Transcript Slide 1

Power Management in SDR
Max Robert, Jeffrey H. Reed
Mobile and Portable Radio Research Group (MPRG)
Virginia Tech
September 14, 2004
1
Overview




Power fundamentals
Overview of approaches
Current state of technology
Power management for SDR



Operation states
Interface descriptions
Conclusion
2
Power Basics
P   CV f
2

Terms:





Fixed attributes



α: switching activity
C: capacitance
V: voltage
f: operating frequency
Switching activity (algorithm-specific)
C is fixed
Attributes open to modification

V, f
3
Software-controlled power

Some attributes are determined at design
time and cannot be changed at run-time



Compiler optimizations
Waveform design
Attributes that can change at runtime




Operating voltage
Operating frequency
Timing control
 Thread management in the case of processors
Active components
4
General Power Management

Power management split into three principal
categories

-
Previous work on each section varies in depth
T h re a d p rio rity a lg o rith m s
D yn a m ic Vo lta g e S ca lin g (D V S ) a lg o rith m s
D yn a m ic F re q u e n cy S ca lin g (D F S ) a lg o rith m s
P o licy se le ctio n a lg o rith m s
A p p lica tio n
- A d va n ce d C o n fig u ra tio n a n d P o w e r In te rfa ce (A C P I)
- O p e ra tin g S yste m P o w e r M a n a g e m e n t (O S P M )
O p e ra tin g S yste m /E n viro n m e n t
H a rd w a re
-
M u lti-vo lta g e H W (C ru so e )
M u lti-fre q u e n cy H W (C P U s, F P G A s, A S IC s)
F le xib le R F
E fficie n t co m p ile rs
A lte rn a te d a ta flo w s
5
Software-Controlled Attributes

Timing management


Thread priority in the case of a GPP or (sometimes) DSP
Bus/message management in system


Algorithm may optimize wait times to cluster work for component
Voltage and Frequency selection are related

Higher voltage will allow higher frequencies

Optimal voltage for frequency not necessarily the best choice


Active component selection is a subset


Voltage switching may be slower than frequency switching
 May desire to maintain operating range for quick response
Set voltage or frequency to zero for that component
Flexible RF

Still unclear what attributes of the RF will be software-controlled


Mixer bias, filter BW, others
Framework- and application-based strategies need to be sufficiently
flexible to allow smooth integration of flexible RF control
6
Application & Hardware

Significant previous research



Adaptive management algorithms
Advanced power hardware-level power
management techniques
Application- and HW-based strategies well
suited for static applications

Current way of developing power-saving
strategies
 Fixed waveform

Can be optimized to specific platform
7
Operating System/Environment

Software structure necessary to support
power management functionality

Standard interface
 Switch between different states



i.e.: sleep (several levels), active
States not necessarily limited to sleep modes
Standard management structure
 Maintain state of all devices in system


State machine for describing system
Unified structure for handling associated
devices
8
State-of-the-Art

Development limited to PC needs


BIOS-based management


Power management for laptops
 Sleep mode management
BPM (BIOS power management)
 Has no awareness of the user’s (or
application’s) needs
Operating-system based management


OSPM (OS power management)
Current de-facto standard
 Most publications today are algorithms for the
efficient switching between states using OSPM
9
ACPI

Advanced Configuration and Power
Interface

State machine used to describe
machine configuration


States associated with different parts of
the system
Common interface provided to enact
changes in the state of the system
10
ACPI States

Basic set of
states

Cx


CPU states
Dx

States for
peripheral
device




Modem
Network card
Screen
Hard drive
11
Interface Descriptions

Multiple standardized interfaces
provided

Example
AcpiEnterSleepStatePrep
 AcpiEnterSleepState
 AcpiLeaveSleepState


Provides common interface for the
change of states for the system
12
OSPM

Operating System Power
Management

Model describing partitioning of power
consumption management
Operating system determines when to
trigger power management features
 BIOS determines how to perform power
management features

13
OSPM/ACPI

OSPM and ACPI
integrated


OSPM provides
mechanism for
selection of
mode
 Kernel
initiates
action
ACPI provides
common
interface to
hardware
14
Power Management For SDR

