Doc.: IEEE 802.11-01/037r1 Submission March 2001 Khaled Turki et. al,Texas InstrumentsSlide 1 Simulation Results for p-DCF, v-DCF and Legacy DCF Khaled.

March 2001

Khaled Turki et. al ,Texas Instruments Slide 1

doc.: IEEE 802.11-01/037r1

Submission

Simulation Results for p-DCF, v-DCF and Legacy DCF

Khaled Turki, Jin-Meng Ho, Sid Schrum, Matthew Shoemake and Don Shaver

Texas Instruments Inc.

March 2001

Khaled Turki et. al ,Texas Instruments Slide 2

doc.: IEEE 802.11-01/037r1

Submission

Simulation Environment

• Simulation platform: OPNET

• MAC schemes simulated: 1. p-DCF as described in 01/139

2. v-DCF as described in 01/132 and 01/133

3. 802.11-1999

March 2001

Khaled Turki et. al ,Texas Instruments Slide 3

doc.: IEEE 802.11-01/037r1

Submission

Parameters

• Buffer size = 2 Mbits

• Packet lifetime = 500 ms

• PHY = 11 Mbps- DSSS

March 2001

Khaled Turki et. al ,Texas Instruments Slide 4

doc.: IEEE 802.11-01/037r1

Submission

Scenarios

• 20 streams• 15 STA• Bi-directional streams• Packet length = 1500 bytes and 250 bytes• Packet length distribution : Constant• Arrival process : Exponential inter-arrival times

March 2001

Khaled Turki et. al ,Texas Instruments Slide 5

doc.: IEEE 802.11-01/037r1

Submission

Contention-Based Access:CSMA with Adaptive Contention -- State Transition Diagram

ESTA active

Medium idle

for DIFS or EIFS?

CalculatePP

Yes

Any TCPP changed?

Transmission permitted?

(X PP?)

Transmita frame

Contentioncontinued?

Stop

Slot idle?

RecalculatePP

No

Yes Yes

No

No

Yes

Yes

No No

March 2001

Khaled Turki et. al ,Texas Instruments Slide 6

doc.: IEEE 802.11-01/037r1

Submission

Scenario 1

• Aggregate load 6 Mbps

• 20 Streams

• 15 STA

• 4 Priorities

• 5 Streams per priority

• Inter-arrival time exponentially distributed

• Packet length: Constant 1500 Bytes

March 2001

Khaled Turki et. al ,Texas Instruments Slide 7

doc.: IEEE 802.11-01/037r1

Submission

Aggregate Offered LoadOffered Load (Mbits/sec)

0

2

4

6

8

10

0 1 2 3 4 5 6 7 8 9 10

Offered Load (Mbits/sec)

March 2001

Khaled Turki et. al ,Texas Instruments Slide 8

doc.: IEEE 802.11-01/037r1

Submission

v-DCF Results

Scenario 1

March 2001

Khaled Turki et. al ,Texas Instruments Slide 9

doc.: IEEE 802.11-01/037r1

Submission

V-DCF Results

• For priority differentiation the following three cases are considered

1. CW_min[ 7,15,31,63], QIFS[DIFS]

2. CW_min[31],QIFS [ DIFS+3 slots, DIFS+2 slots, DIFS+1 slot, DIFS]

3. {CW_min[ 7],QIFS [DIFS]},{CW_min[15], QIFS[DIFS+1 slot]} {CW_min[31],QIFS [ DIFS+2 slots]},{CW_min[63],

QIFS[DIFS+3 slots]}

March 2001

Khaled Turki et. al ,Texas Instruments Slide 10

doc.: IEEE 802.11-01/037r1

Submission

v-DCFv-DCF

0

100

200

300

400

500

600

0 1 2 3 4 5 6 7 8 9 10

Simulation Time (sec)

De

lay

(m

s)

