US 8374358

Here is a concise summary of US Patent 8,374,358.

Title: Method for determining a noise reference signal for noise compensation and/or noise reduction

Assignee: The current assignee of record is Cerence Operating Co. The original assignee was Nuance Communications Inc.

Inventors: Markus Buck, Tobias Wolff, Toby Christian Lawin-Ore, Samuel Ngouoko Mboungueng, Gerhard Schmidt

Filing Date: March 29, 2010

Issue Date: February 12, 2013

Abstract: The invention provides a method for determining a noise reference signal for noise compensation and/or noise reduction. A first audio signal on a first signal path and a second audio signal on a second signal path are received. The first audio signal is filtered using a first adaptive filter to obtain a first filtered audio signal. The second audio signal is filtered using a second adaptive filter to obtain a second filtered audio signal. The first and the second filtered audio signal are combined to obtain the noise reference signal. The first and the second adaptive filter are adapted such as to minimize a wanted signal component in the noise reference signal.

Plain-Language Overview of Independent Claims:

This patent has three independent claims:

Claim 1: A method for determining a noise reference signal.
This claim describes a method to create a clean "noise reference" signal, which is a signal that contains only background noise without a desired sound, like speech. The method involves taking two audio signals from two different paths. Each of these signals is passed through its own adaptive filter (a filter that can adjust its properties). The outputs of these two filters are then combined to produce the final noise reference signal. The key part of this method is that the two adaptive filters are continuously adjusted to remove as much of the desired sound (the "wanted signal") as possible from the final combined signal, leaving primarily the background noise.

Claim 16: A method for processing an audio signal for noise compensation.
This claim describes a full noise-cancellation process that uses the noise reference signal created by the method in Claim 1. It starts by generating the noise reference signal as described above. This noise reference signal is then passed through a third adaptive filter. The output of this third filter, which is an estimate of the noise present in the original audio, is then subtracted from the first original audio signal. The result is a final audio signal with the background noise significantly reduced.

Claim 20: A system for audio signal processing.
This claim describes a physical system designed to perform the method outlined in the previous claims. The system includes a receiver to get the two initial audio signals. It has a first adaptive filter for the first signal and a second adaptive filter for the second signal. A subtractor (or a similar component) is included to combine the outputs of these two filters to create the noise reference signal. The system is designed so that the two adaptive filters can be adjusted to minimize the desired audio signal in their combined output.

A search of the CAFC (Court of Appeals for the Federal Circuit) 2026 dockets for "US patent 8374358" did not yield any specific results for this patent number.

Known Litigation for US Patent 8,374,358

As of May 9, 2026, research has identified one litigation case involving US Patent 8,374,358.

---

Case 1: Cerence Operating Company v. Amazon.com, Inc. et al.

• Plaintiff(s): Cerence Operating Company
• Defendant(s): Amazon.com, Inc., Amazon.com Services LLC, and Amazon Web Services, Inc.
• Jurisdiction: U.S. District Court for the Eastern District of Texas
• Case Number: 2:2026cv00373
• Filing Date: May 4, 2026
• Status: The complaint was recently filed. In the complaint, Cerence Operating Company alleges patent infringement and has included US Patent 8,374,358 as one of seven patents in the suit. A claim chart specifically for US Patent 8,374,358 was attached as an exhibit to the initial complaint.

Proceedings overview

As of May 29, 2026, there are no AIA trial proceedings (Inter Partes Review, Post-Grant Review, or Covered Business Method review) on file for US Patent 8,374,358. This indicates that the patent has not been challenged at the Patent Trial and Appeal Board (PTAB) via these post-grant review mechanisms.

Strategic summary

Currently, all claims of US patent 8,374,358 remain untested in AIA trial proceedings at the PTAB. There are no claims that have been canceled, sustained, or modified through IPR, PGR, or CBM reviews. This means that a defendant currently facing assertion of this patent would not benefit from any prior invalidations at the PTAB. All original claims (1-20) are presumed valid by the PTAB for the purposes of these proceedings, as no challenges have been successfully mounted.

The absence of PTAB activity suggests that either the patent has not been heavily asserted in scenarios that would provoke an IPR filing, or potential challengers have opted not to pursue such avenues. This leaves the entire scope of the patent's claims open for challenge in district court or for future PTAB petitions.

Recommended next steps

Since there is no PTAB activity, a defendant facing assertion of US Patent 8,374,358 would need to consider a de novo validity challenge. This could involve:

• Filing an Inter Partes Review (IPR) petition: If the defendant identifies prior art consisting of patents or printed publications that raise a reasonable likelihood of unpatentability under 35 U.S.C. §§ 102 or 103 for any of the claims. An IPR can be filed after 9 months of the patent grant (February 12, 2013).
• Conducting a thorough prior art search: A comprehensive search for prior art beyond what was considered during prosecution, especially given the "Obviousness" analysis provided earlier, could uncover strong grounds for an IPR or a validity defense in district court.
• Evaluating Post-Grant Review (PGR) or Covered Business Method (CBM) eligibility: While less likely given the technology, a careful analysis should be done. PGR allows challenges on any ground of invalidity under 35 U.S.C. § 282(b)(2) or (3), but must be filed within 9 months of the patent's grant (which has passed for this patent). CBM review applies to patents claiming methods or apparatus for performing data processing used in financial products or services, which this patent does not appear to be, and it also requires the petitioner to have been sued or charged with infringement.
• Monitoring future PTAB filings: Keep an eye on the USPTO's Patent Trial and Appeal Case Tracking System (P-TACTS) for any new petitions filed against this patent.
• Coordinating with ongoing district court litigation: As there is a known district court case (Cerence Operating Company v. Amazon.com, Inc. et al., Case Number: 2:2026cv00373), any decision to file an IPR would need to consider the district court's schedule and the potential for a stay of litigation pending PTAB review.

