Telefonaktiebolaget LM Ericsson Ericsson Inc v. KMizra LLC

1. Case Identification

• Case #: IPR2022-00730
• Patent #: 8,958,819
• Filed: March 18, 2022
• Petitioner(s): Telefonaktiebolaget LM Ericsson, Ericsson Inc., and AT&T Services, Inc.
• Patent Owner(s): K. Mizra LLC
• Challenged Claims: 22, 23, 30, 32, 33

2. Patent Overview

• Title: Femto-assisted location estimation in macro-femto heterogeneous networks
• Brief Description: The ’819 patent discloses methods for locating user equipment (UE) indoors by using a statistical "particle filtering" technique. The technique employs timing information from both large-coverage macro base stations and short-range femto base stations operating together in a heterogeneous network.

3. Grounds for Unpatentability

Ground 1: Claims 22 and 30 are anticipated by Pakzad under 35 U.S.C. §102.

• Prior Art Relied Upon: Pakzad (Patent 8,594,701).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that Pakzad discloses every limitation of independent claims 22 and 30. Pakzad addresses the same problem of indoor UE location tracking by using a particle filter. It explicitly teaches a system that receives timing information (e.g., observed time difference of arrival) from both macro base stations ("base station transceiver") and femto base stations ("local transceiver" or "femtocell"). Pakzad further discloses receiving particle information corresponding to possible UE locations and receiving femto base station position information to determine the UE's location, thereby anticipating the claims.
  • Key Aspects: Petitioner contended that Pakzad's solution to the indoor location problem mirrors the '819 patent's claimed invention by using measurements from various base station types to locate a UE via a particle filter.

Ground 2: Claims 23, 32, and 33 are obvious over Pakzad in view of Ketchum under 35 U.S.C. §103.

• Prior Art Relied Upon: Pakzad (Patent 8,594,701) and Ketchum (Patent 8,600,297).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner asserted that Pakzad provides the foundational system of using a particle filter with macro and femto base station data. For claim 23, Ketchum supplies the teaching of using an estimated or distributed position for femto base stations, as it describes a "self-locating access point" that obtains "accurate estimates of...position." For claims 32 and 33, Petitioner argued Pakzad itself discloses importance sampling through its use of multi-layered "motion models" (a "Dense Routing Graph" for physical constraints and a "Coarse Routing Graph" for likely destinations) to generate successive sets of weighted particles, which are then resampled.
  • Motivation to Combine: A POSITA would combine Pakzad and Ketchum because they address the same technical field (indoor location) and were both assigned to Qualcomm, suggesting a common line of research. Ketchum directly solves the known problem of uncertain femto base station locations—a challenge inherent in the system described by Pakzad. Applying Ketchum's method for estimating femto locations to Pakzad's particle filtering framework would have been a simple substitution of a known element to improve the system's accuracy.
  • Expectation of Success: A POSITA would have had a high expectation of success, as Pakzad’s algorithm was already designed to work with estimations (e.g., estimated UE location), making the use of Ketchum's estimated femto locations a predictable and logical improvement.

Ground 3: Claims 32 and 33 are obvious over Pakzad and Ketchum, further in view of Klepal under 35 U.S.C. §103.

• Prior Art Relied Upon: Pakzad (Patent 8,594,701), Ketchum (Patent 8,600,297), and Klepal (Patent 9,217,788).

• Core Argument for this Ground:

  • Prior Art Mapping: This ground provides an alternative or supplemental basis for the "importance sampling" limitation of claim 32. To the extent the Board might find Pakzad's "motion models" insufficient, Klepal explicitly discloses using "importance sampling" in a particle filter to solve the "degeneracy problem" (where particles acquire near-zero weight) in the specific context of indoor location tracking. Klepal teaches sampling particles from a "related distribution" to generate a new set of particles, directly mapping to the claim limitation.
  • Motivation to Combine: A POSITA seeking to improve the computational efficiency of Pakzad's particle filter would have been motivated to incorporate known optimization techniques. Klepal provides a clear technical disclosure and mathematical basis for importance sampling applied to the same problem (indoor UE tracking), making it a natural and obvious modification to enhance Pakzad's system.
  • Expectation of Success: Success would be highly predictable, as importance sampling was a well-known, established method for improving particle filter performance at the time of the invention.

• Additional Grounds: Petitioner asserted an alternative ground that claims 22 and 30 are obvious over Pakzad alone.

4. Key Claim Construction Positions

• Petitioner argued that the term “importance sampling” (claim 32) should be construed as “generating particles based on a known importance distribution.”
  • This proposed construction was based on the claim language stating that the process "results in a second set of particles being generated" and explicit descriptions in the '819 patent's specification of using an "importance distribution" to generate particles. Petitioner contended this was narrower and more accurate than the Patent Owner's proposed construction of "selecting samples based on importance."

5. Arguments Regarding Discretionary Denial

• Petitioner argued that discretionary denial under 35 U.S.C. §325(d) was inappropriate because none of the asserted prior art references were cited or considered during the original prosecution of the ’819 patent.
• Petitioner also argued against discretionary denial under Fintiv, asserting that: (1) the trial date in the parallel district court litigation was uncertain and distant from the IPR’s projected FWD deadline; (2) Petitioner offered a binding stipulation to not pursue any invalidity grounds in district court that rely on the primary reference, Pakzad; (3) the IPR challenges claims not asserted in the parallel litigation, promoting system efficiency; and (4) the petition demonstrates a strong likelihood of prevailing on the merits.

6. Relief Requested

• Petitioner requests institution of an inter partes review and cancellation of claims 22, 23, 30, 32, and 33 of Patent 8,958,819 as unpatentable.