Case 6:12-cv-00799-JRG Document 143-3 Filed 04/11/14 Page 1 of 11 PageID #: 4258
`Case 6:12—cv—00799—JRG Document 143-3 Filed 04/11/14 Page 1 of 11 Page|D #: 4258
`
`EXHIBIT C
`
`EXHIBIT C
`
`

`
`Case 6:12-cv-00799-JRG Document 143-3 Filed 04/11/14 Page 2 of 11 PageID #: 4259
`
`UNITED STATES DISTRICT COURT
`
`EASTERN DISTRICT OF TEXAS
`
`TYLER DIVISION
`
`
`
`Civil Action No. 6:12-cv-00799-LED
`
`
`
`RODRIGUEZ DECLARATION IN
`SUPPORT OF SUMMARY JUDGMENT
`OF INDEFINITENESS
`
`
`
`§
`§
`§
`§
`§
`§
`§
`§
`§
`§
`§
`§
`§
`§
`§
`§
`§
`§
`§
`§
`§
`§
`§
`§
`
`
`
`
`
`INVENSYS SYSTEMS, INC.,
`
`Plaintiff,
`
`v.
`
`EMERSON ELECTRIC CO. and
`
`MICRO MOTION INC. USA,
`
`Defendant.
`
`MICRO MOTION INC. USA,
`
`Counterclaim Plaintiff,
`
`v.
`
`INVENSYS SYSTEMS, INC.,
`
`Counterclaim Defendant.
`
`
`I.
`
`INTRODUCTION
`I have been retained by DLA Piper LLP on behalf of Invensys Systems, Inc.
`1.
`
`(“Invensys”) as an expert technical consultant in connection with the above-captioned matter.
`
`2.
`
`I understand that Micro Motion Inc. USA (“Micro Motion” or Counterclaim
`
`Plaintiff) has asserted United States Patent Nos. 5,555,190 ( the “’190 Patent”) and 6,505,131
`
`(the “’131 patent”) (collectively, the “Micro Motion Asserted Patents”) against Invensys. I have
`
`been asked to provide my expert technical analysis and opinions concerning certain claim terms
`
`in the ’131 Patent. This report summarizes my opinions and the basis for those opinions.
`
`3.
`rate of $495 per hour. I am also being reimbursed for any reasonable expenses that I incur (e.g.,
`
`I am being compensated for my time spent consulting on this matter at my usual
`
`
`
`
`
`1
`
`EAST\72753549.1
`
`

`
`Case 6:12-cv-00799-JRG Document 143-3 Filed 04/11/14 Page 3 of 11 PageID #: 4260
`
`travel expenses). My compensation is not contingent on the content of any of my analyses,
`
`opinions or testimony or the outcome of this case.
`
`4.
`
`All of the analysis and opinions that I present in this report are based on my
`
`personal knowledge and my professional judgment. I expect to testify at trial regarding the
`
`matters expressed in this report, in any supplemental reports, or in any testimony that I provide
`
`prior to trial. I reserve the right to alter or supplement my technical analyses or opinions in
`
`response to any new information that I discover, including doing so in response to any criticisms
`
`or alternative opinions or information offered by Micro Motion or any of its experts.
`
`II.
`
`QUALIFICATIONS
`I received a B.S. degree in Electrical Engineering from the University of Texas at
`5.
`
`Austin in 1984, a master’s degree in Electrical Engineering from the Massachusetts Institute of
`
`Technology in 1986, and a Ph.D. degree in Electrical Engineering from the University of Texas
`
`at Austin in 1990.
`
`6.
`
`Since 1990 I have been a faculty member in the Department of Electrical and
`
`Computer Engineering at the University of Arizona. At the university, I hold the following
`
`positions:
`
`• Tenured Associate Professor of Electrical and Computer Engineering (1997-
`present)
`• Director of the Signal and Image Laboratory (1990-present)
`• Director of Graduate Studies for the Department of Electrical and Computer
`Engineering (2000-2003, 2005-present)
`
`7.
`
`My teaching and research activities have included signal/image/video processing,
`
`circuit design (incl. analog, digital, and microprocessor systems), and software development. I
`
`have taught a number of undergraduate and graduate courses at the University of Arizona,
`
`including Digital Signal Processing, Advanced Digital Signal Processing, Digital Image
`
`Processing, Signals and Systems, and Electric Circuits.
`
`8.
`
`My current research activities include the development of algorithms for
`
`automated multiphase (gas/liquid/solid) analysis of porous materials, in collaboration with the
`
`
`
`
`
`2
`
`EAST\72753549.1
`
`

