January 2007

Emily Qi et alSlide 1

doc.: IEEE 802.11-07/0058r0

Submission

Paging Power Saving Analysis Update

Notice: This document has been prepared to assist IEEE 802.11. It is offered as a basis for discussion and is not binding on the contributing individual(s) or organization(s). The material in this document is subject to change in form and content after further study. The contributor(s) reserve(s) the right to add, amend or withdraw material contained herein.

Release: The contributor grants a free, irrevocable license to the IEEE to incorporate material contained in this contribution, and any modifications thereof, in the creation of an IEEE Standards publication; to copyright in the IEEE’s name any IEEE Standards publication even though it may include portions of this contribution; and at the IEEE’s sole discretion to permit others to reproduce in whole or in part the resulting IEEE Standards publication. The contributor also acknowledges and accepts that this contribution may be made public by IEEE 802.11.Patent Policy and Procedures: The contributor is familiar with the IEEE 802 Patent Policy and Procedures <http:// ieee802.org/guides/bylaws/sb-bylaws.pdf>, including the statement "IEEE standards may include the known use of patent(s), including patent applications, provided the IEEE receives assurance from the patent holder or applicant with respect to patents essential for compliance with both mandatory and optional portions of the standard." Early disclosure to the Working Group of patent information that might be relevant to the standard is essential to reduce the possibility for delays in the development process and increase the likelihood that the draft publication will be approved for publication. Please notify the Chair < stuart.kerry@philips.com> as early as possible, in written or electronic form, if patented technology (or technology under patent application) might be incorporated into a draft standard being developed within the IEEE 802.11 Working Group. If you have questions, contact the IEEE Patent Committee Administrator at <patcom@ieee.org>.

Date:2007-01-12Name Company Address Phone email Sunghyun Choi Seoul National University Kwanak P.O.Box 34, Seoul 151-600 +82-28801753 schoi@snu.ac.kr Joe Epstein Meru Networks 1309 S Mary Ave, Sunnyvale CA 408-215-5345 jepstein@merunetworks.com Kyunghun Jang Samsung Advanced Institute

of Technology khjang@samsung.com

Moo Ryong Jeong DoCoMo USA Labs 3240 Hillview Avenue, Palo Alto, CA 650-496-4761 jeong@docomolabs-usa.com

Sunggeun Jin ETRI Kwanak P.O.Box 34, Seoul 151-600 +82-28801753 sgjin@mwnl.snu.ac.kr

Youngsoo Kim Samsung Advanced Institute of Technology

kimyoungsoo@samsung.com

Mee-Suk Lee Korea Telecom 17, Woomyenon-dong Seocho-gu Seoul +8225266174 kitlee@kt.co.kr Sudheer Matta Trapeze Networks 5753 W. Las Positas Blvd, Pleasanton,CA 9254742290 matta@trapezenetworks.com R. R. Miller AT&T 180 Park Ave, Florham Park NJ 973-236-6920 rrm@att.com Subbu Ponnuswamy Aruba Networks 1322 Crossman Ave Sunnyvale, CA 408-754-1213 Subbu@arubanetworks.com Henry Ptasinski Broadcom Corp. 190 Mathilda Place Sunnyvale, CA 408-543-3316 henryp@broadcom.com Emily Qi Intel Corp. 2111 NE 25th Ave Hillsboro OR 503-264-7799 Emily.H.Qi@intel.com Dorothy Stanley Aruba Networks Warenville, IL 630-363-1389 dstanley@arubanetworks.com James Tsai Intel Corp. 2111 NE 25th Ave Hillsboro OR 503-712-5509 James.tsai@intel.com Jesse Walker Intel Corp. 2111 NE 25th Ave Hillsboro OR 503-712-1849 Jesse.Walker@intel.com Fujio Watanabe DoCoMo USA Labs 240 Hillview Avenue, Palo Alto, CA 650-496-4726 watanabe@docomolabs-usa.com

Authors:

January 2007

Emily Qi et alSlide 2

doc.: IEEE 802.11-07/0058r0

Submission

Abstract

This document provides a range of power saving analysis in support of Idle Mode and WLAN Paging proposal (doc 11-07-0060r0), based on different use cases and device power consumptions.

