Digital Assistance for Nanoscale Analog Circuits

在过去的40年里，发生了一场众所周知的数字革命，对我们的生活、工作和娱乐方式产生了巨大的影响。随着数字晶体管如摩尔定律所预测的那样缩小，集成电路和系统一年比一年在相同的计算中继续变得成本更低、速度更快、功耗更低。不太为人所知的是，在这40年中，也发生了一场模拟革命，这场革命对提高电子系统的性能至关重要。这些模拟集成电路充当了数字世界和物理世界之间的接口，如果没有模拟电路的性能改进，我们所看到的技术革命就不会发生。最近对模拟电路发展的回顾表明，在过去的10年中，性能已经提高了60倍。虽然这一改进的很大一部分(10倍)是由于晶体管更小、更快，但也有很大一部分(6倍)是由于创造性的电路创新，这是这项研究拨款的重点。 考虑到上述情况，出现了一些问题。给定一个技术节点，模拟电路距离可实现的最佳性能有多近？我们能继续改进模拟电路吗？如何最好地实现这一点？随着数字电路的巨大改进，我们能否利用它们缩小的体积/能量来帮助模拟电路的性能？这些问题推动了模拟集成电路领域的研究。 这项研究计划的目的是使模拟电路在应用于MEMS接口电路、射频前端、用于串行数字通信的高速信令等众多应用时具有更高的功率效率和更高的性能。挑战是开发新的电路、体系结构和理论来改进模拟电路，并通过实验结果证明可以获得的实际优势。

During the past 40 years, there has been a well-known digital revolution that has made a tremendous impact in how we live, work and play. With digital transistors shrinking as predicted by Moore's law, year over year integrated circuits and systems have continued to become lower in cost, higher in speed and lower in power dissipation for the same computation. What is less well known is that during these 40 years, there has also been an analog revolution that has been critical in improving the performance of electronic systems. These analog integrated circuits serve as the interface between the digital world and the physical world and without performance improvements in the analog circuits, the technology revolution we have seen would not have occurred. A recent review of analog circuit advancements has shown that over the past 10 years, there has been a 60 times improvement in performance. While much of this improvement (10 times) is due to transistors being smaller and faster, a large factor (6 times) has been due to creative circuit innovations which is the focus of this research grant. With the above in mind, a number of questions arise. Given a technology node, how close are analog circuits to the best performance achievable? Can we continue to improve analog circuits and how is this best achieved? With the tremendous improvement in digital circuits, can we make use of their decreased size/energy to aid the performance of analog circuits? These questions drive this research in the area of analog integrated circuits. This research program is driven by the desire to enable analog circuits to be more power efficient and have higher performance when applied to numerous applications such as MEMS interface circuits, RF front-ends, high-speed signaling for serial digital communications and others. The challenge is to develop new circuits, architectures and theory that improve analog circuits and through experimental results, demonstrate the practical advantages that can be obtained.