OSBRiDGE 5GXt deployment, Case Scenario 2

The devices were deployed on a 6 kilometers distance. One of the devices is deployed on a 40 meters tall tower, the other one is deployed on the rooftop of a building, 3 meters above the roof. There actually is a line of sight between these two points but fresnel zone is severely obstructed, which seriously impacts performance in one direction - compared to the performance (H-Polarization) in the other direction.

Pictures of the instalation on the tower:

Pictures of the installation on the rooftop, on second side of our link:

Here we can see this big fat builiding that is breaking our v-polarized fresnel zone

Here we can see this big builiding that is breaking our V-polarized Fresnel Zone


For testing purposes we have configured devices to operate using 40MHz channel widths.

System Information screen from the first device (OSBRiDGE 5GXt):

and System Information screen from the second device (OSBRiDGE 5GXt-HP):

Iperf was used to test the link throughput, first using TCP/IP: 

# iperf -c 192.168.20.5 -i 5 -t 30 -r
------------------------------------------------------------
Server listening on TCP port 5001
TCP window size: 85.3 KByte (default)
------------------------------------------------------------
------------------------------------------------------------
Client connecting to 192.168.20.5, TCP port 5001
TCP window size: 16.0 KByte (default)
------------------------------------------------------------
[  6] local 192.168.20.1 port 32818 connected with 192.168.20.5 port 5001
[  6]  0.0- 5.0 sec  14.7 MBytes  24.7 Mbits/sec
[  6]  5.0-10.0 sec  25.2 MBytes  42.3 Mbits/sec
[  6] 10.0-15.0 sec  26.4 MBytes  44.3 Mbits/sec
[  6] 15.0-20.0 sec  26.3 MBytes  44.1 Mbits/sec
[  6] 20.0-25.0 sec  26.5 MBytes  44.5 Mbits/sec
[  6] 25.0-30.0 sec  26.3 MBytes  44.2 Mbits/sec
[  6]  0.0-30.0 sec    145 MBytes  40.7 Mbits/sec
[  6] local 192.168.20.1 port 5001 connected with 192.168.20.5 port 1143
[  6]  0.0- 5.0 sec  32.5 MBytes  54.5 Mbits/sec
[  6]  5.0-10.0 sec  34.1 MBytes  57.1 Mbits/sec
[  6] 10.0-15.0 sec  34.5 MBytes  57.9 Mbits/sec
[  6] 15.0-20.0 sec  33.1 MBytes  55.5 Mbits/sec
[  6] 20.0-25.0 sec  34.5 MBytes  57.9 Mbits/sec
[  6] 25.0-30.0 sec  33.5 MBytes  56.3 Mbits/sec
[  6]  0.0-30.0 sec    202 MBytes  56.5 Mbits/sec

firewall: -root-
#

As you can see the link was capable of running 44 Mbps real TCP/IP traffic in one direction and 56 Mbps real TCP/IP traffic in the other direction.

The full duplex test follows: 

# iperf -c 192.168.20.5 -i 5 -t 30 -d
------------------------------------------------------------
Server listening on TCP port 5001
TCP window size: 85.3 KByte (default)
------------------------------------------------------------
------------------------------------------------------------
Client connecting to 192.168.20.5, TCP port 5001
TCP window size: 16.0 KByte (default)
------------------------------------------------------------
[  6] local 192.168.20.1 port 32821 connected with 192.168.20.5 port 5001
[  7] local 192.168.20.1 port 5001 connected with 192.168.20.5 port 1147
[  6]  0.0- 5.0 sec  12.6 MBytes  21.1 Mbits/sec
[  7]  0.0- 5.0 sec  20.4 MBytes  34.3 Mbits/sec
[  6]  5.0-10.0 sec  10.5 MBytes  17.7 Mbits/sec
[  7]  5.0-10.0 sec  18.5 MBytes  31.0 Mbits/sec
[  6] 10.0-15.0 sec  10.3 MBytes  17.4 Mbits/sec
[  7] 10.0-15.0 sec  15.8 MBytes  26.5 Mbits/sec
[  6] 15.0-20.0 sec  9.25 MBytes  15.5 Mbits/sec
[  7] 15.0-20.0 sec  18.4 MBytes  30.9 Mbits/sec
[  6] 20.0-25.0 sec  10.3 MBytes  17.2 Mbits/sec
[  7] 20.0-25.0 sec  17.0 MBytes  28.6 Mbits/sec
[  7] 25.0-30.0 sec  15.5 MBytes  25.9 Mbits/sec
[  6] 25.0-30.0 sec  10.3 MBytes  17.3 Mbits/sec
[  6]  0.0-30.0 sec  63.3 MBytes  17.7 Mbits/sec
[  7]  0.0-30.0 sec    106 MBytes  29.5 Mbits/sec

firewall: -root-
#

Here we can see first side effects of vertical polarization fresnel zone being obstructed. However we still were able to achieve 18/29 Mbps full duplex real TCP/IP traffic.

