Constant Rate Wifi Manager

[Lukas von Rotz]

Hi all..

In my wifi-test-environment I use the ConstantRateWifiManager to set the datarate of my connection as follows:

wifi.SetRemoteStationManager ("ns3::ConstantRateWifiManager", "DataMode", StringValue ("ErpOfdmRate54MBps"), "ControlMode", StringValue ("ErpOfdmRate54MBps")); where "wifi" is my wifiHelper with 80211g Standard.

I noticed that if I vary the datarate, then the range of my connection changes too. So with 54Mbps I havent connection with a distance of 25m and more. If i set the datarate to 11MBps, the boundary of the connection is at 75m. Has this something to do with different attenuations/dampings?

Many thanks for your help!
Lukas

[Tommaso Pecorella]

Hi,

the more two nodes are far away, the lower the SINR is. Slower transmission rates are more "robust" against low SINR.

T.

[Lukas von Rotz]

Hey Tommaso

Many thanks!
But with 802.11g we are always in the same frequency band (2,400 bis 2,4835 GHz), regardless of the different datarates. Its clear that with a lower frequency we have a wider range, but here we are always in the same frequency band, so I'm a bit confused why these datarates have that much influence..
And I'm wondering me about following: If I set the Datarate to 54 Mbps I have a distance of 25m, like mentioned above. But when I don't set anything (I don't write the line above with the ConstantRateWifiManager), then the connection is standardly set to 54Mbps (so the same datarate), but in that case the distance is 175m. Can you imagine why this big difference comes about?

Lukas

[Tommaso Pecorella]

Hi Lukas,

you need a good book to explain you all these things. WiFi spectrum is divided in channels and each one is about 20 or 40 MHz wide, depending on the particular standard.

The datarate depends on the modulation and coding, which has little or nothing to do with the channel width. Well, there's always the Shannon theorem, but that's another story.

About the last question, if you don't lock the datarate (using ConstantRateWifiManager), the dynamic one will be used. The datarate will change depending on the nodes distance, so the range is larger (and the speed is variable).

Hope this helps,

T

[Lukas von Rotz]

Hi Tommaso

I have one last question to that topic.
Did I understand right, if I would use a mobility model where nodes getting closer to each other, the datarate changes during this movement?

Greets
Lukas

[Tommaso Pecorella]

Hi,

please, read this: http://cutebugs.net/files/mswim04.pdf

Thanks,

T.