`
`Filed on behalf of TQ Delta, LLC
`By: Peter J. McAndrews
`Thomas J. Wimbiscus
`Scott P. McBride
`Christopher M. Scharff
`Andrew B. Karp
`McAndrews, Held & Malloy, Ltd.
`500 W. Madison St., 34th Floor
`Chicago, IL 60661
`Tel: 312-775-8000
`Fax: 312-775-8100
`E-mail: pmcandrews@mcandrews-ip.com
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`_____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`_____________
`
`CISCO SYSTEMS, INC.
`Petitioner
`
`v.
`
`TQ DELTA, LLC
`Patent Owner
`_____________
`
`
`Case No. IPR2016-01020
`Patent No. 9,014,243
`_____________
`
`DECLARATION OF ROBERT SHORT, Ph.D.
`
`
`
`
`
`TQ Delta Exhibit 2003
`Cisco Systems, Inc. v. TQ Delta LLC
`IPR2016-01020
`
`1
`
`CSCO-1103
`Cisco v. TQ Delta, IPR2016-01007
`Page 1 of 49
`
`
`
`Declaration Of Robert Short, Ph.D.
`IPR2016-01020
`
`
`
`
`
`TABLE OF CONTENTS
`
`INTRODUCTION AND SUMMARY OF OPINIONS .................................. 1
`
`I.
`
`II.
`
`PROFESSIONAL BACKGROUND AND QUALIFICATIONS .................. 1
`
`III. COMPENSATION AND MATERIALS CONSIDERED .............................. 3
`
`IV. LEGAL STANDARDS APPLIED .................................................................. 4
`
`A.
`
`Person Of Ordinary Skill In The Art ..................................................... 4
`
`V.
`
`INTRODUCTION TO “MULTICARRIER” COMMUNICATIONS
`TECHNOLOGY AND PEAK-TO-AVERAGE RATIO (“PAR”) ................. 5
`
`A. Multicarrier Systems ............................................................................. 5
`
`B.
`
`C.
`
`Peak-to-Average Power Ratio (“PAR”) ................................................ 7
`
`PAR “Problem” ..................................................................................... 9
`
`D. A Note On Terminology .....................................................................13
`
`VI. CLAIM CONSTRUCTION—“SCRAMBLING…A PLURALITY OF
`CARRIER PHASES” ....................................................................................14
`
`VII. OVERVIEW OF ASSERTED REFERENCES—SHIVELY AND
`STOPLER ......................................................................................................17
`
`A.
`
`B.
`
`Shively .................................................................................................17
`
`Stopler..................................................................................................32
`
`VIII. SHIVELY AND STOPLER COULD NOT BE COMBINED .....................43
`
`
`
`
`
`
`
`2
`
`CSCO-1103
`Cisco v. TQ Delta, IPR2016-01007
`Page 2 of 49
`
`
`
`Declaration Of Robert Short, Ph.D.
`IPR2016-01020
`
`
`
`I, Robert Short, do hereby declare as follows:
`
`
`
`
`I.
`
`INTRODUCTION AND SUMMARY OF OPINIONS
`
`1.
`
`I have been retained by TQ Delta, LLC (“Patent Owner” or “TQ
`
`Delta”) as an expert witness and have been asked to analyze U.S. Patent No.
`
`9,014,243 (“’243 patent”).
`
`2.
`
`I understand that Cisco Systems, Inc. (“Cisco” or “Petitioner”) filed a
`
`petition for inter partes review of the ’243 patent based on certain prior art
`
`references. I understand that the inter partes review proceeding has been docketed
`
`as IPR2016-01020.
`
`3.
`
`I understand that the Patent Trial and Appeal Board (“PTAB” or
`
`“Board”) has initiated review of claims 1–25 of the ’243 patent based on the
`
`following references: (1) U.S. Patent No. 6,144,696 (“Shively”); (2) U.S. Patent
`
`No. 6,625,219 (“Stopler”); and (3) U.S. Patent No. 6,424,646 (“Gerszberg”).
`
`II.
`
`PROFESSIONAL BACKGROUND AND QUALIFICATIONS
`
`4.
`
`I am an expert in the field of digital communications, with a strong
`
`emphasis on the physical layer of communications systems.
`
`5.
`
`I am a practicing electrical engineer with over 35 years of experience.
`
`I have attached as Exhibit A to this declaration a current copy of my curriculum
`
`vitae.
`
`
`
`3
`
`CSCO-1103
`Cisco v. TQ Delta, IPR2016-01007
`Page 3 of 49
`
`
`
`Declaration Of Robert Short, Ph.D.
`IPR2016-01020
`
`
`
`
`
`6.
`
`I received a Bachelor of Science degree in electrical engineering in
`
`1975, graduating with honors. I worked for Motorola until late 1981, designing
`
`circuits, radar, and communication systems. I was employed by the Colorado
`
`Networks Operation of Hewlett-Packard from 1981 until 1984 designing computer
`
`networks, and for Sperry Corporation until 1989 designing military wireless
`
`networks.