Inventors

• Markus Buck: Employee of Nuance Communications Inc. at the time of filing.
• Tobias Wolff: Employee of Nuance Communications Inc. at the time of filing.
• Toby Christian Lawin-Ore: Employee of Nuance Communications Inc. at the time of filing.
• Samuel Ngouoko Mboungueng: Employee of Nuance Communications Inc. at the time of filing.
• Gerhard Schmidt: Employee of Nuance Communications Inc. at the time of filing.

It is common practice for inventors to assign their rights to their employer at the time of filing, which is reflected in the initial assignment to Nuance Communications Inc. on the patent's filing date.

Original assignee

The original assignee named on the issued patent is Nuance Communications Inc.

Nuance Communications Inc. was primarily involved in developing artificial intelligence, speech recognition, and natural language processing software and services. They shipped various products incorporating speech processing technologies, including solutions for healthcare, automotive, and customer engagement.

Nuance Communications Inc. was acquired by Microsoft Corporation in 2022. However, the intellectual property, including US Patent 8,374,358, was spun off into Cerence Inc. in 2019, prior to the Microsoft acquisition.

Assignment timeline

• 2010-03-29 (executed) / recorded 2010-03-30 — Reel 024470/0947

  • Conveyance: Assignment
  • Assignor: BUCK, MARKUS; MBOUNGUENG, SAMUEL NGOUOKO; SCHMIDT, GERHARD; WOLFF, TOBIAS; LAWIN-ORE, TOBY CHRISTIAN (inventors)
  • Assignee: NUANCE COMMUNICATIONS, INC.
  • Correspondent: KENNETH J. MCLEAN, NUANCE COMMUNICATIONS, INC., 1 AVP ROAD, BURLINGTON, MA 01803
  • Context: Initial assignment of patent rights from the inventors to their employer.

• 2019-10-18 (executed) / recorded 2019-10-23 — Reel 050836/0191

  • Conveyance: Assignment
  • Assignor: NUANCE COMMUNICATIONS, INC.
  • Assignee: CERENCE INC.
  • Correspondent: LORNA C. BRYANT, Cerence Inc., 15 Wayside Road, Burlington, MA 01803. This correspondent recurs in subsequent transfers involving Cerence entities.
  • Context: Transfer of intellectual property as part of the spin-off of Cerence Inc. from Nuance Communications Inc.

• 2019-10-28 (executed) / recorded 2019-10-29 — Reel 050860/0313

  • Conveyance: Correction Assignment
  • Assignor: NUANCE COMMUNICATIONS, INC.
  • Assignee: CERENCE OPERATING COMPANY
  • Correspondent: LORNA C. BRYANT, Cerence Inc., 15 Wayside Road, Burlington, MA 01803. This correspondent recurs in subsequent transfers involving Cerence entities.
  • Context: Corrective assignment to accurately reflect the assignee name, transferring to Cerence Operating Company.

• 2019-11-01 (executed) / recorded 2019-11-07 — Reel 050901/0979

  • Conveyance: Security Agreement
  • Assignor: CERENCE OPERATING COMPANY
  • Assignee: BARCLAYS BANK PLC
  • Correspondent: CHIA-LUEI WANG, BARCLAYS BANK PLC, 745 7TH AVE, NY, NY 10019
  • Context: Cerence Operating Company pledges patents as collateral for a financing agreement.

• 2020-06-12 (executed) / recorded 2020-06-12 — Reel 051770/0097

  • Conveyance: Release
  • Assignor: BARCLAYS BANK PLC
  • Assignee: CERENCE OPERATING COMPANY
  • Correspondent: JENNIFER D. RHEIN, SKADDEN ARPS SLATE MEAGHER & FLOM LLP, ONE MANHATTAN WEST, NEW YORK, NY 10001
  • Context: Barclays Bank PLC releases its security interest in the patent.

• 2020-06-12 (executed) / recorded 2020-06-15 — Reel 051770/0256

  • Conveyance: Security Agreement
  • Assignor: CERENCE OPERATING COMPANY
  • Assignee: WELLS FARGO BANK, N.A.
  • Correspondent: JOSHUA LINDSTROM, WELLS FARGO BANK, NATIONAL ASSOCIATION, 800 WALNUT ST., DES MOINES, IA 50309
  • Context: Cerence Operating Company pledges patents as collateral for a new financing agreement with Wells Fargo.

• 2022-04-19 (executed) / recorded 2022-04-19 — Reel 052935/0586

  • Conveyance: Correction Assignment
  • Assignor: NUANCE COMMUNICATIONS, INC.
  • Assignee: CERENCE OPERATING COMPANY
  • Correspondent: LORNA C. BRYANT, CERENCE INC., 15 WAYSIDE ROAD, BURLINGTON, MA 01803. This correspondent recurs in previous transfers involving Cerence entities.
  • Context: Further corrective assignment related to the initial spin-off transfer from Nuance to Cerence.

• 2024-12-23 (executed) / recorded 2025-01-02 — Reel 055106/0424

  • Conveyance: Release
  • Assignor: WELLS FARGO BANK, NATIONAL ASSOCIATION
  • Assignee: CERENCE OPERATING COMPANY
  • Correspondent: MICHELLE G. MOODY, WELLS FARGO BANK, N.A., LAW DEPARTMENT, 800 WALNUT ST, MAC D1053-030, DES MOINES, IA 50309
  • Context: Wells Fargo Bank, N.A. releases its security interest in the patent.