`
`Case 6:12-cv-00799-JRG Document 143-3 Filed 04/11/14 Page 4 of 11 PageID #: 4261
`
`UA Dept. of Soil, Water and Environmental Science. We have been developing algorithms for
`
`automated segmentation and classification of 3-D X-ray computed tomography images of porous
`
`media with applications ranging from theoretical aspects of fluid and interfacial dynamics at the
`
`pore scale to practical applications such as oil recovery and contaminant remediation. The
`
`algorithms we are developing will enable quantitative characterization of solid, liquid, and vapor
`
`phases, and provides boundary conditions for fluid dynamics modeling for prediction of flow
`
`processes. Specifically, we have been implementing and evaluating a 3-D multiphase
`
`segmentation algorithm that uses multiphase k-means clustering for statistical seeding of a
`
`Markov random field image model.
`
`9.
`
`My research activities have also included the development of a digital flow
`
`cytometry system for real-time measurement of biological cells passing through a flow conduit.
`
`A flow cytometer involves a photomultiplier sensor for detecting, classifying, and quantifying
`
`cells passing through a flow conduit. Traditionally, analog electronics have been used in flow
`
`cytometry, but we implemented a digital signal processing system to achieve improved accuracy.
`
`We developed algorithms for processing the digital sensor signal in order to detect cells, quantify
`
`their properties, and classify the cell types. Such techniques can also be used to provide a real-
`
`time digital measurement of the cell flow rate.
`
`10.
`
`In other work, I investigated a digital feedback control algorithm for adaptive
`
`noise cancelation in a voice communication system for the Air Force Office of Scientific
`
`Research. This involved the use of two microphone sensors, one measuring a noisy speech
`
`signal and the other measuring a reference noise signal. The reference noise signal was
`
`processed by an adaptive filter whose gain was automatically generated based on an error signal.
`
`The filtered reference noise signal was then subtracted from the noisy speech signal in order to
`
`achieve noise reduction. Later, I co-authored a research paper regarding VLSI implementation
`
`of an adaptive noise cancelation system.
`
`11.
`
`From 2003 to 2008 I served as Co-Director of Connection One, a National
`
`Science Foundation industry/university cooperative research center of communication circuits
`
`
`
`
`
`3
`
`EAST\72753549.1
`
`

`
`Case 6:12-cv-00799-JRG Document 143-3 Filed 04/11/14 Page 5 of 11 PageID #: 4262
`
`and systems. That work included supervision of a research project involving adaptive control of
`
`voice codec parameters and receiver jitter buffer parameters for real-time, wireless, voice
`
`communication over packet-switched networks.
`
`12.
`
`Other example research projects of mine
`
`include content-adaptive error
`
`concealment for video communication, automated image texture analysis for ovarian cancer
`
`detection using confocal microendoscopy, automated segmentation algorithms for biomedical
`
`applications, and algorithms for automated object tracking in video sequences.
`
`13.
`
`I am a Senior Member of the Institute of Electrical and Electronics Engineers
`
`(IEEE) and the IEEE Signal Processing Society. I have served as a member of the IEEE
`
`Technical Committee on Image, Video, and Multidimensional Signal Processing (2005-2011).
`
`Also, I am General Chair of the 2014 IEEE Southwest Symposium on Image Analysis and
`
`Interpretation. Past appointments include Associate Editor of the IEEE Transactions on Image
`
`Processing, and General Chair of the 2007 IEEE International Conference on Image Processing.
`
`Over the years, I have held positions on many other IEEE committees.
`
`III. MATERIALS CONSIDERED
`14. My opinions are based upon the materials I have considered, as well as my
`
`education, knowledge, and experience. In conducting the technical analysis and in forming the
`
`opinions set forth in this report, I have considered the ’131 patent and its file history. I have also
`
`considered any document or other material cited in this report. My consideration of the materials
`
`is ongoing, and I may amend or supplement this report based on those materials. My anticipated
`
`testimony may be affected by additional information obtained and considered subsequent to my
`
`preparation of this report. If asked to testify at trial or a hearing, I may prepare demonstrative
`
`multimedia presentations and exhibits, such as drawings, schematics, animations, images, and
`
`video in order to explain and illustrate my testimony.
`
`THE PATENT
`A.
`15.
`
`Background
`The ’131 patent, titled “Multi-Rate Digital Signal Processor for Signals from
`
`4
`
`EAST\72753549.1
`
`IV.
`
`
`
`
`
`

