Amazon.com Inc v. Nokia Technology Oy

1. Case Identification

• Case #: IPR2024-00572
• Patent #: 7,724,818
• Filed: April 1, 2024
• Petitioner(s): Amazon.com, Inc. and Amazon.com Services LLC
• Patent Owner(s): Nokia Technologies Oy
• Challenged Claims: 6-9, 11, 15, 21, and 23

2. Patent Overview

• Title: Method for coding sequences of pictures
• Brief Description: The ’818 patent relates to methods for decoding video data from a bitstream by splitting coding parameters into multiple hierarchical structures based on their persistence. The disclosed structures include a "sequence parameter set" for parameters constant across a sequence of pictures, a "picture parameter set" for parameters constant within a single picture, and parameters contained within a "slice header."

3. Grounds for Unpatentability

Ground 1: Obviousness over H.264 Draft - Claims 6-9, 11, 15, 21, and 23 are obvious over JVT-E022d7.

• Prior Art Relied Upon: JVT-E022d7 (an editor’s proposed draft text for the H.264/AVC video coding standard, publicly available October 2002).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that JVT-E022d7, a foundational document for the H.264 standard, discloses all elements of the challenged claims. It explicitly describes splitting parameters into a “sequence parameter set” (containing values that change infrequently) and a “picture parameter set” (containing parameters that remain unchanged within a coded picture). JVT-E022d7 further discloses a “slice header” that contains a frame_num parameter, which serves as a unique identifier for the frame. Petitioner contended this frame_num is a "second picture parameter value" that remains unchanged for all slice headers within a single picture, directly teaching the core limitation of independent claim 6. The dependent claims were allegedly met by JVT-E022d7’s disclosure of identifiers (pic_parameter_set_id and seq_parameter_set_id) that link the hierarchical parameter sets together.
  • Motivation to Combine (for §103 grounds): Not applicable (single reference ground).
  • Expectation of Success (for §103 grounds): Not applicable (single reference ground).

Ground 2: Obviousness over Video Decoding Patent - Claims 6-9, 11, 15, 21, and 23 are obvious over Au.

• Prior Art Relied Upon: Au (Patent 6,646,578).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner asserted that Au teaches a video decoding system with a clear hierarchy of syntax levels corresponding to sequence, picture, and slice levels. Au’s Figure 2 illustrates distinct “Sequence Parameter Sets,” “Picture Parameter Sets,” and “Slice Headers.” The Sequence Parameter Set includes parameters that remain unchanged for an entire video sequence (e.g., frame dimensions). The Picture Parameter Set includes values unchanged within a picture. Critically, Au’s Slice Header is shown to contain a frame number. Petitioner argued this frame number is a picture-level parameter that necessarily remains the same for all slices of a given frame to ensure proper reconstruction, thus teaching the key limitation of claim 6.
  • Motivation to Combine (for §103 grounds): Not applicable (single reference ground).
  • Expectation of Success (for §103 grounds): Not applicable (single reference ground).

Ground 3: Obviousness over VCEG-N52 in view of Sriram

• Prior Art Relied Upon: VCEG-N52 (a 2001 proposal for the H.26L video coding standard) and Sriram (Patent 6,539,059).

• Core Argument for this Ground:

  • Prior Art Mapping: VCEG-N52 taught a general “parameter set concept” for video coding but did not explicitly separate parameters into hierarchical sets for sequences, pictures, and slices. Sriram addressed efficiency in video decoding by teaching the use of smaller, hierarchical data structures for parameters extracted from picture, sequence, and GOP headers, linked by pointers. The combination allegedly rendered the claims obvious. VCEG-N52 provided the foundational parameter set, while Sriram provided the explicit teaching to structure those parameters hierarchically (sequence, GOP, picture) to improve efficiency. This combination would result in distinct parameter sets for different levels, with a picture-level parameter (like a frame number, as suggested by Sriram’s data structures) remaining constant across all slices of a picture.
  • Motivation to Combine (for §103 grounds): A POSITA would combine Sriram’s efficient, hierarchical data structure approach with VCEG-N52’s parameter set concept to achieve the shared goal of increasing bandwidth and processing efficiency. Sriram’s teachings directly addressed how to organize parameters from different video hierarchy levels (sequence, GOP, picture), providing a clear path to improve upon VCEG-N52’s less-structured parameter set.
  • Expectation of Success (for §103 grounds): A POSITA would have had a high expectation of success because both references operate in the same technical field of video decoding, use the same video hierarchy, and aim for efficiency. Applying Sriram’s well-understood data structuring principles to VCEG-N52’s concept was presented as a routine software engineering task.

• Additional Grounds: Petitioner asserted an additional obviousness challenge over Linzer (Patent 7,660,356), which explicitly incorporates the JVT-E022d7 reference in its entirety. This ground was presented as an alternative to Ground 1 should the printed publication status of JVT-E022d7 be successfully challenged by the Patent Owner.

4. Key Claim Construction Positions

• Petitioner argued that no specific claim terms require construction beyond their plain and ordinary meaning.
• For the means-plus-function limitations in claim 23 (e.g., "means for recognizing"), Petitioner identified the corresponding structure disclosed in the ’818 patent as "a decoder as described in the '818 patent specification in figure 4 and equivalents thereof." Petitioner noted this is the same structure identified by the Patent Owner in a co-pending ITC investigation and argued that the asserted prior art references teach this structure.

5. Arguments Regarding Discretionary Denial

• Petitioner argued against discretionary denial under both Fintiv and §325(d).
• Fintiv: Denial was argued to be improper because the co-pending district court proceedings were initiated recently (November 2023), and no trial schedule has been set. Any potential trial would occur well after the Final Written Decision deadline in this IPR. Furthermore, Fintiv does not apply to the co-pending ITC investigation.
• §325(d): Denial was argued to be inappropriate because the asserted grounds rely on art and arguments that are not the same or substantially the same as those considered by the Examiner during prosecution. JVT-E022d7 and Linzer were never before the Examiner. While VCEG-N52 and Au were of record, they were either not substantively analyzed or not applied in the combinations asserted in the petition.

6. Relief Requested

• Petitioner requests institution of an inter partes review and cancellation of claims 6-9, 11, 15, 21, and 23 of the ’818 patent as unpatentable.