Re: [ns] Energy Calculation

[John Heidemann <johnh@ISI.EDU>]

>see http://nesl.ee.ucla.edu/projects/sensorsim/ for
>their implementation and papers on energy model.

That's useful, but that's not the energy model in ns-2.1b8.

>
>- difa
>
>On Thu, 15 Nov 2001, Nevo Transian wrote:
>
>> Hi all,
>>
>> 2 questions:
>>
>> I was trying to do a very simple script, two wireless nodes sending some
>> data with UDP at Constant Bit Rate. I wanted to calculate the amount of
>> energy they are spending in this comunication, but it doesn't seem to be
>> decreased. Could anybody help me to know why it is not decreasing? (I attach
>> my script)
>>
>> Moreover, the Energy Model is very simple, one node is only spending energy
>> when a node is sending or recieving, but radios are also spending energy in
>> idle state and when are changing the state from idle to send or recieve.
>> Does anybody have any more sofisticate model?

Ns should support both send/receive costs and idle (listen) costs.
We've used it that way for several papers (for example, the Xu et al
Mobicom '01 paper).  That should support energy consumption regardless
of traffic type, and it definitely supports idle/send/receieve costs.

Can you fill out a bug report at 
<http://www.isi.edu/cgi-bin/nsnam/reportbug.cgi>?
It sounds like you're either using an old version of ns or something
strange is happening.

   -John Heidemann