Energy Efficient Communication Links for Smart Devices

Nathan Narevsky

EECS Department
University of California, Berkeley
Technical Report No. UCB/EECS-2022-16
May 1, 2022

http://www2.eecs.berkeley.edu/Pubs/TechRpts/2022/EECS-2022-16.pdf

With the continued development of new sensors, connected devices, and continued scaling data rates there is a serious data problem. This large amount of data needs to be communicated efficiently to where it needs to go for processing, logging or actions. There are many applications where the incoming data that needs to be communicated is not a continuous data rate, the required data rate and latency of the link change over time and settings, so a truly efficient system should be able to take advantage of this burst nature and operate accordingly.

For the purpose of developing larger channel count systems for brain machine interfaces and neuroscience research, a compression algorithm is proposed to minimize the amount of data required to send by up to 700x. This would enable fully wireless systems with larger channel counts, with the main energy bottleneck still being the communication link. A low duty cycle burst mode millimeter wave phased array is proposed to further improve the energy efficiency of the entire system, enabling efficient wireless transfer of a scalable data rate over more than 1 to 100 Mbps. Similar design techniques to this communication link are also applied towards a high speed serial link design, tailored towards a large scale phased array system. The rapid on and off operation allows for a low latency communication interface throughout the entire array without overhead of a constantly operating link. This enhancement reduces the overhead power required to synchronize all of the elements, allowing for a more efficient overall system level design.

Advisor: Jan M. Rabaey


BibTeX citation:

@phdthesis{Narevsky:EECS-2022-16,
    Author = {Narevsky, Nathan},
    Editor = {Rabaey, Jan M. and Stojanovic, Vladimir and Olshausen, Bruno},
    Title = {Energy Efficient Communication Links for Smart Devices},
    School = {EECS Department, University of California, Berkeley},
    Year = {2022},
    Month = {May},
    URL = {http://www2.eecs.berkeley.edu/Pubs/TechRpts/2022/EECS-2022-16.html},
    Number = {UCB/EECS-2022-16},
    Abstract = {With the continued development of new sensors, connected devices, and continued scaling data rates there is a serious data problem.  This large amount of data needs to be communicated efficiently to where it needs to go for processing, logging or actions.  There are many applications where the incoming data that needs to be communicated is not a continuous data rate, the required data rate and latency of the link change over time and settings, so a truly efficient system should be able to take advantage of this burst nature and operate accordingly.  

For the purpose of developing larger channel count systems for brain machine interfaces and neuroscience research, a compression algorithm is proposed to minimize the amount of data required to send by up to 700x.  This would enable fully wireless systems with larger channel counts, with the main energy bottleneck still being the communication link.  A low duty cycle burst mode millimeter wave phased array is proposed to further improve the energy efficiency of the entire system, enabling efficient wireless transfer of a scalable data rate over more than 1 to 100 Mbps.  Similar design techniques to this communication link are also applied towards a high speed serial link design, tailored towards a large scale phased array system.  The rapid on and off operation allows for a low latency communication interface throughout the entire array without overhead of a constantly operating link.  This enhancement reduces the overhead power required to synchronize all of the elements, allowing for a more efficient overall system level design.}
}

EndNote citation:

%0 Thesis
%A Narevsky, Nathan
%E Rabaey, Jan M.
%E Stojanovic, Vladimir
%E Olshausen, Bruno
%T Energy Efficient Communication Links for Smart Devices
%I EECS Department, University of California, Berkeley
%D 2022
%8 May 1
%@ UCB/EECS-2022-16
%U http://www2.eecs.berkeley.edu/Pubs/TechRpts/2022/EECS-2022-16.html
%F Narevsky:EECS-2022-16