Power Saving and Power Management in WiFi and Bluetooth Networks

Preview:

DESCRIPTION

Power Saving and Power Management in WiFi and Bluetooth Networks. Prof. Yu-Chee Tseng Dept. of Comp. Sci. & Infor. Eng. National Chiao-Tung University ( 交通大學 資訊工程系 曾煜棋 ). Outline. Power control: - PowerPoint PPT Presentation

Citation preview

tseng:1

Power Saving and Power Managementin WiFi and Bluetooth Networks

Prof. Yu-Chee Tseng

Dept. of Comp. Sci. & Infor. Eng.

National Chiao-Tung University

( 交通大學 資訊工程系 曾煜棋 )

yctseng: 2

Outline Power control:

S.-L. Wu, Y.-C. Tseng, and J.-P. Sheu, "Intelligent Medium Access for Mobile Ad Hoc Networks with Busy Tones and Power Control", IEEE Journal on Selected Areas in Communications, 18(9):1647-1657, Sep. 2000.

Power management: Y.-C. Tseng, C.-S. Hsu, and T.-Y. Hsieh, "Power-Saving Pr

otocols for IEEE 802.11-Based Multi-Hop Ad Hoc Networks", Computer Networks, Elsevier Science Pub., Vol. 43, No. 3, Oct. 2003, pp. 317-337.

WiFi vs Bluetooth: T.-Y. Lin and Y.-C. Tseng, "An Adaptive Sniff Scheduling S

cheme for Power Saving in Bluetooth", IEEE Wireless Communications, Vol. 9, No. 6, Dec. 2002, pp. 92-103.

tseng:3

Introduction: Basic Concept

4

Introduction Battery is a limited resource in any

portable device. becoming a very hot topic is wireless

communication

Power-related issues: PHY: transmission power control MAC: power mode management Network Layer: power-aware routing

5

Transmission Power Control tuning transmission energy for higher

channel reuse example:

A is sending to B (based on IEEE 802.11) Can (C, D) and (E, F) join?

No! Yes!

6

Power Mode Management doze mode vs. active mode example:

A is sending to B (based on 802.11) Does C need to stay awake?

7

Power-Aware Routing routing in an ad hoc network with

energy-saving in mind Example: in an ad hoc network

+

+

+

+

+

+

SRC

N1 N2

DEST

N4N3

wu-ICCCN99:8

WirelessNetTseng

Intelligent Medium Access for Mobile Ad Hoc Networks with Busy Tones and Power

Control

S.-L. Wu, Y.-C. Tseng, and J.-P. Sheu,

IEEE J. of Selected Areas on Communications (JSAC)

wu-ICCCN99:9

WirelessNetTseng

Abstract

A New MAC Protocol based on RTS/CTS with Busy Tones with Power Control

wu-ICCCN99:10

WirelessNetTseng

Power Control Use an appropriate power level to transmit packets.

to increase the possibility of channel reuse to increase channel utilization

Example: (a) without power control:

the transmissions from C to D and from E to F are prohibited.

(b) with power control:all these can coexist.

F

E

DC

BA

F

E

DC

BA

(b )(a )

wu-ICCCN99:11

WirelessNetTseng

How to Tune Power Levels

Assumptions: A mobile host can choose on what power level to transmit a packet. On receiving a packet, the physical layer can offer the MAC layer

the power level on which the packet was received. Suppose Pt and Pr are the power levels a packet is sent and received, re

spectively.

= carrier wavelength n = path loss coefficient (typically 2 ~ 6) d = distance between sender and receiver gt and gr: antenna gains at the sender and receiver sides, respectivel

y

rtn

tr ggd

PP )4

(

wu-ICCCN99:12

WirelessNetTseng

Note: during a short period, the values of n and d can be treated as a constant. This makes power control possible.

Let Pmin be the minimum power level to decode a packet.

Suppose X sends an RTS to Y with power Pt.

If Y wants to reply a CTS to X with a power level PCTS, such that X receives the packet at the smallest power level Pmin, then we have:

Dividing the above formulas, we have:

rtn

tr ggd

PP )4

(

rtn

CTS ggd

PP )4

(min

t

CTS

r P

P

P

Pmin

wu-ICCCN99:13

WirelessNetTseng

General Rules in This Paper

Busy Tone (BT) Senders should send BTt, but gauge any BTr. Receivers should send BTr, but gauge any BTt.

