A particle swarm optimization Steiner minimum tree construction method based on multi-stage transformation and genetic manipulation under an X structure

Abstract

The invention relates to a particle swarm optimization Steiner minimum tree construction method based on multi-stage transformation and genetic operation under an X structure, and provides an effective edge-edge-genetic operation method in order to solve two SMT problems at the same time. Point coding strategy. The invention provides a multi-stage conversion strategy which not only can enlarge thesearch space of an algorithm, but also can ensure effective convergence. By testing three schemes from two stages to four stages and different combinations of the schemes, an optimal combination is given. At the same time, the present invention constructs a discrete particle update operation for discrete VLSI wiring using a genetic operation in combination with union-lookup set segmentation. In addition, in order to introduce uncertainty and diversity, the invention provides a variation operation for improving an edge conversion strategy. According to the method, a wiring scheme with the bestquality can be obtained from a global angle.

Description

technical field [0001] The invention relates to the field of integrated circuit computer-aided design, in particular to a particle swarm optimization Steiner minimum tree construction method based on multi-stage transformation and genetic operation under X structure. Background technique [0002] General routing is an important step in the physical design of very large scale integration (VLSI). The Steiner Minimum Tree (SMT) problem is to construct a minimum-cost routing tree by introducing additional points (Steiner points) connecting a given set of pins. Steiner minimum tree is the best connection model for connecting multi-terminal nets in VLSI general wiring, so the construction of Steiner minimum tree is a key issue in VLSI wiring. [0003] At present, many routing algorithms based on the Manhattan structure have been proposed. With the development of VLSI technology, the interconnect effect becomes the main challenge of chip performance. However, the wiring model ba...

AI Technical Summary

Problems solved by technology

The time complexity of the existing algorithms increases exponentially with the increase of the problem size, and most traditional heuristic algorithms are easy to fall into the local optimal solution due to the greedy strategy

Therefore, the method of constructing the Steiner tree based on the existing algorithm and the traditional heuristic algorithm does not make full use of the geometric characteristics of the non-Manhattan structure, and cannot guarantee the quality of the Steiner tree

In addition, these methods rarely provide suitable topology optimization methods, and have limited optimization capabilities in key indicators such as time efficiency and line length.

Images