Obstacle-Aware Length-Matching Routing for Any-Direction Traces in Printed Circuit Board

TL;DR

This paper addresses length-matching for PCB traces routed in any direction while preserving their original routing and handling obstacles and multiple DRAs. It introduces a two-stage framework: region assignment via linear programming to secure routable space, followed by a DP-based segment extension that meanders traces with obstacle-aware pattern placement. A key contribution is the Multi-Scale Dynamic Time Warping (MSDTW), which converts differential pairs into a median trace and stabilizes their restoration despite decoupling and multi-DRAs. Empirical results show effective length matching and favorable performance compared with commercial tools under more challenging constraints, with reasonable runtime and improved space utilization in obstacle-dense layouts.

Abstract

Emerging applications in Printed Circuit Board (PCB) routing impose new challenges on automatic length matching, including adaptability for any-direction traces with their original routing preserved for interactiveness. The challenges can be addressed through two orthogonal stages: assign non-overlapping routing regions to each trace and meander the traces within their regions to reach the target length. In this paper, mainly focusing on the meandering stage, we propose an obstacle-aware detailed routing approach to optimize the utilization of available space and achieve length matching while maintaining the original routing of traces. Furthermore, our approach incorporating the proposed Multi-Scale Dynamic Time Warping (MSDTW) method can also handle differential pairs against common decoupled problems. Experimental results demonstrate that our approach has effective length-matching routing ability and compares favorably to previous approaches under more complicated constraints.

Figures (16)

• Figure 1: Illustration of routing in various directions and the primary distances restricted in DRC. Solid polygons in the figure denote obstacles.
• Figure 2: Overview of our length-matching approach.
• Figure 3: Four kinds of valid state transitions.
• Figure 4: Illustration of different candidate states with the same value. (a) and (b) contribute the same value to $dp[i][dir]$. Only (a) allows the transition of $p_{local}$, so it has a higher priority than (b) to be retained.
• Figure 5: Illustration of why the patterns are hoped to be connected. The original segment has a capacity of only two patterns, so both cases have the same DP result. However, the former case can provide the capacity of an extra pattern.