General Rules: Data packet and BTt: transmitted with power control. CTS and BTr: transmitted at the normal (largest) power. RTS: at a power level based on how strong the BTr are aroun

d the requesting host. Channel Model:

BTt BTr

controlchannel

datachannel

frequency

wu-ICCCN99:14

WirelessNetTseng

Illustrative Example (I)

A is sending to B. A’s data packet and BTt at t

he minimal level (yellow circle).

B’s BTr at the largest level (white circle).

C intends to send to D. C hears no BTr. D hears not BTt. So the transmission can be

granted (pink circle).

A

DC

B

wu-ICCCN99:15

WirelessNetTseng

Illustrative Example (II)

Now we moe C into A’s circle. A is sending to B.

A’s data packet and BTt at the minimal level (yellow circle).

B’s BTr at the largest level (white circle).

C intends to send to D. C hears no BTr. D hears no BTt. So the transmission can be

granted (pink circle).

A

DC

B

wu-ICCCN99:16

WirelessNetTseng

Illustrative Example (III)

Next we move D into A’s circle. A is sending to B.

A’s data packet and BTt at the minimal level (yellow circle).

B’s BTr at the largest level (white circle).

C intends to send to D. C hears no BTr. D hears A’s BTt. So the transmission can NO

T be granted (pink circle).

A

DC

B

wu-ICCCN99:17

WirelessNetTseng

Illustrative Example (IV)

A is sending to B. A’s data packet and BTt at t

he minimal level (yellow circle).

B’s BTr at the largest level (white circle).

C intends to send to D. C hears A’s BTt and B’s BT

r. D hears no BTt. The transmission can be gra

nted if C controls its transmission power (pink circle).

A

D

C

B

wu-ICCCN99:18

WirelessNetTseng

Illustrative Example (V)

A is sending to B. A’s data packet and BTt at t

he minimal level (yellow circle).

B’s BTr at the largest level (white circle).

C intends to send to D. C hears A’s BTt and B’s BT

r. D hears no BTt. The transmission can be gra

nted if C controls its transmission power (pink circle).

A

DC

B

wu-ICCCN99:19

WirelessNetTseng

Many Transmission Pairs with Power Control and Busy Tones

A

F

ED

C

B

BTt and DATA: yellow circlesBTr: white circles

wu-ICCCN99:20

WirelessNetTseng

The Protocol

Pmax: the maximum transmission power

Pmin: the minimum power to distinguish a signal from a noise

Pnoise: the maximum power at which an antenna will regard a signal as a noise Pmin - Pnoise should be a very small value

Basic “Power” Rules: Data packet and BTt: transmitted with power control. CTS and BTr: transmitted at the largest power Pmax.

RTS: at a power level based on how strong the BTr are around the requesting host.

wu-ICCCN99:21

WirelessNetTseng

Detailed Protocol

On a host X intending to send a RTS to Y, X senses any receive busy tone BTr around it X sends a RTS on the control channel at power level Px:

If there is no BTr, let Px = Pmax.

O/w, let Pr be the power level of BTr that has the highest power among all heard BTr’s.

The RTS should not go beyond the nearest host that is currently receiving a data packet.

Pmax is used because BTr is always transmitted at the maximal power.

r

noixex P

PPP max

wu-ICCCN99:22

WirelessNetTseng

On Y receiving X’s RTS, Y senses any transmit busy tone BTt around it.

If there is any BTt, then Y ignores this RTS.O/w, Y does the following:

reply with a CTS at the maximum power Pmax turn on its receive busy tone BTr at the maximum power Pmax

On X receiving Y’s CTS, X transmits its data packet at power Px.

X turns on its transmit busy tone BTt at power Px.

Pr is the power level at which X receives Y’s CTS. Px is the minimal possible power level to ensure that Y can correctly receive the data packet.

rx P

PPP maxmin

wu-ICCCN99:23

WirelessNetTseng

Many Transmission Pairs with Power Control and Busy Tones

F

ED

C

G H

BTr

BTt

A B

RTS

CTS

wu-ICCCN99:24

WirelessNetTseng

Analysis

Scenario: A is currently sending to B. Another pair, C and D, is intending to communicate.

Goal: We want to find out the probability that C can send to D.

