UNITED STATES PATENT AND TRADEMARK OFFICE
`_______________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`_______________
`
`REALTEK SEMICONDUCTOR CORPORATION
`Petitioner
`
`v.
`
`ANDREA ELECTRONICS CORPORATION
`Patent Owner
`_______________
`
`Case: IPR2015-01392
`
`Patent 6,483,923
`_______________
`
`DECLARATION OF DR. DAVID ANDERSON
`
`RTL923_1025-0001
`
`Realtek 923 Ex. 1025
`
`

`
`TABLE OF CONTENTS
`
`TABLE OF CONTENTS
`
`I.
`
`I.
`
`II.
`II.
`
`Introduction ................................................................................................. 1
`
`Introduction ............................................................................................... .. 1
`
`Qualifications and Compensation ................................................................ 1
`Qualifications and Compensation .............................................................. .. 1
`
`III. Materials Considered ................................................................................... 4
`
`III. Materials Considered ................................................................................. .. 4
`
`IV. Level of Ordinary Skill In The Art ............................................................... 4
`IV.
`Level of Ordinary Skill In The Art ............................................................. .. 4
`
`A.
`
`Burden of Proof ............................................................................... .. 5
`
`V. My Understanding of Patent Law ................................................................ 5
`V. My Understanding of Patent Law .............................................................. .. 5
`A.
`Burden of Proof ................................................................................. 5
`B.
`Anticipation ....................................................................................... 5
`B.
`Anticipation ..................................................................................... .. 5
`C.
`Obviousness ....................................................................................... 6
`D.
`Claim Construction ............................................................................ 7
`
`C.
`
`D.
`
`Obviousness ..................................................................................... .. 6
`
`Claim Construction .......................................................................... .. 7
`
`VI. The ’923 Patent............................................................................................ 8
`
`The ’923 Patent .......................................................................................... .. 8
`
`VI.
`
`VII. Prosecution History of the ’923 Patent......................................................... 9
`VII. Prosecution History of the ’923 Patent ....................................................... .. 9
`
`Claim 1 ................................................................................. ..12
`
`Claim 3 ................................................................................. ..21
`
`Claim 9 ................................................................................. ..25
`
`Claim 11 ............................................................................... ..28
`
`1.
`
`2.
`
`3.
`
`4.
`
`B.
`B.
`
`VIII. Prior Art Analysis .......................................................................................10
`VIII. Prior Art Analysis ..................................................................................... ..10
`A. Kompis In View Of Hoshuyama .......................................................10
`A.
`Kompis In View Of Hoshuyama ..................................................... ..10
`1.
`Claim 1 ...................................................................................12
`2.
`Claim 3 ...................................................................................21
`3.
`Claim 9 ...................................................................................25
`4.
`Claim 11 .................................................................................28
`Kompis In View Of Kates .................................................................28
`Kompis In View Of Kates ............................................................... ..28
`1.
`Claim 4 ...................................................................................29
`2.
`Claim 5 ...................................................................................34
`3.
`Claim 6 ...................................................................................39
`4.
`Claim 8 ...................................................................................39
`5.
`Claim 12 .................................................................................40
`6.
`Claim 13 .................................................................................40
`7.
`Claims 14 and 16 ....................................................................40
`i
`
`1.
`
`2.
`
`3.
`
`4.
`
`5.
`
`6.
`
`7.
`
`Claim 4 ................................................................................. ..29
`
`Claim 5 ................................................................................. ..34
`
`Claim 6 ................................................................................. ..39
`
`Claim 8 ................................................................................. ..39
`
`Claim 12 ............................................................................... ..40
`
`Claim 13 ............................................................................... ..40
`
`Claims 14 and 16 .................................................................. ..40
`
`i
`
`RTL923_1025-0002
`RTL923_1025-0002
`
`

