[net2-wg] link estimator interface

[Omprakash Gnawali]

> > What do people think about abstracting the notion of link quality and
> > saying here is the range of values and the values in between describe
> > link quality in linear proportion of the range? For example,
> > 1/2(max+min) would describe link quality roughly in between the
> > maximum and minimum quality.
> 
> Can you provide a motivating example where this would be needed?


If I write a routing protocol that uses a certain threshold to switch
from one path to another, then we would need some standard on what a
change of 2 means. For, ETX, 1.5 is a huge change but for a
reliability estimator that returns an estimate in the range [1,100], 2
is 2%. If there is a standard, I would not have to configure my
routing protocol when I go from one estimator to the next.


> > Without standardizing the return value, I do not see how different
> > routing protocols can work with a variety of link estimators. The
> > drawback is this will make it harder for protocols that rely on a
> > specific implementation of an estimator - protocols that configure the
> > number of retransmission based on ETX, for example.
> 
> Not true.  If it is "standardized" to ETX, then the value is
> irrelevant--all that matters is that it provides a relative ETX
> metric.
> 
> Certainly there are other routing metrics besides ETX, but the
> question arises--are they metrics that soley the link can determine on
> its own?
> 
> And just to be completely accurate--the link estimator in Boomerang by
> Moteiv is a function of the raw LQI value, not the raw LQI value
> itself.


Yes and this is the approach in the LQI estimator in TinyOS too but I
am not aware of a way to derive ETX from LQI or a function of LQI. I
have seen some statistics but they seem to be valid for some specific
settings.