PTAB

IPR2017-00219

Apple Inc v. California Institute Of Technology

Log In
Key Events
Petition
petition

1. Case Identification

2. Patent Overview

  • Title: Serial Concatenation of Interleaved Convolutional Codes Forming Turbo-Like Codes
  • Brief Description: The ’710 patent discloses a system for error-correcting codes, specifically focusing on irregular repeat-accumulate (IRA) codes. The invention describes an encoder with a first coder (repeater) that duplicates information bits an irregular number of times, an interleaver that scrambles the repeated bits, and a second coder (accumulator) that generates parity bits for improved data transmission reliability.

3. Grounds for Unpatentability

Ground 1: Obviousness over Divsalar and Luby - Claims 1-8 and 11-14 are obvious over Divsalar in view of Luby.

  • Prior Art Relied Upon: Divsalar (a March 1999 conference paper on "turbo-like" codes) and Luby (a 1998 conference paper on irregular graphs for low-density codes).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that Divsalar taught all elements of the challenged claims except for the "irregularity" of the first encoding step. Divsalar disclosed a standard, or regular, repeat-accumulate (RA) code comprising a repeater (first coder), an interleaver, and a rate-1 accumulator (second coder). The combination with Luby supplied the missing element. Luby explicitly taught that introducing irregularity—repeating some bits more than others—results in codes that are "far superior to any regular code." The combination therefore rendered the claimed irregular RA code obvious.
    • Motivation to Combine: A Person of Ordinary Skill in the Art (POSITA) would be motivated to apply the known performance-enhancing technique of irregularity from Luby to the well-known RA code architecture of Divsalar. Petitioner asserted that improving the performance of existing codes was a primary goal in the field, and Luby's teachings provided a clear path to achieving this for the RA codes in Divsalar. The modification was argued to be a simple and predictable implementation.
    • Expectation of Success: A POSITA would have a reasonable expectation of success in combining the references, as Luby demonstrated significant performance gains from irregularity, and applying this principle to the simple and well-understood RA codes of Divsalar would be expected to yield similar benefits.

Ground 2: Obviousness over Divsalar, Luby, and Luby97 - Claims 15-17, 19-22, and 24-33 are obvious over Divsalar in view of Luby, and further in view of Luby97.

  • Prior Art Relied Upon: Divsalar, Luby, and Luby97 (a 1997 conference paper that introduced irregularity and processing data streams).
  • Core Argument for this Ground:
    • Prior Art Mapping: This ground built upon the combination of Divsalar and Luby from Ground 1. Petitioner argued that the additional reference, Luby97, was necessary to teach limitations in claims 15 and 25 related to receiving and processing a "stream of bits." Luby97 explicitly described receiving data as a stream of symbols that is then partitioned into logical blocks for encoding. This supplied the "stream" limitation not explicitly detailed in the primary combination.
    • Motivation to Combine: A POSITA would combine Luby97's teachings because processing continuous data streams is a fundamental and practical requirement for many communication systems. Modifying the block-based encoder of Divsalar and Luby to handle streamed data as taught by Luby97 was presented as a common and obvious design choice to make the system more versatile and applicable to real-world scenarios.

Ground 3: Obviousness over Divsalar, Luby, and Pfister - Claim 10 is obvious over Divsalar in view of Luby, and further in view of Pfister.

  • Prior Art Relied Upon: Divsalar, Luby, and Pfister (a September 1999 conference paper on serial concatenation of rate-1 codes).
  • Core Argument for this Ground:
    • Prior Art Mapping: This ground addressed claim 10, which depends from claim 1 and adds the requirement that the second encoding step "utilize[] two accumulators." While the base combination of Divsalar and Luby taught a single accumulator, Pfister taught a class of codes called "RAA (Repeat-Accumulate-Accumulate)" codes. These codes are identical to Divsalar's RA codes but with an additional accumulator chained in series.
    • Motivation to Combine: Pfister provided an explicit motivation to modify Divsalar's system. Pfister directly compared the performance of Divsalar's RA codes to its own RAA codes and concluded that adding the second accumulator improved performance. This, Petitioner argued, amounted to an express suggestion for a POSITA to incorporate a second accumulator into the irregular RA code of Divsalar and Luby to further enhance its error-correction capabilities.
  • Additional Grounds: Petitioner asserted an additional obviousness challenge for claim 23 (Ground 4) based on Divsalar, Luby, Luby97, and Pfister, which relied on a similar theory as Ground 3 to teach a "second accumulator" in the context of a system processing a stream of bits.

4. Key Claim Construction Positions

  • Petitioner argued that the term "close to one" (recited in claims 1 and 3 in relation to the rate of the second coder) should be construed to mean "within 50% of one." This construction was asserted to be critical for the prior art to meet the claim limitation.
  • Support for this construction was drawn from the ’710 patent's specification, which provided an exemplary embodiment where the inner code "can have a rate that is close to one, e.g., within 50%." Petitioner also pointed to the prosecution history, where the Patent Owner amended a claim to replace "a rate close to one" with "a rate within 50% of one" to overcome a rejection, suggesting the terms were considered equivalent or that "close to one" was even broader.

5. Relief Requested

  • Petitioner requested the institution of an inter partes review and the cancellation of claims 1-8, 10-17, and 19-33 of Patent 7,116,710 as unpatentable under 35 U.S.C. §103.