Virginia Tech researchers Lingjia Liu and Yang (Cindy) Yi are creating a new buzz by using brain-inspired machine learning techniques. This technique is used to increase the energy efficiency of wireless receivers.
This brain-inspired machine technique has high signal quality. Consist of a combination of multiple-input and output techniques with orthogonal frequency division multiplexing.
About Lingjia Liu and Yang (Cindy) Yi
Liu is an associate and Yi is an assistant professor in the Bradley Department of Electrical and Computer Engineering.
They received Best Paper Award from the IEEE Transmission, Access, and Optical Systems Technical Committee on the project of “Realizing Green Symbol Detection Via Reservoir Computing: An Energy-Efficiency Perspective”.
Liu and Yi Collaborations:
- To publish brain-inspired machine learning technique, Liu and Yi are collaborating with researchers from the Information Directorate of the U.S. Air Force Research Laboratory — Jonathan Ashdown, John Matyjas, Michael Medley, and Bryant Wysocki.
- Liu’s Ph.D. student Rubayet Shafin also plays a major role in their findings.
How This Brain-Inspired Machine Learning Technique Works?
The Brain-Inspired Machine Learning Technique uses different ways in the meantime to travel from transmitter to receiver. Brain-inspired high signal quality technique helps minimal interference. This combination gives an inherent advantage for avoiding multipath fading.
While providing an example Liu said, “A combination of techniques and frequency brings many benefits and is the main radio access technology for 4G and 5G networks”.
He continued- “However, correctly detecting the signals at the receiver and turning them back into something your device understands can require a lot of computational effort, and therefore energy.”
In order to make efficient Liu and Yi are using artificial neural networks with computing systems inspired by the inner workings of the brains.
Yi stated “Traditionally, the receiver will conduct channel estimation before detecting the transmitted signals. Using artificial neural networks, we can create a completely new framework by detecting transmitted signals directly at the receiver.”
The technical advisor of AFRL’s Computing and Communications Division and an Air Force Research Laboratory Fellow, Mr. Matyjas also added “This approach can significantly improve system performance when it is difficult to model the channel, or when it may not be possible to set up a straightforward relation between the input and output,”
The Team is using a framework called reservoir computing in order to operate artificial neural network more proficiently on a transmitter-receiver pair.
This reservoir computing particularly a unique design called reverberate state arrange (ESN). An ESN is a sort of recurrent neural network that consolidates elite with low vitality.
The Chief Engineer Of The Air Force Research Laboratory Information Directorate, Mr. Wysocki said “This strategy allows us to create a model describing how a specific signal propagates from a transmitter to a receiver, making it possible to establish a straightforward relationship between the input and the output of the system.”
Testing The Efficiency
Liu, Yi, and their AFRL collaborators compared their discoveries with results and found that their outcomes were more proficient, particularly on the receiver side. They contrasted their findings from more settled training approaches.
“Simulation and numerical results showed that the ESN can provide significantly better performance in terms of computational complexity and training convergence,” said Liu. “Compared to other methods, this can be considered a ‘green’ option.”
We hope that the information provided regarding this new technology “Brain-Inspired Methods to Improve Wireless Communications” will amaze you. You can also grab other amazing stuff and mind-blowing facts by going through our web portal.