`
`C.
`C.
`
`1.
`
`2.
`
`Claim 2 ................................................................................. ..42
`
`Claim 10 ............................................................................... ..44
`
`Kompis In View Of Hoshuyama And Fischer. ..................................40
`Kompis In View Of Hoshuyama And Fischer. ................................ ..40
`1.
`Claim 2 ...................................................................................42
`2.
`Claim 10 .................................................................................44
`D. Kompis In View Of Kates And Fischer .............................................44
`D.
`Kompis In View Of Kates And Fischer ........................................... ..44
`1.
`Claims 7 and 15 ......................................................................45
`
`1.
`
`Claims 7 and 15 .................................................................... ..45
`
`IX. Conclusion ..................................................................................................45
`
`Conclusion ................................................................................................ ..45
`
`IX.
`
`ii
`
`ii
`
`RTL923_1025-0003
`RTL923_1025-0003
`
`

`
`Appendix 1 – Curriculum Vitae of David Anderson
`
`Appendix 2 – List of Documents Considered
`
`Appendix A – Claim Chart For U.S. Patent 6,483,923: Claims 1, 3, 9, And 11 Are
`Unpatentable Under 35 U.S.C. § 103(a) As Being Obvious Over Martin Kompis et
`al., Noise Reduction for Hearing Aids: Combining Directional Microphones with
`an Adaptive Beamformer, J. Acoust. Soc. Am. 96 (3), September 1994 (“Kompis”)
`In View of U.S. Patent No. 5,627,799 To Hoshuyama (“Hoshuyama”)
`
`Appendix B – Claim Chart For U.S. Patent 6,483,923: Claims 4-6, 8, 12-14, And
`16 Are Unpatentable Under 35 U.S.C. § 103(a) As Being Obvious over Kompis In
`View Of James M. Kates et al., A Comparison of Hearing Aid Array-Processing
`Techniques, J. Acoust. Soc. Am. 99 (5), May 1996 (“Kates”)
`
`Appendix C – Claim Chart For U.S. Patent 6,483,923: Claims 2 And 10 Are
`Unpatentable Under 35 U.S.C. § 103(a) As Being Obvious Over Kompis In View
`of Hoshuyama, And Further In View Of Sven Fischer et al., An Adaptive
`Microphone Array for Hands-Free Communication, Proc. IWAENC-95, Røros,
`Norway, June 1995 (“Fischer”)
`
`Appendix D – Claim Chart For U.S. Patent 6,483,923: Claims 7 And 15 Are
`Unpatentable Under 35 U.S.C. § 103(a) As Being Obvious Over Kompis In View
`of Kates, And Further In View Of Fischer
`
`iii
`
`RTL923_1025-0004
`
`

`
`I, David Anderson, hereby declare, affirm and state the following:
`
`I.
`
`1.
`
`Introduction
`
`The facts set forth below are known to me personally, and I have firsthand
`
`knowledge of them.
`
`2.
`
`I make this Declaration in support of a Petition for inter partes review of
`
`U.S. Patent No. 6,483,923 (“the ’923 patent”)
`
`3.
`
`I have been retained by Steptoe & Johnson LLP on behalf of Realtek
`
`Semiconductor Corporation.
`
`4.
`
`I have been asked to provide my technical review, analysis, insights, and
`
`opinions on the materials I have reviewed in this case related to the ’923 Patent,
`
`including the references that form the basis for the grounds of rejection set forth in
`
`the Petition No. IPR2015-01392 for Inter Partes Review of the ’923 Patent
`
`(“Petition”), and the scientific and technical knowledge regarding the same subject
`
`matter at the time of the alleged inventions disclosed in the’923 Patent.
`
`II. Qualifications and Compensation
`
`5.
`
`6.
`
`I am over the age of eighteen and I am a citizen of the United States.
`
`I have summarized in this section my educational background, career
`
`history, and other relevant qualifications. My curriculum vitae, including my
`
`qualifications, a list of the publications that I have authored during my technical
`
`1
`
`RTL923_1025-0005
`
`

