Google LLC v. Uniloc 2017 LLC

1. Case Identification

• Patent #: 6,329,934
• Filed: January 17, 2020
• Petitioner(s): Google LLC
• Patent Owner(s): Uniloc 2017 LLC
• Challenged Claims: 3

2. Patent Overview

• Title: Data-Modifying Assembly and Method
• Brief Description: The ’934 patent describes systems and methods for modifying coded data, particularly within video compression standards like MPEG. The purported invention involves partially decoding a coded data stream, modifying the resulting partially decoded data, and then performing a complementary re-encoding to produce a modified coded data stream, allegedly improving computational efficiency.

3. Grounds for Unpatentability

Ground 1: Anticipation over Liu - Claim 3 is anticipated by Liu under 35 U.S.C. §102(e).

• Prior Art Relied Upon: Liu (Patent 5,907,374)
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that Liu, which describes an apparatus for processing a compressed bitstream, discloses every limitation of claim 3. Liu's "layered decoder" (114) performs the function of the claimed "partial decoder" by selectively decoding an input bitstream to an intermediate state, short of full decoding. Specifically, Liu’s decoder can stop after the inverse discrete cosine transform (IDCT) step, yielding blocks of differences, which Petitioner equated to the claimed "blocks of prediction-error pixels." Liu’s "decoded bitstream processor" (118) performs the function of the claimed "data modifier" by processing this partially decoded data through operations like scaling or filtering. Finally, Liu’s "layered encoder" (124) performs the function of the claimed "complementary coder" by re-encoding the modified data, using only the encoding steps corresponding to the decoding steps that were bypassed. Petitioner asserted that a person of ordinary skill in the art (POSITA) would at once envisage the arrangement in Liu as meeting all claim limitations.

Ground 2: Obviousness over Liu and Chitprasert - Claim 3 is obvious over Liu in view of Chitprasert.

• Prior Art Relied Upon: Liu (Patent 5,907,374) and Chitprasert et al., “Discrete Cosine Transform Filtering,” Signal Processing (1990).
• Core Argument for this Ground:
  • Prior Art Mapping: This ground was presented as an alternative to Ground 1. Petitioner argued that even if Liu were found not to explicitly teach modifying prediction-error pixels in the spatial domain (i.e., after the IDCT step), Chitprasert provides the missing element. Liu discloses a system for partially decoding and re-encoding video data, and its processor can perform filtering on data in the frequency domain (DCT coefficients). Chitprasert taught that filtering operations can be performed in either the frequency domain (via multiplication) or the spatial domain (via convolution) to achieve approximately the same result.
  • Motivation to Combine: A POSITA would have been motivated to modify Liu's system based on Chitprasert's teachings. Chitprasert explicitly discusses the relationship between filtering in the two domains, making it a known design choice. A POSITA seeking to apply a filter to the partially decoded data in Liu would have recognized that applying it in the spatial domain (i.e., to the prediction-error pixels output by the IDCT module) was an obvious alternative to applying it in the frequency domain. This would be a simple modification to combine known elements (Liu's architecture and Chitprasert's filtering principles) according to known methods.
  • Expectation of Success: A POSITA would have had a reasonable expectation of success in combining these teachings. The combination involved applying a known filtering technique (spatial domain filtering) to a known type of data (prediction-error pixels) within a well-understood architecture (Liu’s layered codec). The result—a modified bitstream—was predictable, as Liu's system was already designed to handle processing and re-encoding of modified data streams.

4. Key Claim Construction Positions

• Petitioner dedicated a significant portion of the petition to arguing that three terms in claim 3—"partial decoder (PDEC)," "data modifier (MOD)," and "complementary coder (CCOD)"—should be construed as means-plus-function terms under 35 U.S.C. §112, ¶ 6.
• Petitioner argued these are "nonce" terms, not recognized by a POSITA as having a sufficiently definite structural meaning. Because the terms are functional and the claim recites no structure for performing those functions, Petitioner asserted they invoke §112, ¶ 6.
• Petitioner further contended that the ’934 patent specification fails to disclose adequate corresponding structure (i.e., a clear algorithm) for these functions, which would render the claim indefinite in a different proceeding.
• For the purposes of the IPR, Petitioner proceeded with a dual analysis. First, it adopted an "assumed structure" for each term based on its reading of the specification (e.g., the "partial decoder" is a device performing variable length decoding, inverse quantization, and IDCT, but not motion compensation). Second, it argued that even under a plain and ordinary meaning (which it noted the district court had preliminarily adopted), the prior art still disclosed each limitation. The invalidity arguments were presented to cover both construction scenarios.

5. Relief Requested

• Petitioner requested the institution of an inter partes review and the cancellation of claim 3 of the ’934 patent as unpatentable.