New Asynchronous Pipeline Implementation Method Based on Conventional Function Units

We present a new method to implement QDI asynchronous pipeline stages. Previous implementations of these pipelines were based on the pre-charge logic implementation which imposes some limitations on the size of the logic cell due to the stacking problem. To overcome this limitation, we propose a novel method that uses standard single-rail computational circuits in a dual rail handshake environment. Our method does not have any stacking problem and is not limited in size. The results show that a 4-input lookup table implemented by this method outperforms a previous 3- input lookup table by 16% in speed and 29% in power with a negligible area overhead.

Keywords:

Asynchronous pipelines , pre-charge logic , PCHB