January 2007

Emily Qi et alSlide 3

doc.: IEEE 802.11-07/0058r0

Submission

Agenda

• Motivation • Idle Mode and Paging Concepts • Power Consumption Models• Power Saving Benefit

January 2007

Emily Qi et alSlide 4

doc.: IEEE 802.11-07/0058r0

Submission

• Extend battery standby hours for handhelds and Ultra Mobile PCs in support of applications such as:

- VoIP over WLAN

- Instant Messenger

- Push Email

Motivation

January 2007

Emily Qi et alSlide 5

doc.: IEEE 802.11-07/0058r0

Submission

Power Consumption Issues

• When a standby mode STA is moving, the STA has to wake up to re-associate with the visited APs, which causes:

– Scanning

– BSS transition

• Needs to wake up OS

• Key Management and Derivations

• Exchange of Authentication and Reassociation messages

January 2007

Emily Qi et alSlide 6

doc.: IEEE 802.11-07/0058r0

Submission

Introducing Idle Mode and Paging• Idle Mode: save power and resources

– STA enters “deep sleep” if there is no outgoing/incoming traffic

– If STA roams out of the coverage area of the last known associated AP, STA doesn’t need to reassociate with a new AP when it is in “idle mode”

• Paging: wake STA to receive packet– Infrastructure will “page” the client if there is incoming traffic

– STA wakes up at a longer “sleep” interval to check for incoming frame, and STA will associate with an AP when it receives the “paging”

• Protocol Overview (Beacon/Probe Response)– Idle Mode Request (STAAP) and Response (APSTA) frames

– Enter Idle Mode/Exit Idle Mode

– Paging Indication element– Included in the Beacon frame at Paging Interval

– Included in the Probe Response frame as per STA requests

January 2007

Emily Qi et alSlide 7

doc.: IEEE 802.11-07/0058r0

Submission

Power Saving Analysis Modes Considered and Assumptions

• Legacy Power Saving Mode (PSM)– The STA is on standby (i.e., it has no data to send/receive)

– STAs wake up at Listen Interval (configurable, e.g. 2sec) to receive Beacon and check whether there is any incoming traffic

– Roaming requires scanning and re-association

• Paging Mode (PM)– The STA is on standby (i.e., it has no data to send/receive)

– STAs wake up at paging Interval (configurable, e.g. 2sec) to receive Beacon and check whether there is any incoming traffic

– No required re-association while roaming within the Paging Group, but needs scanning to find an AP in the paging group

Note that any layer 3 and above messages, e.g. Multicast/broadcast/keep-alive, are not considered for either mode.

January 2007

Emily Qi et alSlide 8

doc.: IEEE 802.11-07/0058r0

Submission

Power Consumption Model• Standby Hours (SH)

– SH = Battery Capacity (WATTHOUR)/PC

• Power Consumption (PC), mW– Power Consumption while Receiving Beacons (PC_B)

• Total receiving beacon time over an hour (BT) = awake time receiving beacon(11.7ms or 11.8ms) *(3600s/2s between beacons)

• PC_B = BT/3600000ms * (Device_Idle_PC + RF_Rx_PC)

– Power Consumption while Roaming (PC_R)• Total Roaming Awake Time (RAT)

• P_R = RAT/3600000ms * (Device_Idle_Active_PC + RF_Tx_Rx_PC)

– Power Consumption while Idle (do nothing) (PC_I)• Idle time (IT) = 3600000ms – BT - RAT

• PC_I = IT/3600000ms * (Device_Idle_PC + RF_Idle_PC)

– Total Power Consumption: PC = PC_B + PC_R + PC_I

January 2007

Emily Qi et alSlide 9

doc.: IEEE 802.11-07/0058r0

Submission

Awake Time and PC PSM Mode STA Paging Mode STAScanning time (ST) Active scanning per channel: 12ms to 20ms = 2 ms + 10ms/18ms

Passive scanning per channel: 105ms = 100ms + 5 ms

Active scanning with a perfect .11k support: 36 ms : 12ms x 3 (only .11g)

Active scanning maximum with no 11k support: 480ms : 20ms x (11+13) (dual band, all channels)

