4:24-cv-00795
TurboCode LLC v. Ingenico Group SA
I. Executive Summary and Procedural Information
- Parties & Counsel:
- Plaintiff: TurboCode LLC (Texas)
- Defendant: Ingenico Group SA (France)
- Plaintiff’s Counsel: Direction IP Law
- Case Identification: 4:24-cv-00795, E.D. Tex., 09/02/2024
- Venue Allegations: Venue is alleged based on Defendant committing acts of infringement within the Eastern District of Texas, deriving revenue from Texas, and otherwise conducting business in the state.
- Core Dispute: Plaintiff alleges that Defendant’s point-of-sale terminals, which comply with 3G and 4G/LTE wireless standards, infringe a patent related to the architecture of high-speed turbo code decoders.
- Technical Context: The technology at issue is forward error correction (FEC) using "turbo codes," a method essential for ensuring reliable data transmission in high-speed wireless communication systems.
- Key Procedural History: The patent-in-suit, U.S. Patent No. 6,813,742, underwent an Ex Parte Reexamination, resulting in a Reexamination Certificate issued on February 10, 2009. The asserted claim was amended during this proceeding, which suggests its scope has been scrutinized by the U.S. Patent and Trademark Office post-issuance.
Case Timeline
| Date | Event |
|---|---|
| 2001-01-02 | '742 Patent Application Filing Date |
| 2004-11-02 | '742 Patent Issue Date |
| 2006-07-13 | '742 Patent Reexamination Request Filing Date |
| 2009-02-10 | '742 Patent Reexamination Certificate Issue Date |
| 2024-09-02 | Complaint Filing Date |
II. Technology and Patent(s)-in-Suit Analysis
U.S. Patent No. 6,813,742 - High Speed Turbo Codes Decoder for 3G Using Pipelined SISO Log-Map Decoders Architecture
The Invention Explained
- Problem Addressed: The patent describes prior art turbo code decoders as being too complex, costly, and power-intensive for practical use in consumer mobile devices like 3G cellular phones. These decoders required complex multiplication operations that were difficult and inefficient to implement in semiconductor (ASIC) chips. (’742 Patent, col. 1:45-61, col. 2:16-22).
- The Patented Solution: The invention proposes a more efficient decoder architecture using two serially connected "Soft-In/Soft-Out" (SISO) Log-MAP decoders operating in a pipelined, iterative loop. By performing calculations in the logarithmic domain, the design replaces complex multipliers with simpler, faster binary adder circuits. This architecture enables data to be passed back and forth between the two decoders for multiple iterations, improving decoding accuracy while being simpler, faster, and more power-efficient to implement in an ASIC. (’742 Patent, Abstract; col. 2:31-48).
- Technical Importance: This architectural approach was intended to make the powerful error-correction capabilities of turbo codes practical for mass-market, high-speed mobile communication devices by reducing hardware complexity and power consumption. (’742 Patent, col. 2:52-57).
Key Claims at a Glance
- The complaint asserts independent method claim 6, as amended by the Ex Parte Reexamination Certificate. (Compl. ¶13).
- The essential elements of reexamined claim 6 are:
- A method of iteratively decoding received baseband signals.
- Providing an input buffer with at least three shift registers to generate first, second, and third shifted input signals.
- Providing first and second soft decision decoders in a serially coupled circular circuit, where each decoder processes output from the preceding one, and where the decoders receive the shifted input signals from the buffer.
- Providing at least one memory module coupled to the decoders' outputs, with feedback from the memory associated with the second decoder back to the first decoder.
- Processing systematic and extrinsic information data using a MAP or logarithm approximation algorithm.
- Generating a soft decision based on a MAP or logarithm approximation algorithm.
- Weighing and storing the soft decision information into the memory module.
- Performing iterative decoding for a predetermined number of times, with an output from the last decoder fed back to the first in a circular circuit.
