opportunity for analog memory processing technology,” Miao added. “In the near future, we plan to work on the large-scale integration of the memristive crossbar array into the wireless communication system and push it towards real-world application.”

More information:
Cong Wang et al, Wireless parallel memory computing, Electronics of nature (2023). DOI: 10.1038/s41928-023-00965-5

About the magazine:
Electronics of nature

Nature Nanotechnology

#approach #realizing #wireless #computing #parallel #memory

Leave a Comment