Analyzing noise robustness of wide fan-in dynamic logic gates under process variations
Abstract:
Wide fan-in dynamic logic gates are difficult to design due to the large number of leaky evaluation paths connected to the dynamic node. Designers have to cope with their low noise tolerance further worsened by the effects of process parameter variation. In this paper, a novel analytical model is derived and validated to evaluate the noise robustness of wide fan-in dynamic logic gates taking process variation effects into account. Experiments were performed using a commercial 45-nm 1-V CMOS technology, and the noise robustness in terms of unity noise gain (UNG) was evaluated for 16 and 32-bit OR gates. Obtained results demonstrate that the proposed model is able to pbkp_redict the mean value of the UNG with a maximum error of only 6.8%, whereas the difference between the pbkp_redicted and simulated UNG yield is always lower than five percentage points. Copyright © 2012 John Wiley & Sons, Ltd. Wide fan-in dynamic logic gates are difficult to design due to the large number of leaky evaluation paths connected to the dynamic node. Designers have to cope with their low noise tolerance further worsened by the effects of process parameter variation. In this paper, a novel analytical model is derived and validated to evaluate the noise robustness of wide fan-in dynamic logic gates taking process variation effects into account. Experiments were performed using a commercial 45-nm 1-V CMOS technology, and the noise robustness in terms of unity noise gain (UNG) were evaluated for 16 and 32-bit OR gates. Obtained results demonstrate that the proposed model is able to pbkp_redict the mean value of the UNG with a maximum error of only 6.8%, whereas the difference between the pbkp_redicted and simulated UNG yield is always lower than five percentage points. © 2012 John Wiley & Sons, Ltd.
Año de publicación:
2014
Keywords:
- process variations
- Dynamic logic
- wide fan-in
Fuente:
Tipo de documento:
Article
Estado:
Acceso restringido
Áreas de conocimiento:
- Ingeniería electrónica
- Arquitectura de computadoras
Áreas temáticas:
- Ciencias de la computación