一种定性控制的电路布局优化方法

A Novel Placement Algorithm Based on Qualitative Control Strategy

以总线长为优化目标的优化方法是电路布局优化的重要分支,模拟退火是常用的迭代优化方法.关注优化过程中布局的定性表现,提出一种定性控制的线长优化方案.云模型是一种定性定量转换模型,使用二维云模型可以对元件的布局位置进行建模.迭代过程中,制定布局收缩、布局扩展和布局重置三种策略,通过调整云模型的熵和超熵达到定性控制的目的.使用标准测试电路与模拟退火方法相比,本问题提出的方法能获得更好的优化效果.

Wire length driven placement algorithms is an important branch of the circuit layout optimization. And simulated annealing is the commonly used iterative optimization method. This work focuses on the qualitative performance in the optimizing process and presents a qualitative controlled optimization algorithm. Cloud model is a qualitative and quantitative transformation model. In this paper,we use two-dimensional cloud model to model the component layout. We defined three control strategies which are contraction layout,extension layout and layout reset by adjusting the entropy and hyper entropy of the cloud model. Experiments show that this approach is a little better than the simulated annealing algorithm on four standard testing circuit.