`
`career, and a list of the cases in which, during the previous four years, I have
`
`testified as an expert at trial or by deposition, is attached to this declaration as
`
`Appendix 1.
`
`7.
`
`I earned my Bachelor of Science degree in Electrical Engineering from
`
`Brigham Young University in 1993. In 1994, I earned my Master of Science
`
`degree in Electrical Engineering, also from Brigham Young University. I earned
`
`my Doctorate of Philosophy in Electrical and Computer Engineering from Georgia
`
`Institute of Technology in 1999, with my dissertation on “Audio Signal
`
`Enhancement Using Multi-resolution Sinusoidal Modeling.”
`
`8.
`
`After obtaining my Doctorate of Philosophy degree, I worked as an
`
`Education Specialist at Texas Instruments, Inc. from April 1999 through
`
`September of 1999. In this position, I developed a self-paced course on signal
`
`processing fundamentals and implementation for practicing engineers.
`
`9.
`
`In September of 1999, I joined the faculty of Georgia Institute of
`
`Technology as an Assistant Professor in the School of Electrical and Computer
`
`Engineering. While on this faculty, I taught courses in signal processing and
`
`computer architecture and performed research in signal processing and low-power
`
`implementation of signal processing systems.
`
`2
`
`RTL923_1025-0006
`
`

`
`10.
`
`In June of 2005, I was promoted to the rank of Associate Professor, and in
`
`March of 2012, I became a full Professor, where I continued teaching and research
`
`in signal processing and signal processing systems with an emphasis in low-power
`
`systems, signal enhancement, and signal processing related to human audio
`
`perception.
`
`11.
`
`I have authored and co-authored approximately 200 journal publications,
`
`conference proceedings, technical articles, technical papers, book chapters, and
`
`technical presentations, in a broad array of signal processing technology. I have
`
`also developed and taught over many courses related to digital signal processing
`
`and signal processing systems. These courses have included introductory as well
`
`as more advanced courses.
`
`12.
`
`I have three patents related to the field of audio signal enhancement, U.S.
`
`Patent no. 6,351,731; U.S. Patent no. 6,453,285; and U.S. Patent no. 7,034,603.
`
`Additionally, I have many scholarly publications on enhancing speech signals and
`
`removing noise that are included in my CV including J. M. Hurtado and D. V.
`
`Anderson “FFT-based block processing in speech enhancement: potential artifacts
`
`and solutions,” IEEE Transactions on Audio, Speech, and Language Processing,
`
`19(8):2527-2537, Nov. 2011.
`
`3
`
`RTL923_1025-0007
`
`

`
`13.
`
`I am being compensated for my time at the rate of $350 per hour for my
`
`work in connection with this matter. The compensation is not dependent in any
`
`way on the contents of this Declaration, the substance of any further opinions or
`
`testimony that I may provide, or the ultimate outcome of this matter.
`
`III. Materials Considered
`
`14.
`
`I have carefully reviewed the ’923 Patent and its file history. I have also
`
`reviewed several prior art references.
`
`15.
`
`For convenience, all of the sources that I considered in preparing this
`
`declaration are listed in Appendix 2.
`
`IV. Level of Ordinary Skill In The Art
`
`16.
`
`I have been informed that my analysis of the interpretation of the asserted
`
`claims of the ’923 Patent must be undertaken from the perspective of a person
`
`possessing ordinary skill in the art of the ’923 Patent. In my opinion a person of
`
`ordinary skill in the art at the time of the ’923 Patent would have been an
`
`individual with at least a Master’s degree in Electrical Engineering with a specialty
`
`in digital signal processing. In the alternative, this person would have a Bachelor’s
`
`degree in Electrical Engineering with at least two years of training in signal
`
`processing specializing in adaptive signal processing, and at least two years
`
`practical experience with microphone array processing algorithm development and
`
`4
`
`RTL923_1025-0008
`
`

