Computation power in sensor networks

hi
I need to do an analysis on computation power spent in sensor networks
(general architectures) for computation of certain mathematical
functions for certain number of times, in individual motes/over the
network. I need some good references. Any links? (i dont want those who
just discuss their computation constraints, rather, give some actual
data of power consumed per computation, e.g power consumed for
calculating log functions, X number of multiplication operations, etc
etc) I would appreciate ur input

saad wrote:
> hi
> I need to do an analysis on computation power spent in sensor networks
> (general architectures) for computation of certain mathematical
> functions for certain number of times, in individual motes/over the
> network. I need some good references. Any links? (i dont want those who
> just discuss their computation constraints, rather, give some actual
> data of power consumed per computation, e.g power consumed for
> calculating log functions, X number of multiplication operations, etc
> etc) I would appreciate ur input

I don't know much about the subject, but I can tell you that the term 
"general architectures" isn't specific enough. The power needed to run 
some DSP chips is less than that consumed by the fan in a Pentium. On 
one DSP board, the power is doubled by turning on the LED.

Computations take energy, not power. Halving the clock speed of any 
particular processor just about halves the power it draws, but the 
computation takes twice as long. Energy, the product of power and time, 
remains fairly constant.

Jerry
-- 
Engineering is the art of making what you want from things you can get.
�����������������������������������������������������������������������