Embedded Signal
Processing
Final Project
UNDER CONSTRUCTION
Overview
- ============================
Preliminary !
====================================
- You must follow this website for the latest changes/updates
- Project groups are assigned by instructor
- Do not procrastinate, and do not underestimate the amount of time
needed and complexity of the task
- As a rough guide, the 3-page midsemester project is a prelude to what is expected in final
projects and final reports
- A well-written report/paper is EXPECTED
- STRONGLY RECOMMEND that you read IEEE
authorship series: How to Write for Technical Periodicals
& Conferences
- Final project main phases (see schedule below):
- A a 1-page proposal for faculty-sponsored projects and
special topics
- Design review and 1-page report
- Final walk-through and 1-page report and rough demo
- Final reports (see details below) due on the week before final demonstration
- Final demonstration of circuit on last day of class
- Final powerpoint presentations due during final exam period
(Approx. 15 slides, see below)
IN NO CASE may code or files or data or pictures be exchanged
between student groups, there is to be NO COPYING of group reports!
Also, each student must be able to independently
answer any questions themselves during demos.
All students are expected to learn all aspects of every project.
Nevertheless, students are encouraged to collaborate (not copy)
during the lab sessions.
Rough Schedule (TBD - keep checking for changes)
Schedule and due dates:
- Begin: today
- 11/8 Special projects: see topic0 below, 1-page proposals due for special topics
- Note: each group MUST
propose a Java/NetBeans ethernet
interface to control/display data or a different/unique
hardware display/interface component (LCD text,
graphic, USB, RJ45, bluetooth, etc). So, coordinate with
instructor and get your initial plans
approved by email from instructor before purchasing
hardware!!
- 11/10 Receive purchased parts (LCD, buttons, ADC, DAC, etc)
- 11/15 Design review:
pass/fail, 1-page report showing:
- Use our previous 1-page project report format for the
previous projects, including:
- Summary of technical performance goals/specs
- ADS simulation or Matlab simulation or Mathcad simulation of
basic system functionality as appropriate
- NetBeans mock-up or block
diagram of proposed interface
component with description of technical goals to
allow control and display
- Note: each group MUST
propose a different/unique NetBeans
interface or hardware display/interface component
(LCD text, graphic, USB, RJ45, bluetooth, etc). So,
coordinate with instructor and get your plans
approved before purchasing hardware!!
- Demo your simulation and/or NetBeans
mockup in the lab
- 11/22 Final walk-through:
pass/fail, 1-page report and rough demo showing
- Use our previous 1-page project report format for the
previous projects, including:
- This is your best chance for final
feedback/changes ahead of final report
- Summary of technical performance goals/specs
- FINAL block diagram of interface component with technical
goals to allow control and display
- ADS simulation time domain triangle wave
- ADS frequency domain real and imag parts of Zin
- Preliminary demo your interface component and technical
component in the lab
- 11/29 Final reports
due
- By beginning of class period, you must:
- You must email pdf file
- You must email LaTex source files zipped
- You must bring hardcopies to class period
- Formatting:
- 12/6 Final demonstrations
of hardware in lab during class
- email pdf of any revised final report due
- email zip-file of mbed code due
- 12/13 Final powerpoint
slides
- due by email by noon of final
exam day
- minimum of 13 slides required:
- Title,
- Overview,
- Prior Art/references,
- Theory
- K64F algorithm flowchart
- System block diagram
- Software code snippet of main computation loop
- Photograph of prototypes
- Theoretical results plotted (frequency response, Zin, etc)
- Simulation results and Theoretical values plotted and
compared
- Measured results plotted and compared with theory
- Table of design goals versus measured results and
simulation results
- Summary/Conclusion
- Also: any make-up presentations/materials will be due on
12/13
- email zip-file of any required revised reports, mbed code,
etc
- Professional quality powerpoint is required
- Every member of team must present some 2 slides minimum
- Include: Tech goals, Block diagrams/photos planned,
Theory, Simulation, Measured data, Software code snippet
- Approx. 15 slides and 15 minutes limit per group
Assignments
- Topics are assigned by the instructor
- Each group will be assigned a topic
IN 2016 All Topics Have been Assigned
- Each topic includes 2 components:
- Component 1: A technical topic
component with technical goals for DSP performance,
such as a "-40 pF capacitor at 10 megasample/s"
- Component 2: An interface component
with technical goals to provide both
control and display of technical
parameters, such as a button to increase capacitance and LCD
readout or IoT/NetBeans readout
- Note: each group MUST
propose a different/unique NetBeans
interface or hardware display/interface component
(LCD text, graphic, USB, RJ45, bluetooth, etc). So,
coordinate with instructor and get your plans approved before purchasing
hardware!!