`
`Case 6:12-cv-00799-JRG Document 143-3 Filed 04/11/14 Page 6 of 11 PageID #: 4263
`
`Pick-Offs on a Vibrating Conduit,” was filed on June 28, 1999, as patent Appl. No. 09/344,840;
`
`the patent issued to Denis Henrot on Jan. 7, 2003. With regard to the ’131 patent, I have been
`
`informed and understand that the relevant date for consideration of a person having ordinary skill
`
`in the art (“POSITA”) is on or around June 28, 1999, the filing date for the ’131 patent’s
`
`application.
`
`The ’131 patent describes a digital signal processor system to determine mass
`16.
`flow properties for a Coriolis flowmeter.1 The ’131 patent describes a digital signal processor
`that is allegedly capable of “multi-rate digital signal” processing.2 The patent suggests that its
`claimed inventions have the advantage of “finite arithmetic in lieu of floating point arithmetic”,
`
`thus being implementable on smaller and cheaper digital signal processors using less power
`versus conventional digital signal processing chips. 3
`
`Person Having Ordinary Skill in the Art
`B.
`17. With regard to the ’131 patent, it is my opinion that a POSITA would have at
`
`least the following qualifications: (1) a bachelor’s degree in electrical engineering or the
`
`equivalent education through work experience; and (2) three or four years of experience or post-
`
`graduate education. This experience would include digital signal processing and control theory.
`
`I consider myself to be at least of ordinary skill in the art, and I have done the analysis supportive
`
`of my opinions here in the context of a POSITA.
`
`V.
`
`OPINIONS
`Calculating Dot Products
`A.
`The term “calculating dot products” appears in claims 1, 13 and 26 of the ’131
`18.
`patent. The context of the claim limitation as it appears in claims 1, 13 and 26 includes:4
`
`1 ’131 patent at 1:7-13, 2:43-46.
`2 Id. at 2:38-40.
`3 See Id. at 2:47-67.
`4 See ’131 patent (Certificate of Correction), claims 1, 13 and 26.
`
`
`
`
`
`5
`
`EAST\72753549.1
`
`

`
`Case 6:12-cv-00799-JRG Document 143-3 Filed 04/11/14 Page 7 of 11 PageID #: 4264
`
` “calculating dot products of said normalized pulsation and
`said signals from said first pick-off sensor and said
`second pick-off sensor to translate said signals to said
`center frequency.”
`A dot product is an operation performed on two equal-length sequences of
`
`19.
`
`numbers, for example, two vectors. The resulting value of the dot product operation is the sum
`
`of the products of the corresponding numbers in the two sequences of numbers. Given two
`
`vectors, 𝒂=[𝑎1,𝑎2,…,𝑎𝑛] and 𝒃=[𝑏1,𝑏2,…,𝑏𝑛], the dot product of 𝒂 and 𝒃 is defined as
`𝒏
`𝒂∙𝒃=�𝒂𝒊𝒃𝒊
`=𝒂𝟏𝒃𝟏+𝒂𝟐𝒃𝟐+⋯+𝒂𝒏𝒃𝒏
`𝒊=𝟏
`
`Therefore, a POSITA would have understood that “calculating dot products” means “calculating
`
`a single number from two equal-length sequences of numbers by multiplying the corresponding
`
`components in each sequence and adding together the results.”
`
`B.
`20.
`
`Calculating a Normalized Pulsation
`The term “calculating a normalized pulsation” appears in claims 1, 13 and 26 of
`
`the ’131 patent. The context of the claim limitation as it appears in claims 1, 13 and 26
`includes:5 “calculating a normalized pulsation of said normalized frequency of said signals.”
`The term normalized pulsation did not have a well-defined meaning in the field
`21.
`
`and thus was not a term of art. The ’131 patent discloses the normalized pulsation as a single
`number, i.e. a scalar value, defined by the equation,6
`
`the sampling frequency) of the samples.
`
`22.
`
`Therefore, a POSITA would have understood that “calculating a normalized
`
`𝜔𝑑=2Π(12𝐹𝑑)/𝐹𝑠
`where 𝜔𝑑 is the normalized pulsation, 𝐹𝑑 is the estimated frequency, and 𝐹𝑠 is the frequency (i.e.,
`pulsation” in context of the ’131 patent means “calculating a parameter 𝜔𝑑 using the formula
`
`
`5 See ’131 patent (Certificate of Correction), claims 1, 13 and 26.
`6 ‘131 patent at 9:6-14.
`
`
`
`
`
`6
`
`EAST\72753549.1
`
`