`
`7.
`
`I was awarded a Ph.D. in electrical engineering from the University of
`
`Utah in 1988 with a GPA of 3.95 out of a possible 4.00. My thesis involved
`
`physical layer techniques for sharing wireless channels in a wireless network
`
`environment using CDMA. I was invited, out a pool of over 800 applicants, to join
`
`the faculty of the University of Utah in 1989, where I taught courses at both the
`
`undergraduate and the graduate level, performed research into advanced wireless
`
`networks, and supervised a number of graduate students.
`
`8.
`
`I left the University in 1995 to co-found a company. I have since
`
`been involved in a number of startup companies, all involving the design and
`
`development of communication systems and networks, including designing radios
`
`based various standards and assisting in the design of a wireline communications
`
`system. I was also employed with Alereon in Austin, Texas in 2003 and 2004, as
`
`Chief Engineer of a startup company focused on building integrated circuits for the
`
`multicarrier ultra-wideband personal area network environment. I have designed
`
`
`
`4
`
`CSCO-1103
`Cisco v. TQ Delta, IPR2016-01007
`Page 4 of 49
`
`
`
`Declaration Of Robert Short, Ph.D.
`IPR2016-01020
`
`or assisted in the design of a number of multicarrier systems, including several
`
`
`
`
`UWB systems, and a frequency hopping OFDM system based on the WiMAX
`
`standard. I have also designed or assisted in the design of various spread-spectrum
`
`and CDMA systems.
`
`9.
`
`I am thoroughly conversant with communications standards and
`
`protocols and wrote a chapter in WiMedia UWB by Ghobad Heidari describing the
`
`ultra-wideband (UWB) physical layer. I am currently working with ViaSat in
`
`Carlsbad, California designing random-access communication systems for satellite
`
`networks using CDMA on the return link. I have published a number of seminal
`
`papers in refereed journals, have a number of patents, and am currently in high
`
`demand as a consultant.
`
`III. COMPENSATION AND MATERIALS CONSIDERED
`
`10.
`
`I am being compensated for my time at a rate of $340/hr. In addition,
`
`I am being reimbursed for my reasonable expenses incurred in connection with my
`
`work on this matter. My compensation is not dependent on the opinions that I
`
`offer or on the outcome of this inter partes review proceeding or any other
`
`proceeding.
`
`11.
`
`I am not an employee of TQ Delta, or any of its affiliates, parents, or
`
`subsidiaries.
`
`
`
`5
`
`CSCO-1103
`Cisco v. TQ Delta, IPR2016-01007
`Page 5 of 49
`
`
`
`Declaration Of Robert Short, Ph.D.
`IPR2016-01020
`
`
`
`
`
`12.
`
`In preparing my declaration, I have relied upon my education,
`
`knowledge, and experience. I have also considered the materials and items
`
`described in this declaration.
`
`IV. LEGAL STANDARDS APPLIED
`
`13.
`
`I am not an expert in patent law, and I am not purporting to provide
`
`any opinions regarding the correct legal standards to apply in this proceeding. I
`
`have been asked, however, to provide my opinions in the context of legal standards
`
`that have been provided to me by counsel for Patent Owner.
`
`A.
`
`14.
`
`Person Of Ordinary Skill In The Art
`
`I understand that an analysis of the claims of a patent in view of prior
`
`art has to be provided from the perspective of a person having ordinary skill in the
`
`art at the time of invention of the ’243 patent.
`
`15.
`
`In rendering the opinions set forth in this declaration, I was asked to
`
`consider the patent claims through the eyes of “a person having ordinary skill in
`
`the art.” I was told by counsel for Patent Owner to consider factors such as the
`
`educational level and years of experience of those working in the pertinent art; the
`
`types of problems encountered in the art; the teachings of the prior art; patents and
`
`publications of other persons or companies; and the sophistication of the
`
`technology. I understand that the person of ordinary skill in the art is not a specific
`
`
`
`6
`
`CSCO-1103
`Cisco v. TQ Delta, IPR2016-01007
`Page 6 of 49
`
`
`
`Declaration Of Robert Short, Ph.D.
`IPR2016-01020
`
`real individual, but rather a hypothetical individual having the qualities reflected
`
`
`
`
`by the factors discussed above.
`
`16. For purposes of this declaration only, I have adopted Dr. Tellado’s
`
`definition of a person of ordinary skill in the art. In particular, Dr. Tellado stated
`
`that a person having ordinary skill in the art would have “(i) a Master’s degree in
`
`Electrical and/or Computer or equivalent training, and (ii) approximately five years
`
`of experience working with multicarrier communications systems.” Ex. 1009 at ¶
`
`18. It is through this hypothetical person’s eyes that I have reviewed the prior art
`
`and the ’243 patent.
`
`V.
