Abstract— We report an area-efficient 8bit SAR ADC using
dual capacitor array banks for brain signal interface
microsystems. The proposed ADC consumes 680nW and the
total chip area is 0.035 mm
2
. We reduced the area and power by
a factor of eight when compared with conventional approaches.
If we increase the resolution, the area and power reduction
factor exponentially increases in our architecture (e.g., a factor
of 16 for 10 bit resolution). The measured SNDR, SFDR, THD,
and ENOB are 42.82 ± 0.47 dB, 57.90 ± 2.82dB, -53.58 ± 2.15 dB,
and 6.65 ± 0.07 bits, respectively.
I. INTRODUCTION
ECENTLY, multichannel neural interface systems have
been implemented to monitor neural activities [1-2]. For
the comprehensive analysis of neural activities, it is essential
to realize simultaneous real-time monitoring of multiple sites
in 3D electrode arrays with 64 channels or more. Typically,
neural activities such as spike contain most of information in
the bandwidth below 10 kHz with maximum amplitude of
±500 μV [3]. In these microsystems, the neural signals should
be amplified and converted into digital signals to be
transmitted to wired/wireless communication channels
between the implanted system and the external world.
Simultaneous access of multiple sites requires better noise
immunity in analog-to-digital converters (ADC) in a small
form factor at low power.
A successive approximation register (SAR) ADC is one of
the suitable candidates for neural interface applications due to
its simplicity, low power consumption, and reasonable
resolution. With a gain of 60 dB prior to the ADC, the
quantization noise is required to be less than 5 mV
rms
which
can be achieved by 8 bit or higher resolution capability of
ADC. Fig. 1 shows a conventional 8 bit SAR ADC structure
which typically consists of three parts: capacitor array (for
sample and hold and DAC), comparator, and successive
approximation register (SAR). For relatively lower resolution
ADCs (<6b), the comparator and SAR consume most power
[4]. However, as the resolution of ADCs increases, the power
consumption required for charging and discharging the
capacitor array becomes significant. Also, the total
capacitance required for DAC increases exponentially
This work was partly supported by NSF Project #0735481 and partly by
NSF Center for Wireless Integrated Microsystems.
Sun-Il Chang and Euisik Yoon are with the Department of Electrical
Engineering and Computer Science, University of Michigan, Ann Arbor, MI
48109 USA (email: schang@umich.edu, esyoon@umich.edu).
proportional to the number of bits. In the high resolution
ADCs, the capacitor array takes most of the area and power
consumption. It becomes more important to reduce the total
capacitance and area as the number of bits required in ADC
increases and multiple implementations of ADCs is needed.
Chang, et al. reported an effective switching method to
reduce the power [5]. However, this switching technique can
reduce only half of the power in conventional capacitor arrays.
Yang, et al. proposed an energy-efficient ADC with a small
form factor [6]. However, relatively complex algorithm has to
be implemented for practical use in neural microsystems.
In this paper, we propose an area-efficient 8 bit SAR
ADC using dual capacitor arrays. Using the dual capacitor
array banks, we can reduce the required capacitor array area
by a factor of 2
N/2-1
compared to the conventional approaches.
This feature can not only reduce the area but also the power
consumption by eliminating the power required for
charging/discharging the capacitor array.
II. AREA EFFICIENT SAR ADC ARCHITECTURE
A. Dual Capacitor Arrays
The key idea of the proposed SAR ADC is to perform the
successive approximation in both sides of comparator inputs
using dual capacitor arrays rather than only in one side. As
A Low-Power Area-Efficient 8 bit SAR ADC Using
Dual Capacitor Arrays for Neural Microsystems
Sun-Il Chang and Euisik Yoon
R
Fig. 1: The overall structure of conventional SAR ADC
Fig. 2: The overall structure of the proposed area-efficient SAR ADC
1647
31st Annual International Conference of the IEEE EMBS
Minneapolis, Minnesota, USA, September 2-6, 2009
978-1-4244-3296-7/09/$25.00 ©2009 IEEE