Department of Electrical and Control Engineering

embarrassedlopsidedΤεχνίτη Νοημοσύνη και Ρομποτική

14 Νοε 2013 (πριν από 3 χρόνια και 10 μήνες)

120 εμφανίσεις

Dep
artment

of Electrical and Control Engineering


Title:

Sub_Project IV: A Machine Vision
-
Based Human
-
In
tending Moving Vehicle

Capture
A
nd Tracking (I)

Principal Investigator:

J. Y. Chang,

Sponsor
:

National Science Council

Keywords
:

Image Segmentation, Ob
ject V
isual Capture
a
nd Tracking, Fuzzy

Classifier, View
-
Invariant Classification, Ubiquitous Computing


This project is concerned with machine vision
-
based moving vehicle capture and
tracking. It is known that brain signals are dependent on the excitation
s of vision,
audio, and feeling of a tester. The purpose of this project is to provide various
feedback to a user and then investigate its effect on brain signaling. With an excitation,
subproject III will identify the responsive brain channel and its sign
aling variation.
The brain signaling variations of a person invoking a multi
-
task distraction and spatial
disorder are studied by subprojects II and V, respectively. With these identified
responsive brain signaling, multi
-
mode bio
-
feedback embedded in brai
n
-
computer
interface formulation can reach a new era.

The test
-
bed of moving vehicle to be used in this subproject is an electrical
mini
-
car with rechargeable battery. The first year will dedicate to moving vehicle
capture under various environment conditi
ons. Good and robust image segmentation
of vehicle will be focused on under casting shadow, lighting variation, and color
closeness between object and background. The second year will investigate
vision
-
based positioning and movement recognition of moving
vehicle for BCI
sensing and validation. Through car’s key shape and a learned fuzzy rule base for
movement classification, either linguistic or numerical statement can be outputted to
bio
-
sensing BCI loop. Finally, view
-
invariant movement recognition of ve
hicle will
be studied in the third year. View
-
invariance classification adds more flexibility to
application exploitation, so do to this project. This subproject belongs to the area of
ubiquitous computing, a famous example like the “oxygen” project of MIT
.

NSC 97
-
2221
-
E
-
009
-
140

(
97R456
)

--------------------------------------------------------------------
----------------------------


Title
:
Design
a
nd Fabrication
o
f
Light Emitting Diode
w
ith
High Direction
And

Efficiency
v
ia
Nano
-
Imprinting
o
n
Photonic Cry
stals
o
f
Surface

Principal Investigator
:
Chang
-
Po Chao

Sponsor
:

National Science Council

Keywords
:
Photonic
Crystals, Nani
-
Imprinting, Light Emitting Diode, High


Efficiency,

High Directionality


This proposal is planned to use the method of nano
-
imprintin
g to fabricate

photonic crystals on the top surface of the LED, in order to render high efficiency and

directionality. The developed LEDs with high efficiency and directionality will be

utilized as the light sources in the backlight unit (BLU) with the nan
o
-
imprinted wire

grid polarizer (WGP), leading to a high
-
efficiency BLU.

The concept of photonic crystals was first introduced in 1987 by Yablonovitch
and

John. With significant progress in fabrication technology and precision machine
later,

the technology

for fabricating varied structures of a LED become more complete
with

associated experimetnal validation. The photonic crystals then draw much
attention.

In the 2D or 3D periodic structures of a typical photonic crystals, the
material

refraction indices be
come varied in periodic fashion. This variation mimic
atomic

alignment in solid crystals. Therefore, it is similar to band gaps for the
electrons in

solid crystals. In results, the photonic crstals generate the effects of band
gaps for

photons. There are t
wo mechanisms proposed in the porposal to increase the
emission

efficiency: (1) induce Bragg diffraction by the periodic structure of photonic
crystals

in order to mitigate internal reflections; (2) using the photonic crystals to
induce

guiding mode in ord
er to increase external quantum efficiency. On the other
hand, due

to the band gap effects along the giuding mode caused by the periodic
structres and

photonic crystals, the high directinality of the LED can be achieved. In
the first year

of the project ex
ecution, the LED epitaxy and the dimensions, materials
and structures

of Photonic crystals will be designed. The method finite difference time
domain is

then used to simulate the optics, in order to raise light efficiency and
directionality.


In the second

year of project execution, a process of nano
-
imprinting will be

developed to realize the photonic crystals on LEDs. The method of nano
-
imprinting

was first introduced by Prof. Stephen Chou with merits of low cost, simpleness and

high productivity. Most im
portantly, the nano
-
imprinting is not limited by the

wavelength of light source and then capable of fabricating pattern with resolution less

than 100nm. This nano
-
imprinting is then a potentially high
-
value semi
-
conductor

processing technology. For photo
-
l
ithography, even a lightly
-
curved profile is
difficult

to

fabricate due to the limitation of focusing depth. To nano
-
imprinting,
however, the

patterning by the contact
-
imprinting technology is s better solution. The
first step of

nano
-
imprinting is the man
ufacturing of a mold to realize the pattern. This
proposal

will use the diamond mold fabricated by the sub
-
project 3, and then anlayze
the

process parameters for fabrication quality for achieving optimum photonic crystals.
In

the third year of the project
execution, the Taguchi anf Genetic Algorithm will be
used

to optimize the efficiency, directionality and yield rate.

NSC 97
-
2221
-
E
-
009
-
057
-
MY3

(
97R163
)

--------------------------------------------------------------------
----------------------------


Title
:

Design
a
nd Implementation
o
f A
Novel
Tunable
LC
Lens
i
n
Compact Camera

Modeules
o
f
Mobile Phones

Principal Investigator
:

Chang
-
Po Chao

Sponsor
:

National Science Council

Keywords
:

Liquid
Crystal Lens, Tunable Focal Length Lens, Auto Focusing Lens

Module,

Compact Carema Mudule (
CCM
)


This proposal is to design a novel tunable lens module using the birefringence
property of liquid crystals (LC). The tunability of the designed LC lens in the foucs
length is made possible by adjusting applied filed to turn th
e LCs inside the len to
desired postures, acting as an gradient index lens (GRIN). This LC lens is best
suitable for replacing moving objective lens in the compact cameras modeule (CCM)
for mobile phones to implement the functions of auto
-
focusing and zoom
ing. It is
advantageous to replace conventional lens with lightweight and compact in some
applications, include micro
-
lens module, cell phone component, fixed focal length
web camera, blu
-
ray disk, three
-
dimension (3D) display, and optical tweezer. One can

design positive and negative LC lenses to be combined with CCDs with conventional
lens components for the LC lens module in order to performiong auto
-
focusing and
zooming. In recent years, the on
-
line tunability in the focal length of a common
variable fo
cus lens is achieved by the mechanical motion of the individual lenses in
CCM. The control mechanism for lens motion is generally complicated, and the
movements of the lenses usually need a large space. To remedy the problem, the LC
tunable lens is rpopose
d and used in the CCM herein to save space in mobile phones.
In the first year of project execution, it is planned to establish the optical model of the
LCs and predict the resulted index change distribution and the focus length. The
theoretical prediction
s are also validated by commercial software. With the LC lens
well designed, the lC lenses are optimized to lead to best focusing capability with a
measure defined. Finally, in this year, an inspection machine for measuring the
capability of the designed L
C lens. In the second year, based on the experience gained
in the first year, new LC materials leading to low voltage drives and fast reposnse
offered by sub
-
projects 5 and 6 are adopted for a better tunable LC lens. Also
conducted in this year are combini
ng the developed LC lens with the color
compensation chip developed by sub
-
project 3 to achieve high color gamma. Int eh
last year, the maturally
-
developed LC CMM is integrated with the image stabilizer
developed by sub
-
project 1 to form the novel CCM for
camera phones, as proposed
by the net
-
proposal.

NSC 97
-
2220
-
E
-
009
-
029
-

(
97R341
)

--------------------------------------------------------------------
----------------------------


Title

Design Optimization of Sigma
-
Delta Modulator ADC via Nonlinear and Power

Analysis

Principal Investigator

F.
-
C. Chen

Sponsor

National Science Council

Keywords:

Nonlinear Analysis, Power Consumption Analysis, Design Optimization










NSC96
-
2221
-
E
-
009
-
225
-
MY3 (96N417)

-------------------------------------------------------------------------------------------------------


Tit
le

Current Sensorless Control for Single
-
Phase Interleaved Boost
-
Type SMR

Principal Investigator

Hung
-
Chi, Chen

Sponsor

National Science Council

Keywords

S
witch
-
mode
R
ectifier
,

C
urrent
S
ensorless
C
ontrol,
A
ctive
P
ower
-
factor


C
orrection


With the develop
ment of high
-
efficiency inverter
-
fed technology, more and
more attentions are put on the utilizing quality of input power. Many power factor
correction (PFC) techniques had been proposed to improve the front
-
end power
factor. The bridge circuits cascaded w
ith boost
-
type dc
-
to
-
dc converters (or called
switch
-
mode rectifier, SMR) are mostly used in the power circuits of the PFC
schemes.

Based on the previous result of research, the current shaping performance is able
to be improved. Therefore, a two
-
year pro
ject are proposed. In the first year, we put
attentions on the improvement of performance of current shaping. In addition, the
current s
ensorless

control is preferred in the multi
-
phase interleaved boost
-
type
SMRs for that the number of control inputs is f
ixed regardless the number of
multi
-
phase. Therefore, in the second year, the proposed current sensorless control
will applied to the multi
-
phase interleaved boost
-
type SMRs. To achieve this goal,
the research works are arranged as follows:


First Year
:

(1
)

Analysis of the stagnated current with the current sensorless control

(2)

Establishment of DSP
-
based SMR control, including the set
-
up of the main
power circuits, simulated load and interface circuits and the DSP codes.

(3)

Improvement of the current
S
ensorless

control.

(4)

Simulation of the improved current s
ensorless

control.

(5)

Experimental demonstration of the improved current s
ensorless

control.

(6)

Report preparation.


Second Year
:

(1)

Establishment of DSP
-
based multi
-
phase interleaved SMR contro
l, including the
set
-
up of the main power circuits, simulated load and interface circuits and the DSP
codes.

(2)

Design of the current
Sensorless

control for multi
-
phase interleaved SMR.

(3)

Simulation of the proposed DSP
-
based current sensorless control

(4)

Design of the proposed DSP
-
based current sensorless control

(5)

Implementation of the proposed current sensorless control and performance
evaluation.

(6)

Report preparation.

NSC97
-
2221
-
E
-
009
-
184

(
97R458
)

------------------------------------------------
-------------------------------------------------------


Title:

Design
o
f
t
he Driver
IC
And Algorithm
f
or Zooming
o
f
LC
Lens

Principal Investigator:
Ke
-
Horng Chen

Sponsor:
National Science Council

Keywords:
DC
-
DC
Converter
,
Zoom
,
Heterogenous Integration



The technology is changes with each passing day. The requirements of potable
device

are grow with each passing day. Wireless communication and potable devices
are popular in today’s technology. Low power designs become more and more
important. Owing to th
e growth of low power designs, we even can use the ambient
energy sources for our system. As we know, it is so
-
called green energy. It also means
that this system is a self
-
sustention operating system using green energy as energy
source. Thus, how to conve
rt and manage this tiny energy becomes a major topic in
this system’s design. Our research aims at a smart DC
-
DC converter and adapts a
suitable power management theory in the self
-
sustention system. Based on this
architecture, this self
-
sustention can get

stable supply energy under the optimal energy
transferring efficiency. We can use rectifier and DC
-
DC converter efficiently convert
the unstable voltage to a specified and stable output voltage. The key point of this
project is to supply the driving energ
y of zooming of the cell phone camera. Owing to
the life time of camera battery is finite, power management becomes the most
important issue of power dissipation. Besides, due to the development of technology,
cell phone has become very popular The functio
n requirements are also increasing
each passing day. Therefore, under the trend of technology combination, the functions
of cell phone become more and more. And the combination of digital camera and cell
phone is the most basic equipment. We proposes a “Th
e study of a novel
heterogeneous integrated compact camera module” to solve the cardinal problem. We
integrated LC lens and conventional lens together to make the size of zooming
mechanism smaller without losing the function of zooming. This kind of techni
que
not only provides low power consumption but also extends the standby time of cell
phone.

