I have designed a TDMA protocol to reduce the voice spurts between voice calls in SDRs.

The protocol runs in this manner: control cycles proceeded with data cycles and again control cycles. Control cycles - data cycles - control cycles - data cycles ----

If each control cycle is consist of 12 slots and slot size is 2.24ms then 5 control cycles takes 134.4 ms before the data cycle.

134.4 ms - data cycle - 134.4 ms - data cycle - 134.4 ms - data cycle

I am confused about the term i am using for this time duration of 134.4 ms.

  • Is it jitter or buffering delay?
  • or What is more appropriate term for it?
  • and should i need to consider the delay the source node takes in finding the destination node?
  • and what does this delay call?

Because control slots depend on the node id.


Is it jitter or buffering delay?

Neither. It's a control cycle duration.

Jitter is a random variation in the period of a nominally perfectly periodic system.

Buffering delay is the delay due to buffering. There's no buffering of data in your case, i.e. your data isn't delayed by buffering.

Don't just use terms you come across! These two would've been easy to google!

and should i need to consider the delay the source node takes in finding the destination node?

You use the word "node", which strongly implies we're talking about a network that is not a trivial graph (i.e. not just a chain, a single cycle or an isolated single device).

This question makes no sense at this layer. You're on the Physical medium access layer. The question is a routing question. I don't think there's a term for what you need aside from "time it takes to find a route", it won't be constant, it depends on the size of your net and the algorithm with which you search.

Also, it's extremely uncommon ina real-world netowrks for a source node to "know" the full route to its target nodes. Instead, it typically figures out the next hop (for example, such nodes can be called "gateway", if they lead to access to a group of non-directly connected nodes), and just sends the frame on its way, and the nodes on the way figure out individual next hops themselves.
This is necessary because in most cases, you're either in a scenario where a) routes aren't static (typical example: the internet), or b) routes shouldn't be transparent to the node, as the overhead of communicating the whole route info would put a great overhead on the system without any benefits (e.g. non-IP-based telephony backhaul systems).

  • $\begingroup$ Thank you so much Sir. You have elaborated it so well. Yes sir my node contain SDR and they made a mobile ad hoc network. $\endgroup$
    – MentorGeek
    Nov 23 '17 at 5:32
  • $\begingroup$ If these nodes make such a topology that node[0] one hop neighbor is node[7] and two hop neighbor is node[2]. And i scheduled the control slots in a such a way that node[7] speaks after 20 TDMA slots. Then node[0] gets information about its two hop neighbor node[2]. What should i call this delay? Control slots scheduling delay or routing delay? $\endgroup$
    – MentorGeek
    Nov 23 '17 at 5:36
  • $\begingroup$ Sigh. You wouldn't call it anything. You would find words that describe what it is. See my answer. $\endgroup$ Nov 24 '17 at 8:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.