signal is too weak ? Mixim + Veins

[Arslane Hamza Cherif]

Hi all,

What are the causes of this warning "Signal is too weak" in the output of the simulation ?

I know that is related to many factors including the TwoRayInterferance model, in order to understand the underlying cause, i have several hypothesis :

1- The distance between the two nodes equal or exceed the transmission range, so the signal is too weak to be correctly interpreted.

2- Channel switch of 802.11p (SCH vs CCH) occur juste before such warning, is it possible that the switch causes such thing ? (i read that channel switch causes time guard, where the channel is considered busy)

3- Collision of 2 packets, that are scheduled at the same time : 

with the CSMA/CA broadcast mode of 802.11p, air frames are directly send if the channel is detected as idle, so if two nodes decide to send air frames at the same time, collision can occur and produce "signal is too weak" warning 

Can someone confirm what i'm suggesting ? 

Many thanks,

[Christoph Sommer]

Arslane Hamza Cherif wrote:
> What are the causes of this warning "Signal is too weak" in the output
> of the simulation ?

which module library are you using?

Best,

Christoph



--
Dr. Christoph Sommer
Distributed Embedded Systems Group
University of Paderborn, Germany
http://www.ccs-labs.org/~sommer/

[Arslane Hamza Cherif]

What are you meaning by module library ?

Anyway, currently i'm using Veins 2.0-rc1 with these modules : 

• Mac80211p 
  
• My own version of Mac1609_4 
  
• Mode : brodacast mode (the only one supported by Veins 2.0-rc1, as you explain in the past)

For AnalogueModels :

• TwoRayInterferenceModel
  
• SimpleObstacleShadowing
  

For Decider :

• Decider80211p
  

[Christoph Sommer]

Arslane Hamza Cherif wrote:
>>> What are the causes of this warning "Signal is too weak" in the output of the simulation ?
>> which module library are you using?

> What are you meaning by module library ?
>
> Anyway, currently i'm using Veins 2.0-rc1

thank you.

You can find the source code that outputs "Signal is to weak" in line 52
of Decider80211p.cc (see [1]).

The line is reached when recvPower < sensitivity, that is, when the
signal has less power than was configured as the sensitivity of the
receiver.

Note that this check only deals with the signal strength, not the Signal
to Noise and Interference Ratio (SINR), so packet collisions do not
factor into this decision.

Best,

Christoph


[1]
https://github.com/sommer/veins/blob/veins-2.0-rc1/src/modules/phy/Decider80211p.cc#L52

[Arslane Hamza Cherif]

Thanks Christoph,

I already find this line and note that its check if recvPower < sensitivity, but i don't understand why the recPower is equal or near 0.

So according to what are you saying, it's not related to packet collision or SINR. The only factor is analogues models (TwoRayInterference & SimpleObstacleShadowing)

Now an other question arise : is analogues models simulating also the bit error?

[Christoph Sommer]

Arslane Hamza Cherif wrote:
> is analogues models simulating also the bit error?

This is best explained in the MiXiM technical documentation and the
following papers:

- MiXiM: the physical layer an architecture overview [1]
- Simulating wireless and mobile networks in OMNeT++ the MiXiM vision [2]


Best,

Christoph


[1] http://dl.acm.org/citation.cfm?id=1537713
[2] http://dl.acm.org/citation.cfm?id=1416302