Through complicated calculus, we find that …

wu-ICCCN99:25

WirelessNetTseng

When BC < rmax

INTC(Ra, Rb, AB) = the intersection of the circles centered at a and b Ra = radius of the circle centered at a Rb = radius of the circle centered at b AB = distance of a and b

The probability that C can send to D when A is sending to B:

i.e., the coverage of Rc excluding the coverage of Ra Fig. 6

wu-ICCCN99:26

WirelessNetTseng

wu-ICCCN99:27

WirelessNetTseng

cont...

Integrating over = 0 .. 2, and then over CB = 0 .. rmax

Integrating over AB = 0 .. rmax, we have the final result

On the contrary, the DBTMA has probability of 0.

wu-ICCCN99:28

WirelessNetTseng

When rmax < BC < 3rmax

Main difference: C’s RTS will be sent with max. power.

The probability that C can send to D when A is sending to B:

See Fig. 7:At point C1, node C can always send.At point C2, node C can’t send if D is in A’s range.

wu-ICCCN99:29

WirelessNetTseng

wu-ICCCN99:30

WirelessNetTseng

cont...

Integrating over = 0 .. 2, and then over CB = rmax..3rmax

Integrating over AB = 0 .. rmax, we have the final result

wu-ICCCN99:31

WirelessNetTseng

cont.

On the contrary, the DBTMA has a success probability of

X

change to rmax

wu-ICCCN99:32

WirelessNetTseng

Discrete Power Control

The levels of power provided by hardware may not be infinitely tunable. We may have a discrete number of power levels.

Theorem: Given a fixed integer k, evenly spreading the k power levels

will be the best choice. I.e., (1/k)*Pmax, (2/k)*Pmax, (3/k)*Pmax, …, (k/k)*Pmax.

wu-ICCCN99:33

WirelessNetTseng

Simulation Parameters

Simulation parameters physical area = 8km 8km max transmission distance (rmax) = 0.5 or 1.0 km number of mobile hosts = 600 Speed of mobile hosts 0 or 125 km/hr. length of control packet = 100 bits link speed = 1 Mbps transmission bit error rate = 10-5/bit

wu-ICCCN99:34

WirelessNetTseng

Simulation Results: Channel utilization vs. traffic load

(a) rmax = 0.5 km (b) rmax = 1.0 km

0

2 0

4 0

6 0

8 0

10 0

0 20 0 400 600 8 00 10 00 1200

L oa d (pa ck e t/m s)(a )

Cha

nnel

Util

itiza

tion

O ursD B TM A

0

5

1 0

1 5

2 0

2 5

3 0

0 10 0 200 300 4 00 500 600 700 800

L o ad (p ac k e t/m s)(b )

Cha

nnel

Util

itiza

tion

O urs

D B T M A

wu-ICCCN99:35

WirelessNetTseng

Channel utilization vs. data packet length at various traffic loads

20

25

30

35

40

45

1 2 4 8

Data Packet Length (Kbits)

Cha

nnel

Util

itiza

tion

Load=100 Kbit/msLoad=200 Kbit/msLoad=400 Kbit/msLoad=800 Kbit/ms

wu-ICCCN99:36

WirelessNetTseng

Channel Utilization vs. Number of Power Levels

rmax = 1 km; arrival rate = 200 or 400 packets/ms; packet length = 1 or 2 Kbits So 4 to 6 levels will be sufficient.

0

5

10

15

20

25

0 2 4 6 8 10 12 14 16 18 20

Power Level

Cha

nnel

Util

itiza

tion(

data

pac

ket

time/

ms) 27.34

wu-ICCCN99:37

WirelessNetTseng

Channel Utilization vs. Traffic Load

mobility = 0 km/hr and 125 km/hr The transmission distance rmax = 1.0 km

0

5

10

15

20

25

30

35

40

45

50

55

0 100 200 300 400 500 600 700 800

Load (packet/ms)

Cha

nnel

Uti

liti

zatio

n

packet length=1 Kbits, speed= 0 km/hourpacket length=1 Kbits, speed=125 km/hourpacket length=4 Kbits, speed= 0 km/hourpacket length=4 Kbits, speed=125 km/hour

wu-ICCCN99:38

WirelessNetTseng

Short Conclusion

a new MAC protocol power control on top of RTS/CTS and busy tones

Channel utilization can be significantly increased because the severity of signal overlapping is reduced.