NSC 97
-
2220
-
E
-
009
-
027

(
97R339
)

--------------------------------------------------------------------
----------------------------


Title
:

Development of
M
ulti
-
functi
on MR
C
ontrast
A
gent for
Dru
g
D
elivery and
T
reatment of Parkinson's

Principal Investigator

You
-
Yin Chen

Sponsor

National Science Council

Keywords

Magnetic
R
esonance
I
mage, Contrast
A
gent, Ultra
-
small
P
aramagnetic

I
ron
O
xide, Blood
-
brain
B
arrier


The aim of this sub
-
project is to develop a novel magnetic resonance (MR)
contrast agent with creat
ing significant image contrast and drug delivery in vivo. Due
to its biocompatibility and strong effects on T2(*) relaxation, iron oxide nanoparticles
appear to be the contrast agent of choice, and self
-
assembled monolayer ultra
-
small
paramagnetic iron oxi
de (USPIO) nanoparticles are coated with polyethylene glycol
(PEG).

Furthermore the USPIO provide an active functional group capable of
conjugating with therapeutic agent. In the main project, the image
-
guided focused
ultrasound brain therapy system for P
arkinson’s disease in animal model will be
established. It can conduct localized neural tissue thermal ablation. USPIO can
increase contrast
-
to
-
noise ratios (CNRs) over the perfusion MR imaging that leads to
the evaluation of location and distribution of t
hermal ablation in the brain. The
focused ultrasound brain therapy system also can be used to increase the permeability
of blood
-
brain barrier (BBB) and USPIO labeled drug can have the better penetration
across the BBB. It is a potential method of in vivo
evaluating the progress of
Parkinson’s disease and treatment strategies.

NSC 95
-
2221
-
E
-
009
-
171
-
MY3 (
95R481
-
1
)

-------------------------------------------------------------------------------------------------------


Title
:

Applied
Intelligent Binocular Robo
t t
o Motion Detection, Prediction
a
nd

Description

Principal Investigator
:

Yon
-
Ping Chen

Sponsor
:

National Science Council

Keywords
:

Intelligent Binocular Robot, Motion Detection, Prediction, Description,

Neural

Networks, Grey Prediction Rule


This projec
t proposes an intelligent binocular robot, or called eyeRobot, to detect,
predict and describe the motion of objects in two years. Recently, some related key
technologies have been devloped and listed as below:

1. Machanism design of human
-
like eye movemen
t robot.

2. Image stitching technology.

3. A novel object extraction algorithm in 3D model reconstruction for objects in

arbitrary background.

4. Evolution strategies
-
based learning control system.

5. Discrete time sliding
-
mode control design with grey pre
dictor.

6. A CAM
-
based license plate character recognition system using the pattern

accumulated vector method.

7. DSP
-
based motor control.

Since general image stitching requires a large amount of computation time, this
project instead adopts the neural net
work for the stitching of binocular images, based
on the training process, learning algorithms, and prediction. Furthermore, the control
of eyeRobot would be implemented on the hardware of DSP board. This project
would take two years to achieve the followi
ng goals:

1st year: In order to detect objects, the eyeRobot would learn to stitch binocular
images and construct stereo visual perception based on neural networks.

2nd year:In order to acquire the spatial
-
temporal relationship of objects, the eyeRobot

wo
uld be developed to predict and describe the motion based on learning in neural
networks and prediction with grey theory. Furthermore, the control technology to
track moving objects would be implemented on the DSP board. After this project, the
developed e
yeRobot would be applied to monitoring system that effectively and
efficiently provides the information of the motion of objects under supervision.

NSC 97
-
2221
-
E
-
009
-
089

(
97R450
)

--------------------------------------------------------------------
---------
------------------


Title:


Principal Investigator:

Mu
-
Huo Cheng

Keywords:

Wearable
T
extile
E
lectrode, Baseline
V
oltage, Adaptive
S
ubspace
T
racking, Maximum
L
ikelihood
E
stimation, FPGA
I
mplementation


This project aims to implement a portable real
-
time he
art
-
rate measuring system
in single chip. Since the world population is getting older, the need for high efficiency
and high quality medical system is becoming urgent. Using the tele
-
home healthcare
and comfortable wearing sensors for measuring vital signa
ls has been considered
important milestone in the world. Hence, intensive investment in research projects,
such as the VTAMN project in France, WEALTHY project in Europe, and the
LifeShirt project in US has been made. This project aims to realize a single
-
chip
system which can use ECG signals from wearable textile electrodes to estimate the
real
-
time heart
-
rate of the wearing person. This system, therefore, should include the
analog signal processing circuits as well as the digital signal processing algorit
hms.
Since I have been doing research for many years focusing on the development and
application of digital signal processing algorithms, I still need more experience and
more time to learn on the integration of analog and digital systems. Moreover, we
hav
e been using discrete elements to implement a PCB board for electrode signals
amplification, filtering, and analog/digital converter(ADC).Therefore, this project
focuses on the implementation of digital signal processing algorithms for real
-
time
estimation

of heart
-
rate from wearable ECG signals.


Since the textile electrodes for ECG measurement are worn on the person, the
electrode contact may lose if the wearing is too loose. Also as the wearing person is
moving, or body jerking, the electrodes position
will be moving, making the electrode
contacts shifting, the measured ECG signals will often change dramatically, especially
the abrupt change of baseline voltage in signals. Hence, the digital signal processing
algorithms after the ADC of the measurement E
CG signals are often the key step to
exploit the ECG signals. We have researched and developed an adaptive subspace
tracking algorithm for tracking and removing the baseline voltage. We have also
applied the maximum likelihood estimation approach to develo
p the correlation
method for real
-
time heart
-
rate estimation. The developed algorithms have been
successfully applied to signals measured with many persons who are wearing the
textile electrodes and start with sitting still, then walking steadily, until jo
gging.


Therefore, this project studies to use a single
-
chip to implement an adaptive
subspace tracking algorithm for tracking the baseline voltage and implement the
correlation method for real
-
time heart
-
rate estimation. The architecture for the
real
-
time syste
m will be studied first. Then, the required operations, such as the
square root operation, the inner product, and the addition/subtraction of vectors in the
implementation of the adaptive subspace tracking algorithm will be studied and
realized. The operat
ions in realizing the correlation method such as the fast Fourier
transform(FFT) and finding the maximum from a set of data will be accounted. The
required operation counts and memory size will be also considered. Finally, the
system input to hand
-
shaking
with the ADC output and the system output parameters
for display will be elaborated.

This project will implement the system via FPGA and we shall demonstrate the
developed system. After the completion of this project, we shall continue to design an
applic
ation specific integrated circuit(ASIC) which will merge the digital circuit
blocks with the analog circuit blocks, yielding a real single
-
chip, portable real
-
time
heart
-
rate measurement system.

NSC97
-
2221
-
E
-
009
-
178 (97R417)

-------------------------------
-------------------------------------
----------------------------


Title:

The
Study
o
f
a

Novel Heterogeneous Integrated Compact Camera Module

Principal Investigator:

Jin
-
Chern Chiou

Sponsor:

National Science Council

Keywords:

Heterogeneous
Integration,
CCM
,
Anti
-
Shake,
MEMS
,
IC
Design,

Digital
Image Processing,
Liquid
Crystal Lens,
Zoom
,
Optical
Sys


An advanced and highly integrated research project in compact camera module
(CCM) is proposed in the next three years. The present project consists of six
sub
-
projects that are focused on the research of Anti
-
shake System, Driving IC for
Liquid Crystal Lens, Digital Image Processing, Optical Zoom System, Bistable Liquid
Crystal Lens, and Dynamic Fast Response Liquid Crystal Lens. The main goal of this
project i
s to develop a novel CCM with zoom and anti
-
shake capability. The abstract
of this three years project is: (1) to achieve defined performances and specifications of
each sub
-
project. The research items include designs, simulations and verifications of
rela
ted devices and fabrication processes. Encounter problems and solutions will be
discussed and if necessary, we will modify the designs and specifications; (2) to
accomplish the fabrication, testing and verification are performed to verify original
designs.

An optimal integrated solution will be provided to obtain the CCM; (3) to
integrate all the devices that are fabricated by the sub
-
projects into print circuit board
(PCB) and successfully manufacture the proposed CCM with heterogeneous
integration. Succes
sive tests will be carried out and the final performance will be
reported at the final stage of the project.

NSC 97
-
2220
-
E
-
009
-
026
-

(
97R338
)

--------------------------------------------------------------------
----------------------------


Title:
Sub
-
Projec
t 1 :
The
Design, Fabrication And Test
o
f
MEMS
Image Stabilizer

Principal Investigator:
Jin
-
Chern Chiou

Sponsor:
National Science Council

Keywords:
MEMS
,
Micro
Actuator,
High
Precision Image Stabilizer, Decouple Dual

Actuators


The first sub
-
project entit
led “The design, fabrication and test of MEMS image
stabilizer” of the integrated project plays the role of designing and fabricating and
testing image stabilizer with small form, high precision and low power consumption.
The main target in this project is

achieving image stabilizer by integrating decouple
dual actuator, sensors and other components. This project will be carried out in the
next three years, and abstracts are given as follows.

Abstract of 1st year project:

There are two targets need to be ac
complished in the 1st year consists of selecting the
suitable actuator by comparing driving displacement and decoupling and then
designing MEMS components and verifying the layout by using simulation tools.

Abstract of 2nd year project:

Fabricating a high
precision and large displacement decouple dual actuator is the main
target in 2nd year, in order to achieve the targets, fabricating recipes will be tuned to
optimization. Image sensor and decouple dual actuator will be integrated by using
flip
-
chip techno
logy. Furthermore, PCB will be used to integrate the developed
decouple dual actuator with components from other sub
-
projects.

Abstract of 3rd year project:

The main targets in the 3rd year are designing the controller and integrating with other
components

in a system. After accomplishing controller, the MEMS image stabilizer
will be achieved by integrating controller, decouple dual actuator, image sensor,
driving circuit and inertia sensor. Finally, the characteristics of system will be
measured and analyz
ed to verify the efficacy of the image stabilizer.

NSC 97
-
2220
-
E
-
009
-
044
-

(
97R354
)

--------------------------------------------------------------------
----------------------------


Title:

Sub
-
project 1


The Development of Three
-
Dimensional Recording


Micr
oelectrode Arrays

Principal Investigator:
Jin
-
Chern Chiou

Sponsor:
National Science Council

Keywords:



This sub
-
project is aimed for the development of a three dimensional
multi
-
channel microprobe array (TDMCMA) with the application of novel
Microelectro
mechanical technique. The TDMCMA with multiple sensing /
stimulating electrode and microfluid design can be applied in the field of implantable
biopotential monitoring applications. General two
-
dimension lateral probes can only
sensing in the same plane an
d lack of three dimensionally neural signal discussions,
and existing 3D assembly design is complex with the problem of large volume and
many and diverse parts, which dominates the prime cost. In this sub
-
project, a novel
TDMCMA is proposed with reliable a
ssembly solution and the capability of depth
and angle detection mechanism. By the integration with other sub
-
projects, the overall
system can toward the goal of SoC Platform Technology for Intelligent Prostheses.

NSC 97
-
2220
-
E
-
009
-
043

(
97R476
)

-----------
---------------------------------------------------------
----------------------------


Title
:

Integrated MEMS
Biomedical Senso
r
a
nd Front
-
End Analog Circuit
f
or


EEG
/
EKG
/Fnirs

Principal Investigator
:
Yi Chiu

Sponsor
:

National Science Council

Keywords
:

EEG
(Electroencephalography),
EKG
(Electrocardiogram), Fnirs

(Fuctional Near Infra
-
Red Spectroscopy), Dry Electrodes, Soc (System
On Chip),
ADC
(Analog
-
To
-
Digital Converter), Low
-
Power
Microcontroller, Embedded System


The goal of the project is to design and

fabricate integrated biomedical
dry
-
electrode

sensors and front
-
end analog circuit platforms for detecting and
processing EEG, EKG and fNIRS signals. The sensors and the front
-
end circuits can
be integrated into an embedded system
-
on
-
chip. This chip will
have very large
potential for practical use and commercial products. This three
-
year program is based
on integration technology to implement VLSI (very
-
large
-
scale integration), analog
CMOS integrated circuits and dry electrodes. The scope of this project
includes
CMOS image sensors for fNIR, programmable gain amplifiers (PGA) for biomedical
signals, pre
-
amp instrumentation amplifiers (IA) for EEG and EKG, analog
-
to
-
digital
converters (ADC) for multi
-
biomedical sensors and ultra
-
low power VLSI
microcontroll
ers. The circuits and sensors developed in this subproject will be
integrated in the embedded system of the main project.

