Log Distance Propagation Loss Model

[Caleb Lo]

Hi everybody,
I've been experimenting with the log distance propagation loss model,
and I'm getting a strange result that I hope somebody can clarify for
me.

In particular, I set up a simple example with two nodes (each one uses
the YansWifiPhy) that are exactly 2m apart, and I've used the log
distance propagation loss model. I've kept the default settings for
the YansWifiPhy and the loss model. In this case, the default
transmission power is displayed as 17.0206dBm, which is correct given
a default Tx gain of 1.0dB. The reference attenuation is displayed as
-46.6777dB, which is correct. The reference attenuation is displayed
as -55.7086dB, which is correct given the reference attenuation, the
default path loss exponent of 3, and the inter-node distance of 2m.

This should yield a received power of 17.0206dBm-55.7086dB =
-38.688dBm, yet I'm seeing that the rxPower for the YansWifiChannel's
Send() function is displayed as -94.3966dBm. I'm not sure if this is
a) a bug in ns3 itself, b) a bug in my particular implementation or c)
a flaw in my understanding.

Thank you very much.

Caleb Lo

[Caleb Lo]

Hi everybody,
I did some further investigation, and it turns out that the
YansWifiChannelHelper has the log-distance propagation loss model
already installed by default. In my baseline simulation, I had added
a log-distance loss model, which yielded a "cascaded channel" effect;
thus, I was getting a path loss of -38.688dBm from the default loss
model, and then the loss model that I added yielded a total path loss
of -38.688dBm-55.7086dB = -94.3966dBdBm. When I removed the loss
model from my baseline simulation and only used the default model in
the YansWifiChannelHelper, I obtained the expected path loss of
-38.688dBm. Sorry for the inconvenience.

This does raise another question for me: would it be possible to
disable the default log-distance loss model in the
YansWifiChannelHelper and use a different loss model as the "base loss
model" in my simulation? For example, I'd like to use the three-log-
distance loss model as my "base loss model" and then layer the
Nakagami fading model on top of it. As of now, it seems that by using
the YansWifiChannelHelper, I have to use the default log-distance
model as a part of any compound channel model that I devise.

Thank you very much.

Caleb Lo