Jaclyn Leverett

EECS Department, University of California, Berkeley

Technical Report No. UCB/EECS-2014-206

December 1, 2014

http://www2.eecs.berkeley.edu/Pubs/TechRpts/2014/EECS-2014-206.pdf

Brain Machine Interfaces have shown increasing promise to restore motor function to patients suffering from amputation or paralysis. However, for such systems to be clinically viable, both the implant and the external headstage must be wireless, ultra low power, compact, and inconspicuous. A lightweight, wireless headstage with a long battery life is also necessary to perform meaningful experiments on small animals, such as rats and mice, while they are awake, untethered, and behaving naturally. The elimination of wires allows for a greater number of recording channels as well as interaction between multiple animals in the same environment without the risk of tangling wires, a common problem that inhibits productivity as well as potentially harms the animals.

The focus of this work is on an external wireless neuromodulation system capable of real-time neural recording, on-chip data compression, and dual stimulation on 8 selectable channels, hence offering substantially enhanced functionality over current state of the art (e.g. [7, 17, 3]). Weighing only 4.6 grams and dimensioned at 16mm x 29mm (the same area as a quarter), this wireless headstage achieves a battery life of 10 hours when constantly transmitting data at ranges up to 14 meters.

To achieve such a small form-factor with superior battery life, this headstage has integrated ultra-low power components, including a 65nm CMOS 4.78mm2 neuromodulation Application Specific Integrated Circuit (ASIC) that consumes 417uW from a 1.2V supply while operating 64 acquisition channels with epoch compression at an average firing rate of 50Hz and engaging two stimulators with a pulse width of 250us/phase, differential current of 150uA, and a pulse frequency of 100Hz.

Advisors: Jan M. Rabaey

---

BibTeX citation:

@mastersthesis{Leverett:EECS-2014-206,
    Author= {Leverett, Jaclyn},
    Title= {A Low-Power, Lightweight, Wireless Neural Recording and Stimulating Headstage for Brain Machine Interfaces},
    School= {EECS Department, University of California, Berkeley},
    Year= {2014},
    Month= {Dec},
    Url= {http://www2.eecs.berkeley.edu/Pubs/TechRpts/2014/EECS-2014-206.html},
    Number= {UCB/EECS-2014-206},
    Abstract= {Brain Machine Interfaces have shown increasing promise to restore motor function to patients suffering from amputation or paralysis. However, for such systems to be clinically viable, both the implant and the external headstage must be wireless, ultra low power, compact, and inconspicuous. A lightweight, wireless headstage with a long battery life is also necessary to perform meaningful experiments on small animals, such as rats and mice, while they are awake, untethered, and behaving naturally. The elimination of wires allows for a greater number of recording channels as well as interaction between multiple animals in the same environment without the risk of tangling wires, a common problem that inhibits productivity as well as potentially harms the animals.

The focus of this work is on an external wireless neuromodulation system capable of real-time neural recording, on-chip data compression, and dual stimulation on 8 selectable channels, hence offering substantially enhanced functionality over current state of the art (e.g. [7, 17, 3]). Weighing only 4.6 grams and dimensioned at 16mm x 29mm (the same area as a quarter), this wireless headstage achieves a battery life of 10 hours when constantly transmitting data at ranges up to 14 meters.

To achieve such a small form-factor with superior battery life, this headstage has integrated ultra-low power components, including a 65nm CMOS 4.78mm2 neuromodulation Application Specific Integrated Circuit (ASIC) that consumes 417uW from a 1.2V supply while operating 64 acquisition channels with epoch compression at an average firing rate of 50Hz and engaging two stimulators with a pulse width of 250us/phase, differential current of 150uA, and a pulse frequency of 100Hz.},
}

EndNote citation:

%0 Thesis
%A Leverett, Jaclyn 
%T A Low-Power, Lightweight, Wireless Neural Recording and Stimulating Headstage for Brain Machine Interfaces
%I EECS Department, University of California, Berkeley
%D 2014
%8 December 1
%@ UCB/EECS-2014-206
%U http://www2.eecs.berkeley.edu/Pubs/TechRpts/2014/EECS-2014-206.html
%F Leverett:EECS-2014-206