NSC 97
-
2220
-
E
-
009
-
053
-

(
97R361
)

--------------------------------------------------------------------
----------------------------


Titl
e

Design of GI Image Signal Processor for Capsule Endoscopy

Principal Investigator

Lan
-
Rong Dung

Sponsor

National Science Council

Keywords

Image
C
ompression,
C
apsule Endoscopy
,

Retina
C
ircuit, Power
-
aware

Architecture


The objective of this project is to de
velop a power
-
aware image processor for
capsule endoscopy or swallowable imaging capsules. The image processor consists
three parts: disease target detector, image sensor, and image compressor. In
applications of capsule endoscopy, i
t is imperative to cons
ider
battery
life
/performance trade
-
offs
. Applying state
-
of
-
the
-
art video compression techniques
may significantly reduce the image bit rate by their high compression ratio, but they
all require intensive computation and consume much power from battery.

T
here are many fast compression algorithms for reducing computation load;
however, they may result in distoration of original image, which is not good for the
use of medical care.

Furthermore, the imaging capsules will travel in

alimentary canal
for three
hours.
Enabling all functions of the image processor over the three
-
hour trip will limit the
functionality and resolution of image processing because of insufficient battery power.
Hence, this project aims on three objectives: detection of diagnosing tar
get,
ultra
-
low
-
power g
astrointestinal

(GI) image compression, and power
-
aware
architecture. First of all, this project will apply the retinal processing circuit for the
disease target detection. The retinal processing circuit features high dynamic range,

low power, and high sensitivity to motion object. These features make the
capsule
endoscopy able to operate in the dim canal initially, with part of lights. When
sensing the targets, the endoscopy will turn on all the lights and the image processing
cir
cuits.

Secondly,
this project will simplify traditional video compression algorithms as
per the characteristics of GI image. With a specific scanning on the CMOS sensor,
we will propose a scalable compression algorithm for the later power
-
stepping
techniq
ue. Finally, we will develop a power
-
aware architecture for battery
-
life
extension
. The power
-
aware architecture is an architecture that can properly reduce
the computation load as the battery status changed while the qulity degradation is
little. In th
e project, we will consider not only the minimization of average power
dissipation but
transient characteristics of power dissipation
, such as

peak power and
power gradient or differential
.

First year

1. Petri
-
Net
-
based theorem proving of Architectu
r
al de
sign verification

2.
O
ne conference paper and one journal paper

3.
T
wo students earned master degree

Second year

1. A novel measurement on calculation of property coverage

2. A novel formal verification flow driven by property coverage

3.
T
wo conference
paper and one journal paper

4.
T
hree students earned master degree and one student earned Ph.D. degree

Third year

1. A high
-
level synthesis flow friendly with formal verification

2. Design
-
for
-
formal
-
verification technique

3.
T
wo conference paper and two
journal paper

4.
T
hree students earned master degree and one student earned Ph.D. degree

NSC
95
-
2221
-
E09
-
337
-
MY3 (
95R659
-
1
)

-------------------------------------------------------------------------------------------------------


Title

A
D
igital
C
alibration
S
cheme for the
D
esign of the
L
ow
P
ower
P
ipelined ADC
with
O
pen
-
l
oop
A
mplifiers

Principal Investigator

Hao
-
Chiao Hong

Sponsor

National Science Council

Keywords

Digital Calibration
,
Pipelined ADC
,
Open
-
loop Amplifier



As CMOS technolo
gy keeps shrinking down, the channel length of the MOSFET
is so small

such that there are only hundreds of atoms in it. Consequently, any subtle
difference between

two MOSFETs’ channels will make their electrical characteristics,
such as the threshold

volt
age, severely diverse. That’s the reason why the
design
-
for
-
manufaturability (DFM)

technique becomes an important issue for the
digital circuit design using deep sub
-
micron

CMOS processes.


For analog circuit designers, their situations are even worse than

the digital guys.
First, the

short
-
channel effect limits the intrinsic gain of a deep sub
-
micron MOSFET
down to only

ten or around. This makes designing a high gain amplifier very difficult
in a conventional

way. In addition, the supply voltage currently
is down to 1V for a 90
nm process to avoid

reliability issue. Under such a low supply voltage, many
traditional analog circuit design

techniques such as cascoding are no longer applicable.
The huge deviation of the electrical

characteristics of MOSFETs is
the most horrible
nightmare for analog designers. Just image

that the current ratio of a basic current
mirror now becomes unpredictable even the

MOSFETs are placed very close to each
other. How can we design the circuits which are

much complex than the cur
rent
mirror? Therefore, we have to figure out some novel design

methods to address these
issues in deep sub
-
micron analog circuit design.



An ADC is a basic building block for most mixed
-
signal systems and the
pipelined ADC is a

very popular ADC architect
ure. It is suitable for the application
requiring a sampling rate

around several tens MHz and a resolution around 8 to 10
bits. However, the pipeline ADC is

hard to achieve a resolution as high as 12 bits even
using an old process such as 0.35um. It is

the

mismatch of the devices that limits the
resolution of the ADC.

The digital calibration scheme is a promising approach for advanced ADC design.
The idea

is that if we can digitally estimate the characteristics of the practical circuits
without

interrupting

their normal operations, then we are able to calculate the errors
that the ADC’s

outputs have. By subtracting the digitized error from the digital output,
we can have the

correct digital outputs. That is, we don’t need high performance
analog block any mo
re

thanks to the additional digital calibration circuits. It leads to
another interesting scenario: if

the digital calibration scheme can calibrate both the
linear and non
-
linear errors, we can

replace the original feedback amplifiers in the
pipelined ADC
with the open
-
loop ones. The

open
-
loop amplifier achieves lower
power, small area, and wider bandwidth than the

closed
-
loop one. Similar to the case
of advanced MOSFETs, the major issue of using

open
-
loop amplifiers is the huge
deviation of their electrica
l characteristics. Of course, this is

not an issue by applying
the digital calibration scheme. In addition, digital calibration

schemes are more
suitable in advanced technology since its major hardware overhead is

digital and each
MOSFET in the digital cir
cuits becomes cheaper, smaller, and fasters when

its
channel length is getting shorter.

This project proposes to develop a digital calibration scheme for pipelined ADCs.
With the

new digital calibration scheme, we will realize low power, low cost, high
res
olution, and

high sampling rate pipelined ADCs. The project goals in the following
3 years are:

In the first year, we will develop a digital calibration scheme and verify the scheme by

behavioral simulations. The target is to develop the digital calibratio
n scheme that can

calibrate the pipelined ADC with open
-
loop amplifiers, while it does not suffer from
the

known issues of the state
-
of
-
the
-
art works such as being too complex to
implement and some

unrealistic assumptions on the input signal.

In the second

year, we will realize a 12
-
bits, 200MS/s, 150mW pipelined ADC with

open
-
loop amplifiers using a 0.18um CMOS technology. The test chip will be
analyzed to

find out the difference between practice and theory.

Base on the results of the second year, we will
improve our design and realize a
500MS/s,

12
-
bit, 100mW pipelined ADC using a 90nm CMOS process.

Based on our experiences in ADC testing and design, we believe we can meet the
project

goals on time.

NSC97
-
2221
-
E
-
009
-
169
-
MY3 (97R332)

-----------------------
--------------------------------------------------------------------------------


Title
:

Design
o
f

t
he Analog Front End
IC
For Power Line Communication
o
n

Automobiles

Principal Investigator
:

Hao
-
Chiao Hong

Sponsor
:

National Science Council

Keywords:


The
shortage of our major energy source, petroleum, seriously impacts our
economy. As the gasoline price keeps going ever higher, seeking for an alternative
energy source to replace the petroleum is a must. Besides, the global warming issue
due to the consumpt
ion of petrol fuels also attracts much attention. For the coming era,
we have to look for a cleaner energy source. Electricity is a good candidate. It can be
generated by nuclear, wind, water, and sun. All these energy sources produce no
carbon
-
dioxide. Ve
hicles are one of major power consumption sources of human
activity. Thus, we believe electricity powered vehicles will have a bright future, and it
does happen right now. Some hybrid cars already have been commercialized. Modern
cars have lots of electron
ic systems on them to control the actuation, safety, and
entertainment systems. Currently, it takes quite complex routing wires in the cars. As
a result, the reliability and the maintenance of the electronic systems become issues.
Since all electronic devi
ces must be powered by their power lines, if the signals can
be transmitted through the power lines, the routing wires in cars can be greatly
simplified to address the possible issues mentioned above. Therefore, this joint project
is going to develop a dc
power line communication (PLC) system for vehicles in the
coming 3 years. The system plans to use QAM as its modulation scheme to increase
its communication capacity. A high data rate is substantial for the target system since
some critical commands in the

vehicles must be real
-
time. ADSL is a similar system
that also employs QAM. To implement such a system, a high performance ADC is
mandatory. It is estimated that the ADC should achieve 10
-
bit resolution and be able
to operate at a sampling rate higher tha
n 10MS/s. Due to the limited energy capacity
of batteries, the power consumption of each component in the IC should be as low as
possible. Sigma
-
Delta modulator, pipelined ADC, and successive approximation (SA)
ADC are possible candidates. However, the pip
elined ADC and Sigma
-
Delta
modulator require operational amplifiers whose power consumption is considerable.
Besides, the Sigma
-
Delta modulator needs a moderate over
-
sampling ratio that makes
the power consumption of its operational amplifier even higher.
On the other hand,
the SA ADC does not need any high power amplifier. It may be more suitable for this
application. What this research project proposes are: In the first year, we will design a
10
-
bit, 10 MS/s SA ADC whose power is less than 1mW. In the sec
ond year, we will
roughly estimate the channel model and use it to design the signal conditioning
circuits, the anti
-
aliasing filter. In the third year, we will finish the design of a 10
-
bit,
10MS/s DAC and integrating all the circuits with the results of
the other sub
-
projects.
Finally, we will demonstrate a real PLC system.

NSC 97
-
2220
-
E
-
009
-
048

(
97R356
)

--------------------------------------------------------------------
----------------------------


Title
:

Minimum Energy Control of
i
n
-
W
heel Motors with R
eal Time Road Adaption

Principal Investigator
:

Te
-
Sheng Hsiao

Sponsor
:

National Science Council

Keywords
:

Electric Vehicle, Tire Force Estimation, Vehicle Motion Control


The lack of petroleum in the near future inspires the study of electric vehicles. In
addition to its efficient and clean power usage, the electric vehicle has many other
advantages over the internal combustion engine vehicle. Firstly, the four wheels of the
electric vehicle are driven directly and independently by individual motors; theref
ore
the angular velocities and torques of each wheel are known exactly. The tire force as
well as the road adhesion coefficient can then be estimated with high accuracy.
Besides, motors respond much faster than internal combustion engines; hence the
electr
ic vehicle can move more agilely provided that the motion control system is
designed properly. This project is going to develop an efficient and robust estimation
algorithm for the tire force and the road adhesion coefficient. Taking advantage of the
rich
information about each wheel and the novel tire dynamic model developed
recently (the so
-
called DDT model), we can enhance the accuracy of the estimation,
especially in the low
-
speed and low slip ratio regions. The estimated tire
-
road
information will then

be incorporated into the vehicle motion control system (e.g. the
traction control or the yaw
-
moment control system) such that the controller becomes

“road
-
adaptive,” i.e. the motor outputs are self
-
tuning in accordance with the road
condition. Moreover, t
he energy consumption will also be taken into account in the
design of the control system in order for the electric vehicle to be stable, safe and
power efficient.

NSC 97
-
2220
-
E
-
009
-
049

(
97R357
)

-------------------------------------------------------------
-----------------------------------


Title
:

Design
a
nd Implementation of
T
he Sensor
-
Network Distributed Rescue Robot

Principal Investigator
:
HSU PL

Sponsor
:

National Science Council

Keywords
:

Rescue Robot, Design
a
nd Analysis, Control, Wireless Local Area

Networks,

Wirele
ss Sensor Networks, Vr Training


The rescue in the disaster occurrences in a short period is a critical issue in the
modern technologies. This project will establish a series of intelligent mobile
platforms with different techniques of des
ign, analysis, control, wireless
communication, and artificial intelligence so that the modern sensing techniques can
be employed to achieve the efficient rescue task. It is also important to realize the
wireless networks, 802.11 and ZigBee, efficiently at

the disaster site so that the
communication channel between the rescue base and the person to be rescued can be
established through the wireless networks. This project will integrate the techniques
of design and analysis to develop an agile robot in a dis
tributed form. Furthermore,
the wireless communication network is established so that the image and voice can be
transmitted to the rescue base through the network. The path of the robot history will
be automatically recorded so that water and food can be
transported to the person to
be rescued accurately. On the other hand, this project will train the rescue robot to be
more intelligent in motion with the virtue reality technique. There are four professors
will involve in this project with different expert
ise: solid mechanics, automation,
control, and information science. This project will take three years to integrate the
wireless sensor networks and the distributed robots to establish the platform. The
developed platforms will become a solid base for the
research on remote control and
monitoring systems in the future.

