Abstract
Recent advances in integrated circuit technology
scaling, analog circuitry, and wireless communication
have created the possibility for complex low noise, low
power wireless instrumentation to be deployed in previously
impossible scenarios, particularly in the realm of wireless
sensing for biology research and biomedical applications. The
generation of portable, autonomous devices made possible by
these improvements will allow, for example, unobtrusive longterm
health monitoring by way of fully implantable clinical
diagnostic systems. This paper presents some achievements
made to this end, and describes the circuit design techniques
and system architectures used to realize low power wireless
bioelectrical interfaces.
I. INTRODUCTION
State-of-the-art advances made in the realm of wireless
circuit technology are opening the door to a new breed of
body-worn devices. For example, portable neuroprostheses
promise to improve the quality-of-life for people with
mobility disorders. Wireless EMG technology could enable
remote therapy and advanced human-computer interfaces.
Miniaturized low power wireless technology will allow
previously impossible freely-behaving animal research.
Such systems will require closed loop recording, analysis,
and stimulation performed under a strict power budget.
Size and power reduction necessitated in the rapidly evolving
area of biosignal monitoring are placing increasingly
stringent demands on integrated circuit technology and circuit
designers. This paper discusses the technology behind
and applications of low power, wireless communication
circuitry for bioelectrical interfaces.
The rest of the paper is organized into five sections.
Section II discusses the motivation for and system requirements
of a new generation of wireless biosignal interface
systems. Sections III and IV present two contrasting
paradigms for the realization of such systems, active vs.
passive. Section III introduces an active (battery-powered)
implementation of a wireless neural interface, the ‘Bumblebee’
while Section IV reviews several passive (batteryfree)
realizations. An initial realization of a wirelesslypowered,
functional contact lens is presented in Section V,
and concluding remarks are given in Section VI.
II. APPLICATIONS AND SYSTEM REQUIREMENTS OF AUTONOMOUS BIOELECTRICAL INTERFACES
The deployment of electrical interfaces for biological
research and medical applications places strict requirements
on system power and biocompatibility. For longterm
health monitoring applications, the risk of infection
may be prohibitively high to allow wires or connectors
to penetrate the skin. Hence, in addition to low-power
recording, biosignal interfaces should provide methods
of wirelessly transmitting acquired signals to an external
monitoring device. Robust, multichannel wireless devices
for neural signal acquisition are a critical development that
will permit brain-computer interfaces (BCI) and neuroprostheses
to gain widespread acceptance for the study and
treatment of neurological disorders. The additional burden
of transmission increases the constraint on the power budget.
In addition, chronic recording of bioelectrical signals
in humans, or recording in very small animals and insects,
necessitate a small form-factor for the wireless, miniaturized
biosignal interface. These requirements lead to two,
somewhat competing, paradigms in bioelectrical interfaces,
namely active and passive systems. An active system
employs an on-board battery and enables long-term monitoring
and far-field transmission. However, the inclusion
of a battery necessitates frequent replacement which may
be prohibitive, particularly in human BCI implementations.
In contrast, the passive interface derives its energy from
an energy harvester such as a Thermoelectric Generator
(TEG) and may employ a low-power boost converter or
RF-to-DC rectifier to power the active circuitry [1], [2]. In
the following sections, we discuss prototype systems, both
active and passive, successfully implemented for in-vivo
data acquisition and wireless transmission.
The information in the brain could be recorded as
electrical signals generated by individual neurons in the
form of action potentials, or spikes. As the recording site
is moved away from a single neuron, it captures electrical
activity from multiple sources. Any interface circuit should
be able to acquire the signal in the frequency band of
interest while rejecting the large DC offset resulting from
the electrode tissue interface.


Download full report
http://googleurl?sa=t&source=web&cd=1&ve...496705.pdf%3Farnumber%3D5496705&ei=ltU3TrTiC5HyrQfvuJ33Dw&usg=AFQjCNHECCWzxM3H_E-_p2eeB4bYZ7KQiA&sig2=a7oTCKfdaOmkTnY-3NTPjQ