SDR places challenges different from
classic communications system


Can support application swapping
Needs to support wide set of devices
 Variety of needs and states


Difficult to narrow to small, well-defined set of
states
Requires sophisticated power control
structures

Applications can be more predicable than PC
 Possible to determine “fast enough” speed

Blind throttle for the application may not be
enough
15
State Support

ACPI supports mesh state machine

Assumes basic device states can be throttled
S1

S2
Sn
Linear transitions (throttle) are a subset of the
mesh state machine
S1
S2
Sn
16
Problems with Mesh SM

Assumes that all transitions are fundamentally
“equal”


Does not take into account QoS issues related with
state change
Example:

Voltage and frequency are fundamentally linked

Increased voltage will allow a higher set of
frequency settings to be supported

Throttle transitions based on the assumption that lowest
possible voltage is supported for the desired frequency
 If a change in voltage incurs a higher time delay
than a change in frequency, could lead to unplanned
additional latencies
17
Rate-Change Support in
Communications
Example (802.11b):


Support alternate processing speeds for different
sections of received frame
P L C P P re fix
P L C P P a y lo a d
11 M b p s P S D U
1 M b p s (P re a m b le + H e a d e r)
CW
192us
Tra n s itio n
Tra n s itio n
P ro ce s sin g
ty p ic a lly ~ 4 0 0 u s
Fast
S lo w

Benefits


D e c is io n p o in t: d is c a rd fra m e ?
Minimizes required computing power
Provides ability to discard frame before high-speed
processing is necessary
18
Rate Change and SDR

Waveform takes place of “user” in SDR

Latencies associated with change of state need to be
taken into account

State switching needs to be in order of microseconds
Millisecond-level switches may be too slow for some
waveforms
Ideally, should cluster state changes into transition state



Example:

Crusoe TM5400 automatically controls voltage and
frequency settings




Slow ramp in voltage for up-frequency changes followed by fast
frequency change
Fast down frequency change followed by slow voltage change
Changes performed automatically
 Possible for some equipment to leave change requests up to
the application
Voltage regulator can have a significant impact on the
transition speeds in core operating voltage

May be too slow (ms+) for some waveforms
19
State Machine Description

Break down state
machine into slowchange states and
related fast-change
states

Provides application
with ability to change
states quickly during
waveform operation

Also supports sleep
or standby
operation
F 1 ,1
F 1 ,2
F 1 ,3
V1
F 2 ,1
F 2 ,2
F 2 ,3
V2
F 3 ,1
F 3 ,2
F 3 ,3
V3
20
Sample Operation

Fast operation

Can cycle between 500 and
700 MHz

700
1 .8 V
May choose not to
transition, since change to
600 or 700 MHz expected
soon
300
400
500
1 .5 V
100
200
300
1 .2 V
Can still transition to lower
powers


600
500 MHz may be more
efficient at 1.5V


500
Support significantly lower
power consumption levels
Same concept can apply to
other devices

FPGAs, ASICs, CCMs, DSPs
21
Common Interface

Design of common interface will have to
wait until conceptual framework is
finalized

Will rely on ACPI to determine appropriate
interfaces
 Will also rely heavily on SCA 3.0 interface
specifications


SCA 3.0 concentrates on non-CORBA interface
descriptions
Challenging task
 Generic nature of hardware makes static
definition of interfaces unlikely

Will most likely require a generic structure
 May be able to leverage AML
22
Application-Level Power Management

Algorithm development

Field of research currently has large number of
contributions
 Primarily concentrating on PC-based systems


ACPI/OSPM
Clear from OEPM that SDR will have some
unique characteristics
 Optimization strategies will be based on the
permutations possible by conceptual
framework

This research venue cannot proceed until
conceptual framework is complete
23
Conclusion

Some concepts in power management are
fairly mature




Current state-of-the-art does not cover all
needs of SDR


PC power management
Voltage and frequency scaling
Policies and algorithms
Unique issues related to nature of SDR
Actively developing techniques to resolve
these issues
24
Acknowledgement

This work is funded by the DCI
Postdoctoral Research Fellowship
and the MPRG Affiliates Program
25