Antenna direction of a wifi transmitter

[sai]

Hi every body,

 I am new to ns3, i am working on opportunistic routing protocol, I have few queries regarding the range of a wifi node,

>Is the range depends on direction of the antenna  
 
>Is there any antenna direction setting mechanism in ns3, in the given propagation loss models in ns3, rss strength is not same in all directions(x,y and z), with equal distance but in reality the radio range is same in all directions.

> Is the default antenna direction is unidirectional

please help,     Thanks in advance.

[Konstantinos]

Hi Sai,

On Thursday, June 6, 2013 11:45:47 AM UTC+1, sai wrote:

Hi every body,

 I am new to ns3, i am working on opportunistic routing protocol, I have few queries regarding the range of a wifi node,

>Is the range depends on direction of the antenna  
 

For the directional antennas, yes range depends on the direction

 

>Is there any antenna direction setting mechanism in ns3, in the given propagation loss models in ns3, rss strength is not same in all directions(x,y and z), with equal distance but in reality the radio range is same in all directions.

http://www.nsnam.org/docs/models/html/antenna.html

The antenna moduled can be used with all the wireless technologies and physical layer models that support it. Currently, this includes the physical layer models based on the SpectrumPhy. Please refer to the documentation of each of these models for details.

 

> Is the default antenna direction is unidirectional

WiFi model in NS-3 currently works ONLY with omni antennas (unidirectional) because its PHY model is not based on the SpectrumPhy so the antenna model can't be used. But it could be possible to modify this.

[sai]

Thanks Konstantinos, I have gone thru the documents suggested by you, but i am unable to know
 
> Which Antenna model among the given

Cosine

Isotropic

Parabolic

  correctly models a real wifi node that is its range should be same in all directions for a given transmission power.


>If YansWifiPhy cant be used to set Antenna model in place of it which can be used that is which subclass of SpectrumPhy can be used among

HalfDuplexIdeal

LteSpectrum

RemSpectrum

SpectrumAnalyzer

WaveFormGenerator

Can you please suggest me the solution for the above two


Thanks in advance  

On Thursday, June 6, 2013 5:44:08 PM UTC+5:30, Konstantinos wrote:

Hi Sai,
t

[Konstantinos]

On Thursday, 6 June 2013 16:28:42 UTC+1, sai wrote:

Thanks Konstantinos, I have gone thru the documents suggested by you, but i am unable to know
 
> Which Antenna model among the given

Cosine

Isotropic

Parabolic

  correctly models a real wifi node that is its range should be same in all directions for a given transmission power.

If you want to have the same range in ALL directions, then you do not need that antenna model. 

All the propagation models are depended in the distance (and other parameters specified by each model) but are independent of direction. 

http://www.nsnam.org/docs/doxygen/group__propagation.html

 

>If YansWifiPhy cant be used to set Antenna model in place of it which can be used that is which subclass of SpectrumPhy can be used among

HalfDuplexIdeal

LteSpectrum

RemSpectrum

SpectrumAnalyzer

WaveFormGenerator

All the above can be used with the Antenna module, but none of the above can be used with WiFi.

[sai]

 
 Thanks Konstantinos, Actually i was confused with x,y and z dimensions thats why i was looking for antenna models.Now it is clear that i can use the propagation models with out any antenna model support, thanks for your help konstantinos.

[Adri Ant]

Hello Konstantinos, first of all I wish you a happy new year, and sorry for refloat an old topic.

On Thursday, June 6, 2013 14:14:08 UTC+2, Konstantinos wrot

WiFi model in NS-3 currently works ONLY with omni antennas (unidirectional) because its PHY model is not based on the SpectrumPhy so the antenna model can't be used. But it could be possible to modify this.

 

Could you explain a bit how could be possible to modify the explained above?

I guess that the current wifi waveform is isotropic, i mean, omnidirectional and perfectly spherical. I'm wrong? If the current wifi has a dipole waveform then there is no problem (but i'm not been able to find this information and verify it).

In order to achieve one of my objectives of my final's degree project i need to make the simulation scenario of an opportunistic network more realistic than the current wifi antenna model (if for current wifi antenna model means an isotropic waveform). 

Thanks in advice.

Adrian.