`
`implementation. I possess these qualifications, and I have considered the issues
`
`herein from the perspective of a person of ordinary skill in the art.
`
`V. My Understanding of Patent Law
`
`17.
`
`I am not an attorney but I have had the concept of patentability explained to
`
`me. I understand that a patent claim can be unpatentable under the United States
`
`patent laws for various reasons, including, for example, anticipation or obviousness
`
`in light of the prior art. In arriving at my opinions, I have applied the following
`
`legal standards and analyses regarding patentability.
`
`A.
`
`Burden of Proof
`
`18.
`
`I understand that Petitioner has the burden to prove a proposition of
`
`unpatentability by a preponderance of the evidence. I also understand that this is a
`
`lower standard than the clear and convincing evidence standard that is required to
`
`prove unpatentability in patent litigation before a district court.
`
`B.
`
`Anticipation
`
`19.
`
`I understand that a claim is anticipated by a prior art reference if the prior art
`
`reference discloses every element in the claim. Such a disclosure can be express (it
`
`says or shows it), or it can be inherent (the element must necessarily be there even
`
`if the prior art does not say it or show it). If the claim is anticipated, the claim is
`
`unpatentable.
`
`RTL923_1025-0009
`
`5
`
`

`
`20.
`
`I understand that the first step in an anticipation analysis is to construe the
`
`claim, and the second step is to compare the construed claim to the prior art
`
`reference.
`
`C.
`
`Obviousness
`
`21.
`
`I understand that a patent claim may be unpatentable for obviousness even if
`
`it is not anticipated by the prior art. I understand that a patent claim is obvious if
`
`the differences between the claimed intervention and the prior art are such that the
`
`subject matter of the claimed invention, as a whole, would have been obvious to
`
`one of ordinary skill in the art at the time the invention was made. If the claim is
`
`obvious, the claim is unpatentable.
`
`22.
`
`I understand that before an obviousness determination is made, the level of
`
`ordinary skill in the art must be considered, and the scope and content of the prior
`
`art must be considered, as well. I understand that to determine the scope and
`
`content of prior art, one must determine what prior art is reasonably pertinent to the
`
`particular problem the inventor faced. I understand that prior art is reasonably
`
`pertinent if it is in the same field as the claimed invention, or is from another field
`
`that a person of ordinary skill in the art would look to in trying to solve the
`
`problem.
`
`6
`
`RTL923_1025-0010
`
`

`
`23.
`
`I understand that a patent claim maybe be obvious if the prior art would have
`
`suggested, motivated, or provided a reason to one of ordinary skill in the art to
`
`combine certain prior art references to arrive at the elements of the claim. I also
`
`understand that one can look at interrelated teachings of multiple patents, the
`
`effects of demands known to the design community or present in the marketplace,
`
`and the background knowledge possessed by a person having ordinary skill in the
`
`art—all in order to determine whether there was an apparent reason to combine the
`
`known elements in the fashion claimed by the patent at issue. I further understand
`
`that a person of ordinary skill is a person of ordinary creativity, not an automaton.
`
`This person of ordinary creativity works in the contexts of a community of
`
`inventors and of the marketplace. The obviousness inquiry needs to reflect these
`
`realities within which inventions and patents function. In order to arrive at a
`
`conclusion that an invention is obvious, it can be helpful to identify a reason that
`
`would have prompted a person of ordinary skill in the relevant field to combine the
`
`elements in the way the claimed invention does.
`
`D.
`
`Claim Construction
`
`24.
`
`For the purposes of my opinions related to the issue of patentability of the
`
`’923 Patent, I have been informed that the claims of the ’923 Patent are to be given
`
`their broadest reasonable interpretation in view of the specification from the
`
`7
`
`RTL923_1025-0011
`
`

