PTAB

IPR2025-00691

T-Mobile USA Inc v. Smart RF Inc

Key Events

Petition

petition

Title: Digital Multi-Band Predistortion Linearizer with Non-Linear Subsampling Algorithm in the Feedback Loop
Brief Description: The ’296 patent relates to a system for linearizing multi-band transmitters by using digital signal predistortion to compensate for nonlinearities in power amplifiers (PAs). The system employs an indirect learning architecture that uses concurrent feedback signals from both before and after the PA to model distortion and update the predistorters.

Prior Art Relied Upon: Peroulas (International Publication No. WO 2012/129768).
Core Argument for this Ground:
- Prior Art Mapping: Petitioner argued Peroulas disclosed a nearly identical multi-band transmitter with an indirect learning architecture that meets all limitations of the independent claims. Peroulas’s "MIMO Pre-Distortion Module" was asserted to be the claimed predistorter block that concurrently processes multiple input signals. The system includes a PA to amplify the predistorted signals. Critically, Peroulas’s "MIMO Capture Buffer" and "DSP Block" function as the claimed "analyzing and modelling stage," as they concurrently receive first feedback signals from the PA output and second feedback signals from the predistorter output to model PA nonlinearity and update the predistorter functions.
- Key Aspects: Petitioner contended that the core novelty claimed during prosecution—the use of concurrent feedback signals in an indirect learning architecture—was explicitly taught by Peroulas.

Prior Art Relied Upon: Peroulas (WO 2012/129768) and Cidronali (a 2009 workshop paper titled “A New Approach for Concurrent Dual-Band IF Digital PreDistortion”).
Core Argument for this Ground:
- Prior Art Mapping: This ground addressed dependent claims 8 and 9, which add a "single subsampling-based receiver" to the system disclosed in Peroulas. Cidronali was cited for its teaching of a single subsampling receiver, including an RF filter, a track and hold (T&H) block, and an analog-to-digital converter (ADC), used in a similar dual-band predistortion system.
- Motivation to Combine: Petitioner asserted a POSITA would combine Peroulas with Cidronali’s subsampling teachings for several well-understood reasons. These included reducing the implementation cost and complexity of the feedback loop, improving efficiency for multiband signals where Nyquist-rate sampling is inefficient, and gaining performance advantages from direct down-conversion.
- Expectation of Success: A POSITA would have had a reasonable expectation of success, as the combination involved implementing a known subsampling receiver (Cidronali) into a conventional predistortion feedback loop (Peroulas) using standard components for their established functions.

Prior Art Relied Upon: Posti (Patent 6,999,523).
Core Argument for this Ground:
- Prior Art Mapping: As an alternative to Peroulas, Petitioner argued Posti also disclosed a multi-frequency transmitter that rendered the independent claims obvious. Posti’s system includes a plurality of digital modulators that perform predistortion, a PA, and an "Adaptive Predistorter" that functions as the claimed "analyzing and modelling stage." This stage was alleged to use an indirect learning approach by comparing concurrent feedback signals from the PA output with signals from the digital modulator outputs to calculate and supply updated predistortion coefficients.
- Key Aspects: Petitioner argued that Posti’s teachings were applicable to multi-band systems and that its disclosure of comparing "like versions of the same signal" inherently requires the time alignment recited in claim 5.

Prior Art Relied Upon: Posti (Patent 6,999,523) and Cidronali (a 2009 workshop paper).
Core Argument for this Ground:
- Prior Art Mapping: Similar to Ground 2, this ground argued for the obviousness of combining Posti's predistortion system with Cidronali's subsampling receiver to meet the limitations of claims 8 and 9. Posti provided the base transmitter architecture, while Cidronali supplied the specific "single subsampling-based receiver."
- Motivation to Combine: The motivations were identical to those in Ground 2: a POSITA would have been motivated to replace Posti’s feedback path components with Cidronali’s more efficient and cost-effective subsampling receiver to reduce component count, simplify the architecture, and improve performance.
- Expectation of Success: Success was expected because the modification was a straightforward substitution of one known feedback and sampling architecture for another to achieve predictable benefits.

Petitioner contended that claim terms like "band" and "channel" should be given their plain and ordinary meaning, where a POSITA would understand a "band" is a continuous frequency range and a "channel" is a particular signal path within a band. It was argued that under this construction, the multi-band and multi-channel systems disclosed in Peroulas and Posti map directly to the claimed invention.

§325(d): Petitioner argued against denial because none of the asserted prior art references (Peroulas, Posti, Cidronali) were cited, discussed, or applied during the prosecution of the ’296 patent.
§314(a) (Fintiv Factors): Petitioner argued against discretionary denial under Fintiv, asserting that the parallel district court trial date was uncertain and likely to occur after the Final Written Decision deadline due to the case's complexity and multiple parties. Petitioner also noted that no substantive rulings or claim construction had occurred in the district court and stipulated that, if IPR is instituted, it would not pursue any ground raised or reasonably could have been raised in the parallel litigation.

Petitioner requests institution of an inter partes review and cancellation of claims 1-10 of the ’296 patent as unpatentable.