39

Power Mode Management in IEEE 802.11

Y.-C. Tseng, C.-S. Hsu, and T.-Y. Hsieh, "Power-Saving Protocols for IEEE 802.11-Based Multi-Hop Ad Hoc Networks", Computer Networks, Elsevier Science Pub., Vol. 43, No. 3, Oct. 2003, pp. 317-337 (also in INFOCOM).

40

Power Consumption IEEE 802.11 power model

transmit: 1400 mW receive: 1000 mW idle: 830 mW sleep: 130 mW

41

Power Mode Management Power modes in IEEE 802.11

PS and ACTIVE

Problem Spectrum: infrastructure ad hoc network (MANET)

single-hopmulti-hop ad hoc networks

42

Infrastructure Mode two power modes: active and power-

saving (PS)

43

Ad Hoc Mode (Single-Hop) PS hosts also wake up periodically.

interval = ATIM (Ad hoc) window

Beacon Interval Beacon Interval

ATIM Window

ATIM Window

Host A

Host B

Beacon

BTA=2, BTB=5

power saving state

power saving state

Beacon

ATIM

ACK

active state

data frame

ACK

44

Problem Statement(Multi-Hop MANET)

Clock Synchronization: a difficult job due to communication delays

and mobility Neighbor Discovery:

by inhibiting other's beacons, hosts may not be aware of others’ existence

Network Partitioning: with unsynchronized ATIM windows, hosts

with different wakeup times may become partitioned networks

45

Network-Partitioning Example

Host A

Host B

A

B

C

D

E

F

D

E

F

Host C

Host D

Host E

Host F

ATIM window

Network Partition

46

What Do We Need? PS protocols for multi-hop ad hoc

networks Fully distributed No need of clock synchronization (i.e.,

asynchronous PS) Always able to go to sleep mode, if

desired

47

Features of Our Design Guaranteed Overlapping Awake

Intervals: two PS hosts’ wake-up patterns always

overlap no matter how much time their clocks drift

Wake-up Prediction: with beacons, derive other PS host's

wake-up pattern based on their time difference

48

Beacon Int. (BI)

Act. Win. (AW)

Structure of a Beacon Interval

BI: beacon interval (to send beacons) AW: active window

BW: beacon window MW: MTIM window (for receiving MTIM) listening period: to monitor the

environment

BW MW listening BW MW listening

49

Three Protocols Based on the above structure, we

propose three protocols Dominating-Awake-Interval Periodical-Fully-Awake-Interval Quorum-Based

50

P1: Dominating-Awake-Interval intuition: impose a PS host to stay

awake sufficiently long “dominating-awake” property

BWBIAW 2/

Host A

Host B

Beacon Interval

Beacon Interval

Beacon Interval

Beacon Interval

╳ ╳

51

Problem: only dectectable in ONE direction

Adjustment: odd beacon interval:

Active Window = BW + MW + listening

even beacon interval: Active Window = listening + MW + BW

Host A

Host B

Odd Beacon Interval Even Beacon Interval

Odd Beacon Interval Even Beacon Interval

B

B

B B

╳M

M

M

M

52

Host A

Host B

odd beacon interval

Beacon window MTIM WindowActive window

odd beacon interval

even beacon interval

even beacon interval

Unicast Example

MTIM Data

ACKACK

53

Characteristics dominating awake

wake-up ratio < 1/2 sensibility

A PS host can receive a neighbor’s beacon once every two beacon intervals.

suitable for highly mobile environment

54

P2: Periodical-Fully-Awake-Interval

Basic Idea: In every T intervals, stay awake in one full

interval. wake-up ratio 1/T

compared to 1/2 of protocol 1

Two types of beacon intervals: Low-power interval Fully-awake interval (in every T intervals)

55

Example (T = 3)

Host A

Host B

T(=3) Beacon Intervals

Fully-awake

Fully-awake Low-power

Low-power

Low-power

Low-power

╳ ╳ ╳╳ ╳ ╳

T: Interval between the fully awake periods

A PS host can receive its neighbor’s beacon frame in every T = 3 beacon intervals

56

Definitions of Intervals Low-power interval:

active window + doze window AW = BW + MW

i.e., listening period = 0

Fully-awake interval: no doze window

i.e., AW = BI

very energy-consuming, so only appears once every T beacon intervals

57

P3: Quorum-Based Quorum Sets:

Two quorum sets always have nonempty intersection.

