PTAB

IPR2016-00935

Bungie Inc v. Acceleration Bay LLC

Log In
Key Events
Petition
petition

1. Case Identification

2. Patent Overview

  • Title: Information Delivery Service
  • Brief Description: The ’966 patent describes a computer network for broadcasting information where participants are configured in a non-complete, m-regular graph. The patent claims using a "flooding" technique, where each participant forwards received data to its neighbors to ensure reliable delivery across the network.

3. Grounds for Unpatentability

Ground 1: Obviousness over DirectPlay in view of Lin - Claim 12

  • Prior Art Relied Upon: DirectPlay (a 1998 book on Microsoft's DirectX API) and Lin (a 1999 technical report titled "Gossip versus Deterministic Flooding").
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that DirectPlay discloses a flexible Application Program Interface (API) for developing multiplayer games, which constitutes the claimed "information delivery service" for a "plurality of participants" (claim 1). DirectPlay teaches creating gaming "sessions" which form a "broadcast channel for a topic of interest" (claim 12). However, DirectPlay is network-agnostic and does not teach a specific network topology. Petitioner asserted that Lin discloses the missing network elements: a computer network using "flooding" over a non-complete, m-regular "Harary graph" topology, meeting the structural limitations of claim 1.
    • Motivation to Combine: A POSITA would combine these references for several reasons. First, DirectPlay was designed as a flexible interface to work with various underlying network service providers, and Lin’s reliable and scalable network is an ideal candidate for such a provider. Second, both references address the same technical problem of reliably broadcasting information to multiple participants. Finally, a POSITA implementing a large-scale gaming application as envisioned by DirectPlay would be motivated to use Lin’s scalable and reliable network topology to improve performance and message delivery.
    • Expectation of Success: A POSITA would have a reasonable expectation of success, as the combination involved implementing a known, high-performance network topology (Lin) with a flexible API (DirectPlay) expressly designed to accommodate custom network service providers.

Ground 2: Anticipation by Lin - Claims 1-7 and 16

  • Prior Art Relied Upon: Lin (a 1999 technical report).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner contended that Lin, standing alone, discloses every element of claims 1-7 and 16. Lin expressly describes a "simple broadcast protocol" commonly called "flooding" implemented over a "Harary graph." Petitioner argued this graph is inherently m-regular (e.g., 4-regular in Lin's examples), non-complete (as the number of nodes exceeds m+1), and uses peer-to-peer connections where "no processor has a specific role to play." Lin provides examples of 4-regular networks, meeting the limitations of claims 2, 3, and 16. Furthermore, Lin teaches that Harary graphs are m-connected for reliability, anticipating claims 4 and 5, and that all nodes are peers with peer-to-peer connections, anticipating claims 6 and 7.

Ground 3: Obviousness over Lin and POSITA Knowledge - Claims 6-11 and 17

  • Prior Art Relied Upon: Lin and the general knowledge of a Person of Ordinary Skill in the Art (POSITA).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner presented this ground as an alternative, arguing that if Lin is found not to explicitly teach every detail, any missing elements would have been obvious to a POSITA. Lin provides the core broadcast system over an m-regular, non-complete graph. A POSITA would have found it obvious to implement Lin’s "generic networks" using standard TCP/IP connections (claim 8) and understood that Lin’s "processors" are computers executing a "process" (claim 9).
    • Motivation to Combine: The motivation was simply to implement the network system described in Lin using conventional and well-understood technologies. For instance, using TCP/IP was a standard choice for network protocols. Petitioner also argued that the "ns simulator" used in Lin to simulate Ethernet networks implies a computer hosting multiple participants (claims 10 and 17), an obvious implementation detail. Finally, Petitioner asserted that Lin’s description of forwarding a message "for the first time" inherently teaches sending only one copy to neighbors (claim 11) to avoid redundant flooding, a well-known principle.

4. Key Claim Construction Positions

  • "m-regular": Petitioner proposed this term means "each node is connected to exactly m other nodes."
  • "non-complete graph": Petitioner proposed this term means a "graph in which at least two nodes are not connected to each other."
  • "m-connected": Petitioner proposed this term means "dividing the network into two or more separate parts would require the removal of at least m nodes."

These constructions were argued to be consistent with the patent's specification and essential for mapping the prior art, which explicitly uses graph theory terminology to describe network topologies with these exact properties.

5. Relief Requested

  • Petitioner requested institution of an inter partes review and cancellation of claims 1-12 and 16-17 of the ’966 patent as unpatentable.