Timeline diagram

timeline
    title Ownership of US 8374358
    2010 : Filed by Nuance Communications
    2013 : Issued
    2019 : Assigned to Cerence Inc
         : Assigned to Cerence Operating Co
         : Security Agreement with Barclays
    2020 : Release by Barclays
         : Security Agreement with Wells Fargo
    2022 : Corrective Assignment
    2025 : Release by Wells Fargo
    2026 : Infringement suit filed

NPE / troll-pattern signals

1. Shell-entity transfer — not present. The transfers are between Nuance Communications Inc., Cerence Inc., and Cerence Operating Company, all of which are operating companies. The security agreements are with established banks.
2. Known asserter in the chain — not present. Cerence Operating Company is not listed as a known NPE.
3. Repeat correspondent across the chain — present. Lorna C. Bryant of Cerence Inc. appears as the correspondent for the assignment to Cerence Inc. (2019-10-23, Reel 050836/0191), the corrective assignment to Cerence Operating Company (2019-10-29, Reel 050860/0313), and another corrective assignment to Cerence Operating Company (2022-04-19, Reel 052935/0586).
4. Cascading transfers — not present. While there are multiple transfers/recordings in 2019 and 2020, they are related to a corporate spin-off and subsequent financing agreements, not rapid, consecutive transfers through unrelated shell entities.
5. Pre-litigation transfer — not present. The most recent assignment (release of security interest) was in 2025-01-02, which is more than 6 months before the infringement suit filing date of May 4, 2026.
6. Bankruptcy fire-sale — not present. The transfer from Nuance to Cerence was a corporate spin-off, not a bankruptcy proceeding.
7. Privateering — unclear. While Cerence is an operating company, without further information on the specific nature of its relationship with Nuance post-spin or the strategic intent behind the assertion against Amazon, a definitive call cannot be made. However, the current litigation context suggests a direct assertion by an operating company.
8. Defensive aggregator (anti-NPE) — not present. The chain does not terminate at a known defensive aggregator.

Verdict

Operating-company assertion

The current assignee, Cerence Operating Company, is an operating company that spun off from Nuance Communications Inc. The patent is being asserted in an infringement suit against Amazon.com, Inc. et al., which are other operating companies. The recorded assignments primarily reflect corporate reorganization (Nuance to Cerence, Reel 050836/0191 and Reel 050860/0313) and routine corporate financing (security agreements with Barclays and Wells Fargo, followed by releases). The recurrence of the same correspondent for Cerence entity transfers (Lorna C. Bryant) indicates internal legal handling for an operating entity rather than a multi-entity NPE operation.

USPTO Assignment Center Search for US8374358

Prior Art Analysis for US Patent 8,374,358

This analysis details the prior art references cited by the USPTO examiner during the prosecution of US patent 8,374,358. Each reference is examined for its potential to anticipate the claims of the '358 patent under 35 U.S.C. § 102.

A prior art reference anticipates a patent claim if it discloses, either explicitly or inherently, each and every element of that claim. The '358 patent's independent claims (1, 16, and 20) center on a method and system for generating a noise reference signal by using two adaptive filters on two separate audio signals and adapting the filters to minimize a "wanted" signal component in the combined output.

---

U.S. Patent Citations

1. US Patent 7,248,701 B2

• Full Citation: US Patent 7,248,701 B2, "System and method for adaptively generating a noise reference in a multi-microphone environment," filed by Chen et al. on February 15, 2005, and issued on July 24, 2007. Assigned to Sony Corporation.
• Brief Description: This patent describes a system for generating a noise reference signal for noise cancellation in a multi-microphone setup. It uses a primary microphone signal and one or more secondary microphone signals. An adaptive filter is used to model the transfer function between the desired speech signal at the primary and secondary microphones. This model is then used to subtract the estimated speech component from a secondary microphone signal, thereby creating a noise reference.
• Potential Anticipation: This reference is highly relevant. It discloses receiving multiple audio signals, using an adaptive filter to estimate a wanted signal component, and subtracting it to generate a noise reference. However, a key distinction from claim 1 of the '358 patent is that the '701 patent appears to use a single adaptive filter structure to model the relationship and create the noise reference (as depicted in its figures and description), rather than filtering both the first and second audio signals with separate adaptive filters which are then combined. The '358 patent's use of two distinct adaptive filters (H1 and H2), one for each signal path, which are adapted to model the cross-path transfer functions (G2 and G1 respectively), is a specific implementation not clearly taught in '701. Therefore, while close, it may not anticipate claim 1's requirement of filtering both signals with distinct adaptive filters before combination.

2. US Patent 7,515,708 B2

• Full Citation: US Patent 7,515,708 B2, "Method and apparatus for improving signal quality from a microphone array," filed by Bitzer et al. on August 31, 2005, and issued on April 7, 2009. Assigned to Starkey Laboratories, Inc.
• Brief Description: This patent details a method for improving the signal-to-noise ratio in microphone arrays, particularly for hearing aids. It describes forming a beamformed signal and then using an adaptive filter to remove noise. The system can adaptively switch between different beamforming modes. The noise reference for the adaptive filter is generated by creating a null in the direction of the desired sound source.
• Potential Anticipation: The '708 patent discusses creating a noise reference by steering a null towards the speaker, which is a form of blocking the wanted signal. It uses adaptive filtering to then cancel noise from a primary signal. However, it does not explicitly describe the '358 patent's core novelty: receiving a first and second audio signal, filtering the first with a first adaptive filter, filtering the second with a second adaptive filter, and then combining the two filtered outputs. The '708 patent's method of generating the noise reference appears to be based on beamforming techniques rather than the specific dual-adaptive-filter structure claimed in the '358 patent. Thus, it is unlikely to anticipate the independent claims.