`
`perspective of one skilled in the art. In comparing the claims of the ’923 Patent to
`
`the known prior art, I have carefully considered the ’923 Patent and the ’923 Patent
`
`prosecution history based upon my experience and knowledge in the relevant field.
`
`25.
`
`For the reasons explained below, claims 1-16 of the ’923 patent are
`
`unpatentable as being rendered obvious by the prior art discussed below. I reserve
`
`the right to amend and/or supplement this declaration in light of additional relevant
`
`evidence, arguments, or testimony presented, for example, during discovery for
`
`this IPR.
`
`VI. The ’923 Patent
`26.
`The ’923 patent is directed to a system and method for adaptive interference
`
`cancelling through adaptive beamforming. Ex. 1001, 1:9-12; 2:7-34. The alleged
`
`invention separates a main channel from a reference channel through beamforming
`
`signals received and sampled from an array of sensors. Id. at 4:48-5:18. The
`
`reference channel is adapted to the noise present in the main channel through use
`
`of adaptive filters, and the noise is subsequently subtracted from the main channel
`
`to produce a signal with reduced interference. Id.
`
`27.
`
`The ’923 patent also describes converting the filter weights of the adaptive
`
`filters to the frequency domain and truncating them when these weights adapt to a
`
`level above a predetermined threshold in order to avoid signal leakage. Id. at 9:7-
`
`8
`
`RTL923_1025-0012
`
`

`
`40. It further describes the use of decolorizing filters to flatten the frequency
`
`spectrum of the noise cancelling signal, and an inhibitor which inhibits adaptation
`
`based on a normalized power difference. Id. at 5:6-23.
`
`VII. Prosecution History of the ’923 Patent
`28.
`The application that led to the ’923 patent was filed on August 6, 1998, and
`
`issued on November 19, 2002. Ex. 1001; Ex. 1020 at RTL923_1020-5-40. The
`
`’923 patent derived from a continuation application of U.S. Patent No. 5,825,898
`
`(“the ’898 Patent”). Ex. 1001, Cover.
`
`29.
`
`The Examiner issued a Non-Final Rejection on March 1, 2002 rejecting all
`
`claims as being unpatentable for double patenting in view of the ’898 Patent. Ex.
`
`1020 at RTL923_1020-0066 -74. Specifically, original claims 1-28 were rejected
`
`for statutory double patenting and original claims 29-44 were rejected for
`
`nonstatutory double patenting. Id.
`
`30.
`
`The Applicant filed a response on July 31, 2002 that canceled claims 1-28
`
`and included a terminal disclaimer to overcome the double patenting rejection for
`
`claims 29-44. Ex. 1020 at RTL923_1020-0077-81. In response, the Examiner
`
`issued a Notice of Allowance on August 12, 2002. Ex. 1020 at RTL923_1020-86-
`
`88.
`
`9
`
`RTL923_1025-0013
`
`

`
`VIII. Prior Art Analysis
`
`31.
`
`I now turn to the references applied in the grounds for rejections discussed in
`
`the Petition for Inter Partes Review. In my analysis, I will specifically address the
`
`following references:
`
`No.
`
`1021
`
`Reference
`Martin Kompis et al., Noise Reduction for Hearing
`Aids: Combining Directional Microphones with an
`Adaptive Beamformer, J. Acoust. Soc. Am. 96 (3),
`September 1994
`James M. Kates et al., A Comparison of Hearing Aid
`Array-Processing Techniques, J. Acoust. Soc. Am. 99
`(5), May 1996
`1023 U.S. Patent No. 5,627,799
`Sven Fischer et al., An Adaptive Microphone Array for
`Hands-Free Communication, Proc. IWAENC-95,
`Røros, Norway, June 1995
`
`1022
`
`1024
`
`Referred To As
`
`Kompis
`
`Kates
`
`Hoshuyama
`
`Fischer
`
`Attached hereto as Appendices A-D are claim charts addressing each of the
`
`above references, alone or in combination with other references. I have reviewed
`
`the charts in detail and incorporate the charts herein by reference.
`
`A.
`
`Kompis In View Of Hoshuyama
`
`32.
`
`The article authored by Martin Kompis et al., Noise Reduction for Hearing
`
`Aids: Combining Directional Microphones with an Adaptive Beamformer, J.
`
`Acoust. Soc. Am. 96 (3), September 1994 (“Kompis”), attached to the Petition as
`
`Ex. 1021, in combination with U.S. Patent No. 5,627,799 to Hoshuyama
`
`10
`
`RTL923_1025-0014
`
`

`
`(“Hoshuyama”), attached to the Petition as Ex. 1023, renders obvious at least
`
`claims 1, 3, 9, and 11 of the ’923 Patent.
`
`33. Kompis is generally about reducing noise in audio signals for hearing aids
`
`by using multiple microphones and an adaptive beamformer. Ex. 1021, Abstract
`
`(RTL923_1021-1). In particular, two microphones are used to generate two
`
`beams, wherein a primary beam is directed forward toward the location of a
`
`desired signal source. Ex. 1021 at p. 1911; Fig. 2 (RTL923_1021-0002). The
`
`other beam (secondary beam) has a null response in the desired direction and is
`
`preferentially responsive to signals from other directions. Id. The secondary beam
`
`is filtered with an adaptive filter and then subtracted from the main beam to cancel
`
`interference signals that may be present in the main beam signal. Id.
`
`34. Hoshuyama discloses an adaptive array beamformer that features a
`
`microphone array producing a beam responsive to a desired direction, which
`
`contains a target signal and produces other beams that do not contain the target
`
`signal. Ex. 1023, Abstract (RTL923_1023-0001). The other beams are filtered
`
`using adaptive filters to cancel interference signals from the target signal. Id.
`
`35. A person of ordinary skill in the art would have been motivated to combine
`
`Kompis and Hoshuyama because both references describe similar noise
`
`11
`
`RTL923_1025-0015
`
`

`
`cancellation or reduction systems which have complementary features or aspects as
`
`described below.
`
`1.
`
`Claim 1
`
`1. A method for processing digital input data representing
`signals containing a source signal from a signal source
`on-axis from an array of sensors as well as interference
`signals from interference sources located off-axis from
`the signal source and for producing digital output data
`representing the source signal with reduced interference
`signals relative to the source signal, comprising the steps
`of:
`[a] generating a main channel from the digital input data,
`the main channel representing signals received in
`the direction of the signal source and having a
`source signal component and an interference signal
`component;
`[b] generating at least one reference channel from the
`digital input data, each reference channel
`representing signals received in directions other
`than that of the signal source;
`[c] adaptively filtering said at least one reference channel
`using filter weight values to generate a cancelling
`signal approximating the interference signal
`component in the main channel;
`[d] generating the digital output data by subtracting the
`cancelling signal from the main channel;
`[e] deriving new filter weight values so that the
`difference between the main channel and the
`cancelling signal is minimized; and
`[f] truncating the new filter weight values to
`predetermined threshold values when each of the
`new filter weight values exceeds the corresponding
`threshold value.
`
`12
`
`RTL923_1025-0016
`
`

`
`Ex. 1001 at 11:21-27; 11:28-31; 11:32-35; 11:36-39; 11:40-41; 11:42-
`44; 11:45-47.
`
`36. With respect to the preamble of claim 1, Kompis discloses a method of
`
`processing digital input data representing a source signal is directly incident on an
`
`array of at least two sensors, and a noise signal coming from another direction off-
`
`axis to the source signal interferes with this source signal.
`
`“Directional systems are a successful approach for noise
`reductions in hearing aids. These systems transmit
`signals from acoustic sources lying in front of the
`hearing aid user while suppressing signals from other
`directions, which are assumed to be noise.”
`
`Ex. 1021, Abstract (emphasis added) (RTL923_1021-1).
`
`“A dummy head with two omnidirectional microphones
`in the ears . . . was supplemented by two directional
`microphones placed just above the ears. The directional
`microphones . . . were mounted in behind-the-ear hearing
`aid housings.”
`
`Ex. 1021 at p. 1910, col. 2 (emphasis added) (RTL923_1021-0001).
`
`13
`
`RTL923_1025-0017
`
`

`
`Ex. 1021 at p. 1911, col. 2; Fig. 2 (annotated) (RTL923_1021-0002).
`
`37. Additionally, the Abstract of Kompis discusses the goal of noise reduction,
`
`stating that “[d]irectional systems are a successful approach for noise reductions in
`
`hearing aids,” by “suppressing signals from other directions which are assumed to
`
`be noise.” Ex. 1021, Abstract (RTL923_1021-1). The approach of Kompis is
`
`described as “able to improve the intelligibility of speech in a noisy environment
`
`significantly.” Id.
`
`38. With respect to element 1[a], Kompis discloses a step of generating a main
`
`channel from the input microphones. The microphone signals were digitally
`
`sampled to create “digital input data” as follows:
`
`The microphone signals were processed by a PC-based
`TMS320C30 floating point digital signal processing
`system.
` The sampling rate was 10 kHz for all
`experiments.
`
`Ex. 1021 at p. 1910, col. 2 (emphasis added) (RTL923_1021-0001).
`
`39.
`
`This main channel is generated by calculating the sum of the microphone
`
`signals, as shown in the highlighted portion of Fig. 2 below. The main channel has
`
`source signal component (“Desired signal” in Fig. 2) and an interference signal
`
`component (“interfering noise” in Fig. 2):
`
`“A block diagram of the adaptive beamformer used is
`shown in Fig. 2. The sum and the difference of the
`microphone signals is calculated first. The difference
`
`14
`
`RTL923_1025-0018
`
`

`
`signal, which contains mainly noise, drives the adaptive
`filter which calculates an estimate of the noise in the sum
`signal.”
`
`Ex. 1021 at p. 1911, col. 1 (emphasis added ) (RTL923_1021-0002).
`
`Ex. 1021 at p. 1911, Fig. 2 (annotated) (RTL923_1021-0002).
`
`40. As would be understood by any person of ordinary skill in the art in this
`
`field, the sum of input signals from two spaced microphones as shown in Fig. 2
`
`would create constructive interference in any signals directly incident on both
`
`microphones, such as the “desired signal” shown in Fig. 2 above. Thus, this “main
`
`channel” would represent mostly signals received in the direction of the signal
`
`source.
`
`41.
`
`This main channel will have an interference component as well, as it
`
`enhances the “desired signal,” but still includes components of the “interfering
`
`noise,” as shown in Fig. 2.
`
`15
`
`RTL923_1025-0019
`
`

`
`42. With respect to element 1[b], Kompis also includes a step of generating a
`
`“reference channel,” in which the signals from the two microphones are subtracted
`
`as shown in the highlighted portion of Fig. 2 below:
`
`A block diagram of the adaptive beamformer used is
`shown in Fig. 2. The sum and the difference of the
`microphone signals is calculated first. The difference
`signal, which contains mainly noise, drives the
`adaptive filter which calculates an estimate of the
`noise in the sum signal.
`
`Ex. 1021 at p. 1911, col. 1 (emphasis added) (RTL923_1021-0002).
`
`Ex. 1021 at p. 1911, Fig. 2 (annotated) (RTL923_1021-0002).
`
`43.
`
`The reference channel in Kompis is generated from the “interfering noise,”
`
`which represents signals received in directions other than that of the signal source
`
`(“Desired signal” in Kompis). As noted in the portion of Kompis quoted above, a
`
`“difference signal” contains mostly noise, as a subtraction of the two microphone
`
`signals will create a signal with destructive interference for a signal directly
`
`16
`
`RTL923_1025-0020
`
`

