A. B. Kahng, G. Robins, A. Singh, H. Wang and A. Zelikovsky, "Filling Algorithms and Analyses for Layout Density Control", IEEE Trans. Computer-Aided Design 18(4) (1999), pp. 445-462.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....post CMP topography variation from 767 A to 152 A in the single layer formulation and by avoiding cumulative effect in the multiple layer formulation. The result from single layer formulation compares very favorably both to the rulebased approach widely used in industry and to the algorithm in [3]. The multiple layer formulation has no previously published work. 1. INTRODUCTION Continued aggressive scaling down of VLSI feature size has constrained much of the manufacturing process window so that CMP for inter level dielectric (ILD) planarization has become increasingly important for ....

...., z 1 ,andt are constants for a specific CMP process. As a result, the final topography is determined only by the initial pattern density r 0 #x#y#. The simplest model uses the local spatial pattern density for r 0 #x#y#. An algorithm by Kahng et al. solves the tiling problem based on this model [3]. However, more accurate modeling by D. Ouma et al. considers the deformation of polishing pad during polish [5] The effective local density r 0 #x#y# is no longer directly proportional to local spatial pattern density, but calculated as the summation of weighted spatial pattern density within a ....

[Article contains additional citation context not shown here]

A. Kahng, G. Robins, A. Singh, and A. Zelikovsky. Filling Algorithms and Analyses for Layout Density Control. IEEE Trans. CAD, 18(4):445-462, April 1999.

....to achieve uniformity in long range oxide planarization [1] Post CMP topography is highly related to local spatial pattern density in layout. To change local pattern density, and thus ensure post CMP planarization, dummy features are placed in layout. The only known previously published algorthm [3] for dummy feature placement is based on a very simple and inadequate model. This paper is based on a closed form analytical model for inter level dielectric thickness in CMP process by B. Stine et al. 7] and a model for e ective local layout pattern density by D. Ouma et al. 5] Those two ....

....two models to solve the dummy feature placement problem of a single layer in the xed dissection regime. An experiment, conducted with real industry design data, gives excellent results by reducing post CMP topography variation from 753 A to 169 A, and compares favorably to the algorithm in [3], which only reduced the topography variation to 358 A. 1 Introduction VLSI manufacturing uses chemical mechanical polishing (CMP) to remove excess oxide deliberately deposited on surface of the wafer in order to achieve relative uniformity of inter level dielectric (ILD) thickness. Uniformity ....

[Article contains additional citation context not shown here]

A. B. Kahng et al., \Filling Algorithms and Analyses for Layout Density Control", IEEE Trans. on CAD of Integr. Circ. and Sys., Vol. 18, No. 4, Apr. 1999.

....thus control post CMP topography variation. However, inserting dummy features of a prescribed density ad hoc, wherever there is empty space in layout that is large enough, is neither effective nor efficient. Recent studies on inter layer dielectric (ILD) CMP are based on linear models oxide polish [5,6,8]. Specifically, global density assignment followed by local insertion, proposed by Tian et al. to solve the dummy feature placement problem in the fixed dissection regime with both single layer and multiple layer considerations, gave excellent results by reducing simulated post CMP topography ....

Kahng, A., Robins, G., Singh, A., and Zelikovsky, A. Filling Algorithms and Analyses for Layout Density Control. IEEE Trans. on CAD, 18(4):445-462, April 1999.

No context found.

A. B. Kahng, G. Robins, A. Singh, H. Wang and A. Zelikovsky, "Filling Algorithms and Analyses for Layout Density Control", IEEE Trans. Computer-Aided Design 18(4) (1999), pp. 445-462.

No context found.

A. B. Kahng, G. Robins, A. Singh, H. Wang and A. Zelikovsky, "Filling Algorithms and Analyses for Layout Density Control", IEEE Trans. Computer-Aided Design 18(4) (1999), pp. 445-462.