3. US Patent 7,929,915 B2

• Full Citation: US Patent 7,929,915 B2, "Apparatus and method for canceling noise," filed by Kim et al. on December 22, 2005, and issued on April 19, 2011. Assigned to [LG Electronics Inc.](/litigations/by-plaintiff/LG%20Electronics%20Inc.)
• Brief Description: This patent discloses a noise-canceling apparatus that uses two microphones. The signal from one microphone is delayed and then processed with an adaptive filter. The output is subtracted from the other microphone's signal to cancel out the noise. The system aims to estimate the noise in the primary signal based on the secondary signal.
• Potential Anticipation: This reference describes a classic adaptive noise cancellation setup with two microphones and one adaptive filter, similar to the prior art system shown in FIG. 7 of the '358 patent itself. It does not teach the use of two adaptive filters, one for each input signal, which are then combined. The method described in claim 1, involving filtering the first audio signal with a first adaptive filter and the second audio signal with a second adaptive filter, is absent. Therefore, this patent does not anticipate the independent claims of the '358 patent.

4. US Patent Application Publication 2008/0285786 A1

• Full Citation: US 2008/0285786 A1, "Method and arrangement for noise reduction," filed by Baumgarte et al. on May 16, 2007, and published on November 20, 2008.
• Brief Description: This publication describes a noise reduction system using a microphone array. It focuses on creating a noise reference signal by forming a beam that has a null directed at the desired sound source. This noise reference is then adaptively filtered and subtracted from a main signal beam to reduce noise.
• Potential Anticipation: Similar to the '708 patent, this reference focuses on beamforming techniques to generate a noise reference by creating a null. This is a different approach from the specific two-signal, two-filter architecture claimed in the '358 patent. It does not teach filtering both a first and second audio signal with their own respective adaptive filters and then combining the results. Consequently, it does not anticipate claims 1, 16, or 20.

---

Non-Patent Literature

1. "A robust adaptive beamformer for microphone arrays with a blocking matrix using constrained adaptive filters" by O. Hoshuyama et al., IEEE Transactions on Signal Processing, Vol. 47, No. 10, October 1999.

• Brief Description: This article, cited in the '358 patent's background section, describes a generalized sidelobe canceller (GSC) with an adaptive blocking matrix. The goal is to create noise references that are free of the desired signal. It uses adaptive filters within the blocking matrix to subtract the estimated desired signal from microphone inputs.
• Potential Anticipation: This reference is highly relevant and describes an advanced method for generating a noise reference. The system in Hoshuyama uses a primary beamformed signal and subtracts filtered versions of it from the individual microphone signals. This can be interpreted as having a filter for each channel. However, the architecture is that of a GSC, where a reference signal (the beamformer output) is filtered and subtracted from other signals. The '358 patent claims a more symmetric structure where two arbitrary audio signals are each passed through their own adaptive filters, and the outputs are combined. The adaptation goal in the '358 patent is to make the transfer function of the first filter model the transfer characteristics of the second signal path, and vice versa. This specific approach of creating a pole-free transfer function to block the wanted signal is the novel step argued in the '358 patent. While Hoshuyama is very close, the specific topology and adaptation goal as described in claim 1 may be considered distinct.

---

In summary, while the cited prior art addresses the same general problem of creating a noise reference signal in multi-microphone systems, none of the references appear to explicitly disclose the specific architecture recited in the independent claims of US patent 8,374,358: filtering a first audio signal with a first adaptive filter, filtering a second audio signal with a second adaptive filter, and combining the two filtered signals, where both filters are adapted to minimize the wanted signal component in the final combination. This specific dual-filter structure for creating a blocking matrix without introducing potential instability (poles) appears to be the key point of novelty.

Here is an analysis of the obviousness of US patent 8,374,358 under 35 U.S.C. § 103.

Person Having Ordinary Skill in the Art (PHOSITA)

A person having ordinary skill in the art (PHOSITA) for this patent would have a Master's degree in Electrical Engineering or a related field, with a focus on digital signal processing. This individual would have 2-3 years of practical experience in audio processing, particularly in the areas of adaptive filtering, noise cancellation, and microphone array processing (beamforming). A PHOSITA would be familiar with standard algorithms like the Normalized Least Mean Square (NLMS) algorithm and architectures like the Generalized Sidelobe Canceller (GSC).

Analysis of Independent Claim 1

Claim 1: A method for determining a noise reference signal for noise compensation and/or noise reduction...

This claim would have been obvious over the teachings of Gannot et al. ("Beamforming methods for multi-channel speech enhancement") in view of the knowledge of a PHOSITA regarding common techniques for improving numerical stability in adaptive filters.

1. Gannot et al. Disclosures:
   Gannot teaches a method for multi-channel speech enhancement using a blocking matrix within a GSC framework. The patent under analysis explicitly describes Gannot's approach, stating that it estimates the transfer functions between a wanted signal and the microphone signals (Col. 10, lines 40-44). The '358 patent presents Gannot's blocking matrix, which includes terms like -G2/G1, -G3/G1, etc. (Col. 10, lines 45-58). This structure directly corresponds to the system shown in FIG. 7 of the '358 patent, which aims to create a noise reference signal by filtering a first microphone signal and subtracting it from a second. To do this, the adaptive filter (715) must model the transfer function H = G2/G1.

   Therefore, Gannot teaches the core elements of claim 1:

   • Receiving a first audio signal (e.g., from microphone 1) and a second audio signal (e.g., from microphone 2).
   • Using adaptive filters to model acoustic transfer functions (G1, G2).
   • Combining these signals to obtain a noise reference signal where the wanted signal is minimized or "blocked".

