Hi Swetha,

In both the cases, you can not ensure the packets to follow the same path
in both the directions, so you can not ensure that the time delay you had
is accurate by just divinding the round trip time by 2 and that too in only
one trial.

So you can repeat the process multiple times and take the mean of the
calculated packet delays which should be much better.

As you mentioned the time synchronization, there will be many IEEE papers
available on the time synchronization (ex. TPSN, Time Sync) which may give
you some ideas to calculate the packet delays.

Please let me know if you find any good approaches for the same. :)

Thanks,
Peeyush


On Sun, Feb 3, 2013 at 9:05 PM, Swetha <[email protected]> wrote:

> hello all,
>
>  I am using tinyos 2.1.2 and telosB motes for multihop routing process.
> In order to calculate packet delay in multi-hop transmission, I came up
> with two ways to calculate,
>
>  1- booting all the nodes at same time so that it would provide sort of
> time synchronisation,
>   so by calculating the difference between send time  and received time
> will provide packet delay.
>
>  2- rather than time synchronisation, round trip delay [source to
> destination and destination to source(ACK)], by calculating the difference
> between send time and received time (both the timings are taken at the
> source) will provide round trip delay and by dividing with 2 it gives
> packet delay.
>
>
>  Is this good approach to calculate packet delay or is there any better
> approach?
>
> Swetha
> [email protected]
>
> _______________________________________________
> Tinyos-help mailing list
> [email protected]
> https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
>
_______________________________________________
Tinyos-help mailing list
[email protected]
https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help

Reply via email to