(used here to guarantee detectability) A matrix example:

n

nc1

c2

r1 r2

Host A’s quorum intervals

Host B’s quorum intervals

Non-quorum intervalsintersection

58

Example (2D matrix quorum)

0 1 2 3

4 5 6 7

8 9 10 11

12 13 14 15

0 1 2 3

4 5 6 7

8 9 10 11

12 13 14 15

Host A’s quorum intervals

Host B’s quorum intervals

Non-quorum intervals

Host A’ quorum intervals

1514131211109876543210

31302928272625242322212019181716

Group 1

Group 2

Host B’s quorum intervals

1514131211109876543210

31302928272625242322212019181716

Group 1

Group 2

Overlapping intervals

59

Overlapping Property Overlap no matter how clocks drift

demo ...

0 1 2 3

4 5 6 7

8 9 10 11

12 13 14 15

0 1 2 3

4 5 6 7

8 9 10 11

12 13 14 15

1514131211109876543210

1514131211109876543210

Host A’s quorum intervals

Host B’s quorum intervals

60

Quorum and Non-quorum Intervals

Quorum interval: AW = BI (i.e., fully awake)

Non-quorum interval: no beacon, only MTIM window AW < BI BW = 0, AW = MW

Beacon window MTIM WindowActive window

Beacon Interval Beacon IntervalQuorum Interval

Non-quorum Interval

61

Summary

Protocol Numbers of beacons per inter

val

Active ratio

Neighbor sensitivity

Dominating 1 1/2+BW/BI BI

Periodical 1 1/T T*BI/2

Quorum (2n-1)/n2 (2n-1)/n2 (n2/4) * BIBI: length of a beacon interval

AW: length of an active window

BW: length of a beacon window

MW: length of an MTIM window

T: interval between the fully awake periods

n: length of the square

62

Summary Identify the problems of PS mode in

IEEE 802.11 in multi-hop ad hoc networks. clock drifting, network-partitioning

Propose several PS protocols Connecting this problem to quorum

issue in distributed systems.

63

Sniff Scheduling for Power Saving in Bluetooth

64

Overview (cont.) Addressing

48-bit Bluetooth Device Address (BD_ADDR)

3-bit Active Member Address (AM_ADDR)

8-bit Parked Member Address (PM_ADDR)

Four operational modes: Active Sniff Hold Park

65

Bluetooth Networks Piconet

one master + at most 7 active slaves Scatternet

multiple piconets to form a larger network

Packets Exchange Scenario

MASTER

SLAVE 1

SLAVE 2

SLAVE 3

ACLSCO SCO SCO SCOACLACL ACL

67

Low-Power Sniff Mode A slave can enter the low-power sniff m

ode by setting a parameter (Tsniff, Nsniff_attempt, Dsniff)

in per slave basis

68

LMP_PDUs for Sniff

23 timing control flag Dsniff Tsniff Nsniff_attempt Nsniff_timeoutLMP_sniff_req

LMP_not_accepted

LMP_accepted

LMP_unsniff_req

4 op code reason

3 op code

24

LMP_sniff_req

LMP_accepted / LMP_not_accepted

initiating LM LM

LMP_sniff_req..

LMP_unsniff_req

LMP_accepted

.

.

69

Sniff Scheduling Problem How to determine the sniff parameters?

Goal: balancing power consumption and traffic need

Earlier Works naïvely adjust parameters in an

exponential way double/halve sniff interval or active window

whenever polling fails/succeeds

The placement of active windows of multiple slaves on the time axis is not addressed.

70

Design Goals consider multiple slaves together adaptively schedule sniff parameters

more accurate in determining the sniff-related parameters based on slaves’ traffic loads

include solutions of placing of active windows of sniffed slaves on the time axis

71

Proposed Architecture

LM Evaluator

LM Evaluator

LM Evaluator

Evaluator for Slave1

Evaluator for Slave2

Evaluator for SlavekResource

Pool

Scheduler

S1

SkSearching

S1

S2

Sk

Slave 1

Slave 2

Slave k

Master

.

.

.

.

.