NSC 97
-
2218
-
E
-
009
-
006

(
97R484
)

--------------------------------------------------------------------
----------------------------


Title
:

Research on Remote Control for Distributed Robots thro
ugh Wireless

Networks

Principal Investigator
:

HSU P
.
L
.

Sponsor
:

National Science Council

Keywords
:
Sensor Networks, Distributed Robots, 802.11,
Zigbee


In the area where the catastrophic failure occurs, there is no available network
systems can be utiliz
ed to communicate between the rescuing robot and the base.
Therefore, the rescuing robot can drop some communication nodes during the travel

to establish a sensor network. In available wireless systems, the Wi
-
Fi has a desirable
range. However, its energy
consumption is relatively large. On the other hand, the
blue tooth is a suitable system of which it has been considered as applications in the
cell phone. The ZigBee is a suitable one for sensor networks consuming energy
insignificantly. It will take three

years for this project to establish a sensor network
based on the ZigBee. Besides, it will be integrated with the inertial guidance control
and the image and sound signals so that the robot can function as a communication
channel between the rescued targe
t and the rescuing base. In addition, supportive
water and food can be transferred to the target.

NSC 97
-
2218
-
E
-
009
-
007

(
97
R48
5
)

--------------------------------------------------------------------
----------------------------


Title

A Robust Perception Interface for Human
-
robot Interaction Using Combination
of Speech Source Localization and Human Posture Recognition

Principal Investigator

Jwu
-
Sheng Hu

Sponsor

National Science Council

Keywords:

Intelligent Robot, Intelligent Man
-
machine Interface, Microphone Array,
Sound Source Localization, Sound Source Tracking, Sound Source
Separation, Human Posture Recognition, 3
-
D Image Recognition, Image
Feature Extraction


Among various sensing technologies, it is commonly recognized that i
mage and
voice are till the most straightforward interface when considering the human
-
robot
interaction. Despite many years of research in these two interfaces, the robustness of
using image and voice still remains an important issue for real practice. The

robustness of any interactive interface for robot is the key to its realization. And the
associated theoretical problems need to be studied with a greater depth. This proposal
plans to study the robustness problem based on previous research results. Speci
fically,
the proposal will try to establish the robustness results using high
-
dimensional
microphone array and 3
-
D object recognition technology. The plan covers a 3
-
year
study in the following topics: 1. High
-
resolution full
-
digital microphone array desig
n
and implementation; 2. Multi
-
dimensional microphone array and mobile microphone
array for sound source tracking; 3. Human posture recognition, memorization and
tracking using mono
-
vision; 4. Combining source tracking for object occlusion
recognition.

NSC
96
-
2628
-
E
-
009
-
163
-
MY3 (96N459)

-------------------------------------------------------------------------------------------------------


Title
:

The Research
a
nd Applications of
T
he Image Quality Improvement

Algorithms
for LC Lens

Principal Investigator
:

She
ng
-
Chieh Huang

Sponsor
:

National Science Council

Keywords
:
LC Lens
,
Image Quality Improvement Algorithms
,
Face Diagnosis
Based

On
TCM
,
Emotion Evaluation
,
Intraocular Lens SOC


This plan is the third part of the “Novel Heterogeneous Integrated Compact
Cam
era

Module “. Since the color of the image distorted when it passes through the
lens, the key idea of the image algorithm is based on the human visual system and is
to improve the image quality of the LC lens. In addition, the dedicated LC Image
Pipeline w
ill be further developed for the LC lens. Since the digital image size is so
large, the latest HD Photo image compression JPEG XR standard is used to be built
the efficient compression engine for the high dynamic range 1920 x 1080 30
frames/sec. Therefore,

the dedicated LC image pipeline and the JPEG XR Codec
engine can be combined into the SOC engine for improving the image quality of the

novel heterogeneous integrated compact camera module. For higher profit
applications, this image improvement engine can

be used to the Traditional Chinese
Medicine’s (TCM) face diagnosis and emotion evaluation. We live in a world that the
number of elderly people is increasing faster than ever before. The eyes of elderly
people often need to be replaced the intraocular len
s to recover the sight. Hence, the
intraocular lens SOC will be used to get better sight with the improvement of the
process technology.

NSC 97
-
2220
-
E
-
009
-
028

(
97R340
)

--------------------------------------------------------------------
--------------------
--------


Title
:

The Development of
t
he Embedded Software
a
nd Hardware for Intelligent

Life
-
Assistant Robot

Principal Investigator
:
Yu
-
Lun Huang

Sponsor
:

National Science Council

Keywords:


Life
-
assistant robot is a kind of robot system that can help elde
r people to
improve

their life qualities. A life
-
assistant robot system includes the robot walker,
robot

follower and robot companion. The robot walker can help relieve the
degenerations of

sense organs and movement coordination capability. The robot
follo
wer can help elder people find the specified destinations easily and carry heavy
stuffs. The robot

companion can collect the vital information as the references of
medical diagnosis,

while it can also help elder people perform some essential health
exercis
e at home in

order to develop and sustain their muscle strength and balance. On
developing such kinds of robotic systems, earlier designs took the pure hardware
oriented approaches that almost all required functionalities were implemented via
dedicated har
dware components and integrated circuits instead of general
-
purposed
processors running appropriate software systems. Unfortunately, such pure hardware
based approach has the problem of flexibility and extension. Alternatively, if certain
parts in the syst
em can be redesigned and implemented with the software
-
based
approach, the manufacturing process can be more cost effective than designing the
whole hardware system again. In this project, we plan to take the advantage of the
modern embedded software syste
m that provides multi
-
tasking and scheduling,
parallel and real
-
time processing features. The proposed embedded software approach
can effectively solve the issues that may occur during the robot development and
future extension stages, while providing a ri
cher specification to better fit the robotic
system requirements of each stage. In general, the proposed embedded hardware and
software platform can be considered as the core system of life
-
assistant robots and
serve the purposes of integrating and control
ling various peripherals and sensors. The
embedded platform can also collect all the information and take the appropriate
actions based on what it has known. As a result, researchers in this project will focus
on the scopes of intelligent movement assistan
t and life
-
assistant robots and pursue
the necessary survey of suitable embedded hardware and software platform that can
accomplish the designated goals.

NSC 97
-
2221
-
E
-
009
-
133

(
97R453
)

-------------------------------------------------------------------
----
-------------------------


Titl
e:

Study
o
f Reliable Sliding Mode Control Based
o
n
T
-
S
Fuzzy System Model

Principal Investigator
:

Yew
-
Wen Liang

Sponsor
:

National Science Council

Keywords
:

Nonlinear
Control Systems, Sliding Mode Control,
T
-
S
Fuzzy Model,

Re
liable Control, Stabilizability Analysis, Tracking Performance



Due to the growing demands for system reliability in a highly automated
industrial system and in aerospace missions, where repair and maintenance often can
not be achieved immediately, the st
udy of reliable control has become paramount
importance and has attracted considerable attention. On the other hand, since the
modern control systems are constructed more and more complicated, the employed
control strategy and the time for controller imple
mentation have become extreme
importance. In fact, the two mentioned
-
factors have a strong relation to the quality
and the efficiency of the control mission. In this project, we are about to combine the
T
-
S fuzzy model approach and the Sliding Mode Control

(SMC) scheme for
alleviating the computational burden and promoting associated system reliability
performances. The reason for adopting SMC scheme comes from its own advantages
including responding rapidly and robustness to uncertainties and disturbances,

while
T
-
S fuzzy approach allows one to save lots of on
-
line computational burden, which is
especially important for those systems with highly nonlinear and complicated
dynamics. The combined scheme is expected to save lots of on
-
line computational
burden
while achieve efficiently control objective. In the first year of this project, we
will explore the reliability issues using the combination of SMC scheme and T
-
S
fuzzy approach. For the second year, we will study the application of the first year’s
theore
tical results

to practical systems for efficient control and system reliability.

NSC 97
-
2221
-
E
-
009
-
087

(
97R448
)

--------------------------------------------------------------------
----------------------------


Title:


Principal Investigator
:
Der
-
Cherng Lia
w

Sponsor
:

National Science Counci
l

Keywords
:

Wireless
N
etwork,
R
eliability,
W
ireless
S
ensing


Due to the unexpected natural disaster occurs every day, the people might be
shocked and/or afraid of daily life. For instance, the well
-
known 921
-
earthquake
occ
urred in Taiwan caused the loss of hundreds of people’s lives. How to have an
early warning like a surge signal for the possible appearance of disaster from the civil
structure becomes a very important issue. The major goal of this three
-
year
sub
-
project i
s to find and construct such a possibility. By
the development of wireless
sensor networking with the interfacing from the civil structure sensors, in the
following three year this sub
-
project will cooperate with other two sub
-
projects to
design and verify

the function of the proposed platform. In the second year of the
project, the major tasks had been devoted to the development of the 10 sets of the
prototype for the designed wireless sensor module will be fabricated and tested with
the cooperation from t
he help of other five sub
-
projects. The final functional
validation will be carried on in the proposed third year of the project by using 50 sets
of the built wireless sensor module in the practical civil environment.


NSC97
-
2625
-
M
-
009
-
006 (97R124)

------
--------------------------------------------------------------
----------------------------


Title
:

Development of X
-
by
-
Wire Electronic Braking

Scheme for Vehicles (I)

Principal Investigator
:
Der
-
Cherng Liaw

Sponsor
:

National Science Counci
l

Keywords
:

V
ehic
le, X
-
by
-
wire,
B
rake,

R
eliable,
C
hip
D
esign,
M
onitoring

C
ontrol



In the recent years, due to the fast

growth of electronic devices, the advanced

technologies in communication and computer

engineering have been considered to be

applied to the control of ve
hicle systems.

Among those possible applications,
theso
-
called “X
-
by
-
wire” scheme has been

developed in steering control and braking

control of the vehicles in the labs. However,

due to the requirement of system
reliability,

such designs are not well devel
oped in

commercial usage. The major goal
of this

three
-
year research project is to develop a

FlexRay communication protocol
based

monitoring platform for “brake
-
by
-
wire” in

vehicle control. The project is
consisted of

five sub
-
projects. Three of them are d
evoted

to the development of such a
platform, while

two of five sub
-
projects, respectively, will be

more focused on system
reliability and the

chip
-
design issues. In the first year of the

project, the five
sub
-
projects were cooperated

to develop the system

design specifications.

The
fundamental researches have been

studied in the project such as the application

of
existing FlexRay modules, the preliminary

design of the electromechanical braking

(EMB) device, and the testing platform forvehicle control.

NSC9
7
-
3114
-
E
-
009
-
002 (97R521)

--------------------------------------------------------------------
----------------------------


Title
:

System Development
a
nd SoC Design of A Truely Portable Neuroimaging

System Based on EEG/EKG/fNIRS Multisensors

Principal Inv
estigator
:
Chin
-
Teng Lin

Sponsor
:

National Science Council

Keywords
:

System on Chip
,

EEG interface circuit
,

Analog to Digital Converter
,
Low

P
ower Processor
,

Embedded System


This sub
-
project integrates fNIRS and EEG measurement technology for novel

creati
on on brain research. Further more, considered the the influence of
hemodynamic factor, the project also combines EKG signal measurement into
system.Biomedical identification and application of multi
-
sensors platform for
EEG/EKG/fNIRS. It purpose provide f
or main project for scientific research related to
biomedical and assist to build up other sub
-
project system for biomedical
specifications and identification method. The result can help to setup the biological
safety standard and specification. The sub
-
pr
oject will play the role of key for the
main project to link the role of the biomedical information. The sub
-
project will
construct the EEG/EKG/fNIRS biomedical and electronic system identification
database. At the mean while, the original tool which base
on PC will rewrite to a 3D
neuron
-
image tool which is suitable for embedded platform. It establish its initial safe
standard and specification, and help the forth
-
project to build up the tool of 3D EEG
spectrum usage initial module to proceed a few parts o
f array graph to test

develop
the specification of the biomedical identification. The initial modules do the do animal
and human body research related to experiment. Key factor for every biomedical
interface, for example the initial optical and image syste
m test

the safe identification
of biomedical and electronic device and construct the instrument production behind of
the biomedical identification information and database etc., It will do scientific
identification related to this sub
-
project. In the futur
e use the development tool of this
project, and passing the electronics identification

the experiment of the animal , it can
compare with signal of the EEG/ECG/fMRI. It will application research that duct into
EEG driver status of the Brain Research Center

(BRC).