CW_min=63

CW_min=31

CW_min=15

CW_min=7

Legacy

1 Unstable TC

March 2001

Khaled Turki et. al ,Texas Instruments Slide 11

doc.: IEEE 802.11-01/037r1

Submission

v-DCFv-DCF CW differentiation only

0

50

100

150

200

250

300

350

0 1 2 3 4 5 6 7 8 9 10

Simulation Time (sec)

Ave

rag

e D

elay

(m

s)

CW_min=63

CW_min=31

CW_min=15

CW_min=7

Legacy

March 2001

Khaled Turki et. al ,Texas Instruments Slide 12

doc.: IEEE 802.11-01/037r1

Submission

V-DCF Priorities

• Priority 0 (lowest):CW_min[31] QIFS [DIFS+3 slots]

• Priority 1: CW_min[31] QIFS [DIFS+2 slots]

• Priority 2: CW_min[31] QIFS [DIFS+1slot]

• Priority 3(highest): CW_min[31] [DIFS]

March 2001

Khaled Turki et. al ,Texas Instruments Slide 13

doc.: IEEE 802.11-01/037r1

Submission

v-DCF

v-DCF - QIFS diffrentiation only

0

50

100

150

200

250

300

350

400

450

500

0 1 2 3 4 5 6 7 8 9 10

Simulation TIme (sec)

Dela

y (

ms)

Priority 0

Priority 1

Priority 2

Priority 3

Legacy

March 2001

Khaled Turki et. al ,Texas Instruments Slide 14

doc.: IEEE 802.11-01/037r1

Submission

v-DCF

v-DCF - QIFS differentiation only

0

20

40

60

80

100

120

0 1 2 3 4 5 6 7 8 9 10

Simulation Time (sec)

Avera

ge D

ela

y (

ms)

Priority 0

Priority 1

Priority 2

Priority 3

Legacy

Legacy

March 2001

Khaled Turki et. al ,Texas Instruments Slide 15

doc.: IEEE 802.11-01/037r1

Submission

V-DCF

v-DCF- QIFS+CW

0

20

40

60

80

100

120

140

0 1 2 3 4 5 6 7 8 9 10

Simulation Time (sec)

Dela

y (

ms)

Priority 0

Priority 1

Priority 2

Priority 3

Legacy

March 2001

Khaled Turki et. al ,Texas Instruments Slide 16

doc.: IEEE 802.11-01/037r1

Submission

v-DCF

v-DCF - QIFS + CW differentiation

0

20

40

60

80

100

120

140

0 1 2 3 4 5 6 7 8 9 10

Simulation Time (sec)

Avera

ge D

ela

y (

ms)

Priority 0

Priority 1

Priority 2

Priority 3

Legacy

March 2001

Khaled Turki et. al ,Texas Instruments Slide 17

doc.: IEEE 802.11-01/037r1

Submission

p-DCF Results

Scenario 1

March 2001

Khaled Turki et. al ,Texas Instruments Slide 18

doc.: IEEE 802.11-01/037r1

Submission

p-DCF

p-DCF- 20 ms update interval

0

2

4

6

8

10

12

14

16

18

20

0 1 2 3 4 5 6 7 8 9 10

Simulation Time (sec)

Del

ay (

ms)

TCPP 0

TCPP 1

TCPP 2

TCPP 3

Legacy

March 2001

Khaled Turki et. al ,Texas Instruments Slide 19

doc.: IEEE 802.11-01/037r1

Submission

p-DCF

p-DCF- 20 ms Probability Update Interval

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

1 2 3 4 5 6 7 8 9 10

Simulation Time (sec)

Ave

rag

e D

elay

(m

s)

TCPP 0

TCPP 1

TCPP 2

TCPP 3

Legacy

March 2001

Khaled Turki et. al ,Texas Instruments Slide 20

doc.: IEEE 802.11-01/037r1

Submission

p-DCF

p-DCF- 50 ms probability update interval

0

2

4

6

8

10

12

14

16

18

20

0 1 2 3 4 5 6 7 8 9 10

Simulation Time (sec)

Del

ay (

ms)

