IEEE Solid-States Circuits Magazine - Spring 2023 - 43

of view (FOV) [3]. Most existing RFbased
methods rely on the received
signal strength, time of flight, time
difference of arrival, and angle of
arrival. However, RF signals suffer
very high attenuation when propagating
through human tissue. Severe
multipath effects further limit their
use for in vivo applications [4]. Ultrasound-based
methods are effective
for soft tissues but have poor performance
for bones and gas cavities [5].
Optical methods are limited by their
low imaging depth due to light attenuation
caused by absorption and
scattering [6]. For GI tract monitoring,
the current gold-standard solutions
include invasive techniques
such as endoscopy and manometry
or procedures that require repeated
use of harmful X-ray radiation such
as computed tomography scans and
scintigraphy [2]. Furthermore, these
techniques are not suitable for realworld
ambulatory settings.
In this work, we present a radiation-free
system for high-precision
localization and tracking of miniaturized
wireless devices in vivo,
using magnetic field gradients. Inspired
by magnetic resonance imaging
(MRI), we generate monotonically
varying magnetic fields in three orthogonal
directions, resulting in gradients
that encode each spatial point
uniquely. Highly miniaturized and
wireless microdevices [D1 and D2 in
Figure 1(a)] are designed to measure
their local magnetic field magnitude
and communicate it wirelessly to an
external receiver. The receiver maps
the field data to the respective spatial
coordinates and displays the relative
location of the microdevices in
real time. Our system's overview for
a surgical procedure is shown in Figure
1(b). The patient's leg with a titanium
(Ti) metal rod inside is placed
on top of a bed. The magnetic-sensing
microdevice (shown in green) can
be attached right next to the screw
hole at a known position on the rod,
and another identical device can be
installed on the surgical drill/tool.
The planar gradient-generating coils
are placed beneath the surgical bed.
Highly miniaturized and wireless microdevices
are designed to measure their local magnetic
field magnitude and communicate it wirelessly
to an external receiver.
Using the gradient-based 3D localization
of the microdevices, the surgical
team can maneuver to the screw hole
locations without using any X-ray
fluoroscopy.
In addition to the 3D
position, the system can also detect
orientation information [7].
The wireless and battery-less devices
D1 and D2 in Figure 1(b) consist
of a 3D magnetic sensor, an ASIC
(controller chip) designed in the
65-nm CMOS process, and an inductor
coil wound along the edges of
the device. The sensor (AK09970N)
is based on the Hall effect and has
16-bit data output and high sensitivity
(. .)11 31
-T/LSBn
and measurement
range (±36 mT). The spatial
localization resolution (
by the system, in each of the three
dimensions, is given by (1).
Tx
=+ +
c
T
G
B
eff )1
d i
G
G
m
2
c
d s
G
G
m
2
3
(1)
.
TBeff is the sensor's effective resolution
for magnetic field measurements.
G is the applied magnetic field
gradient, which is determined by the
current in the electromagnets and
their geometrical structure.
d Gs are the errors caused by field interpolation
and supply current variations,
respectively. The errors are
reduced to %11 to get
TT.
xB /Geff
[7]. For applications in surgical navigation,
the desired xT
is 100mn ,
which is achieved by employing
G = 30 mT/m and
T1B 3Teff
n The
.
absolute value of the magnetic field
used in this work for generating the
gradient G is
150 ,mT which is much
lower than the Tesla-level field used
in MRI and is thus safe for human
use. Our gradient switching rate is
also medically safe and is discussed
in detail in [7].
Tx) obtained
The wirelessly powered CMOS chip
d Gi and
consists of a power management unit
(PMU), RF wake-up unit (RFW), and
data acquisition unit (DAU), as shown
in Figure 2(a). The PMU receives wireless
power at 13.56 MHz, and the rectifier
[Figure 2(b)] converts it to dc
power with 82% voltage conversion
efficiency. It uses unbalanced-biased
comparators to allow operation at
low input amplitudes (≥700 mV) and
to minimize leakage current. This
unbalance is created by appropriately
selecting the values of resistors
R1 (30 kΩ) and R2 (0.5 kΩ). Voltage
limiters are connected to the RF inputs
to clamp the voltage to 1.2 V,
which prevents overstressing of the
front-end transistors. The rectified
dc voltage is boosted to 4× by a twostage
interleaved switched-capacitor
dc-dc boost converter [Figure 2(c)],
which achieves a voltage conversion
efficiency of 96% with no load and
65% when peak current is driven by
the regulator loads. This efficiency
is achieved by using nonoverlapping
(NOL) clocks for switching the two
stages of the boost converter to significantly
reduce leakage and chargesharing
losses. The boosted voltage
thus obtained is provided to three
regulators [Figure 2(d)], which produce
stable supply voltages of 2.5 V
for the sensor, 1 V for analog, and
0.5 V for digital circuit blocks. The sensor
regulator has a 100- Fn external
capacitor to store energy for the measurement
phase, during which the
sensor consumes 2 mA at 2.5 V for
1 ms, making it the most power-hungry
phase in the entire operation [8].
Reset and wake-up are transmitted
as ASK signals on a 13.56-MHz carrier
to the chip. An envelope detector in the
RFW detects the presence of low-level
zero signal
pares that with predefined thresholds
IEEE SOLID-STATE CIRCUITS MAGAZINE
SPRING 2023
43
in the carrier and com
IEEE Solid-States Circuits Magazine - Spring 2023

Table of Contents for the Digital Edition of IEEE Solid-States Circuits Magazine - Spring 2023
