One of the challenging problems in the design of electronic circuits is the so-called routing problem. Roughly speaking, the task is to connect so-called terminal sets via wires on a predeened area. In addition, certain design rules are to be taken into account and an objective function such as the wiring length must be minimized. The routing problem in general is too complex to be solved in one step. Depending on the user's choice of decomposing the chip design problem into a hierarchy of stages, on the underlying technology, and on the given design rules, various subproblems arise. We discuss several variants of practically relevant routing problems and give a short overview on the underlying technologies and design rules. Many of the routing problems that come up this way can be formulated as the problem of packing so-called Steiner trees in certain graphs. We consider the Steiner tree packing problem from a polyhedral point of view and present three possibilities to deene an appropriate polyhedron. Weighing their pros and cons we decide for one of these polytopes and sketch some of our investigations.
[1]
D..
Experimental Results for a Linear Program Global Router
,
1986
.
[2]
Thomas Lengauer,et al.
Integer Program formulations of Global Routing and Placement Problems
,
1993,
Algorithmic Aspects of VLSI Layout.
[3]
David S. Johnson,et al.
The Rectilinear Steiner Tree Problem is NP Complete
,
1977,
SIAM Journal of Applied Mathematics.
[4]
Alexander Martin,et al.
Packen von Steinerbäumen: polyedrische Studien und Anwendung
,
1992
.
[5]
Michael Burstein,et al.
Hierarchical Wire Routing
,
1983,
IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems.
[6]
Majid Sarrafzadeh.
Channel-Routing Problem in the Knock-Knee Mode Is NP-Complete
,
1987,
IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems.
[7]
Thomas Lengauer,et al.
Combinatorial algorithms for integrated circuit layout
,
1990,
Applicable theory in computer science.