PTAB

IPR2018-01725

CISCO SYSTEMS, INC. v. TracBeam LLC

Log In

1. Case Identification

2. Patent Overview

  • Title: Method for Locating a Mobile Unit
  • Brief Description: The ’525 patent discloses methods for locating a mobile unit using a wireless communication system. The methods involve obtaining a first location estimate from a first location technique and a second location estimate from a second location technique that uses "pattern recognizers." A resulting location estimate is then determined by either selecting which of the two estimates is preferred or by combining them.

3. Grounds for Unpatentability

Ground 1: Claims 1 and 6 are obvious over Sheffer in view of Tayloe and further in view of Sanderford.

  • Prior Art Relied Upon: Sheffer (Patent 5,844,522), Tayloe (Patent 5,023,900), and Sanderford (Patent 5,717,406).

  • Core Argument for this Ground:

    • Prior Art Mapping: Petitioner argued that the combination of these references taught all elements of the challenged claims. Sheffer was asserted to disclose the foundational method of using multiple independent location techniques to locate a mobile phone and then comparing the results to determine a final position with an associated confidence level. Sheffer’s techniques included azimuth triangulation (Area A), received signal strength indication (RSSI) (Area B), and cell sector analysis (Area C). Petitioner mapped Sheffer’s cell sector analysis to the claimed "first location technique" and its azimuth triangulation or RSSI analysis to the concept of a "second location technique."

    Petitioner contended that Tayloe taught a known method for improving Sheffer’s "first location technique." Specifically, while Sheffer’s cell-based location estimate relied on idealized, predetermined cell boundaries, Tayloe taught a practical method for mapping the actual radio coverage boundaries of cell sites by collecting and analyzing signal strength measurements from mobile units in the field.

    Finally, Petitioner argued that Sanderford taught an improved "second location technique" that met the "pattern recognizer" limitation. Sanderford disclosed using an artificial neural network (an example of a pattern recognizer explicitly mentioned in the ’525 patent) to determine a mobile unit's location. Critically, Sanderford’s neural network was trained to use multipath signal reflections and other signal characteristics as a unique "fingerprint" or "signature" to improve location accuracy in reflection-prone environments. This directly mapped to the claim language requiring the second location estimate to be dependent on "one or more patterns in multipath."

    • Motivation to Combine: Petitioner asserted that a person of ordinary skill in the art (POSITA) would have been motivated to combine the references to create a more accurate and robust location system. The motivation to combine Sheffer and Tayloe was to improve the accuracy of Sheffer's idealized cell-boundary location estimate (Area C) by using Tayloe’s well-understood technique for determining actual, real-world cell coverage. The motivation to combine this improved system with Sanderford was to address the known problem of signal reflection errors, which Sheffer acknowledged but only addressed by assigning a lower confidence level. A POSITA would have looked to Sanderford’s neural network approach not just to identify errors, but to actively exploit multipath "fingerprints" to achieve a more accurate location estimate, particularly in urban areas where such reflections are common.

    • Expectation of Success: Petitioner argued that a POSITA would have had a reasonable expectation of success in combining these teachings. All three references operate in the same technical field of wireless location, address well-known problems like accuracy and multipath interference, and employ compatible and predictable technologies. The combination represented the application of known techniques (Tayloe’s mapping, Sanderford’s neural network) to improve a known system (Sheffer’s) to achieve predictable results.

4. Key Claim Construction Positions

  • "pattern recognizers": Petitioner argued this term, as used in the claims, is a means-plus-function term under 35 U.S.C. §112, ¶ 6, because it fails to recite sufficient corresponding structure. Petitioner proposed the function is "estimating a location of a mobile unit M using a pattern recognition technique." The corresponding structure disclosed in the ’525 patent’s specification was identified as a "location hypothesizing model (FOM) implemented on or by a location center or mobile base station," with specific examples including artificial neural networks and other learning models. This construction was central to Petitioner's argument that Sanderford's neural network supplied the claimed element.
  • "at least one of the following (i) and (ii)": This phrase precedes the two options for determining the final location estimate: (i) determining which estimate is preferred, and (ii) combining the estimates. Petitioner argued that the plain and ordinary meaning of this phrase is disjunctive, meaning at least one of (i) OR at least one of (ii). This interpretation was crucial, as Petitioner’s obviousness argument focused on showing that the prior art taught performing step (i) (preferring an estimate), making it unnecessary to also show that the art taught step (ii) (combining estimates) to prove obviousness.

5. Relief Requested

  • Petitioner requested that the Board institute an inter partes review for claims 1 and 6 of the ’525 patent and cancel these claims as unpatentable.