CE Senior Projects VLSI Research Ken Stevens University of Utah 1
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
CE Senior Projects
VLSI Research
Ken StevensUniversity of Utah
1
Part One:
Senior Projects
2
The Engineering Disciplinel Role
u design and build systemsu change the world around us
n hopefully for the better...u have fund in the process
l Ultimate requirement
u what we build must work!
l Requisite skills
u science: math, physics, chemistry, materials, . . .u engineering: circuit design, testing, simulation, programming. . .u art: creativity, elegance. . .u sociology: team work, presentation skills, technical writing. . .
3
Computer Engineering
l Design and build computer systems
u involves both hardware and software design skills
l System software
u compiler, operating system, device drivers. . .u also interfaces between humans and the hardware
l Hardware
u analog and digital circuit designu board designu FPGA design
4
CE Senior Projects at Utah
l CE program run jointly by ECE and CS departments
l Senior project is capstone of undergraduate program
u apply what you have learnedu team basedu students choose projectu best method to demonstrate abilities to future employers
l Senior Project is year long activity
u Next semester: plan and propose projectu Summer: get parts and start building (optional)u Fall of senior year: build and demonstrate
l Student feedback
u hard, fun, and instructiveu you get what you put into this
5
2005 Projectsl Carputer
u PBDII car data and 802.11g auto-sync to base stationu monitor our car or your kids
l IR Tag
u Paintball without the mess
l Athlete monitor system
u real time tracking of position and heart rate to central coaching stationu GPS, RT, and Heart Rate Monitor on athlete
l Inverted pendulum 2-wheeled robot
l Multi-carrier reflectometry
u finding faults in aircraft wires without tearing the plane apart
l Glider avionics package
u using accelerometers, GPS, and strain sensors
6
2006 Projectsl PEN
u electronic paper – the only paper you’ll ever buy!
l Recipedia
u a cook book that talks and listens to you
l GPS tracker
u track real time location of campus busesu report on cell phone or computer
l OmegaCore
u a DVR that knows how to remove commercials for you
l NoCPR
u bathtub drowning prevention
l Tracking Visor
u virtual reality on your head
7
Current 2007 Projects
l Wireless positioning measurement system
u More than just the Wii - use your whole body!u Sonar arrays using Zigbee wireless
l Unmanned Autonomous Vehicle
u Automatic control for a blimp.u GPS, wireless
l Aquatic Guidance Systems
u Look, ma, I can water ski without a driver!
l Hands Free Music Tablet
u Music scores that change pages for you
l Wi-Fi Clock Radio
u Wake up to your favorite tunes
8
Example from 2004
9
Direction and Speed Control
10
GPS Internals
11
GPS Parts from Motorola Kit
12
Autonomous Anti-Collision System
13
Completed Car
14
View of front and Range Finder
15
GPS Unite connected to µ-Controller
16
GPS Antenna
17
Senior Project Synopsys
l This is just a preview
l Diversity in opportunities and problems
l Have fun with the project
u your chance to do whatever you can dream!u if you can imagine it you can usually build itu your dedication and time are well rewarded
18
Part Two:
VLSI
19
Ubiquitous Nature of VLSI
Ubiquitous
1. existing or being everywhere at the same time2. constantly encountered
VLSI the process of creating integrated circuits by combiningmillions of transistors into a single chip.
integrated circuits (IC, silicon chip, microchip, . . . ) miniaturizedelectronic circuit containing active and passive componentswhich has been manufactured in the surface of a thin substrate ofsemiconductor material.
20
Definitions
transistor A three (or four) terminal solid state semiconductordevice that can be used for amplification, switching, voltagestabilization, signal modulation, and many other functions.
l a “switch” is often referred to as a “gate” when abstracted tothe mathematical form and used in logic equations.
semiconductor A material with electric conductivity that can bothsource and sink electrons, and can operate as an insulator andconductor.
g
d
s
21
Scaling
l Moore’s Law
u transistor counts double every one to two years
l Cost has followed inverse trend
l Imagine this in other scenarios. . .