`
`Case 6:12-cv-00799-JRG Document 143-3 Filed 04/11/14 Page 8 of 11 PageID #: 4265
`
`samples).”
`
`23.
`
`Examining the units for the various terms in this equation is insightful. The
`
`𝜔𝑑=2𝜋(12𝐹𝑑)/𝐹𝑠 (where 𝐹𝑑 is the estimated frequency and 𝐹𝑠 is the frequency of the
`estimated frequency 𝐹𝑑 represents the number of cycles per second, (commonly known as Hertz
`and abbreviated as Hz), and the sampling frequency 𝐹𝑠 represents the number of samples per
`second. The “12” in the equation represents a decimation ratio and has no units, and there are 2𝜋
`pulsation 𝜔𝑑 that is a single value with the following units:
`𝑠𝑒𝑐𝑜𝑛𝑑÷𝑠𝑎𝑚𝑝𝑙𝑒𝑠𝑟𝑎𝑑𝑖𝑎𝑛𝑠𝑐𝑦𝑐𝑙𝑒 ×𝑐𝑦𝑐𝑙𝑒𝑠 𝑠𝑒𝑐𝑜𝑛𝑑 =𝑟𝑎𝑑𝑖𝑎𝑛𝑠𝑠𝑎𝑚𝑝𝑙𝑒
`
`
`
`
`cos(𝜔𝑑𝑘), where the frequency having units of radians per sample is multiplied by the number
`
`radians per cycle (where a full cycle is equal to 360 degrees, which is equal to 2π radians).
`
`24.
`
`Therefore, the multiplication shown in the equation will result in a normalized
`
`25.
`
`In conventional digital signal processing usage, a frequency value with units of
`
`radians per sample sometimes appears as the argument in a sinusoidal expression, such as
`
`of samples to yield units of radians, which gives the proper units for the argument of a sinusoidal
`
`function. Equivalently, the number of samples may be considered to be unitless, in which case
`
`the “samples” would be dropped from the description of units, and the frequency would have
`
`units of just radians. Therefore, a POSITA would understand that the normalized pulsation as
`
`defined in the ‘131 patent would have units of radians per sample, or just radians.
`
`C.
`26.
`
`Demodulating … To Said Center Frequency
`The term “demodulating … to said center frequency” appears in claims 1, 13 and
`
`26 of the ’131 patent. The context of the claim limitation as it appears in claims 1, 13 and 26
`includes:7
`
` “demodulating said signals from said first pick-off sensor and said
`second pick-off sensor to translate said signals to a center
`frequency, wherein said step of demodulating comprises the
`
`7 See ’131 patent (Certificate of Correction), claims 1, 13 and 26.
`
`
`
`
`
`7
`
`EAST\72753549.1
`
`

`
`Case 6:12-cv-00799-JRG Document 143-3 Filed 04/11/14 Page 9 of 11 PageID #: 4266
`
`steps of:
`calculating a normalized pulsation of said normalized
`frequency of said signals; and
`calculating dot products of said normalized pulsation and
`said signals from said first pick-off sensor and said
`second pick-off sensor to translate said signals to said
`center frequency.”
`This claim limitation includes the phrase, “calculating dot products of said
`
`27.
`
`normalized pulsation and said signals.” This requires calculating the dot product of two
`
`sequences of numbers that do not have the same length. A POSITA would have understood that
`
`it is mathematically impossible to calculate such a dot product because there would be numbers
`
`in one sequence that do not have a corresponding number in the other sequence. In other words,
`
`a POSITA would have understood that it is mathematically impossible to compute a dot product
`
`of a scalar value (a single number, such as the “normalized pulsation”) and a non-trivial vector (a
`
`sequence of numbers such as the signals). The ’131 patent does not disclose how to calculate a
`
`dot product of the normalized pulsation, which is a single value, and the signals, which comprise
`
`multiple values, in order to translate the signals to a center frequency, as claimed in the ’131
`
`patent. Even if one could compute some type of product between the normalized pulsation and
`
`the signals, the ’131 patent does not inform a POSITA how such a product calculation could be
`
`used to “translate said signals to said center frequency,” as claimed in the ’131 patent.
`
`28.
`
`For at least these reasons, the meaning of “demodulating … to said center
`
`frequency” is ambiguous.
`
`29.
`
`I have been informed that Micro Motion has proposed a construction for the
`
`phrase “calculating dot products of said normalized pulsation and said signals from said first
`
`pick-off sensor and said second pick-off sensor to translate said signals to said center frequency”
`
`to mean “calculating dot products of a sequence of data representing the normalized pulsation
`
`and sequence of data representing said signal from said first pick off sensor and said second pick
`
`off sensor to shift the frequency content of the signals.”
`
`30.
`pulsation” (the first sequence) that has the same number of components as the “sequence of data
`
`Assuming that one can find a “sequence of data representing the normalized
`
`
`
`
`
`8
`
`EAST\72753549.1
`
`

`
`Case 6:12-cv-00799-JRG Document 143-3 Filed 04/11/14 Page 10 of 11 PageID #: 4267
`
`representing said signal…” (the second sequence), then in principle, it is at least mathematically
`
`possible to calculate such a dot product. However, looking at Micro Motion’s proposed
`
`construction as a whole, the phrase would still be ambiguous for the following additional
`
`reasons.
`
`31.
`
`First, it is unclear how to ascertain whether any given “sequence of data” would
`
`actually “represent” the normalized pulsation in order to meet this limitation. Normalized
`
`pulsation is not a term of art, and therefore, a POSITA would not understand what “representing”
`
`means in this context.
`
`32.
`
`Second, setting aside this problem of not knowing what it means for a sequence of
`
`data to represent the normalized pulsation, performing such a dot product still cannot “translate
`
`said signals to said center frequency” as required by the claims.
`
`33.
`first sequence should represent the normalized pulsation and the data within the second sequence
`
`In attempting to apply Micro Motion’s proposed construction, the data within the
`
`should represent the sensor signal. At a minimum, in order for a data sequence to “represent” a
`
`physical quantity it must at least have the same units. As explained above, the normalized
`
`pulsation has units of radians. Therefore, the data within the first data sequence must have units
`
`of radians.
`
`34.
`
`The sensor signals represent the motion of the flow tubes. Depending on the
`
`design of the meter, position sensors or velocity sensors may be used for this purpose. For the
`
`purpose of the remaining discussion, I will consider position sensors as exemplary, although the
`
`reasoning set forth is equally applicable to velocity sensors or other sensors. Position sensors
`
`provide an electrical signal that measures the position of the flow tube with respect to some
`
`initial position. Therefore the sensor signal represents a length and has units such as inches.
`
`35.
`
`Assuming a POSITA has a first sequence of data representing the normalized
`
`pulsation (with units of radians), and a second sequence of data representing the (position) sensor
`
`signal (with units of inches), calculating the dot product of these two sequences would result in a
`
`single number with units of “radians × inches,” which is not the correct units for a translated
`
`
`
`
`
`9
`
`EAST\72753549.1
`
`

`
`Case 6:12-cv-00799-JRG Document 143-3 Filed 04/11/14 Page 11 of 11 PageID #: 4268
`
`signal (units of “inches”), nor the correct units for a “center frequency” (units of “cycles per
`
`second”).
`
`36.
`
`Therefore, even though POSITA would not understand what is meant by “the
`
`sequence of data representing the normalized pulsation,” it is possible to analyze the proposed
`
`dot product calculation by examining the units in the equation. Having performed this analysis,
`
`it is clear that even if a dot product can be calculated as in Micro Motion’s proposed
`
`construction, the result of the dot product still does not “translate said signals to said center
`
`frequency” as required by the claim.
`
`37.
`
`For these reasons, even under Micro Motion’s proposed construction, the meaning
`
`of “demodulating … to said center frequency” is ambiguous.
`
` I
`
` declare under penalty of perjury that the foregoing is true and correct.
`
`Executed on April 11, 2014 in San Diego, California.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`________________________________
`
` Jeffrey J. Rodriguez
`
`10
`
`EAST\72753549.1