NSC 97
-
2220
-
E
-
009
-
052

(
97R360
)

------------------------------------------------------------------------------------------------


Title
:

Brain Computer Interface
a
nd Biofeedback Study with Moving Vehicle Control

Demonstration

Principal Investigator
:
Chin
-
Teng Lin

Sponsor
:

National Science Council

Keywords
:

EEG Signal, Brain
-
Computer Interface, Moving Vehicle Control

Application, Bio
-
Sensors


Current brain computer interface (BCI) systems provide users with
communication channels that do not depend o
n peripheral nerves and muscles.
Specifically, the BCI system use electroencephalographic (EEG) activities recorded at
the scalp to control or operate external devices. The central processes in each BCI
included two major steps. First, the BCI systems anal
yzed and extracted the
significant features from the electrophysiological inputs, and then BCI systems
encoded those results into commands and express them in device control. However,
there are three limitations at present BCI systems that critically affec
ted the system
performance. First, the bio
-
sensors for BCI systems were designed for a well
controlled recording environment. Second, most BCI systems used the binary control
model to operate external devices. The above two limitations will restrict the
ap
plications of BCI technologies. Third, the performance of BCI was greatly
influenced by the user’s performance and attention since the user was included into
control loops of the BCI system. Therefore, the main goal for this project is
integrating the Micr
o Electro Mechanical systems (MEMS), computer science,
bioinformatics and electrical engineering to develop a new BCI system that can be
easily applied on a moving vehicle. The new system will include high fidelity and
sensitivity EEG sensors with an intel
ligent multi
-
stream physiological signal
recording and analysis system. In addition, the system will control the moving vehicle
with multi
-
comments that extracted from the motor imagery. Furthermore, it will
incorporate a multi
-
dimensional biofeedback cont
rol loop to increase the user’s
navigated performance or attention that can get better
signal to noise ratios of EEG
signals. This project will be the first time to successfully
integrate
the Micro Electro
Mechanical Systems (MEMS), computer informatics, b
ioinformatics and e
ngineering.
Both the concept and its application for control technology are revolutionary. In
addition, such innovation will make a significant contribution to promoting the
biomedical industry, and widespread applications on education,
clinical treatment,
enhancing the elders’ cognitive function and the safe warning system for people who
needed to highly concentrated on their work.

NSC 97
-
2221
-
E
-
009
-
138

(
97R454
)

--------------------------------------------------------------------
--------
--------------------


Title
:
Development of Biofeedback Training System to Enhance Human Spatial

Navigation

Principal Investigator
:
C.T. Lin

Sponsor
:

National Science Council

Keywords
:
Spatial Navigation, Disorientation, Biofeedback, Virtual Reality,

Dyna
mic Platform, Physiological Signals


In this subproject, we focus on solving the problem of human disorientation and
enhance their spatial navigation ability based on the development of virtual
-
reality
(VR) dynamic platform and biofeedback technologies in
the past few years. Once
subjects lose their orientation, they are unable to control the vehicle to reach the
correct target. This kind of disorientation problem is quite important but less
discussed and investigated recently. Therefore, this subproject w
ill investigate the
relation of subjects’ spatial navigation between behavior and physiology responses.
Then, it will enhance their spatial navigation relying on training programs and
biofeedback technologies. While searching the desired target, possessing

well ability
in sense of direction can help human to save more time and power. When we are
determining the direction, human always makes deviation error with the actual
direction due to use different strategies. In the next three years, we will utilize ou
r
development of VR technology to investigate the human difference by using several
of spatial navigation strategies. In the meanwhile, we will also investigate the
correlation between subjects’ behavior data and physiological signals under passive or
acti
ve guidance conditions. At last, biofeedback technology with EEG signal
modulation in active brain area can help training human with disorientation problem
to change their strategies in spatial recognition and expect to enhance their spatial
navigation ab
ilities. During the following three years, this subproject will focus on
three major topics including (1) investigating human strategies of spatial navigation
and correlations with physiological signals. Establishing spatial navigation database
based on ph
ysiological signal and investigating the difference with different strategies
in spatial navigation. (2) Exploring the effect on multi
-
sensory activation in spatial
cognition and navigation, such as visual perception, body sensation and motor control.
(3)
Finally, we can modulate the physiological signal to train some well strategies to
enhance human spatial navigation ability. We can expect to improve the performance
via biofeedback technology because subjects after training programs of spatial
navigation
can effectively control the moving vehicle to reach the correct target.

NSC97
-
2221
-
E
-
009
-
141
-

(
97R457
)

---------------------------------------------------------------------------------
---------------


Title
:
Ecology
-
City: Network Platform for Lifestyle of
Healthy and Sustainability

Principal Investigator
:
Lin Chin
-
Teng

Sponsor
:

National Science Council

Keywords
:
Life of Health and Sustainability
,

Humanity

and Social Science,



Technology
,

Networking
,

Industry
,

Hospitals
,

“Eco
-
City: Health



and Sustainability” Regional Integration Center

for Intelligent


Mobile Technology
,

C
o
-
enrichment of

H
uman and
T
echnoscience for


the
U
ser
-
oriented
T
echnology
,

S
cenario
-
oriented

I
nteractive



D
emonstration and

E
xperience
P
latform”
and

I
nteractive
N
etworking


P
latform

,

Co
-
life
,

T
hree
-
stage
M
odel



While the mass production technology of industrial revolution in the 19
th
century
has

satisfied the basic needs of human, and the spatial and temporal tran
sformations
brought

about by electrical and information technology in the 20
th
have re
-
shapen our
lives and turns

the world into a global village, how to further realize the potential of
science and technology

so as to meet the human needs of living the
Li
fe of Health
and Sustainability
(LOHAS) has

been an important issue in the 21
th
century. With
this aim in mind, our interdisciplinary team

wants to explore the lifestyles and the
needs of elderly and high
-
stressed populations in order

to realized science a
nd
technology for our users’ better quality of lives.


Think globally and act locally. In Hsin
-
Chu, the elderly and high
-
stressed
working

populations have grown rapidly along with the decrease of birth rate and the
development of

high tech industry. These
are common in the industrialized countries,
Taiwan included. These

two populations have been suffering different dis
-
abled
situations. For elderly people, they

have been dis
-
abled by the losses of essential
cognitive and body functions due to the aging

cor
poreality. For high
-
stressed working
people, they have been dis
-
abled by the stress and

less active lifestyles due to the
temporal and spatial compression caused by the demanding

workload.


This project will integrate the interdisciplinary teams and expert
ises of: (1)
Humanity

and Social Science,
including Social science (Humanity and Social
Science College, NTHU),

Cognition (Cognitive Psychology teams, NYMY and NCU)
and Ergonomics (Applied Arts

Institute, NCTU);
Technology,
including Engineering
Technology

(Electrical and

Information Colleges, NCTU and NTHU) and Technology
on Architectural

Space(Architecture and Urban Design Institute, CCU);
Networking
(National Center for

High
-
Performance Computing), (2)
Industry,
including
Hsin
-
Chu Science Park (HCSP)

Adm
inistration and companies in HCSP, and (3)
Hospitals,
including Hsin
-
Chu General

Hospital, Ton
-
Yen General Hospital, and
Taipei Veterans General Hospital. We are working

together to develop
“Eco
-
City:
Health and Sustainability” Regional Integration Center

for Intelligent Mobile
Technology
to alleviate their dis
-
abled situations and realized the

LOHAS for elderly
and high
-
stressed users.

The substantial aims of this center are:

1. Alleviate the dis
-
abled situations of the ta
rget populations and create the

in
frastructure

for healthy and sustainable lifestyles.

2. Creating an environment for interdisciplinary innovation and construct a flexible
and

sustainable chain for realizing the innovations.

3. Embody and multiply the benefits and effects of living science

and living tech

industry.


In order to achieve the aims, we will firstly embody the ideas of
co
-
enrichment
of

human and technoscience for the user
-
oriented technology
to integrate our
achievements

of interdisciplinary and national projects by facilitating

the flow in the
chain of knowledge

supplies from collaborative teams. Secondly, we will induce user
driven open innovation by

constructing a sustainable loop for innovation. Finally,
construct a
“scenario
-
oriented

interactive demonstration and experience
platform”
and
“interactive networking

platform”
to create the interdisciplinary
communications among industry, government, and

educational and research fields.

In terms of the interactive demonstration and experience platform, we have an
Open
-
Lab,

an Open
-
House and the outdoor campus space, we will also develop an
Open Cargo and a

concept car that demonstrate a whole day living scenario, including
home, traffic, office, and

the exercise and leisure space, for high
-
stressed and elderly
populations. It is wor
th noticing

that our mobile Open Cargo will go to other sites and
enable the users to experience our

platform in their neighborhood.


For the interactive networking platform,
Co
-
life
of National Center for

High
-
Performance Computing will be the backbone of

this platform for integrating
the

educational platform, encouraging mechanism, and commercialization model in
order to

cerate value form an sustainable chain for innovation.


With these teams’ contributions, this center will develop in a
three
-
stage model

in terms

of its user participation and user
-
driven innovation and value creation. In the
first stage, we

will engage academics and industry and users to participate in open
discussions and concept

sharing activities. In the second stage, we will enlarge t
he
benefits and effects of the

knowledge chain by including more participants through
the mechanism of knowledge

sharing and securing their originality. In the third stage,
apart from the products, we will

create profit through ‘pay for service’ and ‘shari
ng
mechanism’ to support the operation of

this center.



Hsin
-
Chu is the best city to realize our center; the percentage of elders in it
population is

above the average of our country, and it host the high tech working
population of the first

science park
in Taiwan. Collaborating with the NSC project
office and other regional centers,

it is believed that this project we are proposing will
realize and popularize the model of

Eco
-
City for not only alleviating the target
populations’ dis
-
abled situations, but
also enable

them to enjoy LOHAS.

NSC97
-
2218
-
E
-
009
-
026 (97R027)

---------------------------------------------------------------------------------
---------------


Title
:

I
nterceptor And Target Engagement: System Design, Simulation And

Performance Evaluation

Principal Investigator
:
Ching
-
An Lin

Sponsor
:

National Science Council

Keywords:


The proposed three year research will study the guidance and control aspects of a
missile defense system operating under hit
-
to
-
kill scenario. The interceptor consists of
a b
ooster stage and a kill vehicle. The kill vehicle, separated after the booster phase, is
equipped with an infrared seeker for target tracking and on
-
off thrusters for divert and
attitude control. The tasks include simulation program development, guidance l
aw
design, divert and attitude control system (DACS) design, and Monte Carlo
simulations. The simulation program implemented using Matlab/Simulink software
will be the key tool for the evaluation of performance of the interceptor. The
engagement will inclu
de a six degree
-
of
-
freedom (6DOF) interceptor dynamics and a
three degree
-
of
-
freedom (3DOF) target. Both classical proportional navigation based
guidance laws and the state space formulated modern optimal guidance laws will be
implemented and their perform
ance evaluated. We will consider both threshold firing
and pulse width modulation (PWM) for the implementation of divert control system
employing on
-
off divert thrusters. Under the constraint of on
-
off attitude thrusters, we
will consider two types attitud
e control system design: the bang
-
bang optimal control
and linear control with PWM implementation. We will conduct extensive parametric
Monte Carlo simulations to evaluate the effects of subsystem parameters on miss
distance and kill probability. The resul
ts obtained from this research will help
establish key guidance and control technology and determine specifications of the
subsystems in a missile defense system.