- Control may be by use of pushbuttons, sliders, webpages,
laptop USB app, or cellphone apps (TTY not acceptable)
- Control must be sufficient to vary parameters as required
- Readout may be LCD text, LCD graphics, LED numeric,
webpage, laptop USB app, or cellphone apps, etc. (blinky and
TTY not acceptable)
- Readout must be reasonable, such as readout in nF, or
ohms, or whatever
- Example: push a button and each time the resistance
increases 50 ohm, cycling through 10 values, displayed on a
4-digit LCD
- Example: enter a resistance on a webpage or NetBeans,
write through RJ45 internet port, display values on
webpage
- USB-TTY is not an adequate
display/interface, you must use ethernet or new hardware
for display
- For groups working on RC and RL and any similar projects
- Please note that the simulation results in ADS may give
the FALSE impression that impedance measurements are easy.
- Do not underestimate the time required to measure the
frequency domain response of impedance,
- … and do not underestimate the need for each group to
review methods for measuring impedance using an oscilloscope
at frequencies between 1 KHz and the Nyquist limit (half
sample rate). NOTE: network analyzers cannot
measure your circuits, since these instruments do not
measure below 30 KHz.
- Your plots should have at least 20 points showing real and
imaginary parts of Z(s) from 1 KHz to 40 KHz.
- It would be best for one member in each group to
begin measurements of a simple known circuit such as 200
ohms in series with 10 nF in the lab, … so you are prepared
to measure your embedded project later. Measurement of
this known circuit will ensure confidence in you chosen
measurement technique (do not skip this).
- Some example measurement methods:
- Technical topics, specifications, and goals
- Topic0: Custom
faculty-supported project
- The objective of the final project is to produce quality
research and write a research paper ready for IEEE
conference submission. Therefore, other
faculty-sponsored research projects using embedded signal
processing can be considered, if a
faculty follow-on paper submission is planned
- However, it is incumbent on the student to submit a 1-page
project description before the specified deadline
- Generally, such a project should include:
- Technical component with
specifications
- Interface component with
description
- Note: each group MUST
propose a different/unique
interface hardware display/interface component
(LCD text, graphic, USB, RJ45, bluetooth, etc). So,
coordinate with instructor and get your plans approved before purchasing
hardware!!
- The 1-page description should follow the format for the
"Topic1" below, including:
- Name of lead/sponsor faculty
- Names of all team members
- Block diagrams/photos planned
- Theory planned
- Simulation plans
- Measured data plan
- Software code snippet plan
- A primary reference paper such
as my digital non-Foster paper at ISCAS2015
- Topic1: digital series RL
- Using the FRDM K64F board, design a digital series-RL
(digital resistor in series with inductor) as described in
the digital series RL
theory writeup
- Technical specifications:
- Base specification: Sampling rate 80,000 sample/s
(T=12.5 μs), Rs = 50 Ω, Rser= -100, and L= -0.02 H
(Rser is the series resistance)
- For comparison, also show time domain L di/dt ramp
waveform for a 20 mH inductance, in addition to -20 mH
- Beware of dc current
leading to infinite values: you will likely need to set
some artificial ground point at some voltage
- Control interface:
- Vary L from -50 mH to 50 mH in steps of 5 mH or
smaller
- Vary Rser from -250 to 250 ohms in steps of 50 or
smaller
- Note: each group MUST
propose a different/unique
interface hardware display/interface component
(LCD text, graphic, USB, RJ45, bluetooth, etc).
So, coordinate with instructor and get your plans approved before purchasing
hardware!!
