MDU cable network channel S parameter measurements - Tested in Wuxi city, Jiangsu province

(1)

MDU cable network channel S

parameter measurements

- Tested in Wuxi city, Jiangsu province

Zhao Cui - Cable TV technology research institute,

ABS, SARFT

Gao Xiaojun – Jiangsu Cable Network

1 IEEE P802.3bn> EPoC – October 2012

(2)

Contents

•

Test items

•

Test program

•

Test topology and environment

•

Test results

(3)

Test items

•

Scenario: Node+0 passive cable network at MDU

•

Test items: channel S parameters

•

Test equipment: Network analyzer ( Agilent E5070B )

•

Selected 3 sites in Wuxi, and tested on 2012.10.18-10.19

–

Test site 1 --

畅舜苑小区

–

Test site 2 --

广丰二村

–

Test site 3 --

金河湾家园

•

Considerations on test sites selection:

– mainly according to upstream CNR measured on the CMTS side. Dividing

the network transmission performance into excellent/good/bad 3 classes.

– Tall building (32 floors) and multi-floor building (5-6 floors)

– Mainly test passive tree topology network (with cascaded tap/splitter)

(4)

Test program

•

Full spectrum bandwidth test & multi- trunk test (divide

whole spectrum into several segments)

–

Full spectrum test:

•

1MHz~2GHz; 1601 sampling points

–

multi- trunk test

：

•

1MHz ~1.5GHz

•

64MHz segment, 1601 points/per 64MHz

•

Network analyzer calibration

Network analyzer First stage tap/splitter Last stage tap/splitter Distribution network Passive cable network at MDU Calibration plane 电子校准件 Short cable Long cable

Note: Test results described herein can only be used for the analysis of coaxial network characteristics trends, does not apply to the channel modeling system design

(5)

Topology - test site1 ( building 11#)

•

Building infos

：

–

11# is a 6-floors building,

4households per floor

–

N31

°

33'40",E120

°

17'30

“

–

Altitude of 11m

•

Test time/environment

–

2012.10.18 morning

•

Start time - 10:35

•

End time - 12:50

–

environment

•

Temperature: 15

°

C

•

Humidity : 42%

220 218 216 212 210 206 Open 2F 1F 4F 3F 6F 5F 601 602 Cut off VNA （E5070B） -5 501 502 401 402 201 202 301 302 101 102 220 218 216 212 210 206 604 603 -5 504 503 404 403 204 203 304 303 104 103 open Cut off 108 -5 -5 -5 -5 -5 -5 -5 -5 -5 -5 Notes:

1. all coaxial cable are 75-5 cable; 2. Imperial F -type connectors -5 -5 5 IEEE P802.3bn> EPoC – October 2012

(6)

Topology - test site2 (building 22#)

•

Building infos

：

–

22# is a 5-floor building , 4

households per floor

–

N31

°

35'56",E120

°

19'3"

–

Altitude: 9m

–

Only one subscriber connects

with network, other ports are all

open.

•

–

2012.10.18 afternoon

• Start time: 14:20 • End time: 16:33

–

environment

• Temperature: 16°C • Humidity : 28% Cut off 2F 1F 4F 3F 5F open open open Impe dance matc hed open 204 818 814 414 open open open -5 -5 -5 21# open VNA （E5070B） Notes:

1. all coaxial cables are 75-5 cable

2. Imperial F-type connector

(7)

Topology - test site2 (building 31#)

•

Building infos

：

–

31# is 32-floor building, 3households per floor

–

N31

°

36'4",E120

°

18'24"

–

Altitude 5m

•

–

2012.10.19 morning

•

Start time: 10:25

•

End time: 12:00

–

environment

•

Temperature: 14

°

C

•

Humidity : 52%

19F用户 18F用户 20F用户 21F用户 23F用户 22F用户 26F用户 25F用户 106 620 618 616 614 Load 18F 28F用户 27F用户 29F用户 30F用户 33F用户 31F用户 35F用户 33F用户 620 618 616 614 Load 17F用户 16F用户 13F用户 15F用户 12F用户 11F用户 10F用户 9F用户 620 618 616 614 Load 8F用户 7F用户 5F用户 6F用户 4F用户 3F用户 2F用户 Load 620 618 616 814 Load Load Cut off VNA （E5070B） Notes:

1. all coaxial cables are 75-5 cable 2. imperial F-type connectors 106

75ohm

(8)

Test results - S21 & S12

S21 _S12

(9)

Test results – micro-reflection1

0 0.5 1 1.5 2 2.5 3 x 10-6 -120 -100 -80 -60 -40 -20 0 X: 1.735e-007 Y: -33.11 Time(s) A m p lit u d e (d B ) 0 0.5 1 1.5 2 2.5 3 x 10-6 -120 -100 -80 -60 -40 -20 0 X: 1.262e-007 Y: -25.27 Time(s) A m p li tu d e (d B ) Test site 2

•

769~833MHz

；

•

1601 points

•

Blackman windows

+IFFT

Test site 3 Test site 1 9 0 0.5 1 1.5 2 2.5 3 x 10-6 -90 -80 -70 -60 -50 -40 -30 -20 -10 0 X: 8.008e-008 Y: -27.96 Time(s) A m p lit u d e (d B )

(10)

Test results – micro-reflection2

•

961~1025MHz

；

•

1601 points

•

Blackman window

+IFFT

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 x 10-6 -120 -100 -80 -60 -40 -20 0 Time(s) A m p lit u d e (d B ) 0 0.5 1 1.5 2 x 10-6 -120 -100 -80 -60 -40 -20 0 Time(s) A m p li tu d e (d B ) X: 7.886e-008 Y: -15.64 0 0.5 1 1.5 2 x 10-6 -80 -70 -60 -50 -40 -30 -20 -10 0 X: 6.406e-008 Y: -17.15 Time(s) A m p lit u d e (d B ) Test site 2 Test site 3 Test site 1 10

(11)

Some conclusions

• Bi-directional characteristics of passive coax networks are basically the same (S21=S12)

• Channel transmission loss is gradually stepping down under 1.5GHz, and

decreased rapidly at 1.5G ~ 1.8GHz according to the test results at test site 2.

• The channel response is quite flatten below 1GHz spectrum. But above 1GHz, the channel response has significant fluctuations.

• Parts of network exist strong micro-reflections

• According to test results , some views on passive coaxial cable network channel characteristics

– Multi path (micro reflection) do exist in passive cable network

• Poor connections between cable/splitter/tap can cause serious reflections.

• Splitter/tap port impedance unmatch/poor connections will cause frequency selective fading.

• Poor network consistency: channel performances differ significantly on different spectrum

• Cable channel changes slowly over time.