InterJournal Complex Systems, 40
Status: Accepted
Manuscript Number: [40]
Submission Date: 963011
Comparing the bit-energy of continuous and discrete signal representations
Author(s): Paul Furth , Andreas Andreou

Subject(s): CX.07

Category: Brief Article

Abstract:

A canonical model of a communication channel is employed to quantify the performance of physical computational systems. Using information theory we investigate the efficiency as measured in bit-energy (joules/bit) of continuous/discrete-value and continuous/discrete-time circuits. We employ physical models of noise and power dissipation in the circuit elements in order to derive the signal-to-noise ratio, average power consumption, and channel capacity given a constraint on the maximum range of the processed signals. Results of this comparison indicate that analog systems outperform digital systems only at very low signal-to-noise ratios.

Retrieve Manuscript
Submit referee report/comment


Public Comments: