`
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`____________________
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`___________________
`
`LG ELECTRONICS, INC.,
`
`PETITIONER,
`V.
`BELL NORTHERN RESEARCH, LLC,
`PATENT OWNER.
`___________________
`
`Case No. IPR2020-00108
`U.S. Patent No. 8,416,862
`___________________
`
`
`DECLARATION OF EDWIN A. HERNANDEZ-MONDRAGON, PHD
`
`
`
`
`
`Bell Northern Research, LLC, Exhibit 2004, Page 1 of 46
`
`
`
`TABLE OF CONTENTS
`
`
`INTRODUCTION ........................................................................................ 1
`I.
`BACKGROUND AND QUALIFICATIONS .............................................. 2
`II.
`LEGAL PRINCIPLES .................................................................................. 7
`III.
`IV. A PERSON OF ORDINARY SKILL IN THE ART .................................. 10
`
`CLAIM CONSTRUCTION ........................................................................ 11
`V.
`VI. OVERVIEW OF THE FIELD AND CITED REFERENCES ................... 13
`VII. PRIOR ART REFERENCES ...................................................................... 17
`Ground 1: Claims 9-12 – Li-748 in view of Tong and Mao ............................ 18
`Ground 2: Claims 9-12 – Tong in view of Mao .............................................. 26
`Ground 3: Claims 9, 11-12 – Li-054 in view of Mao ...................................... 28
`VIII. SECONDARY CONSIDERATIONS OF NON-OBVIOUSNESS ............ 30
`
`
`Bell Northern Research, LLC, Exhibit 2004, Page 2 of 46
`
`
`
`I, Edwin A. Hernandez-Mondragon, PhD, hereby declare as follows:
`
`I.
`
`INTRODUCTION
`
`1.
`
`I am over the age of eighteen (18) years and otherwise
`
`competent to make this declaration.
`
`2.
`
`I have been retained as an expert witness on behalf of Bell
`
`Northern Research, LLC for the above-captioned inter partes review (“IPR”). I
`
`understand that the petition for inter partes review involves U.S. Patent No.
`
`8,416,862 (“the ’862 patent”), which was filed on September 28, 2005. The
`
`’862 Patent names Carlos Aldana and Joonsuk Kim as co-inventors. The ’862
`
`Patent issued on May 2, 2006.
`
`3.
`
`I make this declaration based on my personal knowledge,
`
`educational background and training, consideration of the materials I discuss
`
`herein, and my expert opinions.
`
`4.
`
`I am being compensated at a rate of $600 per hour for my time
`
`in this matter. My compensation does not depend on the outcome of this
`
`proceeding and I have no financial interest in its outcome.
`
`5.
`
`In preparing this Declaration, I have reviewed and considered
`
`the ’862 Patent, the ’862 Patent prosecution history, and each of the documents
`
`cited herein, and I have considered them in light of general knowledge in the art
`
`in the time frame of April 21, 2005, the earliest priority date of the ’862 Patent.
`
`
`
`1
`
`Bell Northern Research, LLC, Exhibit 2004, Page 3 of 46
`
`
`
`In formulating my opinions, I have relied upon my experience, education and
`
`knowledge as they relate to the relevant art. I also have considered the
`
`viewpoint of a person of ordinary skill in the art in the time frame of the filing
`
`date of the ’862 patent.
`
`II. BACKGROUND AND QUALIFICATIONS
`
`6.
`
`I am an owner and founder of EGLA CORP, an intellectual
`
`property, engineering consulting, and startup accelerator incubator in the fields
`
`of a) wireless communications, 4G and 5G, b) media streaming, and c) health
`
`technologies.
`
`7.
`
`I have a Ph.D. in Computer Engineering in Mobile Computing
`
`in 2002 and obtained a Masters in Science in Electrical and Computer
`
`Engineering in 1999, both from the University of Florida. Prior to that, I
`
`graduated with my B.S. in Electronics Engineering from Costa Rica Institute of
`
`Technology in 1995. A copy of my curriculum vitae, which includes a more
`
`detailed summary of my background, experience, patents, and publications, is
`
`marked as Exhibit 2005.
`
`8.
`
`I have previously been retained as an independent expert
`
`consultant in the fields of cable television systems and broadcasting,
`
`multimedia streaming, mobile devices and systems, air-interface and Long-
`
`Term Evolution (LTE), cloud storage and data synchronization, wireless
`
`
`
`2
`
`Bell Northern Research, LLC, Exhibit 2004, Page 4 of 46
`
`
`
`communications, block-chain technology, power management, personal area
`
`networking, and smart phones and wireless embedded software development.
`
`9. My experience in Wireless Communications, modulation, RF
`
`propagation models, and antenna engineering dates back to the 1990s, where I
`
`was an engineer working on AX.25 radio node in from 1996 – 1997, working
`
`with civil band radio, VHF bands (140 MHz), and FSK (Frequency-Shift
`
`Keying) modulation.
`
`10.
`
` I founded COMPUNET in 1997 and was its lead engineer from
`
`1997 to 2009. While at COMPUNET, I was a lead developer for authentication
`
`services, security services, web services, and networking configuration services.
`
`11.
`
`I worked for Microsoft from 2001 to 2003 as a Technical
`
`Program Manager. In that position, I was responsible for driving architecture,
`
`design, test automation, and security analysis for Bluetooth Personal Area
`
`Networking (PAN). I also drove testing over networking protocols, such as
`
`IPv4 networks and IPv6 networks, including early versions of 802.11b and
`
`candidates in the 2.4GHz bands that used Frequency Hopping Spread spectrum.
`
`12. As part of my PhD, I became an expert in RF system
`
`simulation, including propagation models, and selection of appropriate
`
`modulation techniques based on SNR (Signal to Noise Ratios) and other
`
`parameters experimented with using 802.11b technologies and SISO Antennas.
`
`
`
`3
`
`Bell Northern Research, LLC, Exhibit 2004, Page 5 of 46
`
`
`
`13.
`
`I worked for Motorola, Inc. from 2003 to 2010 as a Principal
`
`Staff Software Engineer. In that position, I was responsible for application
`
`development for Google and Android platforms. I participated in kernel-level
`
`prototyping, data support, and digital rights management (DRM).
`
`14. While at Motorola, I received training on RF system testing, RF
`
`modulation techniques, specially trained on WIMAX technologies, and other air
`
`interfaces that were used by Motorola, namely iDEN, CDMA, and WIMAX.
`
`15. As a result of my experience, I am very familiar with RF
`
`modulation, system testing, and have been involved in this field since 2001.
`
`16. From 2009 to 2019, I have been working with Software-defined
`
`Radios and GNU Radio-based systems including Ettus Research boxes, and
`
`recently with Blade RF in the creation of a Soft\ware & hardware LTE emulator
`
`based on the US Patent 7,231,330.
`
`17. Blade RF 2.0 provides a 2x2 MIMO interface that I have used
`
`for testing and experimentation at my technology incubator in Boca Raton, FL.
`
`18.
`
` Starting in 2010, I founded EGLA CORP. There, I created
`
`MEVIA applications, such as Clout to Cable. MEVIA is a “software-as-a
`
`service” and a cloud-based platform that enables “MEVIA Music,” which is
`
`currently in operation in several countries including Brazil, Honduras, and the
`
`United States.
`
`
`
`4
`
`Bell Northern Research, LLC, Exhibit 2004, Page 6 of 46
`
`
`
`19. Cloud to Cable is a patented platform that merges cloud and
`
`cable television systems and simplifies music and video distribution to different
`
`platforms. Cloud to Cable is servicing operators, such as CABLE COLOR in
`
`Honduras.
`
`20. As part of my experience in EGLA, I have worked for cable TV
`
`systems in multiple operators: CABLEVISION Mexico, Axtel TV, CLARO,
`
`Direct TV, SKY Brazil, and many others. Hence, my technical experience and
`
`training covers cable TV systems, STBs, video-on-demand (VOD) systems, and
`
`several broadcasting methodologies.
`
`21. Additionally, over my career, my research has involved aspects
`
`of network security, wireless communications, network and communications
`
`reliability, artificial intelligence, multimedia streaming, and software
`
`engineering.
`
`22.
`
`I am a named inventor on eleven patents issued by the United
`
`States Patent and Trademark Office, including the following:
`
`•
`
`U.S. Patent No. 7,564,810 – Method and System for Managing Power
`
`Consumption of a Network Interface Module in a Wireless Computing
`
`Device
`
`•
`
`U.S. Patent No. 7,231,330 – Rapid Mobility Network Emulator Method
`
`and System
`
`5
`
`
`
`Bell Northern Research, LLC, Exhibit 2004, Page 7 of 46
`
`
`
`•
`
`•
`
`•
`
`•
`
`•
`
`•
`
`•
`
`•
`
`•
`
`U.S. Patent No. 7,697,508 – System, Apparatus, and Method for
`
`Proactive Allocation of Wireless Communication Resources
`
`U.S. Patent No. 8,213,417 – System, Apparatus, and Method for
`
`Proactive Allocation of Wireless Communication Resources
`
`U.S. Patent No. 7,269,388 – Bluetooth PAN Driver
`
`U.S. Patent No. 8,788,715 – Rules-based Network Selection Across
`
`Multiple Media
`
`U.S. Patent No. 7,996,505 – Rules-based Network Selection Across
`
`Multiple Media
`
`U.S. Patent No. 8,024,487 – Smart Scan for Bluetooth PAN Services
`
`U.S. Patent No. 8,707,337 – Java-based Push to Talk
`
`U.S. Patent No. 7,331,793 – Magnetic Connector
`
`U.S. Patent No. 10,123,074 and 10,524,002– Method, System, and
`
`Apparatus for Multimedia Content Delivery to Cable TV and Satellite
`
`Operators
`
`
`
`6
`
`Bell Northern Research, LLC, Exhibit 2004, Page 8 of 46
`
`
`
`23.
`
`I have assisted several IPRs and provided testimony on
`
`handover patents for LTE and other wireless technologies.
`
`24.
`
`I am not an attorney and offer no legal opinions, but in my
`
`work, I have had experience studying and analyzing patents and patent claims
`
`from the perspective of a person skilled in the art.
`
`III. LEGAL PRINCIPLES
`
`25.
`
` Claim Construction: I understand that the first step in
`
`performing a validity analysis of the patent claims is to interpret the meaning
`
`and scope of the claims by construing the terms and phrases found in those
`
`claims. I understand that a determination of the meaning and scope of the
`
`claims of the Patents-in-Suit is a matter of law. I have been informed that to
`
`determine the meaning of the claims, courts consider the intrinsic evidence,
`
`which includes the patent’s claims, written description, prosecution history,
`
`materials incorporated by reference in the patent, and prior art cited in the
`
`patent or its prosecution history. Courts give claim terms their ordinary and
`
`accustomed meaning as understood by one of ordinary skill in the art at the time
`
`of the invention in the context of the entire patent. A patentee may also define
`
`his or her own terms or disclaim claim scope. The intrinsic record may also
`
`resolve ambiguous claim terms where the ordinary and accustomed meaning of
`
`the words used in the claims lack sufficient clarity to permit the scope of the
`
`
`
`7
`
`Bell Northern Research, LLC, Exhibit 2004, Page 9 of 46
`
`
`
`claim to be ascertained from the words alone. However, particular embodiments
`
`and examples appearing in the specification will not generally be read into the
`
`claims. A term’s context in the asserted claims can also be helpful. Differences
`
`among the claim terms can also assist in understanding a term’s meaning. For
`
`example, when a dependent claim adds a limitation to an independent claim, it
`
`is presumed that the independent claim does not include the limitation.
`
`26.
`
`I have been informed that extrinsic evidence can also be useful
`
`in determining the meaning of claim terms, however, it is less significant than
`
`the intrinsic record. Technical dictionaries may be useful to show the manner in
`
`which one skilled in the art might use claim terms, but technical dictionaries
`
`may provide definitions that are too broad or may not be indicative of how the
`
`term is used in the patent.
`
`27.
`
` Standard for Obviousness: I have been instructed by counsel
`
`and understand that a combination of prior-art references may render a claim
`
`obvious if, at the time of the invention, a person of ordinary skill in the art
`
`would have selected and combined those prior-art elements in the normal
`
`course of research and development to yield the claimed invention.
`
`28.
`
` I understand that in making the obviousness inquiry, one
`
`should consider the Graham factors: the scope and content of the prior art; the
`
`differences between the claimed inventions and the prior art; the level of
`
`
`
`8
`
`Bell Northern Research, LLC, Exhibit 2004, Page 10 of 46
`
`
`
`ordinary skill in the art; and certain secondary considerations, identified below.
`
`I further understand the obviousness analysis is to be performed on a claim-by-
`
`claim basis. I understand that a person of ordinary skill in the art is a person of
`
`ordinary creativity, not an automaton.
`
`29.
`
` I have been instructed by counsel and understand that
`
`obviousness requires more than a mere showing that the prior art includes
`
`separate references covering each separate limitation in a claim under
`
`examination. I understand obviousness requires the additional showing that a
`
`person of ordinary skill at the time of the invention would have been motivated
`
`to combine those references, and, in making that combination, a person of
`
`ordinary skill in the art would have had a reasonable expectation of success.
`
`30.
`
` I also understand that an obviousness analysis must be
`
`conducted with awareness of the distortion caused by hindsight bias and with
`
`caution of arguments reliant upon ex post reasoning. Counsel has instructed me
`
`that when considering obviousness, I should not consider what is known today
`
`or what was learned from the asserted patents. Instead, I should put myself in
`
`the position of a person of ordinary skill in the field at the time of the invention.
`
`31.
`
`In particular, I understand that it is improper to use the
`
`invention as a roadmap to find its prior-art components because such an
`
`
`
`9
`
`Bell Northern Research, LLC, Exhibit 2004, Page 11 of 46
`
`
`
`approach discounts the value of combining various existing features or
`
`principles in a new way to achieve a new result.
`
`32.
`
`I understand that an invention would not have been obvious to
`
`try when one would have had to try all possibilities in a field unreduced by
`
`direction of the prior art. Stated another way, when what would have been
`
`“obvious to try” would have entailed varying all parameters or trying each of
`
`numerous possible choices until one possibly arrived at a successful result,
`
`where the prior art gave either no indication of which parameters were critical
`
`or no direction as to which of many possible choices would have been likely to
`
`be successful, an invention would not have been obvious. Further, an invention
`
`is not obvious to try where prior art does not guide one toward a particular
`
`solution.
`
`33.
`
`It is my understanding that I must also consider certain
`
`objective evidence of no obviousness if present, which includes, among others,
`
`the prior art as a whole teaching away from the invention, a long-felt need for
`
`the invention, the failure of others, copying of the invention, and industry
`
`recognition/praise by others.
`
`IV. A PERSON OF ORDINARY SKILL IN THE ART
`
`34.
`
`I understand that the claim terms of a patent are to be construed
`
`from the point of view of a hypothetical person of skill in the art (POSA) in the
`
`
`
`10
`
`Bell Northern Research, LLC, Exhibit 2004, Page 12 of 46
`
`
`
`time frame of the patent’s priority date. (At times I may refer to a person of skill
`
`in the art as simply “a skilled person.”) For the purpose of assessing the level of
`
`ordinary skill of such an individual, I have considered the types of problems
`
`encountered in the art, the prior solutions to those problems found in the
`
`literature, the speed with which innovations were made in the time relevant time
`
`frame, and the level of education and expertise of active workers in the field.
`
`35.
`
`I understand that Dr. Wells has opined that a Person of
`
`Ordinary Skill in the Art would have had a Bachelor’s degree in Electrical
`
`Engineering, Computer Engineering, Computer Science, or a related field, and
`
`at least 2–4 years of experience in the field of wireless communication, or a
`
`person with equivalent education, work, or experience in this field. (Ex. 1003, ¶
`
`23.) For the purposes of this declaration, I do not dispute Dr. Wells’ definition
`
`of a POSA, though I reserve the right to offer my own at a later date, should it
`
`be necessary.
`
`V. CLAIM CONSTRUCTION
`
`36.
`
`It is my opinion that the term “decompose the estimated
`
`transmitter beamforming unitary matrix (V) to produce the transmitter
`
`beamforming information” would be understood by a person of ordinary skill in
`
`the art at the time of the invention, and that no construction would be necessary.
`
`To the extent that the Board believes construction would be necessary or
`
`
`
`11
`
`Bell Northern Research, LLC, Exhibit 2004, Page 13 of 46
`
`
`
`helpful, a person of skill in the art would understand that “decompose the
`
`estimated transmitter beamforming unitary matrix (V) to produce the
`
`transmitter beamforming information” means “factor the estimated transmitter
`
`beamforming unitary matrix (V) to produce a reduced number of coefficients or
`
`angles.”
`
`37.
`
`In the context of this term, it is important to understand that the
`
`goal of the claimed approach, as described in the specification, is to reduce the
`
`size of the feedback required. See Ex. 1003, 13:25-36. Citing one example of
`
`the decomposition that may be done, the patent states that “with the condition
`
`of V*V=VV*=I, some of angles of the Givens Rotation are redundant. With a
`
`decomposed matrix form for the estimated transmitter beamforming matrix (V),
`
`the set of angles fed back to the transmitting wireless device are reduced.” Ex.
`
`1003, 13:64-14:3.
`
`38. Further, the “coefficients of Givens Rotation and the phase
`
`matrix coefficients serve as the transmitter beamforming information that is
`
`sent from the receiving wireless communication device to the transmitting
`
`wireless communication device.” See Ex. 1003, 15:34-39. In the example at the
`
`top of column 15, one can see an example of the decomposition, which shows
`
`how the reduction can reduce the number of elements within a matrix but
`
`importantly also retain the dimension of the matrix. The Givens rotation
`
`
`
`12
`
`Bell Northern Research, LLC, Exhibit 2004, Page 14 of 46
`
`
`
`operates to reduce the set of coefficients of the estimated transmitter
`
`beamforming matrix (V). See Ex. 1003, 14:48-15:8. This reduction permits
`
`transmitting fewer coefficients back.
`
`39. Thus, it is my opinion that a person of skill in the art would
`
`understand that the plain and ordinary meaning of “decompose the estimated
`
`transmitter beamforming unitary matrix (V) to produce the transmitter
`
`beamforming information” requires a matrix decomposition that reduces the set
`
`of elements (be the called coefficients or angles) of the estimated transmitter
`
`beamforming unitary matrix (V). So, it is my opinion that the plain and ordinary
`
`meaning of this term is “factor the estimated transmitter beamforming unitary
`
`matrix (V) to produce a reduced number of coefficients or angles.”
`
`VI. OVERVIEW OF THE FIELD AND CITED REFERENCES
`
`40. All cited references, Li-748 (Ex. 1004), Tong (Ex. 1005), Mao
`
`(Ex. 1006), Li-054 (Ex. 1007), and Poon (Ex. 1009) refer to different
`
`mechanisms used by MIMO feedback channels to send parameters to the
`
`transmitter to adjust transmission based on the noise of the channel. The main
`
`dilemma that many systems are designed to perform is, how to determine the
`
`value of H assuming that N, which is the Noise, at the receiving antenna can be
`
`measured.
`
`
`
`13
`
`Bell Northern Research, LLC, Exhibit 2004, Page 15 of 46
`
`
`
`41. There are several ways to solve this equation in MIMO
`
`systems, especially when m, or the number or transmitters, is greater than one
`
`and the number of receivers is also greater than one, which we refer to mxn
`
`MIMO systems.
`
`42. Signal processing engineers apply standard linear algebra and
`
`can solve for X by finding the inverse of matrix H. The inverse of H is a
`
`difficult task, and in many cases cannot be easily implemented in a chipset or
`
`receiver that is supposed to handle Megabits per second or even Gigabits per
`
`second.
`
`43. Hence determining the fastest way to solve for H is valuable
`
`and highly important.
`
`44. The research literature defines multiple ways to solve H and
`
`determine its value, including a method called “Singular Value Decomposition”
`
`(SVD). SVD is a method by which two matrices, U and V, are used, where one
`
`of the matrices is known at the transmitter and the other one known at the
`
`receiver, by feeding back the receiver matrix to the transmitter. Dr. Wells
`
`presents a mathematical example in Ex. 1003, ¶¶51-52.
`
`45.
`
`In essence, each modulation technique may or not work
`
`correctly as the lesser number of matrices to multiply and less complexity to be
`
`used the better.
`
`
`
`14
`
`Bell Northern Research, LLC, Exhibit 2004, Page 16 of 46
`
`
`
`46. On the other hand, Petitioner prior art can be summarized as
`
`follow:
`
`a. Tong (Ex. 1005):
`
`i. Presents a system that is inefficient in the way to solver
`
`SVD requiring complex multiplication of several Givens
`
`matrices,
`
`ii. Uses real numbers that makes it more inefficient and
`
`non-bit friendly,
`
`iii. Complexity of O(n2) which increases given the amount
`
`of computations derived from more multiplications and
`
`complex operations.
`
`b. Li-748 (Ex. 1004)
`
`i. Presents a way to use Hermitian matrices to solve SVD
`
`that will only work for nxn systems as it is unclear how
`
`mxn systems would function using Hermitian matrices.
`
`ii. Hence, complexity for this solution remains O(n2) times
`
`the number of sub-carriers,
`
`c. The combination of Tong (Ex. 1005) and Li-748 (Ex. 1004)
`
`(has several issues, specially how the Hermitian matrix
`
`processing would be combined with Tong, plus Tong uses CQI-
`
`
`
`15
`
`Bell Northern Research, LLC, Exhibit 2004, Page 17 of 46
`
`
`
`messages as part of the feedback channel to determine the type
`
`of coding that an AMC (Adaptive Modulation Coding) module
`
`should utilize. The combination of both prior-arts is highly
`
`unlikely and, the use of CQI-messages and higher payload
`
`makes Tong a highly inefficient prior-art.
`
`d. Mao (Ex. 1006) on the other hand is introduced by Petitioner,
`
`however Mao uses a Maximal-Ratio Combiner (MRC) which is
`
`a method of diversity combination where the gain of each
`
`channel is made proportional to the rms signal level and
`
`inversely proportional to the noise. MRC is depicted in FIG 1
`
`and FIG 2 of Id. reflect the use of a signal quality measurement
`
`or power for the Beamforming function. Hence, given prior
`
`PTAB order, Mao is not applicable as prior art at it operates in
`
`the RF amplitude domain and does even incorporate an FFT
`
`components, at all.
`
`e. Li-054 (Ex 1007) on the other hand is essentially the same as
`
`Li-748 but uses codebooks and other techniques that will be
`
`covered in this report. Hence, the same deficiencies found in
`
`Li-748 apply to Li-054. In summary, all combinations of
`
`systems where signal processing or software tools operate in the
`
`
`
`16
`
`Bell Northern Research, LLC, Exhibit 2004, Page 18 of 46
`
`
`
`frequency domain, are not combinable with anything that
`
`operates with power-levels.
`
`47. Additionally, Yang (Ex. 1008) and Poon (Ex. 1009) are brought
`
`as prior art in combination with others.
`
`a. Yang (Ex. 1008) is nothing but one of the older research
`
`publications describing SVD processing from 1991.
`
`b. Poon (Ex. 1009) is another patent that operates with the
`
`“Weakest N Spatial Channel” which means it works with
`
`amplitude or power-levels.
`
`48. Hence,
`
`a. All combinations of Ex. 1004-Ex. 1009 suggested by Petitioner
`
`does not make obvious Claims 9-12 of the ’862 patent.
`
`VII. PRIOR ART REFERENCES
`
`49.
`
` I have reviewed the arguments and prior-art references
`
`included in the Petition to support Grounds 1-5. In the paragraphs below, I offer
`
`limited opinions and observations about those references. If required or asked to
`
`do so, I will offer additional opinions regarding the grounds asserted in the
`
`Petition at a later time.
`
`
`
`17
`
`Bell Northern Research, LLC, Exhibit 2004, Page 19 of 46
`
`
`
`Ground 1: Claims 9-12 – Li-748 in view of Tong and Mao
`50.
` The Li-748 (Ex. 1004) reference presents a method of using a
`
`Hermitian matrix for Id. at 5:40-52. The Hermitian matrix is a “square” matrix
`
`of n2 elements as presented in Id. at Eq. 3. Additionally, a square matrix is one
`
`where the number of rows equal the number of columns. Below are a sample of
`
`the generator matrices disclosed in Li-748:
`
`Ex. 1004, 11:10-26.
`51. Clearly, Li -748 is limited to nxn systems, as Hermitian Matrix
`
`
`
`is necessarily A=AT requiring a square number of antennas, not asymmetrical
`
`like an mxn MIMO system.