2. The Problem with Gannot's Approach:
   The '358 patent itself identifies the critical weakness in the Gannot approach: "a blocking matrix comprises an inverse of a first transfer function modeling the transfer between the wanted signal and the first microphone signal, undesired artifacts in the noise reference signal may occur if the first transfer function approaches zero" (Col. 10, lines 60-64). This phenomenon, known as the "comb-filter" effect in room acoustics, causes the transfer function G1 to have zeros at certain frequencies. Attempting to compute 1/G1 at these frequencies leads to division by zero, causing instability and a poor-quality noise reference signal. This was a well-understood problem in the field of signal processing at the time of the invention.

3. Motivation to Modify Gannot:
   A PHOSITA, when implementing the system taught by Gannot, would immediately recognize the potential for instability due to the 1/G1 term. The motivation to solve this known problem would have been high, as it directly impacts the performance and reliability of the noise reduction system.

   The solution proposed in claim 1 of the '358 patent is to replace the single-filter structure which calculates S*G1 * (G2/G1) with a two-filter structure that calculates S*G1*H1 - S*G2*H2. The system then adapts H1 to model G2 and H2 to model G1 to cancel the wanted signal S.

   This modification represents a standard engineering technique for avoiding division and improving numerical stability. Reformulating an equation A/B into a cross-multiplication form (A*D - B*C = 0) is a common method taught in introductory signal processing and control theory to handle potential divisions by zero. A PHOSITA, motivated to solve the known instability of Gannot's approach, would have found it obvious to apply this standard technique. By replacing the problematic division with a more robust cross-multiplication structure using two adaptive filters instead of one, the PHOSITA would arrive at the method described in claim 1.

Analysis of Independent Claim 16

Claim 16: A method for processing an audio signal for noise compensation...

This claim would have been obvious over the combination of Gannot et al. (as modified above) and the teachings of Widrow et al. ("Adaptive noise cancellation: Principles and applications") or the established Generalized Sidelobe Canceller (GSC) framework (e.g., as described in Van Veen and Buckley).

Claim 16 adds the following steps to the method of claim 1:

• Filtering the generated noise reference signal with a third adaptive filter.
• Subtracting this filtered signal from the first audio signal to produce a noise-reduced output.

This is the classic structure of an adaptive noise canceller.

• Widrow is the seminal work in this field and teaches precisely this method: use a noise reference signal as the input to an adaptive filter, and subtract the filter's output from a primary signal to remove correlated noise.
• The GSC architecture (shown in FIG. 4 of the '358 patent) inherently includes this structure. The entire purpose of the blocking matrix (412) is to create noise reference signals that are then fed into an interference canceller (413), which is an array of adaptive filters. The output is then subtracted (414) from the main signal path.

A PHOSITA, having developed the improved noise reference signal generation method from claim 1 (by modifying Gannot), would have been motivated to use it for its intended purpose: noise cancellation. The most well-known and direct way to do this would be to integrate it into the standard Widrow or GSC framework. It would have been entirely obvious to take the superior noise reference signal and apply it within the established noise cancellation architecture, which is precisely what claim 16 describes.

Analysis of Independent Claim 20

Claim 20: A system for audio signal processing...

This system claim recites the hardware/software components necessary to carry out the method of claim 1: a receiver, two adaptive filters, and a subtractor. As the method of claim 1 is rendered obvious by Gannot in view of a PHOSITA's knowledge, a system claim that does little more than implement this obvious method in standard components would also be obvious. The motivation to create the system is the same as the motivation to perform the method: to generate a more stable and accurate noise reference signal. A PHOSITA would know how to implement adaptive filters and subtractors using common components like Digital Signal Processors (DSPs), FPGAs, or general-purpose computers, as listed in the patent's own description (Col. 14, lines 5-29).

Term, Adjustments, and Application History for US Patent 8,374,358

Based on a review of the USPTO's public records for US patent 8,374,358, the following details regarding its term, related applications, and family members have been determined.

Patent Term and Projected Expiration:

• Filing Date: March 29, 2010

• Issue Date: February 12, 2013

• Standard Term: A U.S. patent term is typically 20 years from the earliest effective filing date. For this patent, the 20-year term would normally end on March 29, 2030.

• Patent Term Adjustment (PTA): The USPTO has granted a Patent Term Adjustment to this patent. According to the information on the face of the issued patent, a total of 335 days of adjustment was calculated due to delays in prosecution by the USPTO.

• Projected Expiration Date: The standard 20-year term is extended by the granted PTA.

  • Standard Expiration: March 29, 2030
  • PTA: + 335 days
  • Adjusted Expiration Date: February 27, 2031

• Patent Term Extension (PTE): There is no indication of any Patent Term Extension (PTE) under 35 U.S.C. § 156, which is typically granted for delays caused by pre-market regulatory review by agencies like the FDA.

Continuity and Divisional Applications:

• Continuation Applications: A review of the continuity data shows that US patent 8,374,358 is the parent application for at least one continuation application. A continuation application allows an inventor to file a new application based on the parent, often to pursue a different set of claims.
  • Child Application: Application number US 13/748,264 was filed on January 23, 2013, claiming priority to this patent. This continuation application later issued as US Patent 9,280,965.
• Divisional Applications: There is no record of any divisional applications having been filed from this patent. A divisional application would be filed if the original application was found by the examiner to contain more than one distinct invention.

Patent Family:

• Priority Application: This U.S. patent claims priority to European Patent Application No. 09004609.5, which was filed on March 30, 2009.
• Family Members: This patent is part of a family of related patents and applications filed in different countries, all claiming priority to the original European application. This includes the corresponding U.S. patent application publication US 2010/0246851 A1. The existence of a patent family indicates that the assignee sought protection for this invention in multiple international jurisdictions.