`
`incident to the microphones, i.e., the “desired signal.” Thus, the output of the
`
`subtraction will be a signal containing mostly noise, received in directions other
`
`than the source signal.
`
`44.
`
`Regarding element 1[c], Kompis discloses an adaptive filter as follows, as
`
`shown in the highlighted portion of Fig. 2 below:
`
`The coefficients of the FIR-structured adaptive filter
`are updated continuously in real time by a least-mean-
`squares (LMS) adaptation algorithm.
`
`Ex. 1021 at p. 1911, col. 1, ¶ 3 (emphasis added) (RTL923_1021-
`0002).
`
`The difference signal, which contains mainly noise,
`drives the adaptive filter which calculates an estimate
`of the noise in the sum signal.
`
`Ex. 1021 at p. 1911, col. 1, ¶ 1 (emphasis added) (RTL923_1021-
`0002).
`
`Ex. 1021 at p. 1911, Fig. 2 (annotated) (RTL923_1021-0002).
`
`17
`
`RTL923_1025-0021
`
`

`
`45. Kompis teaches that the difference signal from its reference channel “drives
`
`the adaptive filter,” which shows that the Kompis adaptive filter receives signals
`
`from the reference channel. Using the difference signal, the adaptive filter
`
`“calculates an estimate of the noise in the sum signal,” thereby “generating a
`
`cancelling signal” that approximates the interference signal component of the main
`
`channel. Further, Kompis discloses that an adaptive filter that has coefficients
`
`(“adaptive filter weights”) that are updated continuously. Thus, Kompis discloses
`
`an FIR adaptive filter that is continuously updated, and calculates an estimate of
`
`noise in the main channel.
`
`46.
`
`Regarding element 1[d], Fig. 2 of Kompis shows a subtractor which
`
`generates digital output data by subtracting the noise estimate (“cancelling
`
`signal”), outputted from the adaptive filter, from the delayed main channel, as
`
`shown in the highlighted portion of Fig. 2 below:
`
`18
`
`RTL923_1025-0022
`
`

`
`Ex. 1021 at p. 1911, Fig. 2 (annotated) (RTL923_1021-0002).
`
`47. With respect to element 1[e], as discussed above with respect to element
`
`1[c], Fig. 2 shows that digital output data from the subtractor is fed back to the
`
`adaptive filter that updates the filter weights (coefficients) “continuously in real
`
`time.” Ex. 1021 at p. 1911, col. 1, ¶ 2 (RTL923_1021-0002). Kompis also
`
`discloses that the adaptive filter calculates an estimate of noise in the sum signal
`
`(“cancelling signal”), which, in conjunction with the step of updating the adaptive
`
`filter with new filter weights, is an alternative way to state that it minimizes the
`
`difference between the cancelling signal and the main channel in order to subtract
`
`the noise from the main channel. Ex. 1021 at p. 1911, col. 1, ¶ 1(RTL923_1021-
`
`0002). Thus, Kompis discloses deriving new filter weights values so that the
`
`difference between the main channel and the cancelling signal is minimized.
`
`48. With respect to element 1[f], Hoshuyama discloses the step of truncating the
`
`new filter weight values to predetermined threshold values. Specifically,
`
`Hoshuyama discusses in numerous places how it constrains tap weight values in
`
`the adaptive filter 14 to maximum and minimum values:
`
`A third embodiment of the present invention is shown in
`FIG. 8 which is a further modification of the first
`embodiment.
` In
`this modification, a coefficient-
`constrained adaptive filter 14 is used instead of each
`leaky adaptive filter 8 of FIG. 4. As illustrated in detail
`in FIG. 9, each coefficient-constrained adaptive filter 14
`19
`
`RTL923_1025-0023
`
`

`
`has a memory 89 in which maximum tap weight values
`(cid:2264)(cid:2777) (cid:1533) (cid:2264)(cid:2170