III. The Accused Instrumentality
Product Identification
The accused instrumentalities are Defendant's point-of-sale (POS) terminals, including but not limited to the Move 3500, Move 5000, APOS A8, Link 2500, iWL 250, and iWL 220 models. (Compl. ¶12).
Functionality and Market Context
- The accused products are handheld and mobile POS terminals designed for merchants. Their core accused functionality stems from their compliance with 3G and/or 4G/LTE wireless communication standards. (Compl. ¶12). The complaint alleges that to comply with these standards, the devices must necessarily implement a turbo decoder for error correction when processing received wireless signals. (Compl. ¶14, ¶17).
- The complaint includes a product datasheet for the APOS A8, describing it as a handheld terminal providing "full-spectrum wireless connectivity (4G, 3G, GPRS and WIFI)." (Compl. p. 5).
IV. Analysis of Infringement Allegations
'742 Patent Infringement Allegations
| Claim Element (from Independent Claim 6) | Alleged Infringing Functionality | Complaint Citation | Patent Citation |
|---|---|---|---|
| providing an input buffer comprising at least three shift registers, for receiving an input signal and generating first, second, and third shifted input signals; | The accused products allegedly include an input buffer structure with three component shift registers (or a software equivalent) to provide time-aligned inputs (e.g., systematic, parity 1, parity 2 data) for decoding computations, as required by 3GPP standards. A diagram from an academic paper shows distinct input buffers for a turbo decoder architecture. | ¶20, ¶22, p. 18 | col. 4:55-65 |
| providing first and second soft decision decoders serially coupled in a circular circuit, wherein each decoder processes soft decision from the preceding decoder output data... | Alleged to be met by implementing the standard turbo decoder architecture, which contains two constituent SISO decoders (e.g., "SISO 1" and "SISO 2") coupled serially and operating in an iterative, circular fashion. A diagram from the 3GPP standard shows two constituent decoders. | ¶23-24, p. 10 | col. 4:8-14 |
| providing at least one memory module coupled to an output of each of the first and second soft decision decoders, wherein the output of the memory module associated with the second soft decision decoder is fed back as an input of the first soft decision decoder; | The accused products allegedly use memory modules, such as interleavers and deinterleavers, coupled to the decoders. The output of the second decoder's memory module ("soft output 2") is fed back to the input of the first decoder to enable iterative processing. | ¶27-28, p. 23 | col. 4:10-14 |
| processing systematic information data and extrinsic information data using the maximum a posteriori (AP) probability algorithm, and/or logarithm approximation algorithm; | The accused products allegedly use the Bahl-Cocke-Jelinek-Raviv (BCJR) algorithm, which the complaint identifies as a MAP probability algorithm, to process systematic and extrinsic data as part of the turbo decoding process required by the standards. | ¶30, ¶32 | col. 6:40-54 |
| generating soft decision based on the maximum a posteriori (MAP) probability algorithm, and/or logarithm approximation algorithm; | The BCJR algorithm used in the accused products allegedly generates soft decision outputs (e.g., "soft output 1" and "soft output 2") based on its MAP calculations. | ¶35 | col. 6:25-33 |
| weighing and storing soft decision information into the corresponding memory module; | The soft decision information is allegedly "weighed" (normalized) before being stored in memory modules (interleaver/deinterleaver). The complaint points to source code examples from the 3GPP standard showing normalization of soft decision values. | ¶36-37, ¶39 | US 6,813,742 C1, col. 2:42-44 |
| performing, for a predetermined number of times, iterative decoding from the first to the last of multiple decoders, wherein an output from the last soft decision decoder is fed back as an input to the first soft decision decoder...and propagate to the last decoder in a circular circuit. | The accused products allegedly perform a predetermined number of decoding iterations (e.g., 8 iterations as defined by "FEC_ITERATIONS" in a reference source code) in a circular feedback loop, as is inherent to turbo decoding. The complaint includes a standard "Turbo decoder" diagram illustrating this iterative feedback loop. | ¶40, ¶42, p. 41 | col. 9:35-49 |
- Identified Points of Contention:
- Scope Questions: The complaint's infringement theory rests heavily on the accused products' compliance with 3G/4G/LTE standards. A primary point of contention may be whether compliance with a functional standard is sufficient to prove infringement of the patent's more specific architectural claims (e.g., "at least three shift registers," "decoders serially coupled"). This raises the question of whether the standards mandate the claimed structure or merely a functional outcome that could be achieved through non-infringing implementations.