Active scanning (assumed): 240ms= 20ms x 12 (dual band) *

Passive scanning : 1856ms = 20ms x 11 + 105ms x 8 (dual band)

Passive scanning (assumed) : 500ms =~520ms=20ms x 5 + 105ms x 4 (dual band)

Number of triggered scans/roams over an hour N N/2 **

System resume time (SRT) Phone: 50ms; UMPC: 500ms 0

Transition time (TT) 11r: 50ms; 0

Waiting time after Assoc. (WT) Average 50ms 0 ***

Total Roam Power Consumption (PC-R):

Note: RF_Rx_Tx_PC = 0.4 * RF_Tx_PC + 0.6 * RF_Rx_PC

1/3600000 x N

(ST * (Device_Idle_PC +RF_Rx_Tx_PC) + SRT * 2 * (Device_Active_PC + RF_Rx_PC) + (TT+WT) * (Device_Active_PC + RF_Rx_Tx_PC))

1/3600000 * N/2 * ST * (Device_Idle_PC + RF_Rx_Tx_PC)

Roam Awake time (RAT) and Power Consumption

*“An Empirical Analysis of the IEEE 802.11 MAC Layer Handoff Process” by Arunesh Mishra et al. **Paging scan is less frequent than a legacy device roaming scan, at the ratio of cell size and beacon coverage. Ratio could be: 1/1 ~ 1/4. The more optimized and highly dense the network, the larger difference. Ratio of ½ is assumed here. See backup slides.***Paging Indication received in Probe Response during the active scanning

January 2007

Emily Qi et alSlide 10

doc.: IEEE 802.11-07/0058r0

Submission

Use Cases and Estimated BSS Transition Rate• Assume AP Cell Radius is ~40ft and Walking speed is ~ 4 ft/s.• Our statistical data shows Average Time in Cell (ATC) is ~30s (conservatively)• Carpeted Enterprise

– Assume an engineer moves 1 time per hour for a meeting or break, 5 minutes walking per office hour.

– Estimated total transitions over an office hour: 5*60s/30s = 10• Manufacturing (e.g. Semiconductor Fabrication)

– Assume a Fab technician walks 30 minutes per hour wearing a WiFi phone, although sometimes those movements might be within the same cell (25%)

– Estimated total transitions per working hour: 30*60/ 30s *75%= 45• Hospital

– Best case: Assume a hospital nurse walks 30 minutes per working hour wearing a WiFi phone or carrying a UMPC. At times, though, those movements might be within the same cell (25%).

– Estimated total transitions per working hour: 30*60/ 30s*0.75 = 45– Note that a lot of interference in a hospital setting may cause more scans and roams.

Based on the various BSS transition rates, the following power consumption comparison is given in a range of [0 roams, 60 roams]

January 2007

Emily Qi et alSlide 11

doc.: IEEE 802.11-07/0058r0

Submission

Power Consumption Comparison based on various Data Sets

We have calculated power consumption comparisons based on: – Roam rate in a range of [0 roams, 60 roams]

– Scanning time at: 36ms, 240ms and 500ms

– 3 Phone devices and 3 UMPC devices with various Tx/Rx//Idle Power Consumption data

– Phone - 400mW/200mW/3mW

– Phone - 550mW/350mW/5.5mW

– Phone - 800mW/500mW/5.5mW

– UMPC - 500mW/300mW/5.5mW

– UMPC - 800mW/500mW/5.5mW

– UMPC - 1200mW/800mW/5.5mW

January 2007

Emily Qi et alSlide 12

doc.: IEEE 802.11-07/0058r0

Submission

UMPC Standby Hours Comparision Scanning Time @240ms

80

85

90

95

100

105

110

115

120

0 10 20 30 40 50 60 70

The Number of Roams per Hour

Th

e S

tan

db

y H

ou

rs

PS Mode

Paging Mode

UMPC Standby Hours – Scanning Time @ 240ms

Battery Capacity = 11840 mWH // 3200mA x 3.7 VRF_Tx_PC = 800mW RF_Rx_PC = 500mW RF_Idle_PC = 5.5mW Device_Active_PC = 1200mW Device_Idle_PC = 95mW