....objectives, then proposes several new approaches to a flat and hierarchical density control for CMP. Section II defines the single layer filling problem for both flat and hierarchical layouts. Both formulations are based on the practical industry standard fixed dissection density analysis regime [12]. Relevant objectives include the Min Var and Min Fill objectives. Though hierarchical filling can speed up verification of filled layout and decrease data volume, there is an obvious conflict between honoring the layout hierarchy and achieving high quality filling results. The filling problem for ....

.... that either: ffl (Min Var Objective) the variation in window density (i.e. maximum window density minus minimum window density) is minimized while the window density does not exceed the given upper bound U ; or any window remains in the given range (L;U) The Min Var objective, introduced in [12], captures the manufacturing side of fill synthesis, which seeks the most uniform density distribution possible. The Min Fill objective, recently proposed in [25] models the design side in that it seeks to minimize the coupling capacitance and the uncertainty caused by filling. Algorithms for ....

[Article contains additional citation context not shown here]

A. B. Kahng, G. Robins, A. Singh, H. Wang and A. Zelikovsky, "Filling Algorithms and Analyses for Layout Density Control", IEEE Trans. Computer-Aided Design 18(4) (1999), pp. 445-462.

....variation, and minimizing the total amount of inserted fill. Comparisons with previous filling methods show the advantages of our new iterated MonteCarlo and iterated greedy methods. We achieve near optimal filling with respect to each of the objectives and for both density models (spatial density [3] and effective density [8] Our new methods are more efficient in practice than linear programming [3] and more accurate than non iterated Monte Carlo approaches [1] 1 Introduction As the manufacturing process increasingly constrains physical layout design and verification [5] one particular ....

....show the advantages of our new iterated MonteCarlo and iterated greedy methods. We achieve near optimal filling with respect to each of the objectives and for both density models (spatial density [3] and effective density [8] Our new methods are more efficient in practice than linear programming [3] and more accurate than non iterated Monte Carlo approaches [1] 1 Introduction As the manufacturing process increasingly constrains physical layout design and verification [5] one particular requirement is to control manufacturing variation due to chemical mechanical polishing (CMP) 4] 6] ....

[Article contains additional citation context not shown here]

A. B. Kahng, G. Robins, A. Singh, H. Wang and A. Zelikovsky, "Filling Algorithms and Analyses for Layout Density Control", IEEE Trans. Computer-Aided Design 18(4) (1999), pp. 445-462.

No context found.

A. B. Kahng, G. Robins, A. Singh, H. Wang and A. Zelikovsky, "Filling Algorithms and Analyses for Layout Density Control", IEEE Trans. Computer-Aided Design 18(4), 1999, pp. 445-462.

....methods on area fill synthesis also focused exclusively on the fixed dissection context, including: ffl Linear Programming (LP) methods based on rounding relaxation of the corresponding integer linear program formulations. The LP formulations for filling were first proposed by Kahng et al. in [6] and adapted to other objectives and CMP models in [12, 13] ffl Greedy methods which iteratively find the best tile for the next filling geometry to be added into the layout. These methods were first used in [3] for ILD thickness control, and also used for shallow trench isolation (STI) CMP ....

....fixed grid, which underestimates the actual gridless density variation, but has been justified on grounds that gridless analysis is impractical. In this paper, we show for the first time the viability of gridless or floating window analyses, originally developed for the spatial density model [6], and extend it for the more accurate effective density model [9] Second, previous research in layout density control concentrated on the global uniformity achieved by minimizing the window density variation over the entire layout. However, the density variation between locations which are far ....

[Article contains additional citation context not shown here]

A. B. Kahng, G. Robins, A. Singh, H. Wang and A. Zelikovsky, "Filling Algorithms and Analyses for Layout Density Control", IEEE Trans. Computer-Aided Design 18(4) (1999), pp. 445-462.

