Title :
A Non-Linear Neural D/A Converter for Direct Digital Frequency Synthesizers
Author :
Sadati, Nasser ; Sedighi, Behnam
Author_Institution :
Sharif Univ. of Technol., Tehran
Abstract :
In this paper, a new non-linear D/A converter (DAC) is presented. This D/A converter uses neural network to approximate a sine function. It is targeted for direct digital frequency synthesizers in which the phase to sine amplitude conversion is usually performed using a ROM. Utilizing the proposed DAC, the power hungry ROM is removed. A design example is presented and the performance of the DAC is compared to the conventional method. The impact of non-idealities on the performance of the system is also investigated.
Keywords :
digital-analogue conversion; direct digital synthesis; neural nets; ROM; direct digital frequency synthesizers; neural network; nonlinear D/A converter; Communication switching; Digital-to-frequency converters; Energy consumption; Frequency synthesizers; Image converters; Instruments; Neural networks; Phase locked loops; Read only memory; Table lookup;
Conference_Titel :
Neural Networks, 2006. IJCNN '06. International Joint Conference on
Conference_Location :
Vancouver, BC
Print_ISBN :
0-7803-9490-9
DOI :
10.1109/IJCNN.2006.247147