- For this project, you should read also the notes on latency included in the
the digital series RL
theory writeup, and adjust accordingly
- Required report contents:
- Block diagrams/photos: 1)
a block diagram for digital non-Foster as in Fig. A1 of
digital RL writeup, 2) simple block diagram of control
interface scheme (buttons+ LCD +FRDM?), 3) photograph of
complete system assembled
- Theory for: 1) difference
equation, 2) H(z), 3) stability, and 4) impedance Z(s)
- Simulations showing: 1)
time domain L di/dt as in Fig. A6b of digital RL
writeup, 2) frequency domain real and imaginary
parts of Zin as in Fig. A7a of digital RL writeup,
- Measured data showing: 1)
time domain L di/dt as in A6a of digital RL writeup (but
not a photo!), 2) micosoft excel frequency domain
real and imaginary parts of Zin over specified frequency
range as in Fig. A7a of digital RL writeup including
dotted theoretical real and imaginary as example in Fig.
A7b, 3) time domain L di/dt as in A6a, but for positive
inductance
- Tabular data: use 5
columns: data-item, theoreticalValue, MeasuredValue,
SimulatedValue, units. In rows, place the
following data-items: di/dt in A/s, peak voltage in mA,
computed L from di/dt and peak voltage, Zin(s) at 1 KHz
- Software code snippet
showing difference equation for base specification
- Topic2: high-speed digital
series RL
- Same as topic1, except:
- Must present formal 1-page proposal as for Topic0 above
- Demonstrate some observable negative inductance value
at/above 1 megasample/s (possibly external ADC)
- Must see sawtooth ramp in time domain
- Interface component can be reduced/eliminated to
compensate for increased difficulty
- Latency will become a large issue
- Topic3: floating RC
- Same as topic1, except:
- Floating RC instead of grounded RL
- Must present formal 1-page proposal as for Topic0 above
- Demonstrate a floating (not grounded) negative
capacitance value at or above 80,000 samples/s
- Must see sawtooth ramp in time domain
- Interface component can be reduced/eliminated to
compensate for increased difficulty
- Topic4: floating RL
- Same as topic3, except:
- Floating RL instead of floating RC
- See each topic above for minimum
required data content for your reports and your demos
Report Formatting TBD
- ============================
WARNING !! ====================================
- **** WARNING **** YOU MUST ADHERE
TO REQUIRED IEEE FORMAT Here:
- FORMAT TBD
- Format: 5 or 6 page along lines of 3-page mid-semester
report, but add simulations and add interface design
components
- Must have higher quality images
- MUST be suitable quality for
IEEE publication
- The final report format is likely to be a 5-6 page version
of the midsemester 3-page report
template ( embDspMidtermProjTemplateSec.docx )
for this project (see instructor for password)
- You MUST make full use of all required number of pages,
empty space will be penalized
- A well-written report/paper is
EXPECTED
- STRONGLY RECOMMEND that you read IEEE
authorship series: How to Write for Technical Periodicals
& Conferences
- Clearly describe everything, including:
- varibles in block diagrams
- variables in formulas
- units of variables KHz, pF, nH, m, s,
- all traces on oscilloscope plots
- pictures of display/interface or NetBeans
- For NetBeans A block diagram of whole system from
laptop/netbeans through microrcontroller showing DSP block
and ADC/DAC
- all curves on plots
- all results in any tables
- Project Demos
- Be prepared to demonstrate and discuss items such as:
- Demonstrate sine wave response
- Demonstrate triangle wave response
- Demonstrate triangle wave response for
- Change the sampling period and sampling rate
- Change C, L, R for a digital device
- Set breakpoints
- Start and stop the debugger
- Demonstrate a full clean/build
- Be prepared to answer questions such as:
- What is frequency response of the devices?
- What is z-transform of the difference equations?
- What port number is ADC (PT???)
- What port number is DAC(PT???)
- How can the value of the digital device be changed?
- What happens if the sample rate changes for the code?
- Which line of code sets up the interrupt/timer?
- How does myDsp() subroutine code get executed, if it is
not called inside of main{}?
- What is an interrupt?
- Explain stability criteria.
- Where are stable poles in the z-plane?
- What is the starred transform?
Copyright 2015 T. Weldon
Freescale, Kinetis are registered trademarks of Freescale
Semiconductor, Inc. Texas Instruments, TI, C2000, and
Code Composer Studio are a registered trademarks of Texas
Instruments Incorporated. Adafruit is a registered
TradeMark of Limor Fried. ARM, Cortex and Keil are
registered trademarks of ARM Limited (or its subsidiaries).
Apple, Yosemite, Mac OS, iPhone, iPad, MacBook, Mac, iTunes, and
Xcode are trademarks of Apple inc. All other product or service
names are the property of their respective owners. Arduino
is a registered trademark of Arduino.