Netflix Inc v. Avago Technologies Intl Sales Pte Ltd

1. Case Identification

• Case #: IPR2021-01293
• Patent #: 9,402,098
• Filed: August 10, 2021
• Petitioner(s): Netflix, Inc.
• Patent Owner(s): Avago Technologies International Sales Pte. Limited
• Challenged Claims: 1-5, 7

2. Patent Overview

• Title: System and Method for Reduced Latency in a Video Signal Processing System
• Brief Description: The ’098 patent describes a method to reduce video streaming latency. The system transmits an initial portion of video at an accelerated rate, which is determined in real-time based on available network bandwidth, before switching to a slower, steady-state rate for transmitting the remainder of the video content.

3. Grounds for Unpatentability

Ground 1: Obviousness over Patrick - Claims 1-2 are obvious over Patrick.

• Prior Art Relied Upon: Patrick (Application # 2004/0128396).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that Patrick taught every limitation of independent claim 1. Patrick disclosed a method of streaming media by first receiving a request, then determining an initial "accelerated rate" that is higher than a subsequent "steady state rate." This accelerated rate was based on the available communication bandwidth. Patrick further taught transmitting an initial portion of the content at this accelerated rate to quickly fill a client buffer, followed by transmitting the remaining content at the steady-state rate. Petitioner also asserted that Patrick taught adapting the transmission rate in response to changes in available bandwidth during the initial transmission period, as recited in claim 1.
  • Motivation to Combine (N/A): This ground was based on a single reference.
  • Key Aspects: For dependent claim 2, Petitioner argued it would have been obvious to implement the content request as a "channel change request," as this was a well-known and common method for selecting video streams at the time of the invention.

Ground 2: Obviousness over Patrick and Barrett - Claim 2 is obvious over Patrick in view of Barrett.

• Prior Art Relied Upon: Patrick (Application # 2004/0128396) and Barrett (Application # 2004/0034864).
• Core Argument for this Ground:
  • Prior Art Mapping: Patrick was asserted to teach the base accelerated streaming method of claim 1. Barrett was asserted to explicitly teach a system for "seamless channel changing" where video is streamed from a server to a client in direct response to a "channel change request," thus supplying the missing limitation of claim 2.
  • Motivation to Combine: A POSITA would combine Patrick’s streaming system with Barrett’s content selection method to improve Patrick’s system with a familiar and user-friendly navigation feature. Applying a channel change mechanism to a streaming video system was described as choosing from a finite number of identified, predictable solutions to enhance usability.
  • Expectation of Success: A POSITA would have had a reasonable expectation of success because both references were directed to video streaming and the combination represented a simple application of a known technique to a known system ready for improvement.

Ground 3: Obviousness over Patrick and Gringeri - Claims 3-5 are obvious over Patrick in view of Gringeri.

• Prior Art Relied Upon: Patrick (Application # 2004/0128396) and Gringeri (Patent 6,233,226).
• Core Argument for this Ground:
  • Prior Art Mapping: Patrick provided the foundational accelerated streaming method. Gringeri taught a video transmission system that used a "decoder model" (such as the MPEG standard’s Video Buffer Verifier, or VBV) in an analysis phase to determine traffic control parameters, including transmission rates. Gringeri specifically taught using the decoder model to determine when a buffer is full to prevent overflow, which directly maps to the limitations of claims 3-5 concerning using a decoder model to determine the initial rate and/or time period to avoid overflowing an input buffer.
  • Motivation to Combine: A POSITA would combine the teachings to gain complementary benefits. Incorporating Gringeri's decoder model analysis into Patrick’s system would allow for a more sophisticated determination of the initial transmission rate and duration, balancing bandwidth consumption with user experience and preventing buffer overflow—a known issue in video streaming.
  • Expectation of Success: Success was expected as both references used packet-based protocols compliant with the MPEG-2 standard, which specified the use of a decoder model like the VBV. The integration was therefore a straightforward application of standard-compliant techniques.

Ground 4: Obviousness over Patrick and Mahdavi - Claim 7 is obvious over Patrick in view of Mahdavi.

• Prior Art Relied Upon: Patrick (Application # 2004/0128396) and Mahdavi (Patent 6,801,964).
• Core Argument for this Ground:
  • Prior Art Mapping: Patrick’s method determined the initial accelerated rate based primarily on available bandwidth. Mahdavi taught a method to "fast fill" media player buffers by determining an increased transmission rate based on a "desired startup latency." This directly addressed the limitation in claim 7 of determining the initial transmission period using a "predetermined latency goal."
  • Motivation to Combine: A POSITA would be motivated to modify Patrick’s system with Mahdavi’s technique to improve and customize the user experience. This modification would allow for direct control over startup latency—a key performance metric—rather than indirectly influencing it by relying on available bandwidth.
  • Expectation of Success: The combination involved a simple substitution of one known method for another (determining rate based on a latency goal vs. available bandwidth). Because the mathematical relationship between transmission rate, data amount, and latency was well-known, this modification would have produced a predictable result.

4. Key Claim Construction Positions

• "transmission rate": Petitioner proposed this term be construed as "a rate at which data is transmitted from a server to a receiver." This construction was intended to distinguish the claimed rate from the output rate of a video encoder.
• "steady-state transmission rate": Petitioner proposed this term encompasses "a transmission rate that is similar to a rate at which video data is processed or presented." This supported mapping prior art references that described the steady-state rate as being equivalent to the content's encoded bit rate.

5. Arguments Regarding Discretionary Denial

• Petitioner argued that discretionary denial under 35 U.S.C. § 314(a) based on Fintiv factors would be inappropriate. The core arguments were that the parallel district court cases were in their infancy, with no significant proceedings such as claim construction briefing or invalidity contentions having occurred. Petitioner also asserted the petition was strong on the merits and that none of the relied-upon prior art had been considered by the USPTO during the original prosecution of the ’098 patent.

6. Relief Requested

• Petitioner requested the institution of an inter partes review and the cancellation of claims 1-5 and 7 of the ’098 patent as unpatentable.