`
`
`
`18
`
`Bell Northern Research, LLC, Exhibit 2004, Page 20 of 46
`
`
`
`52. Attempting to combine Li-748 and Tong reference on the other
`
`hand fails for several reasons:
`
`a. Tong Relies on CQI-messages (Channel Quality Indicator) to
`
`determine the type of constellation used by the receiver,
`
`b. A POSA would not seek to combine Tong with Li-748 for
`
`reasons I will explain below.
`
`53. Petitioner concedes that Tong uses n2-n on pp. 21 of Petition
`
`and that Li-748 uses n2-1. (Pet. at 30 (citing 1004, 2:63-67).)
`
`54. Further, Petitioner identifies FIG 43, shown below, as support
`
`for its position:
`
`
`
`19
`
`
`
`Bell Northern Research, LLC, Exhibit 2004, Page 21 of 46
`
`
`
`Id. FIG 43.
`
`55. As shown in Fig. 43, Element 320 or Beamformer VT uses each
`
`stream of the AMC or Adaptive Modulation Coding that is generated from the
`
`CQI Feedback Channel or Channel Quality Indicator (Element 342) which was
`
`generated from the AMC/Eigenvector Assignments or Element 343.
`
`56. The CQI Feedback Channel shown in FIG. 43 comes from the
`
`802.16 WIMAX specification, clearly at [0072] Tong defines CQI as CQI-
`
`Channel Quality Indicator.
`
`57. Further, a person of skill in the art would be well aware that
`
`WIMAX 802.16 includes a Channel Quality Indicator (CQI) as part of the
`
`CQICH layer.
`
`58. Tong describes in [00085] how the CQI Feedback channel is
`
`used
`
`
`
`20
`
`Bell Northern Research, LLC, Exhibit 2004, Page 22 of 46
`
`
`
`Id. at 48; at [0085].
`
`
`59. The degree of modulation is chosen base on the CQI for a
`
`
`
`particular mobile terminal.
`
`60. CQI-messages does not exist in 802.11 nor in LTE. Notably,
`
`Dr. Wells omits in his report all references to 802.16d presented by Tong on
`
`throughout the reference.
`
`61. Therefore, although Tong presents a Givens operation, it also
`
`identifies two feedback channels: one AMC -> CQI Feedback -> AMC ->
`
`Beamformer; and another where V is passed along to the Beamformer module.
`
`A POSA would not attempt to combine the two, and it is not clear how a
`
`combination could be accomplished, in any event. First, Li-748 uses a single
`
`parameter as feedback, the Matrix V, a POSA would not be motivated to
`
`
`
`21
`
`Bell Northern Research, LLC, Exhibit 2004, Page 23 of 46
`
`
`
`combine Tong’s path with Givens (Elements 330, 334, 336, 338, 340) to
`
`replace Li-748’s method using Hermitian Matrices.
`
`62.
`
`Instead, Tong at [0195] states that “SVD generally provides
`
`optimal performance when Adaptive Modulation Coding (AMC) can be
`
`performed on each individual layer”
`
`63. The Mao reference on the other hand presented in Ex. 1006 and
`
`cited by Petitioner in pp. 23 citing FIG. 1. Presented herein:
`
`
`
`Id., FIG. 1.
`
`
`
`22
`
`
`
`Bell Northern Research, LLC, Exhibit 2004, Page 24 of 46
`
`
`
`64. As presented in FIG 1, a Maximal Ratio Combiner (MRC) or
`
`Maximum Radio Combiner is the generator of the Output sequence, y(n) from
`
`all the different delayed signals after passing the Multipath Delay Profile
`
`Estimator.
`
`Id. at 9:15-24
`65.
`
` A Maximal Ratio Combiner is defined by1:
`
`
`
`
`66. Since the Maxima-radio Combining technique that “weights the
`
`output of each signal by the amplitude of that channel before adding them
`
`together,” instead of measuring Phase or angles, indeed constellation, Mao uses
`
`
`1 Mischa Schwartz, Mobile Wireless Communications, Cambridge University, 2005, Pp. 54
`
`
`
`
`23
`
`Bell Northern Research, LLC, Exhibit 2004, Page 25 of 46
`
`
`
`voltage or power of a signal. Power can be computed from the amplitude or to
`
`voltage through well-known relationships.
`
`67. Similarly, on Id. FIG 2, it is clear that signal quality
`
`measurement or amplitude is applied at element 710, 715, 720, and combined to
`
`MRC (Maximal-Ratio Combiner) at element 480.
