On 6/19/08, Henning Rogge <<a href="mailto:rogge@fgan.de">rogge@fgan.de</a>> wrote:<br><br>>I discovered that if I calculate the LQ values for each timeslice, I get bad<br>>results because some slices don't have enough packages.<br>
>that was the reason for me to think of larger time slots,..<br><br>But how is the detection of lost packets, now?<br>When do you count how much packets as lost, when not receiving any packets?<br><br>> The linkquality value is calculated by<br>
>LQ = sum (received[1..7]) / ( sum (received[1..7]) + sum (lost[1..7]) )<br><br>this formula wonn`t react fast on dead links,.. )-;<br>if not counting more than the lost hellos, a communicative link will neeed nearly full history duration to get an bad ETX if the link dies (gets totally disconnected)<br>
<br>The old approach gave each time slot the same weight, which was better for detection of dead links(as long as there are at least some packets per time slot) <br>now it`s again based on the number of packets, which i dislike,.. <br>
<br>but the history has still fixed duration, which is good (-;<br><br>Markus<br>