Defensive Disclosure Document for US Patent 8,374,358

Publication Date: May 9, 2026
Subject: Derivative Works and Obvious Variations of a Method for Determining a Noise Reference Signal for Noise Compensation and/or Noise Reduction.
Reference Patent: US 8,374,358 B2 ("the '358 patent")

This document discloses a series of methods, systems, and applications that build upon, vary, or represent alternative embodiments of the core inventive concept described in the '358 patent. The purpose of this disclosure is to place these variations into the public domain, thereby establishing them as prior art for future patent applications. The core concept of the '358 patent involves using two separate adaptive filters on two audio signals and combining their outputs to generate a noise reference signal, where the filters are adapted to minimize a wanted signal component.

---

Axis 1: Material & Component Substitution

Derivative 1.1: Analog Implementation with MEMS Piezoelectric Adaptive Filters

• Enabling Description: The digital signal processor (DSP) and discrete adaptive filters of the '358 patent are replaced with an analog, low-power implementation using Micro-Electro-Mechanical Systems (MEMS). Two MEMS microphone inputs feed their signals to two corresponding MEMS adaptive filters. Each filter comprises a silicon cantilever array coated with a piezoelectric film (e.g., lead zirconate titanate, PZT). A low-power microcontroller executes the adaptation logic by applying a variable DC bias voltage to the cantilevers. This voltage modifies the stiffness and resonant frequency of each cantilever, thereby changing the filter's overall transfer function. The analog outputs from the two MEMS filters are then differenced by an operational amplifier to produce the noise reference signal. The microcontroller calculates the adaptation criterion (e.g., minimizing output energy during speech) and updates the control voltages accordingly. This architecture significantly reduces power consumption and latency, making it suitable for battery-powered hearables and edge devices.
• Mermaid Diagram:

  graph TD
      subgraph System Architecture
          Mic1[MEMS Mic 1] --> AF1[MEMS Piezoelectric Filter 1];
          Mic2[MEMS Mic 2] --> AF2[MEMS Piezoelectric Filter 2];
          AF1 --> Sub[Analog Subtractor];
          AF2 --> Sub;
          Sub --> NRS[Noise Reference Signal];
          NRS --> MCU[Microcontroller Unit Adaptation Logic];
          MCU -->|Control Voltage V1| AF1;
          MCU -->|Control Voltage V2| AF2;
      end
  

Derivative 1.2: Optical Correlator for Ultrafast Filter Adaptation

• Enabling Description: The digital adaptation algorithm (e.g., NLMS) is replaced by an optical processing unit for near-instantaneous computation of filter coefficients. The two input audio signals are converted to optical signals via acousto-optic modulators (AOMs). These modulated light beams are passed through spatial light modulators (SLMs) which physically represent the filter coefficients H1 and H2. The resulting beams are interfered on a photodetector array, yielding the noise reference signal. A portion of the input and output light is diverted to an optical correlator, which uses Fourier-transforming lenses to compute the cross-correlation between the noise reference and input signals. The resulting optical pattern provides the error gradient used to update the SLMs, thereby adapting the filters. This method is suited for extremely wideband signals where digital adaptation would be a computational bottleneck.
• Mermaid Diagram:

  flowchart LR
      subgraph Signal Path
          A[Audio In 1] --> AOM1[Acousto-Optic Modulator 1]
          B[Audio In 2] --> AOM2[Acousto-Optic Modulator 2]
          AOM1 --> SLM1[Spatial Light Modulator H1] --> BeamCombiner
          AOM2 --> SLM2[Spatial Light Modulator H2] --> BeamCombiner
          BeamCombiner --> Photodetector --> NRS[Noise Ref Out]
      end
      subgraph Adaptation Path
          NRS --> OptCorr[Optical Correlator]
          AOM1 --> OptCorr
          AOM2 --> OptCorr
          OptCorr -->|Control Signal| SLM1
          OptCorr -->|Control Signal| SLM2
      end
  

---

Axis 2: Operational Parameter Expansion

Derivative 2.1: Cryogenic Superconducting Implementation for Quantum-Level Sensing

• Enabling Description: For applications requiring extreme sensitivity, such as quantum computing or deep-space radio astronomy, the system is implemented using superconducting electronics operating at cryogenic temperatures (e.g., 4 Kelvin). The inputs are detected by superconducting sensors (e.g., SQUIDs). The adaptive filters are constructed from Josephson junction arrays, with coefficients stored as magnetic flux quanta in superconducting loops. The adaptation algorithm is performed by a quantum annealing processor that adjusts the flux quanta to find the global minimum of the wanted signal power. The subtraction is executed by a Superconducting Quantum Interference Filter (SQIF). This implementation eliminates thermal noise, achieving signal-to-noise ratios impossible at room temperature.
• Mermaid Diagram:

  sequenceDiagram
      participant Sensor1 as Superconducting Sensor 1
      participant Sensor2 as Superconducting Sensor 2
      participant JJA1 as Josephson Junction Array (H1)
      participant JJA2 as Josephson Junction Array (H2)
      participant SQIF as Superconducting Subtractor
      participant QAP as Quantum Annealing Processor
  
      Sensor1->>JJA1: Signal 1
      Sensor2->>JJA2: Signal 2
      JJA1->>SQIF: Filtered Signal 1
      JJA2->>SQIF: Filtered Signal 2
      SQIF-->>QAP: Noise Reference Signal (U)
      QAP->>JJA1: Update Flux Quanta (Coefficients)
      QAP->>JJA2: Update Flux Quanta (Coefficients)
  