`
`Id. at FIG 3. Pg. 3.
`68. Mao’s patent also describe at 3:63-66 states that, “… tuning the
`
`
`
`beam based on the best measurement of quality metric for the received signal
`
`such as instant power, SINR, or BER…. .” further confirming that Mao’s
`24
`
`
`
`Bell Northern Research, LLC, Exhibit 2004, Page 26 of 46
`
`
`
`technique is an evolution of US Pat. No. 6,049,307 as presented in Id. at 3:8-17,
`
`which was analog now, a similar mechanism is performed digitally using a DSP
`
`or Digital Signal Processor.
`
`69. Since, an appeal process existed and as cited by petitioner on
`
`Pp. 9.
`
`“In its Appeal Brief, Applicant argued that Kim
`discloses “a method of determining the ‘transmission
`power’ to be allocated to each of the transmit
`antennas,” and not “any mechanism or determining
`transmitter beamforming information.’”
`
`70. Therefore, Mao cannot be used in combination with Tong
`
`and/or Li, as Mao is related to power not to beamforming as defined by this
`
`court in the past.
`
`71. Dr. Wells also recognizes all components in ¶ 70 for Mao but
`
`omits the MRC element, which is the one that generates ultimately the signal,
`
`y(n). However, Dr. Wells manages to combine Mao that uses amplitude and
`
`power, with Li-748 uses angles and phases with Tong.
`
`72. First, although my prior opinions clearly explain why these
`
`references are not valid as prior art and how arguments for obviousness are
`
`hence false. I will now address Petitioner’s arguments as follows:
`
`a. Simply because Li-748 uses SVD and determines the values of
`
`matrix V and sends it to the transceiver, it is clear that the
`
`
`
`25
`
`Bell Northern Research, LLC, Exhibit 2004, Page 27 of 46
`
`
`
`proposed method is highly inefficient, complex, and it does not
`
`take advantage of constellation mapping and de-mapping, and
`
`therefore the methods provided by Li-748 to obtain matrix V
`
`does not make the ‘862 patent obvious,
`
`b. Although Tong makes use of the Given’s transform, Tong’s
`
`solution uses two paths to provide feedback to the beamformer,
`
`one using CQI-CH (part of WiMAX protocol) and the
`
`computation of a value of VT using QR Factorization.
`
`c. Finally, Mao uses a Maximal Ratio Combiner (MRC) that
`
`operates at the power-level of an RF signal. Just because an RF
`
`signal is converted to a base band, doesn’t make it a prior-art or
`
`provides and obvious argument as suggested by Petitioner on
`
`Pg. 24.
`
`Ground 2: Claims 9-12 – Tong in view of Mao
`73. First, a POSA will not be motivated to combine Tong’s and
`
`Mao cannot be combined, as a hybrid of amplitude and phase will have to be
`
`created, further increasing the complexity of the system, and it may not be
`
`combinable at all, as the MRC output will contain a set of weights and will
`
`definitely alter an SVD process handled by Tong. For example, Tong’s
`
`equations work in the frequency domain, in fact Tong’s FIG. 43 on pp. 38
`
`
`
`26
`
`Bell Northern Research, LLC, Exhibit 2004, Page 28 of 46
`
`
`
`element 326 and element 328 depict “SVD H=UDV” on elements in the
`
`frequency domain, whereas Mao’s at 10:42-52 explains how at different phases
`
`or beam directions, e.g. 15o, 45o each will be processed at weight banks
`
`referring to amplitudes (real-time signal power estimation in Id. at 11:18-23).
`
`74. Hence, it is unclear for a POSA will be motivated to combine
`
`the teachings of Tong using SVD methods that compute V matrices with Mao
`
`that uses a weight-based system and power levels and does not even convert a
`
`sequence x(n) from the time to domain to the frequency domain using an FFT
`
`(Fast Fourier Transform). Further, in Mao the term FFT is not in the patent at
`
`all.
`
`75.
`
`Indeed, both methods convert Analog to Digital, but Mao in
`
`particular implemented in a DSP and will not work on frequency domain, a
`
`POSA will not be motivated to combine Mao and Tong in any way as
`
`processing at the Digital Signal Processor for Tong will requ