Ericsson Inc v. Sol IP LLC

1. Case Identification

• Case #: IPR2020-00244
• Patent #: 8,971,168
• Filed: December 10, 2019
• Petitioner(s): Ericsson Inc.
• Patent Owner(s): Electronics and Telecommunications Research Institute
• Challenged Claims: 18-23 and 25

2. Patent Overview

• Title: Carrier Aggregation in Wireless Communication Systems
• Brief Description: The ’168 patent discloses a method for wireless communication where data symbols are transmitted in two time slots using Discrete Fourier Transform Spread Orthogonal Frequency Division Multiplexing (DFT-S-OFDM). The method involves multiplying data symbols with specific orthogonal sequences, using length-5 DFT sequences for a first slot and length-4 Walsh sequences for a second, shortened slot.

3. Grounds for Unpatentability

Ground 1: Claims 18-23 and 25 are obvious over Ericsson-2 in view of Baldemair and 36.211 v8.7.0.

• Prior Art Relied Upon: Ericsson-2 (3GPP Contribution R1-103506), Baldemair (Patent 8,638,880), and 36.211 v8.7.0 (3GPP Technical Specification Release 8).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that the combination of references taught every limitation of the challenged claims. Independent claim 18 recites a method of receiving a first slot with five DFT-S-OFDM symbols and a second slot with four DFT-S-OFDM symbols. Data symbols in each slot are multiplied by scrambling and orthogonal sequences. The first orthogonal sequence is selected from a specific set of length-5 DFT sequences, the second from a specific set of length-4 Walsh sequences, and the sequence index for both is the same.
    • Petitioner asserted that Ericsson-2, a 3GPP contribution for a new Physical Uplink Control Channel (PUCCH) format, taught nearly all of these elements. It disclosed a two-slot transmission scheme with five DFT-S-OFDM symbols in the first slot and a "punctured" second slot with four symbols. Ericsson-2 also taught using length-5 DFT sequences for the first slot and length-4 DFT sequences for the second, as well as applying cell-specific scrambling sequences to the data bits before modulation.
    • To address the claim limitation of applying scrambling sequences to data symbols (post-modulation), Petitioner relied on Baldemair. Baldemair taught that for better interference randomization, scrambling could be applied at the symbol level rather than the bit level, and that this was "mathematically equivalent" to pre-modulation scrambling.
    • To meet the requirement of using length-4 Walsh sequences for the second slot (instead of the DFT sequences taught by Ericsson-2), Petitioner cited 36.211 v8.7.0. This LTE standard disclosed the use of the exact length-4 Walsh sequences recited in the ’168 patent for PUCCH transmissions, making their substitution for the similar DFT sequences in Ericsson-2 a predictable design choice.
    • For the final limitation—that the sequence index for the first and second slots is the same—Petitioner argued this was the simplest and most obvious implementation. Since Ericsson-2 did not suggest using different indices, a person of ordinary skill in the art (POSITA) would default to this most straightforward approach.
  • Motivation to Combine: A POSITA would combine these references because they all address the same specific technical problem: defining a new PUCCH format for carrier aggregation in LTE Release 10. Petitioner argued that Ericsson-2 and Baldemair were contemporaneous proposals that built upon the foundational 36.211 v8.7.0 standard. A POSITA would look to Baldemair to improve the interference randomization of the system proposed in Ericsson-2 by moving scrambling to the symbol level. The POSITA would then consult the existing 36.211 v8.7.0 standard for well-known and workable orthogonal sequences, such as the specified Walsh sequences, to implement in the second slot of the Ericsson-2 framework, as they offered equivalent performance and were a known alternative to DFT sequences.
  • Expectation of Success: Petitioner contended a POSITA would have a high expectation of success. The proposed modifications were characterized as slight, involving the substitution of known, equivalent components within a standard framework. Baldemair explicitly taught that symbol-level scrambling was mathematically equivalent to bit-level scrambling, ensuring predictable results. The interchangeability of DFT and Walsh sequences for providing orthogonality was also well-understood in the art, as evidenced by their alternative use in the 36.211 v8.7.0 standard itself.

4. Arguments Regarding Discretionary Denial

• Petitioner filed this petition concurrently with a second petition (IPR2020-00243) against the same patent and argued against discretionary denial under §314(a). Petitioner asserted that filing two petitions was justified because the grounds relied on different primary references (Ericsson-2, a 3GPP submission, versus Nazar, a §102(e) patent in the other petition), each requiring distinct and lengthy analysis. The complexity of the technology, coupled with a necessary discussion of the patent’s priority date due to alleged misrepresentations by the Patent Owner during prosecution, were presented as further reasons justifying the separate filings.

5. Relief Requested

• Petitioner requested institution of an inter partes review and cancellation of claims 18-23 and 25 of Patent 8,971,168 as unpatentable under 35 U.S.C. §103.