Low Power Computing

Teknologi Sistem Cerdas
CCE62461
Embedding Determination and Innovation in Tradition 1

Agenda kuliah
• Introduction + Overview
• Klasifikasi dan klastering pada embedded system
• Review paper terkait
• Kendali dan kontrol pada embedded system
• Wearable computing
• -----------------------------------UTS-------------------------------------
• Common Sense Computing
• Ubiquituous Comp, Context Aware, Intelligent Interface
• Biomedical system application
• low power computing

• -----------------------------------UAS-------------------------------------
8 September 2022 Slide 2

Low Power optimization on
embedded system
• Basics of Embedded Systems: low power design. Prof. Dr. Kalle
Tammemäe, et.al
• Low-Power Optimization
• MSP430 Microcontroller Basics, John H. Davies
• High-Level Optimization for Low Power Consumption on
Microprocessor-Based Systems, Ortiz & Santiago
• Power and Energy Efficiency in Embedded Systems
• Design techniques for low power systems, Havinga & Smit
• EMBEDDED SYSTEM POWER CONSUMPTION: A SOFTWARE
OR HARDWARE ISSUE? Matalon, et.al
Introduction
• Why low power?

▪ Portable and mobile devices are getting popular, which have limited power
sources, e.g., battery
▪ Energy conservation for our planet
▪ Power generates heat low carbon
▪ Power optimization becomes a new dimension in system design,
besides performance and cost
4
Motivation – why worry about power?
Wearables
More computation
▪ Video decoding
▪ Augmented reality processing
▪ Speech recognition and synthesis
▪ Complex communication protocols
Autonomous units
▪ Sea and ocean technology
▪ Space technology, etc.
Battery capacity and weight limits (10%
improvement per year)
Energy harvesting limits (solar panel, etc.)
5
Energy and Power
• Energy: ability to do work

• Most important in battery-powered systems
• Power: energy per unit time
• Important even in wall-plug systems---power becomes heat
Power draw increases

with…
• Vcc
• Clock speed
• Temperature
6
Power is related to performance
Design Choices
Power issues areas
▪ Display
This hardware is a major power drain. It is essential that the driving software
monitors utilization very carefully and dims or shuts down displays when they
are not in use.
▪ Wireless Peripherals
In almost all cases, the software may turn off the wireless interface when it is
not in use or optimize the transmission power for current circumstances
▪ CPU Utilization
The execution performance of the code may be quite significant – minimizing
the number of instructions that need to be executed to perform a given task,
reduces the number of Watt-Hours required. Low power consumption is closely
related to algorithm performance
▪ Memory
Although it’s unlikely to be a significant factor, memory does consume power,
and minimizing the memory footprint of the software may be desirable.
▪ USB
If the device is a USB peripheral, its software should respond correctly to
requests from the host to shut down or suspend. An embedded system may
also be a USB host, in which case the host software must monitor USB
peripheral utilization and shut down or put to sleep unused devices
General Strategies
• Put the system in low-power modes and/or use low-power modules
as much as possible
• How?
❑ Provide clocks of different frequencies frequency scaling
❑ Turn off clocks when no work to do clock gating
❑ Use interrupts to wake up the CPU, return to sleep when done (another reason
to use interrupts)
❑ Switched on peripherals only when needed
❑ Use low-power integrated peripheral modules in place of software
11
Low-power efficient coding techniques:
▪ Optimize program flow
▪ Use unsigned data types where possible
▪ Avoid floating point operations
▪ Power control in software:
▪ Sleep/suspend
▪ Dynamic voltage and frequency scaling
http://processors.wiki.ti.com/index.php/ULP_Advisor
12
Tips for estimating
battery life accurately
[Beningo engineering]

Tips 1:traditional battery budget analysis
▪ To estimate the required battery size

▪ To anticipate battery life
▪ For each component of the system:
• Determine minimum, typical, maximum current consumption
▪ Estimate percentage of time in each state
[http://www.eetimes.com/author.asp?doc_id=1324160] 14
Tips 2: Software Rate Monotonic Analysis (RMA)
▪ Identify different tasks performed by software

▪ Identify the time of each task execution
▪ Can improve traditional battery budget “guestimations“
15
Tips 3: Chip vendor tools
▪ There are many variables concerning how microcontroller

consume energy
▪ Vendor datasheets have pages about power consumption based
on temperature, voltage, peripheral set, etc.
▪ Still, how are these numbers obtained and how accurate they are?
▪ E.g. ARM CortexA9 processor has 6 hardware counters which can
be configured to monitor 62 local events
▪ Chip vendors starting to supply developers with tools with ability
to simulate current consumption
▪ Example: STMCubeMx from STMicroelectronics
16
Tips 4: Bench-top experiments
▪ Test assumptions on the bench!

▪ As early as possible in the design cycle, prototype parts be tested for
real-world behavior!
▪ This should not be clean test, but verifies basic assumptions
▪ Find a quick and inexpensive way to prove power portion of the
design
▪ Data taken from the bench can be used to refine the model of power
analysis
17
Tips 5: Battery life-cycle analysis
▪ Concerning rechargable batteries:

Manufactured battery good lifetime may decrease dramatically after
a few month of usage
▪ Peak current draw from embedded system:
When the battery is partially discharged, it is possible that peak
current can cause the battery voltage to dip into or below its brown
out value => dead embedded system
18
Measuring power consumption
Test setup for measuring average current consumption:

• oscilloscope for measuring current with respect to time;
• for voltage probe use relatively small value resistor
o Small enough to not affect low power circuitry
o Large enough to provide voltage to measure with decent precision
o Easy to calculate
19
Result example: Current consumption versus time
State Time (µs) Current (mA)
Wake up 496 6.1
Rx 288 22.3
Tx 104 29.3
Processing 1180 8.1
20
Formulas and calculations
Average current during event =
[(State 1 time)*(State 1 current) + (State 2 time)*(State 2 current) + …] / (Total awake time)
Average current =
[(Interval – Total awake time)* (Average sleep current) + (Total awake time)* (Average current during event)] / (Interval)
Expected battery life running continuously =

(Battery capacity) / (Average current)
21
Issues to Discuss
• Which saves more energy?

❑ Use a higher frequency to run a program faster so as to sleep longer
❑ Use a lower frequency to run a program to save power, but system may be
active longer
22
Ada Pertanyaan ?


Low Power Computing

Uploaded by

Copyright:

Available Formats

Low Power Computing

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Low Power Computing

Uploaded by

Copyright:

Available Formats

Teknologi Sistem Cerdas

Embedding Determination and Innovation in Tradition 1

• low power computing

8 September 2022 Slide 2

• Why low power?

• Energy: ability to do work

Power draw increases

Embedding Determination and Innovation in Tradition 13

▪ To estimate the required battery size

▪ Identify different tasks performed by software

▪ There are many variables concerning how microcontroller

▪ Test assumptions on the bench!

▪ Concerning rechargable batteries:

Test setup for measuring average current consumption:

State Time (µs) Current (mA)

Wake up 496 6.1

Processing 1180 8.1

Expected battery life running continuously =

• Which saves more energy?

Embedding Determination and Innovation in Tradition 23

You might also like