Very Large Scale Integration (VLSI) for Digital Signal Processing

对超大规模集成电路（VLSI）设计的这些方面进行了研究，特别是与数字信号处理相关的方面。本研究涉及在定制VLSI电路技术中实现重要的数字信号处理算法，如线性滤波和傅立叶变换。这些算法具有许多固有的结构规律性，本文的目标是利用这种规律性来提高吞吐量，使信号处理芯片的设计和测试更加简单和容易。在这项研究中，自动叶单元（在许多情况下使用相同的单元）优化，规则结构的整体性能和规则结构的测试正在进行中。这项工作的长期目标是减少算法结构，以最快速和有效的方式在芯片上实现。

Research is conducted on these aspects of very large scale integrated (VLSI) design that are especially relevant to digital signal processing. This research deals with the implementation of important digital signal processing algorithms, such as linear filtering and Fourier transformation, in custom VLSI circuit technology. These algorithms have a great deal of inherent structural regularity and it is the objective of this work to exploit this regularity to increase throughput and to make design and testing of signal processing chips straightforward and easier. In this investigation, automatic leaf-cell (use of same cell in many instances) optimization, global performance of regular structures, and testing of regular structures are being pursued. The long-range goal of this effort is to reduce the algorithmic structure to implementation on a chip in the most expeditious and efficient manner.