Derivative 2.2: Industrial-Scale Application for Seismic Wave Cancellation

• Enabling Description: The principle is scaled up for geophysical applications. A primary seismometer array (first signal) is placed at a monitoring site, and a secondary array (second signal) is placed near a known noise source (e.g., a highway). The multichannel time-series data from these arrays are processed on a high-performance computing (HPC) cluster. The dual adaptive filters are high-order IIR filters designed to model the complex transfer functions of seismic waves through heterogeneous geological strata. The system adapts the filters to generate a "noise reference wavefield," effectively canceling the correlated noise from the urban source to enhance the detection of faint earthquake signals or clandestine underground tests. The adaptation uses a block-based Recursive Least Squares (RLS) algorithm to handle the long impulse responses inherent in seismic data.
• Mermaid Diagram:

  graph TD
      A[Seismometer Array 1 (Primary)] --> HPC{High-Performance Cluster};
      B[Seismometer Array 2 (Secondary)] --> HPC;
      subgraph HPC
          A_data[Data from Array 1] --> H2[Adaptive Filter H2];
          B_data[Data from Array 2] --> H1[Adaptive Filter H1];
          H2_out[Filtered Primary] --> Sub[Combiner];
          H1_out[Filtered Secondary] --> Sub;
          Sub --> NRS[Cleaned Seismic Signal];
          NRS --> Adapt[Adaptation Logic];
          Adapt --> H1;
          Adapt --> H2;
      end
      HPC --> Output[Geophysical Analysis];
  

---

Axis 3: Cross-Domain Application

Derivative 3.1: AgTech - Filtering EMI from Soil Moisture Sensor Data

• Enabling Description: In precision agriculture, the method is used to remove electrical noise from soil moisture probe readings. A primary sensor (S1) is a capacitance probe whose signal contains the true moisture value plus electromagnetic interference (EMI) from irrigation pumps. A secondary sensor (S2) is a simple antenna or induction coil placed to primarily capture this ambient EMI. The dual adaptive filter algorithm, running on the sensor node's embedded controller, processes the signals from S1 and S2. It adapts filters H1 and H2 to generate a noise reference signal U where the probe's own operating signal is nulled out. This pure EMI reference is then subtracted from the primary probe reading to yield a more accurate soil moisture measurement.
• Mermaid Diagram:

  graph TD
      subgraph In-Field Sensor Node
          Moisture[Soil Moisture] --> Probe[Capacitive Probe S1]
          EMI[Electrical Noise] --> Probe
          EMI --> Antenna[EMI Antenna S2]
          Probe -- Capacitive Coupling --> Antenna
          Probe --> FilterH2[Adaptive Filter H2]
          Antenna --> FilterH1[Adaptive Filter H1]
          FilterH1 -- (-) --> Subtractor
          FilterH2 -- (+) --> Subtractor
          Subtractor --> NoiseRef[EMI Reference]
          Probe --> FinalSub(-)
          NoiseRef -- Adaptively Filtered --> FinalSub(+)
          FinalSub --> CleanData[Accurate Moisture Reading]
      end
  

Derivative 3.2: Consumer Electronics - Isolating Fetal Heartbeat

• Enabling Description: The technique is applied in an at-home fetal heartbeat monitor to separate the faint fetal heartbeat from the overpowering maternal heartbeat. Two acoustic sensors are placed on the abdomen. Sensor 1 (S1) is positioned to best capture the fetal heartbeat. Sensor 2 (S2) is positioned to capture the maternal heartbeat as strongly as possible. The dual-filter system is adapted to treat the maternal heartbeat as the "unwanted" signal to be canceled from a reference. The filters H1 and H2 adapt to create a "maternal heartbeat reference signal" (U) by minimizing the fetal component within it. This clean reference of the mother's heartbeat is then adaptively subtracted from S1 to isolate and enhance the much fainter fetal heartbeat.
• Mermaid Diagram:

  sequenceDiagram
      participant MH as Maternal Heart
      participant FH as Fetal Heart
      participant S1 as Sensor 1 (Fetal Focus)
      participant S2 as Sensor 2 (Maternal Focus)
      participant Processor
      participant Output
  
      MH->>S1: Strong Signal
      MH->>S2: Very Strong Signal
      FH->>S1: Faint Signal
      FH->>S2: Very Faint Signal
  
      S1->>Processor: Process with Filter H1
      S2->>Processor: Process with Filter H2
      Processor->>Processor: Combine to create Maternal Heartbeat Reference (U)
      Note right of Processor: Filters adapted to minimize Fetal Heartbeat in U
      Processor->>Processor: Subtract U from S1
      Processor->>Output: Isolated Fetal Heartbeat
  

---

Axis 4: Integration with Emerging Tech

Derivative 4.1: AI-Driven Contextual Adaptation Control

• Enabling Description: A machine learning model, such as a convolutional neural network (CNN), is used to supervise the adaptation process. The raw audio signals are fed into the CNN, which is pre-trained to classify the acoustic environment ('car', 'cafe') and signal content ('speech', 'music', 'transient'). The CNN's output is a control vector that dynamically adjusts the parameters of the dual-filter system's adaptation logic. For example, it can change the adaptation step-size, increase filter length in reverberant conditions, or freeze adaptation entirely during double-talk or when only stationary noise is present. This AI supervisor makes the noise cancellation far more robust to real-world conditions.
• Mermaid Diagram:

  flowchart TD
      Mic1 --> DualFilter[Dual Adaptive Filter System];
      Mic2 --> DualFilter;
      Mic1 --> CNN[Context-Aware CNN];
      Mic2 --> CNN;
      CNN -- Control Vector (μ, Filter Length, etc.) --> DualFilter;
      DualFilter --> NoiseRef;
  

