SDCLR: Scalable Dual Control-Layer Routing for continuous-flow microfluidic biochips with minimized control ports

Recent advancements in continuous-flow microfluidic biochips have led to highly integrated lab-on-a-chip systems with complex functionality and minimized control ports. Despite these advancements, achieving efficient control-layer routing under strict valve synchronization remains a significant challenge. Clustering more valves into the same group can reduce the usage of control ports but necessitates building a larger routing tree to meet the length-matching requirements of synchronized valve groups, significantly reducing the routing resources available for other valves and diminishing routing feasibility. To strike an optimized balance between control port usage and routing feasibility, this paper proposes a scalable dual-control layer routing method called SDCLR, aimed at optimizing routing feasibility while minimizing the number of control ports. The main features of SDCLR include a DBSCAN-based valve readdressing to optimize routing area allocation, a conflict-aware layer assignment to minimize resource contention, and a synchronized-driven routing framework to ensure precise synchronized valve operations. Experimental results show that SDCLR performs excellently in terms of routability, control port number, and timing requirements.