Loading ...
Sorry, an error occurred while loading the content.

16795Re: [tracker2] 1-persistence

Expand Messages
  • owenvk
    Aug 18, 2014
      Just to reply to the 1-persistence thing.The way p persistence works is you specifiy probability p, and when the TNC wants to tx it waits for /CD, calculates a pseudo random number from 0 to 1, and if the number is less than p, then it transmits, othewise it waits for slottime and does it all again.If you specify p=1, then it always transmits immediately there is /CD... so 1-persistence is describing no wait (other than for /CD).So if there are several digis that are hearing and will repeat a mobile with 100% probability as sood as /CD, you have an effective 1-persistence. It might seem an obscure way of naming it, but I didn't invent it!So, if a two digis used p=25% and they are both going to repeat a mobile's packet, there is a 6% chance on that they will collide. One of them will transmit in this or a following slot, but it remains unlikely they both tx simultaneously in which case the other one now waits for /CD and goes through the routing again, eventually transmitting.None of this discussion above mentioned whether they are UI frames or session frames, I cannot see how connectionless packets work any differently in terms of multiple access to the channel.Owen
    • Show all 9 messages in this topic