22
Manufacturing
Size of wafers (single silicon crystal)
23
Size Perspective
Proposition: What you pay for a product is proportional to it’s weight
24
Part Three:
Research
25
Learn the rules so you know how to break them properly.Dalai Lama
26
Research Cycle
Do
1. learn technology or application
2. build something novel and cool: a rule breaker!
3. automate your learnings for productivity
Forever (or until we get alzheimers. . . )
27
Why VLSI is Cool
l Artistic and creative pursuit (quite figuratively and literally. . . )
l Satisfaction in doing something new or better
l General satisfaction in products that improve our existence
u Pentium 4’s all had circuits due to my work!
28
Circuits and CAD
1. take this simple transistor
2. replicate it 108 times
3. connect instances in ways that break traditional rules
4. write software to support this
Transists and translates into all sorts of fun!
g
d
s
29
Synopsis of my Research
These are the particular directions I’ve taken VLSI
1. Asynchronous chips and design
2. Relative Timing
3. Transistor and circuit research
4. CAD for VLSI
5. Networks
6. Biological designs
7. Streaming video chips
8. Reliable and Tamper Resistant Circuits
30
Asynchronous Design
31
Asynchronous Design
32
Learn the Rules: Multiple Input Switching (MIS)
$
%j $
%j
cb
a
observe effects at this node�
��
l average pushout of 21% (r00 lib)
l average speedup of 47%
l several sources of effects
l this configuration shows a 28.8%delay pushout
33
CAD for automatic MIS vector generation
34
Rule Breaker: Relative Timing (RT)A huge difference in performance and power is derived by usingsimple timing assumptions that are easy to represent as a logicalconstraint and easy to validate in a design.
lo↓
li↓
lo↑
li↑
� �?
x?
?
?
PPPPPPPPPPq
����������)
x
ro↓
ri↑
ro↑
ri↓
� �?
?
x?
? lo↓
li↓
lo↑
li↑
� �?
x?
?
?
PPPPPPPPPPq
����������)
x
ro↓
ri↑
ro↑
ri↓
� �?
?
x?
?
RTA9�
RTA8j
� � b
��HH b �
� b��HH b � b
��HH b
����b
HH��b HH
��b
HH��b
li
lo
ro
ri
y1 y2r
�rr
r
rrr
r
r
-rr
rr
Speed Independent FIFO Controller (SI)
li rri ��
HH a r��a
��a
��HH a r lo
��HH a r ro
axr
Relative Timed FIFO Controller (RT-BM)
35
Apply and Automate: On Chip Networking
20mm
20m
m
x x
Start with your basic20×20mm
integrated circuit ina 65nm process
-
∃ blocks of 100kgates which need to
communicate
XXXXXXXXXXXXz
Studycommunication link
between pair of logicblocks
��������������������������1
36
Networking Problem Formulationl The critical repeater distance of this process is 600 microns
(optimal power/performance point)
l This nominal wire will therefore contain ≈ 17 repeated segments
l Each segment can be flopped
l Bandwidth depends on pipelining
Total distance = 10,000 microns, minimum delay = 30 FO4
���H
HH���H
HH s s s���H
HH���H
HH
���
���
���
���
HHH
HHH
HHH
HHH s s s
���
���
���
���
HHH
HHH
HHH
HHH
600µ 1.8 FO4 delays
37
Representative Results10,000µ 32-bit bus in a 65nm
process with low bit activity factor
and moderate bus utilization rate:
4-cycle least efficient. Clocked
and 2-cycle comparable, source-
synchronous better at high
frequency.
10,000µ 32-bit bus in a 65nm
process with low bit activity factor
and light bus utilization rate:
clocked protocols least efficient.
2-cycle and source-synchronous
clearly the best from energy
perspective.
38
Post Office: On Chip Network Implementations
39
Post Office: On Chip Network Implementation
“The good news is we’ve created a highly sophisticated, multi-processor
computer that’s the size of a doughnut. The bad news is . . . Fitsimmons just
dunked it in his cocoa.”
40
We cannot solve our problems with the same thinkingwe used when we created them.
Albert Einstein
This is what VLSI is all about.
Let’s think about our problems differently and use this rich canvas tosolve them in ways never previously dreamed.
41
Related Documents