UDP throughput test was performed too: 

# iperf -c 192.168.20.5 -i 5 -t 30 -r -u -b 100M 
------------------------------------------------------------
Server listening on UDP port 5001
Receiving 1470 byte datagrams
UDP buffer size:   105 KByte (default)
------------------------------------------------------------
------------------------------------------------------------
Client connecting to 192.168.20.5, UDP port 5001
Sending 1470 byte datagrams
UDP buffer size:   105 KByte (default)
------------------------------------------------------------
[  6] local 192.168.20.1 port 32768 connected with 192.168.20.5 port 5001
[  6]  0.0- 5.0 sec  56.1 MBytes  94.1 Mbits/sec
[  6]  5.0-10.0 sec  56.1 MBytes  94.1 Mbits/sec
[  6] 10.0-15.0 sec  56.0 MBytes  94.0 Mbits/sec
[  6] 15.0-20.0 sec  55.9 MBytes  93.7 Mbits/sec
[  6] 20.0-25.0 sec  55.8 MBytes  93.7 Mbits/sec
[  6]  0.0-30.0 sec    336 MBytes  93.9 Mbits/sec
[  6] Sent 239515 datagrams
[  6] Server Report:
[  6]  0.0-30.3 sec  76.5 MBytes  21.2 Mbits/sec  0.475 ms 184911/239505 (77%)
[  6]  0.0-30.3 sec  1 datagrams received out-of-order
[  5] local 192.168.20.1 port 5001 connected with 192.168.20.5 port 1024
[  5]  0.0- 5.0 sec  35.8 MBytes  60.0 Mbits/sec  0.166 ms 13867/39393 (35%)
[  5]  5.0-10.0 sec  36.3 MBytes  60.9 Mbits/sec  0.093 ms 14751/40630 (36%)
[  5] 10.0-15.0 sec  35.3 MBytes  59.2 Mbits/sec  0.147 ms 15335/40526 (38%)
[  5] 15.0-20.0 sec  34.1 MBytes  57.3 Mbits/sec  0.125 ms 16359/40713 (40%)
[  5] 20.0-25.0 sec  36.3 MBytes  60.8 Mbits/sec  0.145 ms 14766/40637 (36%)
[  5] 25.0-30.0 sec  35.9 MBytes  60.3 Mbits/sec  0.141 ms 14885/40506 (37%)
[  5]  0.0-30.2 sec    215 MBytes  59.6 Mbits/sec  5.149 ms 90550/243803 (37%)
[  5]  0.0-30.2 sec  1 datagrams received out-of-order

firewall: -root-
#

Results: 21.2 Mbps (Fresnel Zone!) in one direction and 59.6 Mbps in the other direction UDP throughput.

Full Duplex UDP test: 

# iperf -c 192.168.20.5 -i 5 -t 25 -d -u -b 70M  
------------------------------------------------------------
Server listening on UDP port 5001
Receiving 1470 byte datagrams
UDP buffer size:   105 KByte (default)
------------------------------------------------------------
------------------------------------------------------------
Client connecting to 192.168.20.5, UDP port 5001
Sending 1470 byte datagrams
UDP buffer size:   105 KByte (default)
------------------------------------------------------------
[  6] local 192.168.20.1 port 32768 connected with 192.168.20.5 port 5001
[  5] local 192.168.20.1 port 5001 connected with 192.168.20.5 port 1024
[  6]  0.0- 5.0 sec  41.7 MBytes  70.0 Mbits/sec
[  5]  0.0- 5.0 sec  18.1 MBytes  30.3 Mbits/sec  0.132 ms 16121/29008 (56%)
[  6]  5.0-10.0 sec  41.7 MBytes  70.0 Mbits/sec
[  5]  5.0-10.0 sec  33.0 MBytes  55.4 Mbits/sec  0.103 ms 6200/29747 (21%)
[  6] 10.0-15.0 sec  41.7 MBytes  70.0 Mbits/sec
[  5] 10.0-15.0 sec  33.0 MBytes  55.3 Mbits/sec  0.154 ms 6233/29761 (21%)
[  6] 15.0-20.0 sec  41.7 MBytes  70.0 Mbits/sec
[  5] 15.0-20.0 sec  33.1 MBytes  55.6 Mbits/sec  0.159 ms 6139/29768 (21%)
[  6]  0.0-25.0 sec    209 MBytes  70.0 Mbits/sec
[  6] Sent 148810 datagrams
[  5] 20.0-25.0 sec  32.6 MBytes  54.6 Mbits/sec  0.162 ms 6373/29595 (22%)
[  5]  0.0-25.2 sec    151 MBytes  50.1 Mbits/sec  2.909 ms 41361/148806 (28%)
[  5]  0.0-25.2 sec  1 datagrams received out-of-order
[  6] Server Report:
[  6]  0.0-25.3 sec  71.0 MBytes  23.5 Mbits/sec  0.469 ms 98186/148809 (66%)
[  6]  0.0-25.3 sec  1 datagrams received out-of-order

firewall: -root-
#

Results: 55/23 Mbps full duplex UDP throughput. And again, performance of one of the polarizations is affected by obstructions on the path though the results are quite good.

Please note that all these tests are performed using uncompressable data, therefore they are pretty close to the real life traffic. This has nothing to do with so-called 'test' traffic which is highly compressable, may give better visual effects, but is hardly ever spotted in real-life deployment scenarios.

Retrieved from "http://www.osbridge.com/wiki/index.php/Main_Page/5GXt_FAQ/5GXt_deployment_chojnice2"
Last modified October 31, 2007 1:49 am