4.92% 22.10%

January 2007

Emily Qi et alSlide 13

doc.: IEEE 802.11-07/0058r0

Submission

Phone Standby Hours Comparision Scanning Time @240ms

150

155

160

165

170

175

180

185

190

0 10 20 30 40 50 60 70The number of Roams per Hour

Th

e S

tan

db

y H

ou

rs

PS Mode

Paging Mode

Phone Standby Hours - Scanning time @ 240ms

Battery Capacity = 3240 mWH // 900mA x 3.6 V RF_Tx_PC = 550mWRF_Rx_PC = 350mWRF_Idle_PC = 5.5mWDevice_Active_PC = 150mW Device_Idle_PC = 10mW

2.36%

10.65%

January 2007

Emily Qi et alSlide 14

doc.: IEEE 802.11-07/0058r0

Submission

Power Saving Percentage- Scanning Time @36ms

For a “less-mobile” user (10 roams per hour): 1.19% - 5.57%For a “more-mobile” user (45 roams per hour): 7.01% - 25.16%

Power Saving: Scanning Time @36ms

-5

0

5

10

15

20

25

30

35

40

0 10 20 30 40 50 60 70

The Number of Roams per Hour

Po

we

r S

av

ing

Pe

rce

nta

ge

(%

)

Phone-400mW-200mWPhone-550mW-350mWPhone-800mW-500mWUMPC-500mW-300mWUMPC-800mW-500mWUMPC-1200mW-800mW

January 2007

Emily Qi et alSlide 15

doc.: IEEE 802.11-07/0058r0

Submission

Power Saving Percentage- Scanning Time @500ms

For a “less-mobile” user (10 roams per hour): 2.73% - 6.12%For a “more-mobile” user (45 roams per hour): 12% - 27.08%

Power Saving: Scanning Time @500ms

-5

0

5

10

15

20

25

30

35

40

0 10 20 30 40 50 60 70

The Number of Roams per Hour

TH

e P

ow

er

Sa

vin

g P

erc

en

tag

e (

%)

Phone-400mW-200mWPhone-550mW-350mWPhone-800mW-500mWUMPC-500mW-300mWUMPC-800mW-500mWUMPC-1200mW-800mW

January 2007

Emily Qi et alSlide 16

doc.: IEEE 802.11-07/0058r0

Submission

Summary

With what we have done today, you can consider the power consumption of your devices and your use cases, and determine power saving benefits of WLAN Paging and Idle Mode proposal.

Thank You!

January 2007

Emily Qi et alSlide 17

doc.: IEEE 802.11-07/0058r0

Submission

802.11g AP Coverage Area*• Coverage area

• Transmit power: 15dBm• OFDM

9095100

140180210

250300ft

54Mbps48Mbps36Mbps24Mbps18Mbps12Mbps9Mbps6Mbps

Sufficient to cover 20,000 sq. ft

Sufficient to cover 60,000 sq. ft

cover 306,000 sq. ft

January 2007

Emily Qi et alSlide 18

doc.: IEEE 802.11-07/0058r0

Submission

Paging Mode Scanning at 2.4GHz

Beacon coverage AP-2 (ch 6)

Cell size

Beacon coverage

AP-3 (ch 11)

AP-1 (ch 1)

AP1-RSSI-PagingRoaming-Scanning Threshold

Active scan and find (AP-4) in the same channel, stop scanning.

Paging Scanning time

Device moves from left to right

Cell size- 40ft

AP-4 (ch 1)

January 2007

Emily Qi et alSlide 19

doc.: IEEE 802.11-07/0058r0

Submission

Battery Standby Hours Calculation

• Office/Working hours only shall be considered for standby hours calculation

• For a dual mode phone, if mobile user is not in the office and WiFi network is not available, device will connect to the cellular network. So battery life shall not be considered as part of WiFi standby hours.

• For a pure WiFi, if mobile user is not in the office, there are two possibilities:

– Leave it at the office and power off – for energy efficiency concerns

– Bring it home. WiFi network discovery on the way back home consumes a lot of power

– In either case, battery life shall not be considered as part of WiFi standby hours.