TCPP 0

TCPP 1

TCPP 2

TCPP 3

Legacy

March 2001

Khaled Turki et. al ,Texas Instruments Slide 21

doc.: IEEE 802.11-01/037r1

Submission

p-DCF

p-DCF 50 ms Probability Update Interval

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

1 2 3 4 5 6 7 8 9 10

Simulation time (sec)

Ave

rag

e D

elay

(m

s)

TCPP 0

TCPP 1

TCPP 2

TCPP 3

Legacy

March 2001

Khaled Turki et. al ,Texas Instruments Slide 22

doc.: IEEE 802.11-01/037r1

Submission

Simulation Resultsp-DCF and v-DCF

Scenario 2

March 2001

Khaled Turki et. al ,Texas Instruments Slide 23

doc.: IEEE 802.11-01/037r1

Submission

Scenario 2

• Aggregate load 3 Mbps

• 20 Streams

• 15 STA

• 4 Priorities

• 5 Streams per priority

• Inter-arrival time exponentially distributed

• Packet length: Constant 250 Bytes

March 2001

Khaled Turki et. al ,Texas Instruments Slide 24

doc.: IEEE 802.11-01/037r1

Submission

Aggregate Offered LoadAggregate Load (Mbits/sec)

0

2

4

6

8

10

0 1 2 3 4 5 6 7 8 9 10

Simulation Time (sec)

Wireless LAN.Load (Mbits/sec)

March 2001

Khaled Turki et. al ,Texas Instruments Slide 25

doc.: IEEE 802.11-01/037r1

Submission

Simulation Results v-DCF

Scenario 2

March 2001

Khaled Turki et. al ,Texas Instruments Slide 26

doc.: IEEE 802.11-01/037r1

Submission

v-DCFv-DCF

0

500

1000

1500

2000

2500

3000

0 1 2 3 4 5 6 7 8 9 10

Simulation Time (sec)

Dela

y (

ms)

CW_min=63

CW_min=31

CW_min=15

CW_min=7

Legacy

2 Unstable TC’s

March 2001

Khaled Turki et. al ,Texas Instruments Slide 27

doc.: IEEE 802.11-01/037r1

Submission

v-DCFv-DCF

0

200

400

600

800

1000

1200

1400

0 1 2 3 4 5 6 7 8 9 10

Simulation Time (sec)

Avera

ge D

ela

y (

ms)

CW_min=63

CW_min=31

CW_min=15

CW_min=7

Legacy

March 2001

Khaled Turki et. al ,Texas Instruments Slide 28

doc.: IEEE 802.11-01/037r1

Submission

Simulation Results p-DCF

Scenario 2

March 2001

Khaled Turki et. al ,Texas Instruments Slide 29

doc.: IEEE 802.11-01/037r1

Submission

p-DCFp-DCF

0

500

1000

1500

2000

2500

3000

0 1 2 3 4 5 6 7 8 9 10

Simulation Time (sec)

Dela

y (

ms)

TCPP 0

TCPP 1

TCPP 2

TCPP 3

Legacy

1 Unstable TC

March 2001

Khaled Turki et. al ,Texas Instruments Slide 30

doc.: IEEE 802.11-01/037r1

Submission

p-DCF p-DCF

0

100

200

300

400

500

600

700

800

900

1000

0 1 2 3 4 5 6 7 8 9 10

Simulation Time (sec)

Ave

rag

e D

ela

y (

ms

)

TCPP 0

TCPP 1

TCPP 2

TCPP 3

Legacy

March 2001

Khaled Turki et. al ,Texas Instruments Slide 31

doc.: IEEE 802.11-01/037r1

Submission

Conclusions

• Several simulations have been performed in order to evaluate the performance of two EDCF proposals (v-DCF and p-DCF)

• The probabilistic approach (p-DCF)shows performance superiority over v-DCF at higher loads and large STA population.– Delay and delay variance

– Dropped frame rate

– Stability in terms of frame delay

• V-DCF achieves differentiation by introducing additional delay among traffic categories (TC’s)

• Starvation effect is also observed in v-DCF