MiXiM 2.1 example WSNRouting transmission range and thermal noise

[jimena]

Hi,

I'm running convergecast simulations with only 2 nodes, and I find
that if I disable thermal noise, nodes are always in range, for
example at a distance of 350 m they still are in range, I assume
interference range, as all packets are received.
I left phy paremeters as default in the example, using:

**.node[*].nic.mac.txPower = ${txPower=0.1} mW

I thought thermal noise would affect BER.
How can I set the 100m range when disabling thermal noise?

Thanks,

[jimena]

I did some new tests using convergecast example, and I found that if I
set:

**.connectionManager.sat = -80dBm

I get an interference of 100 m, testing with 2 nodes.

Is this a valid configuration for connectionManager?

Thanks in advance,

On Sep 17, 4:24 am, jimena <jim...@gmail.com> wrote:
> Hi,
>
> I'm running convergecast simulations with only 2 nodes, and I find

> that if I disablethermalnoise, nodes are always in range, for

> example at a distance of 350 m they still are in range, I assume
> interference range, as all packets are received.
> I left phy paremeters as default in the example, using:
>
> **.node[*].nic.mac.txPower = ${txPower=0.1} mW
>

> I thoughtthermalnoisewould affect BER.

[Laura Marie Feeney]

Hi Jimena,

The connection manager parameters are used to define the maximum
distance over which a node's transmissions can cause interference.

It's basically a simulation filter: for a given receiver, nodes further
away than the distance defined by connection manager parameters are
assumed to "not exist". That is, even if they are transmitting at max
power, they are assumed to be unlikely to cause enough interference at
the receiver to affect its behavior.

The analog model and decider parameters (usually in config.xml) and the
nic.phy parameters (usually in omnetpp.ini) are the ones that actually
model radio behavior and determine the (combined) signal strength of
various transmissions and whether the frame is received correctly.

So, no, it is never correct to attempt to control the nic.phy behavior
using connection manager parameters. The connection manager parameters
should always be more "permissive" than the nic.phy ones.

It would be VERY nice if MiXiM were able to make a sanity check of the
two sets of configuration parameters. I think the variety of analog
models makes this rather complicated, but perhaps it is something that
the MiXiM developers could discuss.

Laura

[jimena]

Thanks Laura,

I don't quite understand why is kind of difficult to control the
transmission range, having a mote datasheet that states a range of
75-100m.
http://www.memsic.com/support/documentation/wireless-sensor-networks/category/7-datasheets.html?download=148%3Amicaz

I'm still trying the example WSNRouting, which uses
Decider802154Narrow , BreakpointPathlossModel and LogNormalShadowing
analogue models

I've tried to adjust the range with nic.phy parameters now,
**.node[*].nic.phy.sensitivity <- doesn't seem to affect anything

And with txPower 0.1 mW packets are received at a range of 250 m.

If I disable thermalNoise, packets are received at the interference
range.
Is this the expected behavior?


In BaseDecider.cc I see frames are discarded if recvPower <
sensitivity. But I can't figure out if the same happens with
Decider802154Narrow.

Any suggestions will be highgly appreciated.

Thanks,