....density obtained via dummy fill can mitigate macroscopic process proximity effects such as contact etch variation in reactive ion etch, and nonuniformity of chemical vapor deposition. Dummy fill creates a number of critical flow issues, including: 1 The Min Var objective was introduced in [5], and captures the manufacturing side of the Filling Problem by seeking the most uniform density distribution possible. The Min Fill objective was introduced in [12] and captures the design side by seeking to minimize total coupling capacitance and uncertainty caused by dummy fill. Minimizing ....

.... will be worse for hierarchical solutions than flat solutions, simply because the former are more constrained; and ffl hierarchical filling explodes the number of constraints in linear programming formulations, and thus cannot use the LP techniques which have been successful for flat filling [5] [12] 2 e.g. Cadence and Avant gridded routers are often restricted to well defined pin availabilities at points of the routing grid. The main contribution of this paper is a new proposed hierarchical filling algorithm which mitigates these drawbacks. Our approach is based on hybridizing ....

[Article contains additional citation context not shown here]

A. B. Kahng, G. Robins, A. Singh, H. Wang and A. Zelikovsky, "Filling Algorithms and Analyses for Layout Density Control", IEEE Trans. Computer-Aided Design 18(4) (1999), pp. 445462.

No context found.

A. B. Kahng, G. Robins, A. Singh and A. Zelikovsky, "Filling Algorithms and Analyses for Layout Density Control", IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, to appear. 31

....variation, and minimizing the total amount of inserted fill. Comparisons with previous filling methods show the advantages of our new iterated MonteCarlo and iterated greedy methods. We achieve near optimal filling with respect to each of the objectives and for both density models (spatial density [3] and effective density [8] Our new methods are more efficient in practice than linear programming [3] and more accurate than non iterated Monte Carlo approaches [1] 1 Introduction As the manufacturing process increasingly constrains physical layout design and verification [5] one particular ....

....show the advantages of our new iterated MonteCarlo and iterated greedy methods. We achieve near optimal filling with respect to each of the objectives and for both density models (spatial density [3] and effective density [8] Our new methods are more efficient in practice than linear programming [3] and more accurate than non iterated Monte Carlo approaches [1] 1 Introduction As the manufacturing process increasingly constrains physical layout design and verification [5] one particular requirement is to control manufacturing variation due to chemical mechanical polishing (CMP) 4] 6] ....

[Article contains additional citation context not shown here]

A. B. Kahng, G. Robins, A. Singh, H. Wang and A. Zelikovsky, "Filling Algorithms and Analyses for Layout Density Control", IEEE Trans. Computer-Aided Design 18(4) (1999), pp. 445-462.

....we are in the fixed dissection regime. Solving the Filling Problem in the fixed dissection regime consists of two steps: i) finding the amount of fill to be embedded in each tile, and (ii) embedding the corresponding fill amount into each tile. B. Previous Approaches Previous papers [3] 4] 5] [6] gave the first formulations of the filling problem. These works also developed a num tile windows n w w r X Y Figure 1: The layout is partitioned using r 2 (r = 4) distinct dissections (each with window size w Theta w) into nr w Theta nr w tiles. Each w Theta w window (darker) ....

....nr w tiles. Each w Theta w window (darker) consists of r 2 tiles. Note that a pair of windows from different dissections may overlap. ber of algorithms for density analysis, along with filling synthesis algorithms in the fixed dissection regime for flat and hierarchical designs [3] 4] 5] [6]. In [4] we proposed the first optimal solution to the minvariation filling formulation in the fixed dissection regime. The approach, based on linear programming (LP) assumes that we can fill the slack area of each cell independently and uniformly, as is the case when the size of fill geometries ....

A. B. Kahng, G. Robins, A. Singh, and A. Zelikovsky, Filling Algorithms and Analyses for Layout Density Control, IEEE Trans. Computer-Aided Design, 18 (1999), pp. 445--462.

No context found.

A. B. Kahng, G. Robins, A. Singh and A. Zelikovsky, "Filling Algorithms and Analyses for Layout Density Control", IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, to appear.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC