PTAB

IPR2014-01160

ATopTech Inc v. Synopsys Inc

Log In
Key Events
Petition
petition

1. Case Identification

2. Patent Overview

  • Title: Static Timing Analysis in the Presence of Crosstalk
  • Brief Description: The ’348 patent discloses methods for performing static timing analysis on an integrated circuit design to account for signal delay caused by crosstalk. The method calculates the worst-case delay on a primary ("victim") net by generating a "bump-envelope waveform" representing the noise effect from a nearby "aggressor" net over a switching window.

3. Grounds for Unpatentability

Ground 1: Claims 1-28 are obvious over Gross in view of Dartu and further in view of Shepard.

  • Prior Art Relied Upon: Gross (a 1998 conference paper titled “Determination of Worst-Case Aggressor Alignment for Delay Calculation”), Dartu (a 1997 conference paper titled “Calculating Worst-Case Gate Delays Due to Dominant Capacitance Coupling”), and Shepard (a 1997 conference paper titled “Global Harmony: Coupled Noise Analysis for Full-Chip RC Interconnect Networks”).
  • Core Argument for this Ground:

    • Prior Art Mapping: Petitioner argued that Gross teaches a nearly identical method for determining worst-case victim delay as claimed in the ’348 patent. Gross disclosed accessing a netlist, identifying victim and aggressor nets, generating a primary (victim) waveform, generating a bump (aggressor) waveform, super-positioning them to create a composite waveform, and determining the composite waveform’s threshold voltage crossing point to find the worst-case delay. Petitioner asserted that Gross teaches every limitation of the independent claims except for the specific step of "generating a bump-envelope waveform." Instead, Gross combined the actual aggressor bump waveform with the primary waveform. Dartu was presented as reinforcing the core technical principle shared by both Gross and the ’348 patent: that the maximum delay on the victim net occurs when the effect of the maximum noise from the aggressor is aligned with the victim waveform's voltage threshold.

    • Motivation to Combine: Petitioner contended that the sole alleged point of novelty in the ’348 patent—the "bump-envelope waveform"—was an obvious design choice to a person of ordinary skill in the art (POSITA). This waveform is created by taking the maximum voltage value of the aggressor's bump waveform and "stretching" it across a predetermined "switching window." Shepard was argued to teach the benefits of using such "switching windows" (defined by earliest and latest signal arrival times) to constrain timing analysis, thereby reducing pessimism and improving accuracy. A POSITA, knowing from Gross and Dartu that the peak noise value is the critical factor for determining maximum delay, would combine this with Shepard’s teaching of using switching windows. The most direct and efficient way to do so would be to apply that peak noise value as a constant (or DC offset) across the entire switching window, which is precisely what the patent’s "bump-envelope waveform" accomplishes. This avoids more complex calculations while capturing the worst-case scenario within the relevant timeframe.

    • Expectation of Success: A POSITA would have a high expectation of success in combining these references because all three address the same well-known problem of crosstalk delay in static timing analysis for integrated circuits. The references use compatible methodologies, and combining them represented a predictable path toward a more efficient and accurate analysis tool. The combination simply applied known principles (using peak noise for max delay) within a known, bounded context (a switching window) to achieve a predictable result.

4. Key Claim Construction Positions

  • "Bump-Envelope Waveform": Petitioner argued this term should be construed according to its explicit definition in the specification as "A bump waveform that has been stretched along a predetermined aggressor switching window." This construction was central to the obviousness argument, as it framed the "stretching" not as a complex new technique but as a simple, obvious application of a peak noise value across a time window taught by the prior art.

5. Relief Requested

  • Petitioner requests institution of an inter partes review (IPR) and cancellation of claims 1-28 of the ’348 patent as unpatentable.