Samsung Electronics Co Ltd v. Red Rock Analytics LLC

1. Case Identification

• Patent #: 7,346,313
• Filed: February 5, 2018
• Petitioner(s): Samsung Electronics Co., Ltd.
• Patent Owner(s): Red Rock Analytics, LLC
• Challenged Claims: 7, 11-21, 44, 48-58

2. Patent Overview

• Title: I-Q Imbalance Calibration in Transceivers
• Brief Description: The ’313 patent describes a system and method for balancing gain between in-phase (I) and quadrature (Q) channels in a radio frequency (RF) transceiver. The technology involves injecting a calibration RF signal from the transmit chain output back to the receive chain input to measure and correct for I-Q gain imbalance independently in each chain.

3. Grounds for Unpatentability

Ground 1: Claims 7, 15, 16, 21, 44, 52, 53, and 58 are obvious over Yellin in view of Su.

• Prior Art Relied Upon: Yellin (Patent 6,898,252) and Su (Patent 6,272,322).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that Yellin discloses the core inventive concept: a communication device that corrects for I-Q mismatch by feeding a signal through a transmit path, injecting it into a receive path, and using a "mismatch trainer" to adjust gain and phase parameters based on the output. However, Yellin does not detail a complete transceiver. Su was argued to cure this deficiency by disclosing a conventional direct-conversion transceiver architecture, including elements recited in the claims but not detailed in Yellin. Petitioner contended that in the combined system, Yellin's input vectors (Vm) serve as the 'baseband transmit signal,' Su’s attenuators provide the 'baseband I-Q amplification subsystem,' and Su’s mixers and combiner provide the 'direct-conversion subsystem.' For the receive chain, Yellin’s IQ demodulator acts as the direct-conversion subsystem, and its mismatch trainer (377) functions as the claimed 'processor' that creates an observable indicator of I-Q imbalance (a cost function) and acts as the 'channel gain adjuster' by independently varying transmit and receive parameters to correct the imbalance.
  • Motivation to Combine: A POSITA would combine Yellin and Su because both address improving performance in wireless communication devices. It would have been an obvious application of a known technique (Yellin's I-Q mismatch correction) to a known, suitable device (Su's transceiver) to yield the predictable result of a transceiver with reduced I-Q gain imbalance.
  • Expectation of Success: Petitioner asserted a high expectation of success, arguing the combination involves implementing a known calibration technique on a standard transceiver platform without altering the fundamental principles of either reference.

Ground 2: Claims 11, 17, 48, and 54 are obvious over Yellin and Su in view of Faulkner.

• Prior Art Relied Upon: Yellin (Patent 6,898,252), Su (Patent 6,272,322), and Faulkner (an Electronics Letters article from 1991).
• Core Argument for this Ground:
  • Prior Art Mapping: This ground builds on the Yellin/Su combination to address dependent claim 11, which requires the calibration signal to include "a sequence of pulses taking on purely real or imaginary values." Yellin generally refers to a "special transmission" for calibration without providing specifics. Petitioner asserted that Faulkner explicitly teaches using purely real (A, 0) and purely imaginary (0, A) test vectors to measure and correct for I-Q gain mismatch.
  • Motivation to Combine: A POSITA would be motivated to look to Faulkner to implement Yellin's generic "special transmission." Faulkner provides a well-known, specific, and computationally simple method for isolating and correcting gain mismatch, making it an obvious and advantageous choice for refining the Yellin/Su system.

Ground 3: Claims 12, 13, 18, 19, 49, 50, 55, and 56 are obvious over Yellin and Su in view of Sewerinson.

• Prior Art Relied Upon: Yellin (Patent 6,898,252), Su (Patent 6,272,322), and Sewerinson (Patent 4,613,976).
• Core Argument for this Ground:
  • Prior Art Mapping: This ground addresses claim 12, which requires the calibration signal to include a "sampled phasor." While Yellin is agnostic to modulation type, Petitioner argued that Sewerinson discloses a Quadrature Phase Shift Keying (QPSK) modulation system where input data is mapped to specific I-Q vectors on a constellation diagram. Each of these vectors represents a discrete, sampled "phasor" with a specific phase and constant magnitude.
  • Motivation to Combine: A POSITA would find it obvious to use the QPSK modulation method from Sewerinson in the Yellin/Su transceiver. Yellin's calibration works with various modulations, and QPSK was a widely used, well-known industry standard. It represented a routine design choice for a POSITA seeking to implement a complete communication system.

Ground 4: Claims 14, 20, 51, and 57 are obvious over Yellin, Su, and Sewerinson in view of Haykin.

• Prior Art Relied Upon: Yellin (Patent 6,898,252), Su (Patent 6,272,322), Sewerinson (Patent 4,613,976), and Haykin (a 2000 textbook on communication systems).
• Core Argument for this Ground:
  • Prior Art Mapping: This ground addresses claim 14's requirement for a discrete phasor "comprising j^n or j^-n." The QPSK constellation in Sewerinson has phasors at 45-degree offsets. Haykin, a foundational textbook, was cited to show two "commonly used" QPSK constellations. Petitioner argued that one of these known alternatives places the phasors directly on the real and imaginary axes (0°, 90°, 180°, 270°), which correspond mathematically to values of 1, j, -1, and -j (i.e., powers of j, or j^n).
  • Motivation to Combine: A POSITA would have recognized the on-axis constellation from Haykin as a simple, known design alternative to the one shown in Sewerinson. Petitioner argued that substituting one known and functionally equivalent QPSK constellation for another was an obvious design choice that would produce the same expected result of representing modulated data.

4. Arguments Regarding Discretionary Denial

• Petitioner argued that institution should not be denied under 35 U.S.C. § 325(d). It acknowledged a previously denied IPR petition against the same patent (IPR2017-01490) but contended that the instant petition relies on different primary prior art (Yellin) than the denied petition (which relied on a Warner patent). Therefore, Petitioner asserted that the "same or substantially the same prior art or arguments" were not previously presented to the Office for consideration.

5. Relief Requested

• Petitioner requested institution of an inter partes review and cancellation of claims 7, 11-21, 44, and 48-58 of Patent 7,346,313 as unpatentable under 35 U.S.C. §103.