.

Link Manager(LM)

S2

sniff related LMP packets

72

Tk,Nk,Dk: current sniff parameters for slave k. Uk: the slot utilization of slave k. Bk: the buffer backlog of slave k. Wk: a weighted value to indicate the current requirement of s

lave k.

Bmax is the maximum buffer space

Sk: the desired slot occupancy of slave k, which is the expected ratio of Nk / Tk.

0 < δ < 1 (to tolerate some unexpected traffic)

CalculatorX

Uk

Bk

CalculatorY

Sk Wk

the evaluator

73

Resource Pool (RP) Although time slots are an infinite seque

nce, we represent them as a sequence of 2-D matrices. each matrix M is of the size 2u × T time slots are viewed in a “row-major” way

The availability of M:

74

RP Example M’s size = 23 × 15 = 120

1 0 0 0 0

1

0

1

1

1

0

1

1 1 0 0

0 0 0 0

1 1 0 0

0 0 0 0

1 1 0 0

0 0 0 0

1 1 0 0

1 1 1 1 1

0

0

0

1

0

0

0

0 0 0 0

0 0 1 1

0 0 0 0

1 1 1 1

0 0 0 0

0 0 1 1

0 0 0 0

1 0 0 0 0

1

1

0

1

1

1

0

1 1 1 1

1 1 1 0

0 0 0 0

0 0 0 0

1 1 1 1

1 1 1 0

0 0 0 0

column

row

0

1

2

3

4

5

6

7

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

75

Example: to allocate a slot occupancy of 16/120(** Note: 16/120 = 8/60)

76

Example: to allocate a slot occupancy of 16/120(** Note: 16/120 = 4/30 = 2/15)

77

A Running Example 5 slaves Each slave initially has an equal occupa

ncy of 1/5 of the matrix M. We discuss two strategies:

longest sniff interval first shortest sniff interval first

78

Scheduling Policies:Longest Sniff Interval First(LSIF)

a) initial state (equal shares)

b) reduce S2 to 2/60

c) reduce S3 to 3/120

d) increase S4 to 6/30

S1 S1 S1 S2 S2S1S1S1S1S1S1S1

S1 S1 S2 S2S1 S1 S2 S2S1 S1 S2 S2S1 S1 S2 S2S1 S1 S2 S2S1 S1 S2 S2S1 S1 S2 S2

S2 S3 S3 S3 S4S2S2S2S2S2S2S2

S3 S3 S3 S4S3 S3 S3 S4S3 S3 S3 S4S3 S3 S3 S4S3 S3 S3 S4S3 S3 S3 S4S3 S3 S3 S4

S4 S4 S5 S5 S5S4S4S4S4S4S4S4

S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5

Round0 N1/T1 = N2/T2 = N3/T3 = N4/T4 = N5/T5 = 3/15

S1 S1 S1S1S1S1S1S1S1S1

S1 S1S1 S1S1 S1 S2 S2S1 S1S1 S1S1 S1S1 S1 S2 S2

S3 S3 S3 S4S3 S3 S3 S4S3 S3 S3 S4S3 S3 S3 S4S3 S3 S3 S4S3 S3 S3 S4S3 S3 S3 S4S3 S3 S3 S4

S4 S4 S5 S5 S5S4S4S4S4S4S4S4

S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5

Round1 W2 = 0.18 < rlb => S2 = [(3/15)*0.18]/0.8 = 0.045 0.045 = 5/120 = 2/60 = 1/30 = 0/15 => (Ok' , Tk' , Nk' ) = (3, 60, 2)

S1 S1 S1S1S1S1S1S1S1S1

S1 S1S1 S1S1 S1 S2 S2S1 S1S1 S1S1 S1S1 S1 S2 S2

S4

S3

S4S4S4S4S4S4

S3 S3 S4

S4 S4 S5 S5 S5S4S4S4S4S4S4S4

S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5

Round2 W3 = 0.11 < rlb => S3 = [(3/15)*0.11]/0.8 = 0.028 0.028 = 3/120 = 1/60 = 0/30 = 0/15 => (Ok' , Tk' , Nk' ) = (5, 120, 3)