NSC 97
-
2221
-
E
-
009
-
046
-
MY3

(
97R314
)

----------------------------------------------------------
----------
----------------------------


Titl
e:

A
Study
o
f
3D
Depth
-
Image Based Rescue
a
nd Exploration Multi
-
Robot

System

Principal Investigator
:

Sheng
-
Fuu Lin

Sponsor
:

National Science Council

Keywords
:

3D
Reconstruction, Rescuers
-
Searching Robot, Localiz
ation

o
f Robot,


Path Planning
o
f Robot


Recently, the design of rescuer robots has become a popular research field.
Among these researches, rescuers
-
searching robot is mainly used to decide the search
spot in the collapsed building. This kind of robot c
an memorize the information of
location and identify the explored or unexplored region. One advantage of applying
robots to rescue is that robots can search the victims effectively and efficiently.
Another advantage is that the inner condition of the colla
psed building can be
understood through the exploration of robots, so as to aid rescuers enter to take action.
As a result, this proposal proposes a study of 3D depth
-
image based rescue and
exploration multi
-
robot system. It provides function of searching
the trapped people
and helping rescuers enter to relive the victims. This project mainly includes 3D
reconstruction

localization of robot

motion planning of robot and path planning. By
3D reconstruction, rescuers can be informed of the inner condition of c
ollapsed
building; however, localization helps us record moving direction and angle of robots
and build the paths. Motion planning helps robots moving toward the right direction
and finally mapping out the shortest path suitable for rescuers to enter throu
gh path
planning. If many robots are used, re
-
exploration for same section can be avoided by
information transmission and cooperation among robots, and the performance can be
improved effectively.

NSC 97
-
2221
-
E
-
009
-
091
-

(
97R452
)

---------------------------
-----------------------------------------
----------------------------


Titl
e:

A Study of 3
-
D Object Tracking and Recognition Based on Lidar Imagery

Principal Investigator
:

Sheng
-
Fuu Lin

Sponsor
:

National Science Council

Keywords
:

3D
I
mage
R
ecognition
, 3D
P
ose
-
independent F
eature,
S
pin
I
mage
,
3D


C
ueing
,
I
terative
C
losest
P
oint (ICP)
A
lgorithm



As the laser measurement system is developed continuously,
laser range finder
technology is used in many domains
, like distance measurement, t
errain

monitor

and
3
-
D object measurement.

Both the range image and the intensity image

of object will
be acquired by the 3
-
D image laser

scanner. The range image can
provide 3
-
D
information about object
.
Compares

with

depth information which
is
produce
d by

the
traditio
nal stereoscopic vision image
, the advantages of using
3
-
D image laser

scanner
are avoiding camera
calibrat
ion and obtaining accurate result. Hence, the 3
-
D surface
contour is obtained by
3
-
D image laser

scanner can be used for 3
-
D object recognition.
In o
rder to develop an algorithm of 3
-
D object recognition, the most important thing
in this project is to extract a pose
-
independent

feature.


The
purpose
s of this 3
-
D object recognition project are: (1)pose
-
independent
feature extraction, (2)ROI detection in

a 3
-
D scene, and (3)
automatic
3
-
D object

recognition algorithm
. First of all, the concept of
spin image

is introduced in
pose
-
independent feature extraction. The
spin image uses local coordinate system to
encode all
points on the surface of object. In add
ition,

the feature
is extracted by

spin
image can
not

vary with
different 3
-
D poses
.

In short, all pose
-
independent features of
3
-
D
models
in this project are generated based on this

characteristic
. After that, a 3
-
D
cueing is employed to detect ROIs in a 3
-
D scene.
This

can help to find interested
objects in a large
-
scale scene.

Then use correlation process to
match

points of the
model among ROI
s

and compute plausible transformation.

Finally, the
iterative
closest point
(ICP) algorithm

is applied
to verify t
ransformation

and to obtain the
value of
recognition goodness of fit
.

As a result, the
higher
value of
recognition
goodness of fit
, the
higher confidence index
of
scene
-
to
-
model

is. Indeed, this is the
aut
omatic
3
-
D object

recognition

approach in this proj
ect.

NSC97
-
2623
-
7
-
009
-
002
-
D

(
97R012
)

--------------------------------------------------------------------
----------------------------


Title
:

Multi
-
level Ordinal Optimization for Constrained Combinatorial Optimization



Problems

Principal Investigator
:
Sh
in
-
Yeu Lin

Sponsor
:

National Science Council

Keywords:


In this project, we propose a multi
-
level ordinal optimization algorithm to solve
constrained combinatorial optimization problems with huge solution space for a good
enough solution using limited comp
utation time. There are two parts in this project.
The first part regards the theory. The second part is application. We will select two
types of constrained combinatorial optimization problems and use the proposed

algorithm to solve them. The keys of mult
i
-
level ordinal optimization are (i) how to
handle the constraints,(ii) how to develop the approximate models in each level,

and
the design of methods to pick estimated good enough solutions from the candidate
solution set in each level. At the end, we wil
l use statistics based analysis to analyze
the goodness of the good enough solution obtained by the multi
-
level ordinal
optimization algorithm using extensive simulations. The constraints of combinatorial
optimization problem may vary depending on the natu
re of considered problems. The
same logic applies to the approximate model needed in each level. Therefore, in the

application part of this project, we will provide different constraint handling technique,
different approximate models and different methods

for selecting estimated good
enough solutions from the candidate solution set for different problems. The work of
our three
-
year project can be outlined as follows: First year: (i) Study the constraint
handling technique for constrained combinatorial opti
mization problem with huge
solution space. (ii) Proposing approximate model in each level. (iii) Designing
methods to select estimated good enough solutions. (iv) Proposing the multilevel
ordinal optimization algorithm (v) Using statistics based analysis t
o analyze the
goodness of the good enough solution obtained by the multi
-
level ordinal
optimization algorithm using extensive simulations. Second year: (i) Proposing a
multi
-
level ordinal optimization algorithm to solve the adaptive subcarrier assignment
a
nd bit allocation problem for multiuser OFDM system. The proposal includes: a.
technique for handling constraints b. approximate models in each level, and c. method
to pick good enough solutions in each level. (ii) Use the statistics based analysis to
anal
yze the goodness of the obtained good enough solution. Third year: (i) Proposing
a multi
-
level ordinal optimization algorithm to solve the resource allocation problem
in grid computing system for maximizing the reliability. The proposal includes: a.
techni
que for handling the constraints in grid computing system, b. approximate
models in each level, and c. method to pick good enough solutions in each level.

(ii) Use the statistics based analysis to analyze the goodness of the obtained good
enough solution.

NSC 97
-
2221
-
E
-
009
-
088
-

(
97R449
)

------------------------
------------------------------------------------------------------------


Tit
le:

Design
a
nd Realization of
a
n Innovative Micro
-
transducer with High
-
Energy


Density

Principal Investigator
:
Shir
-
Kuan L
in

Sponsor
:

National Science Council

Keywords
:

Hybrid
Energy Scavenging System,
Hula
-
Hoop Motion Transformer,




Micro
-
Generator,
Energy
Transducer,
Converter
Circuit


This proposal plans to design and implement a hybrid energy scavenging system.
The

cont
ains three sub
-
proposals, They are (1) Design and implementation of a hybrid
energy scavenging system; (2) Design, analysis and implementation of a novel
micro
-
generator; (3) Design and chip realization of a customized converter circuit for
the system cont
aining a hula
-
hoop motion transformer and a micro
-
generator. The
objectives of this proposal are to combine the two different functions of energy
scavenging devices: a hula
-
hoop and a piezo
-
system, in one hybrid energy scavenging
system. In addition to rea
lize this hybrid system, the main features of the proposal is
the utilization of the hula
-
hoop motions and manufacture feasibility of the
micro
-
generator. The hula hoop motion transformer proposed by the sub
-
proposal one
first convert linear reciprocating
vibrations to near constant speed rotation. The
micro
-
generator proposed by sub
-
proposal two is then responsible for convert

the
energy of mechanical rotations to electrical energy in near
-

AC signals with higher

efficiency than other existing micro
-

energ
y scavenger. The AC signals are then
converted by the high
-
efficiency converter circuit designed and implemented by
sub
-
proposal 3 to almost DC signals to be saved in a terminal energy reservoir; e.g., a
battery. The novelty of the present proposal lies in

the
combined usage of a hula hoop
motion transformer, a micro
-
generator and a customized converter circuit. This idea
can efficiently convert the energy in irregular vibrations to electrical energy.
The
efficiency, based on preliminary calculation, could
be up to ten times of those offered
by the existing micro
-

energy scavenger.
This proposal is centered around
sub
-
proposal one; then the other two sub
-

proposals are responsible for energy
conversion and saving. The jobs of this proposal includes: (1) fost
ering

the total
framework of the overall energy
-
scavenging system in order to ensure the effective
results and future applicabilities of the proposed scavenger; (2) coordinating and
interfacing the efforts specifications of the sub
-
proposals; (3) supportin
g
administration and accounting of the sub
-
proposals; (4) arranging possible
interaction
with ITRI and industry; (5) organizing related seminars and meetings; (6) controlling
progress, results, problem solving for each sub
-
proposal; (7) Purchasing and
mana
ging expensive equipments for all sub
-
parties; (8) applying and managing
resulted patents.

NSC 97
-
2221
-
E
-
009
-
085
-
MY3

(
97R428
)

--------
----------------------------------------------------------------------------------------


Title
:
Electronic Converter Desi
gn
a
nd Chip Implementation of
A

Hula Hoop



Energy Harvester

Principal Investigator
:
Shir
-
Kuan Lin

Sponsor
:

National Science Council

Keywords
:

Hybrid
Energy Scavenging System,
Power
Converter Circuit,
Adaptive




Control


Energy harvesting is the proces
s of capturing minute amounts of energy from
one or

more of naturally
-
occurring energy sources, such as light energy, wind energy,
heat energy,

mechanical vibration, sound energy, electromagnetic energy etc. The
cavenger energy is accumulated and stored fo
r later use, e.g., the application of
wireless communication or portable type products. To collect, to transmit, and to
control the scavenger energy are the basic concept of the energy harvesting
technology. In this project, the range of the induced energy

from the hula
-
hoop
transducer is investigated. This will be used to design the matching circuit for the
converter. Moreover, the high efficiency power converter topology circuit will be
proposed. And the different operation mode in the power transmit proc
ess are
analyzed for the strategy of the power control. The main goal is to achieve maximum
power transfer. One

direction is to construct a high efficiency resonant converter.
Estimating the output power of hula
-
hoop transducer is used to generate a
freque
ncy
-
varying control strategy on the resonant converter. After simulation
verification, the control law will be implemented on an FPGA development platform.
Finally, the control chip is developed with steps of design, layout, simulation, and
product. The ex
periment circuit for the manufactured chip will be set up for the
overall system verification.

NSC 97
-
2221
-
E
-
009
-
086
-
MY3

(
97R429
)

-----------
-------------------------------------------------------------------------------------


Title

Transmitting and Receiving Window Designs for Multicarrier Systems

Principal Investigator

Yuan
-
Pei Lin

Sponsor

National Science Council

Keywords:

Multicarrier Systems, Discrete Multitone Systems, DMT, OFDM,

Window Design, RFI, Spectral Leakage


Windo
wing is often applied to multicarrier systems to improve frequency
separation among the subcarriers. At the transmitter side better frequency separation
leads to a smaller out
-
of
-
band spectral leakage and also less interference to radio
frequency transmiss
ion. At the receiver side better separation gives more suppression
of radio frequency interference. As these are frequency based characteristics, a
filterbank representation renders a natural and useful framework for formulating the
problem.

In this resear
ch project, we propose a unified filterbank framework to the design
of windows for multicarrier systems. The approach used here will be more general:
we will consider the so
-
called subfilters. The use of subfilters have the potential of
enhancing the frequ
ency selectivity of the transmitting/receiving filters while
maintaining the orthogonality among the subchannels. The characterization of the
desired transmitter and receiver
--
good frequency separation
--
lies in the frequency
domain. The filterbank approach

lends itself naturally to frequency domain
formulation. Correspondingly, for the transmitter side spectral leakage can be reduced
and for the receiver side RFI can be further suppressed.

In the first year, we have established the condition for the subfilt
ers that maintain
the orthogonality among the subchannels. The condition is essential and is to be
incorporated in the design of transceivers later. We have also derived the transmitted
spectrum in terms of the transmitting filters. The result will be usef
ul in considering
egress emission and the design of transmitting filters. In the second and third year, we
will continue on the design of transmitting and receiving filters based on the results
obtained in the first year.

The results from this research pla
n will have practical applications in the design
of transmitters and receivers for multicarrier systems. We can develop useful
transceivers that have less spectral leakage and RFI interference, and achieve a higher
transmission rate.

NSC 96
-
2628
-
E
-
009
-
005
-
MY2 (96N430)

-------------------------------------------------------------------------------------------------------


Title
:

Time
-
Varying
Spatio
-
Spectral
Characteristics
o
f
Meditation EEG

Principal Investigator
:
Pei
-
Chen Lo

Sponsor
:

National Science Counci
l

Keywords
:

Electroencephalograph
(
EEG
),
Zen
Meditation,
Time
-
Varying




Spatio
-
Spectral Characteristics,
Heart
Rate Variability (
HRV
),




Meditation
Brain Mapping (
MBM
),
Meditation
Brain Mapping Scenario



(
MBMS
),
Fuzzy C
-
Means
,
Wavelet

Analysis
,
Clustering
,
Stress
Study


Since 1988, the principal investigator has been devoted to the research on

physiological and mental/conscious phenomena under Zen meditation. A number of

important results have been reported, of which we mainly focus on investig
ating the
EEG (electroencephalograph) characteristics based on the scientific approach.
Subjects of the experiment practice the Zen
-
Buddhist meditation. From a large
amount of meditation EEG signals acquired, we characterized their temporal and
spectral fe
atures by a number of advanced DSP methodologies. Some particular
findings aroused our attention of further exploring the spatial foci that generate such
kind of
Zen
brain dynamics. It has been well noticed that ‘stress’ significantly affects
modern health

both in physiology and psychology. Meditation has been proved in
clinics to be one of the best stress
-
manipulation methods. Our research group has
observed the stress
-
releasing effects of Zen meditation based on galvanometric skin
resistance (GSR), electr
ocardiograph (ECG), and heart
-
rate variability (HRV). It
would be an urgent task to develop a scientific, quantitative principle for evaluating
the meditation effects on stress reduction. To understand the time
-
varying
spatio
-
spectral phenomena of brain el
ectrical activities under Zen
-
meditation state, we
designed the experimental protocol and environment for collecting EEG and other
physiological signals in the first
-
year study (August 2007 to July 2008). In addition,
methods and algorithms have been devel
oped for quantitative analysis of time varying
spatio
-
spectral characteristics of 30
-
channel meditation EEG. Research plan for the
following year will then be aimed at: (1) establishment of meditation brain mapping
(MBM), (2) investigation of time
-
varying
MBMs, or the MBM Scenario (MBMS),
and (3) study of meditation effect on stress manipulation based on correlation between
MBM and other physiological parameters. In addition to the large amount of
editation EEG recordings, this study requires the developme
nt of multi
-
faceted DSP
methods to characterize complex MBMS.

NSC 97
-
2221
-
E
-
009
-
093

(
97R377
)

----------------------------
--------------------------------------------------------------------


Title

Study of Image
-
Based Mobile Manipulation

Principal Investig
ator

Kai
-
Tai Song

Sponsor

National Science Council

Keywords:

Mobile Manipulator, Intelligent Robot, Real
-
Time Robotic Vision,

Image Recognition


The recent trend of robotic research is to enhance the robotic technology for
more assistance and service to hu
man beings. One possibility of increasing the skills
of a robot is to provide a robotic arm to an autonomous mobile robot. In other words,
a mobile manipulation with robotic vision will greatly enhance the ability of a robot.
However, it is a challenge for

a mobile manipulator to work in an unstructured and
dynamic changing environment such in a home setting. Advanced hard and software
design needs to be considered for a safe and reliable operation of a robot equipped
with a manipulator. This project aims t
o study vision
-
based mobile manipulation. In
the first year, we will develop image processing algorithms for object recognition
based on scale invariant feature transform (SIFT). The target object will be located
automatically by the robot for further appr
oaching and grasping. The second year will
be focused on the design and construct the mobile manipulator with robotic vision.
We plan to adopt an omni
-
directional platform for giving smoother motion in a
narrow space such as in a home setting. A three
-
degr
ee
-
of
-
freedom robotic arm will
be equipped onboard the platform. In the third year, integrated autonomous grasping
of a specified object will be investigated. Image
-
based visual servoing techniques will
be applied for the perception
-
based mobile manipulati
on design. We hope in this
study a practical assistant robot will be demonstrated through free motion, grasping
and bring an object back to its master.

NSC 96
-
2628
-
E
-
009
-
152
-
MY3 (96N454)

---------------------------------------------------------------------
----------------------------------


Title

Design of Mobile Platform and Intelligent Behavior Controller of a Walking
Helper

Principal Investigator

Kai
-
Tai Song

Sponsor

National Science Council

Keywords:

Intelligent Robots, Robot Control, Robot Locali
zation
, Human
-
Robot
Interaction


Many countries world
-
wide such as Japan, United States of America, European
countries and Taiwan are facing an aged society. The low birth rate in Japan and
Taiwan will cause lacking of manpower in the future. It is well recogniz
ed that elderly
care will become a serious problem for our society. Some researchers proposed to
develop elderly
-
care robots to assist the elderly solve this problem. The main
objective of this project is to develop a walking helper robot to assist the eld
erly. The
walking helper can also be used by impaired people for walking rehabilitation. The
basic idea for developing the robotic platform is “to design a walking helper that is as
easy as a cane; but it is intelligent to take care of its master.” We plan

to complete the
project in three years. In the first year, the main task will be the design of a flexible
motion platform of the walking helper and implement a compliant motion controller.
In the second year, we will focus on the self
-
localization of the
robot in an indoor
environment. Wireless Zigbee modules will be adopted to position an object which
contains a Zigbee tag in the working space. A navigation system will be designed for
the robotic system. The integration of the navigation system with the c
ompliant
motion control will be investigated. At this stage, the walking helper can have some
autonomy to find a path to the assigned place for the elderly. In the third year,
intelligent elderly
-
care functions such as fall detection, health
-
condition moni
toring
will be integrated to the walking helper. An evaluation procedure will be carried to
test the applicability of the designed hardware and software modules through field
tests. We are confident that these objectives will be completed in the scheduled
research period.

NSC 96
-
2628
-
E
-
009
-
162
-
MY3 (96N458)

-------------------------------------------------------------------------------------------------------


Title

BIST Design, Implementation, and Ana
lysis for High Speed Serial I/O

Principal Investigator

Chau
-
Chin Su

Sponsor

National Science Council

Keywords

High Speed Serial I/O, Built
-
in Self Test, Phase Locked Loop, Tranceiver,
Mixed Signal Circuit


The purpose o
f this research is to design, implement, and analyze a built
-
in self
test (BIST) circuits for high speed serial I/O. A high speed serial I/O contains three
major building blocks, a phase
-
locked loop, a transmitter, and a receiver. For PLL
BIST, the major i
ssue is the jitter measurement. The difficulty is to achieve PS
resolution. Here, we already have some experiences and will be presented later. For
Transmitter BIST, the major issue is the eye diagram measurement. An eye opening
includes the jitter at the
eye edge transition and height of the eye. For the jitter
measurement, we will take a digital approach to minimize the hardware overhead. For
the height measurement, we will modify the technique proposed in an ASPDAC 2003
invited paper to be used in a high

speed environment. For the Receiver BIST, the
major issue is the jitter injection and attenuation filter. One must be able to control
precisely the attenuation and jitter in order to test the receiver

s tolerance on the signal
integrity. For the Tranceive
r BIST, the most cost effective method is the loop
-
back bit
error rate test method. However, this method can not provide the information on how
much the test margin. The margin is an important information for the reliability
consideration. Here, we would l
ike to derive the relationship between the above three
physically measured data and theoretical aspect of the BER. With which, one is able
to apply BER test with confidence.

This three
-
year research project is expected to produce a 5~6Gbps
transceiver

wit
h BIST. For PLL, transmitter, and receiver, we will design a BIST circuit for the
jitter, eye opening, and signal integrity tolerance measurement. For the
transceiver
, we
will derive
mathematically

the relationship between BER and physically measured
data.

Basically, we would like to derive
mathematical

model based on physical
measurement instead of pure
mathematical

derivation. After all, when a timing
approaching PS range, the circuit will encounter all sort of nonlinearities, just like the
nanometer circ
uit design.

NSC 95
-
2221
-
E
-
009
-
328
-
MY3 (
95R658
-
1
)

-------------------------------------------------------------------------------------------------------


Title
:

Automobile Power SoC with
E
mbedded Power Line Communication

Principal Investigator
:
Chau
-
Chin S
u

Sponsor
:

National Science Council

Keywords
:

SoC, Power Line Communication, Power IC, Motor Actuator


This project integrates power IC and power line communication into an SoC. It
provides the needed power management, power driving, and data communication

at
the same time. With such an technology, a car requires only one power line to
integrates all the driving and communication needs. Such that, the design and
manufacturing cost can be minimized significantly. There are two distinguishing
features in this

project. First, the power line communication we use is different from
the PLCs used in the home, such as HomePlug. It is battery powered DC power line.
It consists of only one line, the return loop is made up by the framework of the car. So
that, we needs

to measure and build the channel model, which has never been
investigated before. With the channel model, the system design parameters can then
be determined. Second, the power driving IC is able to produce the optimal voltage or
current profile according

to the command sent through the PLC. The project
technology has the following two advantages. First, due to the use of power line for
communication, all the wires, except power lines, can be removed. The assembly cost
is significantly reduced. Second, wit
h the programmable divers on chip, the driven
nodes no long need its own driving circuitry. As a result, the component cost is also
reduced. This project contains four subprojects. Subproject 1: This subproject will
survey a channel modeling for the automo
bile DC power lines, derive the
communication algorithm, and implement the baseband circuitry. Subproject 2: This
subproject will design the front
-
end analog and AD/DA converting circuitry
according to the specification derived in subproject 1. Subproject
3: This subproject
will investigate the driving algorithm for the automobile motors in order to meet the
driving requirement and minimize the power consumption. Subproject 4: This
subproject will design and implement an programmable power driving IC that c
an
produce needed voltage/current profile according to the command.

NSC 97
-
2220
-
E
-
009
-
047

(
97R355
)

-----------
----------------------------
---------------------------------------------------------


Title
:
Multi
-
F
unction Automobile Power IC

Principal Investi
gator
:

Chau
-
Chin Su

Sponsor
:

National Science Council

Keywords
:
Power IC, SoC, Automobile Electronics, Motor Driver, DC Converter


This project proposes an universal power driving circuit for automobile
applications. It can produce the desired voltage or c
urrent driving profile, according to
the command. The main purpose is to provide multiple driving needs from a single
DC power source. This project will use 0.35um technology to design, manufacture,
and measure the chip. The tentative goal is to provide 2A

of driving current for the
feasibility study purposes. The distinguishing feature is the programmability.
According to the command, it can produce the driving signal to driver the driven
devices. Not only the flickering frequency and brightness of the lig
ht but also the
motor driving power profile can be provided by the circuit. As a result, the cost of the
device can be reduced significantly. The proposed programmable power driving
circuit is able to produce the following outputs. The first one is the vol
tage
programmable DC power source for electronic equipment, The second one is
frequency and voltage programmable for light control. The third one is the power
driving profile programmable motor driver. With the above three functions, the
majority of the el
ectrical parts in automobile can be driven by this circuitry.

NSC 97
-
2220
-
E
-
009
-
050

(
97R358
)

----------
----------------------------
----------------------------------------------------------


Title
:
A Novel Spatial Precoding with Limited Channel Phase Feedb
ack
a
nd Its



ASIC Design

Principal Investigator
:
Shang
-
Ho Tasi

Sponsor
:

National Science Council

Keywords:


In this proposal, we propose a novel precoding scheme for MIMO (multiple
input and multiple output) systems to overcome the drawbacks of orthogona
l
space
-
time block code (OSTBC) and beamforming techniques. For OSTBC, the
coding scheme that achieves full diversity order for complex symbols with full rate
exists only for two transmit antennas, i.e. Alamouti scheme. Also, to implement
OSTBCs, we need t
o store consecutive symbols in both transmitter and receiver sides
for encoding and decoding purposes, which requires extra memory and leads to
latency. The

memory and latency increase as the OSTBCs applied to OFDM
(orthogonal frequency domain multiplexing
) systems, where the number of
subchannels can be up to several thousand, e.g. up to 2048 in standards of IEEE
802.16 family. For beamforming, the receiver needs to compute the
eigendecomposition of MIMO channel. Moreover, it requires feedback to send back

the eigendecomposed information. The amount of feedback increases as the numbers
of transmit and

receive antennas increase. When beamforming is used in OFDM
systems, the feedback amount is in general very large and hence may somewhat be
impractical when t
he channel environment varies rapidly. Moreover, the
eigendecompositions for all subchannels lead to high computational burden in the
receiver. Furthermore, the transmitter and receiver need to perform matrix
multiplications based on the eigendecomposition
. All those drawbacks may limit
beamforming techniques be used

in portable devices such as mobile phone, where
simple transceiver design is more appreciated. In this proposal, we propose a novel
spatial precoding scheme that can provide additional coding g
ain as well as achieve
full diversity order for complex symbols with full rate. In systems with 2 transmit
antennas, the proposed precoding scheme can outperform the Alamouti scheme with
only 1
-
bit feedback information due to the additional precoding gain.

Moreover,
unlike orthogonal space
-
time block code, this precoding scheme can be used for
arbitrary number N of the transmit

antennas. We found that using (N
-
1)
-
bit feedback
information, the proposed scheme can achieve a full diversity order while also pro
vide
additional precoding gain in systems with N transmit antennas. The precoding gain
increases as N increases. Unlike orthogonal space
-
time block codes, this precoding
scheme does not need extra memory to store consecutive symbols and hence does not
resu
lt in latency in both transmitter and receiver sides. Furthermore, we propose a fast
algorithm to compute the feedback information for precoding to avoid the use of
exhaustive search.

The proposed spatial precoding scheme and the algorithm to generate prec
oding
codewords can be directly used in standards that use MIMO techniques such as IEEE
802.11n and IEEE 802.16 family. In this research, we will investigate the
performance and complexity of the proposed precoding scheme and that of OSTBC
and beamforming.

Also, we will show how to use the proposed techniques in IEEE
802.11n and IEEE 802.16 family. Finally, we will implement the proposed
architecture with IC.

NSC 97
-
2221
-
E
-
009
-
071
-
MY2

(
97R227
)

---------
----------------------------
---------------------------
--------------------------------


Title

Development of System
-
On
-
a
-
Chip Digital Control Technology for Sensorless
Control of

Permanent
-
Magnet Synchronous Motor Drives with Sensorless
Power Factor Correction Control

(1/3)

Principal Investigator


Ying
-
Yu Tzou

Sponsor

National Science Council

Keyw
ords

PMSM
M
otor
S
ensorless
C
ontrol,
M
otor
C
ontrol IC,
S
ensorless PFC,
S
ystem
-
on
-
a
-
chip,
P
ower
L
ine
C
ommunication


In forecasting the future development of wide applications of sensorless
adjustable brushless permanent
-
magnet motors in home automation and i
nformation
appliances, this project focusing on the development key technology needed in the
design of system
-
on
-
a
-
programmable
-
chip (SOPC) for modern smart motor drives.
This project makes a research on the design and implementation of a FPGA
-
DSP
based se
nsorless PFC control sensorless speed control of adjustable brushless motor
drives. The combination of the power factor control and PWM inverter control can
greatly simplify the control circuitry for a motor drive with unit power factor and high
efficiency
. The abstract of this three
-
year project is described as follows:


The 1st
-
year: Design and Realization of a Digital Sensorless Variable
-
Output
Power Factor Correction Control IC for Sensorless PMSM Drives
.

During the first
year of this project we will fo
cus on the realization of the SOPC design platform for
digital motor control and make a study on control architectures suitable for the PFC
AC
-
DC converter and PWM DC
-
AC inverter combination control scheme of a
sensorless motor drive. A digital programmabl
e PFC
-
PWM control scheme will be
proposed in applications to single
-
phase adjustable speed sensorless brushless PM
motor drive with unit power factor. Numerical realization issues for the sensorless
power factor control algorithm will be studied. Simulatio
n
-
oriented analysis will be
carried out to study the feasibility of the proposed control scheme. Experimental
verification will be carried out using a designed FPGA&DSP
-
based programmable
digital controller.


The 2nd
-
year: Design and Implementation of a

Digital Sensorless Control IC for
Permanent
-
Magnet Synchronous Motors

In the second year of this project we will
develop an FPGA realizable digital control scheme for the sensorless speed control of
permanent magnet synchronous motor with nonsailent poles
. The low
-
speed
sensorless control techniques developed for PMSM can be classified into two
categories: one is the model
-
based rotor position estimation scheme and the other is
the signal injection rotor position estimation scheme. The measured phase curre
nt and
back emf will be used for the estimation of rotor flux synchronization signals. A
digital phase
-
locked loop technique will be used to synchronize the
phase
-
commutation signals for rotor position control. Fixed
-
point realization of the
sensorless est
imation scheme will be studied. Computer simulation and FPGA&DSP
realization will be carried out to verify the proposed sensorless position estimation
scheme.

The 3rd
-
year: Design and Implementation of a SOPC Sensorless Control IC with
Sensorless Unit Pow
er Factor Control
.
This research will make an integration of the
previous two
-
year research results to develop a systematic design approach for the
design of SOPC sensorless motor control IC with sensorless unit power factor. The
designed SOPC will also pr
ovide power line communication interface to facilitate the
interface for home automation. Simulation oriented analysis and design will be used
in the synthesis of this SOPC based digital power and motor control IC. A systematic
design procedure will be dev
eloped for the design and implementation of SOPC IC
design for sensorless motor drives.

In summary, this project will develop SOPC based systematic design platform
and methodology for the design of digital power and motor control chips for the
sensorless c
ontrol of PM synchronous motor drives with unit power factor control. We
try to unify the design of digital power control and digital motor using the advanced
system
-
on
-
a
-
programmable
-
chip (SoPC) technology. The developed technology will
be used for the de
velopment of an intelligent integrated motor drive. The carry out of
this project is beneficial to the development of advanced digital motor and power
control techniques.

NSC 95
-
2221
-
E
-
009
-
338
-
MY3 (
95R660
-
1
)

------------------------------------------------
-------------------------------------------------------


Tit
le:

Design
a
nd Implementation of Robust Controller via Hopfield Neural Network

Principal Investigator
:

Chi
-
Shu Wang

Sponsor
:

National Science Council

Keywords
:
Hopfield
Neural Network, Cellular Ne
ural Network, Controller Design


This project is to explore the theoretical and practical issues of off
-
line training
and real
-
time applications of Hopfield neural network (HNN) as a robust controller
for control systems. The controller for a control syste
m could have malfunctioned due
to the aging problem, where the details of the controller design are usually not
available. Considering the difficulties in reconstructing the controller, we prefer to
replace the existing controller by a new one trained by t
he input and output data of the
control system. Hence, this project proposes a new approach of the design of robust
controller by using HNN. We will train a HNN using the training data from the
existing controller. This trained HNN can then be used to repl
ace the existing
controller with better robust performance than that of the existing controller. Both
training algorithm with robustness analysis will be performed in the first year by
extensive computer simulations. However, if more HNN neurons are requir
ed for a
complex system, its messy feedback architecture will be infeasible for hardware
implementation. Thus a Cellular Neural Network (CNN) architecture for HNN as a
robust controller will be investigated. The above issues will be conducted in the first
and second years using computer simulations. Finally, the hardware implementation
of the trained HNN

will be realized by controlling several plants, such as an inverted
pendulum system and a ball
-
and
-
beam system, in the third year.

NSC 97
-
2221
-
E
-
009
-
090
-

(
97R451
)

-------------------------
-----------------------------------------------------------------------


Title
:

Intelligent Omnidirectional Detection of Vehicle Surroundings
a
nd Driving




Safety System

Principal Investigator
:
Bing
-
Fei Wu

Sponsor
:

Natio
nal Science Council

Keywords:


This subproject is a three
-
year project, which focuses on the development of the

automatic vision
-
based detection system for the environment surrounding a vehicle

and the research of the vehicle driver’s behaviors.

The conten
t of this project is separated into three parts: (1) Establish the 3.5G/3G

wireless embedded platform; develop the parking assistance algorithm; and draw the

corresponding map using the GPS data. (2) Derive the front view pedestrian detection

technique; tr
ansfer the videos of the intelligent intersection systems, built in the main

project, to the embedded platform and the mobile devices through the efficiency video

processing algorithms, provided by the subproject seven; and port the parking

assistance algo
rithm to the transportation robots which are implemented in the main

project. (3) Design the blind spot pedestrian detection algorithm and the advanced

burglarproof system. The final objective of this project is to accomplish an intelligent

vehicle which c
an automatically detect the lanes, the obstacles, and the pedestrians

around it. We hope to improve the technologies in the autotronic fields, and better the

international images of Taiwan.

NSC 97
-
2221
-
E
-
009
-
072

(
97R281
)

------------------------
-----------
-------------------------------------------------------------


Title

Development of an Intelligent Life
-
Assistant Robot System

Principal Investigator

Young, K.Y

Sponsor

National Science Council

Keywords:

Intelligent Life
-
Assistant Robot System, Robot Walke
r, Robot Follower,
Robot Companion, Aging Society Investigation


Taiwan is going to be a highly aging society. The pressure of the lack of
health
-
caring personnel is growing fast. In this three
-
year integrated project, we
propose developing an Intelligent
Life
-
Assistant Robot System. We aim on the elderly
people who can be on their own and only need partial assistance. During the
development, we will start from the robot walker, which is designed to passively
provide the elderly people with force informatio
n, support, and guidance during
walking. Later on, the robot walker will be upgraded to be a semi
-
autonomous robot
follower, which can follow the user to go outside and provide the necessary
information and load sharing. Finally, the robot follower will be

again upgraded to be
a fully autonomous robot companion, which can serve as an interface between the
elderly people and the medical center. Meanwhile, several health
-
enhancing computer
games will be included in the robot companion for a better accompany o
f the elderly
people in daily life. This integrated project consists of five sub
-
projects, focusing on,
respectively, robot design and behavior control, embedded firmware platform,
audio
-
video environmental perception interface, force management and
health
-
enhancement system, and multi
-
lingual/multi
-
modal spoken dialog
-
based
interface. After the establishment of the entire integrated system, we will verify its
feasibility via an extensive field study. A social analysis on the distribution, health
condition
, and medical resource of Taiwan aging society will also be executed.
Meanwhile, we have already invited medical doctor in rehabilitation to be our
consultant. We intend to develop an assistant robot system that meets the demand and
is ready for market. Th
e developed system is expected to highly enhance the life
quality and dignity of the elderly people.

NSC 96
-
2628
-
E
-
009
-
161
-
MY3 (96N457)

-------------------------------------------------------------------------------------------------------


Titl
e

Force Management and Health Enhancement for an Intelligent Life
-
Assistant
Robot System

Principal Investigator

Young, K.Y.

Sponsor

National Science Council

Keywords:

Intelligent Life
-
Assistant Robot System, Force
-
Feeling Representation,
Health
-
Enhancement

Computer Game, Sensory
-
Motor Coordination


Taiwan has entered the era of the aging society. We therefore propose an
integrated project for the development of an Intelligent Life
-
Assistant Robot System.
Because there will be force signals flowing back and
forth between the proposed robot
system and the elderly people to assist, it is very crucial to manage the interaction in
between successfully. The purpose of this project is to formulate force feeling
properly, develop salient force control strategies, an
d build the force
-
sensing
mechanism. On the other hand, from the biomedical research, the external training can
help the elderly people in maintaining their ability in sensory
-
motor coordination. We
will thus apply this force management system to develop a

health
-
enhancement
computer game. Through the game playing, the elderly people can improve their
motor plasticity and coordination ability in a joyful mood. In this three
-
year project,
in responding to the tasks of walking assistance, outdoor assistance,
and
health
-
enhancement computer game, respectively, we will analyze the force
interaction, develop the manipulation strategies, and build the entire system. We will
also link the force management system with the physiological signal measurement
system for
synchronous observation of the performance of EEG and EMG during the
health
-
enhancement computer game. The effect of the force training on the recovering
of the sensory
-
motor coordination ability can then be evaluated. With a theoretical
suppor
t and not so

complicated mechanism and program in implementation, this
computer game provides a salient alternate for the elderly people in health
enhancement. The developed system will be verified via an extensive field study. The
completion of this project will be m
uch helpful on building practical assistant robots,
and highly enhance the life quality and dignity of the elderly people.

NSC 96
-
2628
-
E
-
009
-
164
-
MY3 (96N460)

--------------------------------------------------------------------------------------------------
-----