PTAB

IPR2019-00795

Metaswitch Networks Ltd v. RIBBON COMMUNICATIONS INC.

1. Case Identification

2. Patent Overview

  • Title: User and Session Identification Based on Connections, Protocols and Protocol Fields
  • Brief Description: The ’536 patent describes systems and methods for applying a specific service level to data packets from multiple users being transferred over a single connection. The service level for a given set of packets is determined based on a connection identifier and the protocol of the packets.

3. Grounds for Unpatentability

Ground 1: Claims 1-10 and 13-24 are obvious over Maher in view of Heller and Kloth.

  • Prior Art Relied Upon: Maher (Application # 2003/0118029), Heller (Patent 6,888,807), and Kloth (Patent 6,643,260).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that the combination of these references teaches all limitations of the challenged claims. Maher was asserted to disclose the foundational system: a network device that enforces service level agreements by implementing Quality of Service (QoS) policies on traffic flows (or "sessions"). Maher's system scans packet headers to associate a data packet with a subscriber, creates a session ID based on information like source/destination ports and protocol, and applies a QoS policy. However, Maher determines QoS on a per-packet basis. Heller was asserted to cure this inefficiency by teaching the association of a specific QoS or class of service with an entire session or flow ID. This allows a single QoS determination to be applied to all subsequent packets in that flow. Kloth was argued to provide a further, known optimization. While Maher teaches using Layer 4 information (ports) to identify connections, Kloth teaches using simpler identifiers, such as Layer 2 addresses, to distinguish flows. The combination, therefore, results in a system that applies a pre-determined QoS to an entire session (from Heller) which is identified using an efficient Layer 2 connection identifier (from Kloth), as recited in the challenged claims.
    • Motivation to Combine: A person of ordinary skill in the art (POSITA) would combine Maher and Heller to improve the efficiency of Maher's system. Applying a QoS level to an entire session, as taught by Heller, is more efficient than re-analyzing each individual packet as suggested by Maher. A POSITA would have been further motivated to incorporate Kloth's teachings to modify the combined Maher/Heller system. It was known that processing Layer 4 header information was complex and a limiting factor for router performance. A POSITA would have recognized that using a simpler Layer 2 identifier, as taught by Kloth, would simplify header processing and improve the efficiency of Maher's system.
    • Expectation of Success: Petitioner asserted that a POSITA would have had a reasonable expectation of success. The combination involved applying a known technique (Heller’s per-flow QoS) to a known system (Maher’s packet-network system) to achieve a predictable improvement in efficiency. Similarly, substituting a known, simpler identifier (Kloth's Layer 2) for a more complex one (Maher's Layer 4) was a predictable design choice to improve performance.

4. Key Claim Construction Positions

  • "storing the service level" (claims 3, 10): Petitioner argued a POSITA would understand this to mean storing the association between the service level and the user device. This is distinct from merely storing the existence of a service level. This construction was important because it supports the argument that the prior art teaches associating a specific QoS profile with a specific user/session for later use, a core element of the efficiency-based motivation to combine.
  • "wherein session identifiers for different data protocols ... are determined differently" (claim 8): Petitioner contended that this limitation is taught by the prior art regardless of its interpretation. If construed to mean the session identifiers themselves are different, Maher's teaching of creating a "unique" session ID for each new traffic flow would meet the limitation. If construed to mean a different method is used to determine the identifiers for different protocols (e.g., IP vs. ATM), this would have been an obvious, known technique to a POSITA familiar with handling various network protocols.

5. Arguments Regarding Discretionary Denial

  • Arguments against §325(d) Denial: Petitioner argued that denial under §325(d) was inappropriate because the core prior art references—Maher, Heller, and Kloth—were never considered by the Patent Office during the original examination of the ’536 patent. Therefore, the arguments and art presented in the petition were neither old nor cumulative.
  • Arguments against Fintiv Denial: Petitioner argued against discretionary denial based on co-pending district court litigation in the Eastern District of Texas. The petition asserted that the litigation was in an early phase, with a trial date over a year away and subject to rescheduling. It was also argued that the IPR addresses a much broader set of claims (1-10, 13-24) than the district court case (1, 8, 10, and 19), and that parallel IPR proceedings would be more efficient for resolving the patentability of the challenged claims.

6. Relief Requested

  • Petitioner requested institution of an inter partes review and cancellation of claims 1-10 and 13-24 of the ’536 patent as unpatentable.