深亚微米条件下的多层VLSI／MCM层分配的遗传算法

本文提出了一个深亚微米条件下的多层ＶＬＳＭＣＭ有约束分层层分配的遗传算法。该算法分为两步：首先进行超层分配，使各线网满足Ｃｒｏｓｓｔａｌｋ约束，且超层数目最少；然后进行各超层的通孔最少化二分层。与目前的层分配算法相比，该遗传算法具有目标全面，全局优化能力强等特点，是一种可应用于深亚微米条件下的ＩＣ ＣＡＤ的有效分层方法。

Deep Sub-micro-meter Multilayer VLSI/MCM Layer Assignment Based on Two Genetic Algorithms

This paper presents an algorithm for deep sub- micron- meter multi-layer VLSI/ MCM layer assignment based on two genetic algorithms(GAs). The first GA, which is called LAGA, is used to assign each net to a superlayer (X-Y plane pair) as to obtain minimal crosstalk and the least number of layers. The second GA, called VAGA, can be employed to design the layout on each superlayer with minimized vias. Compared to traditional algorithms, these GAs are more powerful in global optimization with multiple-objective function. They are efficient for the CAD of deep sub- micron-meter ICs.