S1 S1 S1 S4 S4S1S1S1S1S1S1S1

S1 S1S1 S1 S4 S4S1 S1 S2 S2S1 S1 S4 S4S1 S1S1 S1 S4 S4S1 S1 S2 S2

S4 S4 S4 S4

S4

S4

S4S3

S4 S4 S4

S4 S4 S4

S4 S4 S4S3 S3

S5 S5 S5S5 S5 S5S5 S5 S5S5 S5 S5S5 S5 S5S5 S5 S5S5 S5 S5S5 S5 S5

Round3 W4 = 0.9 > rub => S4 = [(3/15)*0.9]/0.8 = 0.23 0.23 = 27/120 = 13/60 = 6/30 = 3/15 => (Ok' , Tk' , Nk' ) = (18, 30, 6)

79

S1 S1 S1 S2 S2S1S1S1S1S1S1S1

S1 S1 S2 S2S1 S1 S2 S2S1 S1 S2 S2S1 S1 S2 S2S1 S1 S2 S2S1 S1 S2 S2S1 S1 S2 S2

S2 S3 S3 S3 S4S2S2S2S2S2S2S2

S3 S3 S3 S4S3 S3 S3 S4S3 S3 S3 S4S3 S3 S3 S4S3 S3 S3 S4S3 S3 S3 S4S3 S3 S3 S4

S4 S4 S5 S5 S5S4S4S4S4S4S4S4

S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5

Round0 N1/T1 = N2/T2 = N3/T3 = N4/T4 = N5/T5 = 3/15

S1 S1 S1S1S1S1S1S1S1S1

S1 S1 S2S1 S1S1 S1 S2S1 S1S1 S1 S2S1 S1S1 S1 S2

S3 S3 S3 S4S3 S3 S3 S4S3 S3 S3 S4S3 S3 S3 S4S3 S3 S3 S4S3 S3 S3 S4S3 S3 S3 S4S3 S3 S3 S4

S4 S4 S5 S5 S5S4S4S4S4S4S4S4

S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5

Round1 W2 = 0.18 < rlb => S2 = [(3/15)*0.18]/0.8 = 0.045 0.045 = 5/120 = 2/60 = 1/30 = 0/15 => (Ok' , Tk' , Nk' ) = (3, 30, 1)

S1 S1 S1S1S1S1S1S1S1S1

S1 S1 S2S1 S1S1 S1 S2 S3S1 S1S1 S1 S2S1 S1S1 S1 S2 S3

S4S4S4S4S4S4S4S4

S4 S4 S5 S5 S5S4S4S4S4S4S4S4

S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5S4 S5 S5 S5

Round2 W3 = 0.11 < rlb => S3 = [(3/15)*0.11]/0.8 = 0.028 0.028 = 3/120 = 1/60 = 0/30 = 0/15 => (Ok' , Tk' , Nk' ) = (4, 60, 1)

S1 S1 S1S1S1S1S1S1S1S1

S1 S1 S2S1 S1S1 S1 S2 S3S1 S1S1 S1 S2S1 S1S1 S1 S2 S3

S4 S4 S4S4S4S4S4S4S4S4

S4 S4S4 S4S4 S4S4 S4S4 S4S4 S4S4 S4

S5 S5 S5S5 S5 S5S5 S5 S5S5 S5 S5S5 S5 S5S5 S5 S5S5 S5 S5S5 S5 S5

Round3 W4 = 0.9 > rub => S4 = [(3/15)*0.9]/0.8 = 0.23 0.23 = 27/120 = 13/60 = 6/30 = 3/15 => (Ok' , Tk' , Nk' ) = (5, 15, 3)

Scheduling Policies:Shortest Sniff Interval First(SSIF)

a) initial state (equal shares)

b) reduce S2 to 1/30

c) reduce S3 to 1/60

d) increase S4 to 3/15

80

Conclusions Proposed:

Power-saving protocols for IEEE 802.11-based multi-hop ad hoc networks

Sniff-scheduling schemes for Bluetooth-based piconets

References:1. T.-Y. Lin and Y.-C. Tseng, “An Adaptive Sniff Scheduling S

cheme for Power Saving in Bluetooth”, IEEE Personal Communications (to appear).

2. Y.-C. Tseng, C.-S. Hsu, and T.-Y. Hsieh, “Power-Saving Protocols for IEEE 802.11-Based Multi-Hop Ad Hoc Networks”, IEEE INFOCOM, 2002.

Recommended