- Technical Questions: What evidence does the complaint provide that the accused products' actual, physical chip implementations map onto the reference diagrams and source code from standards bodies and academic papers? The analysis will likely focus on whether the highly integrated System-on-a-Chip (SoC) designs in modern devices contain the discrete "decoders," "shift registers," and "memory modules" as claimed, or if they use a unified, reconfigurable processing block that achieves a similar result.
V. Key Claim Terms for Construction
The Term: "weighing...soft decision information"
Context and Importance: This term was added to claim 6 during reexamination, suggesting it was critical for establishing patentability. Its construction will be central to the infringement analysis. The complaint alleges this limitation is met by "normalizing" soft decision values. Practitioners may focus on this term because the defense could argue that "normalization" or the value-clipping functions cited in the complaint's evidence do not constitute "weighing" as understood by a person of ordinary skill in the art.
Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: The term itself is not explicitly defined in the patent, which may support an argument that it should be given its plain and ordinary meaning, potentially encompassing any form of scaling or adjustment like normalization.
- Evidence for a Narrower Interpretation: The absence of the term "weighing" in the original patent specification could be used to argue its meaning should be limited. A defendant may argue that since the specification does not provide an explicit definition or example, the term should be narrowly construed to require a specific mathematical operation distinct from the "normalization" shown in the complaint's extrinsic evidence.
The Term: "input buffer comprising at least three shift registers"
Context and Importance: This is a specific structural limitation at the front-end of the claimed method. Infringement will depend on whether the accused devices, likely built on integrated SoCs, possess a structure that meets this definition. The complaint alleges this can be met by hardware or an equivalent software implementation (Compl. ¶22).
Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: Plaintiff may argue the term should be construed functionally to cover any hardware or software structure that provides the three separate, time-aligned data streams (e.g., systematic, parity 1, parity 2) needed for the decoders to operate.
- Evidence for a Narrower Interpretation: The patent includes Figure 5, which explicitly depicts three distinct "N-BIT SHIFT REGISTER" blocks. A defendant may use this figure to argue for a narrower structural interpretation requiring three physically or logically distinct hardware registers, as opposed to a partitioned general-purpose memory buffer. (’742 Patent, FIG. 5).
VI. Other Allegations
- Willful Infringement: The complaint seeks a finding of willful infringement and enhanced damages. (Compl. ¶V.c). However, the complaint's factual allegations do not specify a basis for this claim, such as alleging that the Defendant had knowledge of the '742 Patent prior to the lawsuit being filed.
VII. Analyst’s Conclusion: Key Questions for the Case
- A core issue will be one of Standard Compliance versus Structural Infringement: Can the plaintiff prove infringement of the patent’s specific hardware-oriented claims primarily by demonstrating the defendant's compliance with the functional requirements of the 3G/4G/LTE standards? The case will test whether the standards mandate the claimed architecture or merely a result that could be achieved through other means.
- A second central issue will be one of Definitional Scope: The case may turn on the construction of "weighing," a term added during reexamination. The key question is whether the "normalization" and value-clipping functions, which the complaint alleges are performed by the accused devices, fall within the scope of this term, or if "weighing" requires a more specific mathematical operation not present.
- A key evidentiary question will be one of Implementation versus Abstraction: Does the evidence presented—consisting of high-level block diagrams from standards documents and academic papers—accurately reflect the actual, physical implementation of the turbo decoder within the defendant's proprietary, highly-integrated chipsets?