Derivative 4.2: Distributed Noise Cancellation via IoT Network

• Enabling Description: The two audio inputs are sourced from spatially separate IoT devices. A user's wearable device provides the first microphone signal (S1), containing their voice. A network of stationary IoT sensors (e.g., smart speakers) in the environment provides the second audio signal (S2). An edge or cloud server receives both streams, using Network Time Protocol (NTP) for synchronization. The server selects the IoT sensor closest to the user as S2 and executes the dual-filter algorithm. This use of spatially diverse signals allows for superior noise reference generation, which is then used to clean the user's voice for a command or call.
• Mermaid Diagram:

  graph TD
      subgraph Edge_Cloud
          Processor(Dual Filter Processor)
      end
      subgraph Smart_Environment
          User[User with Wearable Mic S1] --> |Audio Stream 1| Processor
          IoT1[IoT Device Mic S2] --> |Audio Stream 2| Processor
          IoT2[IoT Device]
          IoT3[IoT Device]
      end
      Processor --> CleanAudio[Cleaned User Speech]
  

Derivative 4.3: Blockchain for Verifiable Noise Cancellation Provenance

• Enabling Description: For secure or forensic applications, a blockchain provides an immutable audit trail of the noise cancellation process. At each time block, a hash is generated from the raw input signals, the filter coefficient vectors for H1 and H2, and the final output signal. This hash, along with a timestamp and device identifiers, is committed as a transaction on a private blockchain. This allows any third party to independently verify that the noise cancellation was performed correctly using the recorded coefficients and that the wanted signal was not tampered with, which is critical for the admissibility of recorded evidence.
• Mermaid Diagram:

  stateDiagram-v2
      [*] --> Processing
      Processing --> Hashing: At each time block (k)
      state Hashing {
          direction LR
          S1_k: Raw Signal 1
          S2_k: Raw Signal 2
          H1_k: Filter 1 Coeffs
          H2_k: Filter 2 Coeffs
          --> SHA256: Hash Generation
      }
      Hashing --> Blockchain: Commit Transaction
      Blockchain --> Processing: Next time block (k+1)
      state Blockchain {
          direction LR
          Block_N: [Hash_k, Timestamp]
          --> Block_N_1
      }
  

---

Axis 5: The "Inverse" or Failure Mode

Derivative 5.1: Failsafe Divergence Detection and Bypass Mode

• Enabling Description: The system includes a monitoring module to ensure stability. This module continuously calculates the short-term energy of the output signal and checks the L2-norm of the filter coefficient vectors. If the output energy exceeds the input energy for a sustained period, or if the coefficient norms exceed a threshold, the filters are presumed to be unstable. Upon detection, a digital multiplexer immediately bypasses the entire noise cancellation block, routing the original primary signal directly to the output. An indicator flag is set, and the filter coefficients are reset before adaptation is carefully re-initiated.
• Mermaid Diagram:

  graph TD
      S1[Primary Mic In] --> MUX[Bypass Mux]
      subgraph NC_Processor
          S1 --> DualFilterSystem
          S2[Secondary Mic In] --> DualFilterSystem
          DualFilterSystem --> NoiseRef
          S1 --> Subtractor
          NoiseRef --> Subtractor
          Subtractor --> CleanOut[Clean Audio Out]
          CleanOut --> DivergenceMonitor
          DualFilterSystem -- Filter Coeffs --> DivergenceMonitor
          DivergenceMonitor -- Instability_Detected! --> MUX
      end
      MUX -- select --> SystemOut[Final Audio Out]
      CleanOut --> MUX
  

Derivative 5.2: Low-Power Mode with Frozen Coefficients

• Enabling Description: For battery-constrained devices, the system features a dual-mode operation. It begins in a full-power 'Training Phase' where the dual adaptive filters converge. Once the rate of change of the filter coefficients drops below a delta threshold, the system enters a 'Low-Power Phase.' In this mode, the adaptation logic is power-gated, and the converged filter coefficients for H1 and H2 are frozen. The system now acts as a fixed spatial blocker, providing moderate noise reduction without the computational cost of continuous adaptation. A significant change in signal statistics or a user command can trigger a return to the 'Training Phase'.
• Mermaid Diagram:

  stateDiagram-v2
      [*] --> Training
      Training: Full-power adaptation of H1, H2
      Training --> LowPower: Convergence criteria met
      LowPower: Adaptation logic disabled. H1, H2 are fixed.
      LowPower --> Training: Re-train trigger
  

---

Combination Prior Art Scenarios with Open-Source Standards

1. Combination with WebRTC Standard for Echo-Robust Noise Suppression:

• Enabling Description: A web browser implementing the WebRTC standard integrates the dual-filter method into its audio processing pipeline. The user's local microphone provides the first signal (S1). The audio stream being played out to the user's speakers (the far-end audio) is used as the second signal (S2). The dual-filter algorithm, implemented in WebAssembly, adapts to create a reference signal U that contains only the ambient background noise, having canceled out both the near-end user's speech and the far-end echo. This high-fidelity, echo-free noise reference is then used by a subsequent Wiener filter to clean the user's speech before transmission.

2. Combination with VAD from the Opus Codec to Gate Adaptation:

• Enabling Description: The adaptation logic of the dual-filter system is gated by a Voice Activity Detection (VAD) module using the open-source algorithm from the Opus codec. The VAD analyzes the primary input signal and provides a binary output indicating speech presence. The adaptation step-size μ for filters H1 and H2 is set to its operational value only when the VAD indicates speech is present. When VAD indicates silence, μ is set to zero, freezing adaptation. This prevents the filters from erroneously adapting to changes in the background noise, significantly improving the system's robustness.

3. Combination with SOFA (Spatially Oriented Format for Acoustics) Standard for Fast Initialization:

• Enabling Description: In an augmented reality headset with a known microphone array geometry, the dual adaptive filters are initialized using Head-Related Transfer Functions (HRTFs) from a pre-loaded SOFA file. Instead of starting with zero coefficients, the filters H1 and H2 are initialized with coefficients derived from the HRTFs corresponding to a default "look direction." This provides the adaptation algorithm with a highly accurate starting point based on the known acoustics of the device, enabling dramatically faster convergence and more effective cancellation of off-axis noise.