|
|
 |
| |
|
|
|
Modified Analog-to-Information Conversion and Reconstruction for Approximately Sparse Signal |
|
|
|
PP: 261-265 |
|
|
|
Author(s) |
|
|
|
Tian Wen-Biao,
Rui Guo-Sheng,
Fu Zheng,
Liu Yu,
|
|
|
|
Abstract |
|
|
| The framework for analog-to-information conversion (AIC) as an alternative to conventional ADC is inspired by the recent
theory of Compressed Sensing (CS). Both of them require that the input signal has a sparse representation in some domain. But
mostly, the signals in our daily lives are approximately sparse ones. The modified analog-to-information conversion (MAIC) and a new
approximately sparse signal reconstruction algorithm are proposed. The contour of the input real-time streaming signal is pre-extracted
and the details are compressed, and then adaptive piece-wise basis pursuit (APBP) algorithm is used to reconstruct the input precisely.
The validity of the MAIC framework and APBP algorithm is demonstrated. As is shown in the simulation, the mean square error of
APBP reconstruction in the MAIC framework is of the order of about 10−14 |
|
|
|
|
|
|
|
