Unified Patents LLC v. DivX

1. Case Identification

• Case #: IPR2021-01476
• Patent #: 10,326,987
• Filed: September 3, 2021
• Petitioner(s): Unified Patents, LLC
• Patent Owner(s): DivX, LLC
• Challenged Claims: 1 and 10

2. Patent Overview

• Title: Adaptive Bitrate Streaming
• Brief Description: The ’987 patent describes methods for adaptive bitrate streaming of media content. The technology involves a playback device obtaining an index file of alternative video streams at different bitrates, measuring network conditions, and selecting an appropriate stream to prevent buffer underflow during playback.

3. Grounds for Unpatentability

Ground 1: Claims 1 and 10 are obvious over Biderman in view of Gigliotti.

• Prior Art Relied Upon: Biderman (WO 2010/078281) and Gigliotti (Application # 2009/0307367).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that Biderman, which discloses an adaptive streaming system using multiple bitrates, taught most of the challenged claim limitations. Biderman’s system obtains a “playlist file” (an index file) that identifies variant streams with different bandwidth requirements. The client device measures network speed to select an initial stream and can later switch to a higher bitrate stream if bandwidth improves. However, Petitioner asserted Biderman did not explicitly teach using a “minimum buffer level criterion” to manage stream switching. To remedy this, Petitioner introduced Gigliotti, which taught monitoring buffer fill levels (measured in playback time) and using a predetermined threshold or "watermark" to decide when to switch to a lower bitrate stream to prevent interruptions.
  • Motivation to Combine: A Person of Ordinary Skill in the Art (POSITA) would combine Gigliotti’s buffer management with Biderman’s streaming system to improve performance. Gigliotti expressly stated its solution ensures an "uninterrupted or substantially uninterrupted stream of media." A POSITA would have recognized that applying Gigliotti’s technique of ensuring the buffer is sufficiently full before switching streams would prevent the stalling and interruptions that could occur in Biderman's system, thereby achieving a known benefit.
  • Expectation of Success: A POSITA would have had a reasonable expectation of success because using buffer capacity measurements to determine when to switch between stream variants was a well-known technique in quality-adaptive streaming systems at the time of the ’987 patent.

Ground 2: Claims 1 and 10 are obvious over Sood, Myers, Casalena, Nilsson, and Ma.

• Prior Art Relied Upon: Sood (Application # 2010/0235472), Myers (Patent 8,190,677), Casalena (Patent 9,324,375), Nilsson (Application # 2009/0116551), and Ma (WO 2010/111261).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner presented this combination as a step-by-step assembly of known components to arrive at the claimed invention. Sood provided the foundational adaptive streaming system that selects an initial media encoding and switches between streams based on a Quality of Service (QoS) metric. Myers was combined with Sood to provide specific details of the "manifest file" (index file) used in the Microsoft Smooth Streaming system that Sood referenced, teaching how bitrates and stream locations are specified. Casalena was added to improve upon Sood’s default method of selecting the "lowest available bit rate" for the initial stream. Casalena taught performing a "network connection speed test" during startup to select the optimal quality stream based on available bandwidth. Finally, Nilsson and Ma were introduced to supply the missing "minimum buffer level criterion" for preventing buffer underflow. Nilsson taught monitoring the amount of playback time stored in a buffer (e.g., 15 seconds) and switching streams when the buffer level crossed a threshold. Ma disclosed a predictive algorithm that estimates future buffer occupancy to proactively switch streams and prevent "under-running" (buffer underflow).
  • Motivation to Combine: A POSITA would have been motivated to combine these references to build a more robust and efficient streaming system. Combining Myers with Sood was logical because Myers detailed the specific manifest file implementation referenced by Sood. Adding Casalena’s teachings was motivated by the desire to improve the user experience by starting playback with the highest possible quality stream, a stated goal of Casalena. Incorporating Nilsson and Ma was motivated by the need to prevent stalling, a common problem in streaming. Nilsson and Ma provided explicit solutions for maintaining a stable buffer, which would predictably improve the Sood/Casalena system by making it more resilient to network fluctuations.
  • Expectation of Success: A POSITA would have expected success in combining these elements. The combination represented the integration of known techniques (initial stream optimization, detailed manifest files, predictive buffer management) into a conventional adaptive streaming framework to solve well-understood problems and achieve predictable improvements.

4. Arguments Regarding Discretionary Denial

• Petitioner argued that the Board should not exercise discretionary denial under §325(d) because none of the asserted prior art was previously cited or considered during the original prosecution of the ’987 patent.
• Petitioner further contended that discretionary denial under §314(a) based on the Fintiv factors was inappropriate. It argued that the parallel district court litigation was in its early stages, with a motion to stay pending. Key factors weighing against denial included that the Petitioner is not a party to the litigation, there is no overlap in arguments, and the petition presents a particularly strong case on the merits for unpatentability, which favors institution to promote patent quality.

5. Relief Requested

• Petitioner requested the institution of an inter partes review and the cancellation of claims 1 and 10 of the ’987 patent as unpatentable.