Google LLC v. Geoscope Technologies Pte Ltd

1. Case Identification

• Case #: IPR2023-01209
• Patent #: 7,561,104
• Filed: August 10, 2023
• Petitioner(s): Google LLC
• Patent Owner(s): Geoscope Technologies Pte. Ltd.
• Challenged Claims: 1-3, 11

2. Patent Overview

• Title: Method to modify calibration data used to locate a mobile unit
• Brief Description: The ’104 patent discloses methods for determining the location of a mobile station, particularly indoors, by modifying observed network measurement data (e.g., signal strength) before comparing it to a database of pre-gathered calibration data to improve accuracy.

3. Grounds for Unpatentability

Ground 1: Anticipation by Shkedi - Claims 1-3 and 11 are anticipated by Shkedi under 35 U.S.C. §102.

• Prior Art Relied Upon: Shkedi (Patent 7,706,811).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that Shkedi’s "Signal-Comparison Based Location-Determining Method" discloses every element of the challenged claims. Shkedi teaches creating a database of signal strength measurements from multiple transmitters at known locations (the claimed "database of previously-gathered calibration data"). It then describes collecting current signal strength measurements at the mobile station’s unknown location ("collecting observed network measurement data"). Critically, Shkedi discloses normalizing these measurements by using the "strongest signal" as a unit to modify the other, "less strong" signals, for example by "dividing the 3 less strong signals by the strongest signal." Petitioner asserted this directly teaches the claimed steps of "determining which of said...signal characteristics has a greater magnitude" and "modifying said observed network measurement data using the greater magnitude signal characteristic." Finally, Shkedi compares these modified (normalized) measurements to the database to determine the mobile station's location.
  • Key Aspects: Petitioner contended that Shkedi’s method for improving location accuracy in poor-signal areas like "dense multi-story urban centers" directly anticipates the ’104 patent's purported solution for indoor location finding.

Ground 2: Obviousness over Shkedi and Spain - Claim 2 is obvious over Shkedi in view of Spain under 35 U.S.C. §103.

• Prior Art Relied Upon: Shkedi (Patent 7,706,811), Spain (Application # 2003/0064735).
• Core Argument for this Ground:
  • Prior Art Mapping: As an alternative position for claim 2, Petitioner argued that if Shkedi were found not to expressly teach a database with "non-uniform grid points," Spain remedies this. Spain explicitly discloses improving the accuracy of a fingerprint database by using "intelligent algorithms that interpolate the fingerprints at all locations in between the sampled locations." Petitioner contended that this interpolation inherently generates non-uniform grid points to "reconstruct the grid," directly teaching the limitation of claim 2.
  • Motivation to Combine: A Person of Ordinary Skill in the Art (POSITA) would combine Spain’s interpolation technique with Shkedi's location system to enhance the accuracy and reduce the computational requirements of Shkedi’s database. Both references share the common goal of improving location determination accuracy using fingerprinting techniques, particularly in areas with degraded GPS capability, making the combination a predictable solution to a known problem.
  • Expectation of Success: A POSITA would have a high expectation of success because both references operate in the same technical field of RF fingerprinting and address congruent problems. Applying Spain’s known method for generating a more robust grid to Shkedi’s system would have been a routine modification.

Ground 3: Anticipation by Zhu - Claims 1-3 and 11 are anticipated by Zhu under 35 U.S.C. §102.

• Prior Art Relied Upon: Zhu ("Indoor/outdoor location of cellular handsets based on received signal strength," 2005 IEEE 61st Vehicular Technology Conference).

• Core Argument for this Ground:

  • Prior Art Mapping: Petitioner argued Zhu’s method for locating indoor cellular handsets anticipates all challenged claims. Zhu teaches creating a calibrated RF Map Database (RFMD), collecting network measurement reports (NMRs) from a handset, and then selecting a subset of the "N strongest" signal strength values from the NMR. This selection process was argued to be the same as "determining which...signal characteristics has a greater magnitude." Zhu then modifies the observed data by normalizing the NMR vector based on these strongest signals to generate relative signal strength values. Finally, Zhu compares this modified data to the RFMD by calculating the Euclidean distance to find the best match and determine the handset's location.
  • Key Aspects: Petitioner highlighted that Zhu’s database, constructed from measured points that "tend to cluster along paths" and enhanced with interpolation, inherently comprises "non-uniform grid points" (claim 2). Zhu also explicitly states that the measurements are collected at the handset (claims 3 and 11).

• Additional Grounds: Petitioner asserted additional obviousness challenges based on combinations of Zhu in view of Laitinen, and Zhu and Laitinen in view of Spain, relying on similar theories of improving RF fingerprinting databases for GSM networks and generating non-uniform grid points.

4. Key Claim Construction Positions

• "calibration data": Petitioner proposed the construction "modified or unmodified network measurement data associated with a geographic location." This construction was argued to be critical because it encompasses the normalized or ratio-based data taught by the prior art and is consistent with the patent’s own disclosure that calibration data can be modified.
• "observed network measurement data": Petitioner proposed the construction "measurement data from a network measurement report (i.e., a report used in cellular networks which provides the results of a measurement from a mobile device on one or more cells)." This construction was important to ground the claims in standard cellular network terminology used in the prior art.
• "grid point": Petitioner proposed the construction "a point associated with representative calibration data for an area." This broad construction was argued to be necessary to cover the various ways the prior art, like Shkedi and Zhu, establishes location databases using non-uniform, user-selected, or interpolated points.

5. Arguments Regarding Discretionary Denial

• Petitioner argued against discretionary denial under Fintiv factors, contending that the parallel district court litigation was in an early stage. Key arguments included that a trial date had not been scheduled, expert discovery was incomplete, and the petition was filed diligently within three months of Petitioner being made aware of the full set of asserted claims.
• Petitioner also argued against denial under §325(d), asserting that the primary prior art references (Shkedi and Zhu) were never considered by the USPTO during prosecution or reexamination. This failure to consider highly material, anticipatory art constituted a "material error," which Petitioner argued weighs strongly in favor of instituting review to correct the record.

6. Relief Requested

• Petitioner requests institution of an inter partes review and cancellation of claims 1-3 and 11 of Patent 7,561,104 as unpatentable.