`
`INTRODUCTION TO “MULTICARRIER” COMMUNICATIONS
`TECHNOLOGY AND PEAK-TO-AVERAGE RATIO (“PAR”)
`
`A. Multicarrier Systems
`
`17. The ’243 patent discloses a system that communicates using
`
`multicarrier signals. Ex. 1001 at 1:26–29. A multicarrier signal includes a number
`
`of carrier signals (or carriers) each operating at a different frequency. Each carrier
`
`is modulated to encode one or more bits (i.e., “1” or “0”). Each carrier effectively
`
`serves as a separate sub-channel for carrying data. The carriers are combined as a
`
`group to produce a transmission signal, which is transmitted across a transmission
`
`medium (e.g., phone lines, coaxial cable, the air, etc.) to a receiver.
`
`
`
`7
`
`CSCO-1103
`Cisco v. TQ Delta, IPR2016-01007
`Page 7 of 49
`
`
`
`Declaration Of Robert Short, Ph.D.
`IPR2016-01020
`
`
`
`
`
`18.
`
`In an example illustrated below, four carriers—Carrier 1, Carrier 2,
`
`Carrier 3, and Carrier 4—are combined simultaneously into one transmission
`
`signal.
`
`
`
`
`
`19. Multicarrier systems may use the phase of carriers to encode different
`
`bit values. In the example above, a carrier with a phase of zero represents a bit
`
`value of “0”; conversely, a carrier with a phase-shift of π (or 180°) represents a bit
`
`value of “1”. In this example, Carriers 1, 2, and 4 have a phase-shift of π, and
`
`therefore each represent a “1”. Carrier 3 has as phase of zero, and therefore
`
`represents a “0”. Together, these four carriers encode input bits having binary
`
`values of 1, 1, 0, and 1. This information is transmitted as a single transmission
`
`signal—that is, the irregular waveform shown above on the right side of the figure.
`
`In practice, a multicarrier
`
`transmission signal will
`
`typically comprise a
`
`
`
`8
`
`CSCO-1103
`Cisco v. TQ Delta, IPR2016-01007
`Page 8 of 49
`
`
`
`Declaration Of Robert Short, Ph.D.
`IPR2016-01020
`
`combination of many more than four carriers (e.g., hundreds or even thousands of
`
`
`
`
`carriers) and in this way can substantially increase the “speed” or data carrying
`
`capacity of the system.
`
`B.
`
`Peak-to-Average Power Ratio (“PAR”)
`
`20. A multicarrier transmission signal can be characterized by a metric
`
`known as “PAR,” which stands for peak-to-average ratio or peak-to-average power
`
`ratio. Ex. 1001 at 1:60–65. As the ’243 patent explains, “The PAR of a
`
`transmission signal is the ratio of the instantaneous peak value (i.e., maximum
`
`magnitude) of a signal parameter (e.g., voltage, current, phase, frequency, power)
`
`to the time-average value of the signal parameter.” Id. at 1:65–2:2 (emphasis
`
`added). References to PAR herein relate to PAR for the power of a transmission
`
`signal.
`
`21.
`
`In the following illustration, a signal (blue) has a peak power (red
`
`line) and an average power (black line). The ratio of the peak power to the average
`
`power of the signal is the PAR.
`
`
`
`9
`
`CSCO-1103
`Cisco v. TQ Delta, IPR2016-01007
`Page 9 of 49
`
`
`
`Declaration Of Robert Short, Ph.D.
`IPR2016-01020
`
`
`
`
`
`
`
`22. A high PAR can occur when a large number (or percentage) of the
`
`carriers have the same phase. The ’243 patent recognized that: “If the phase of the
`
`modulated carriers [in a transmission signal] is not random, then the PAR can
`
`increase greatly.” Id. at 2:15–16. The phases of the carriers would not be
`
`“random,” for example, when the underlying data being modulated is repetitive
`
`(e.g., a long string of 0s or a long string of 1s), or where the same data is purposely
`
`sent in a redundant manner on multiple carriers. In the example below, all 25 of
`
`the carriers have the same phase of zero. Because the carrier signals are “in-
`
`phase,” their amplitudes will add together to create a transmission signal
`
`(illustrated on the right side of the figure below) having large spikes in amplitude
`
`and, therefore, a high PAR.
`
`
`
`10
`
`CSCO-1103
`Cisco v. TQ Delta, IPR2016-01007
`Page 10 of 49
`
`
`
`Declaration Of Robert Short, Ph.D.
`IPR2016-01020
`
`
`
`
`
`
`
`C.
`
`PAR “Problem”
`
`23. Because a multicarrier transmission signal is the sum of many carrier
`
`signals, the transmission signal is expected to have a significant PAR.
`
`Conventional multicarrier systems, therefore, were designed to accommodate PAR.
`
`There is only a “problem” with PAR, however, when an undue amount of PAR-
`
`induced distortion occurs in a multicarrier transmitter, creating additional errors in
`
`the receiver above the normal noise-induced errors.
`
`24. Electronic components in a multicarrier transceiver are ideally
`
`designed to process multicarrier signals without distortion. Distortion occurs when
`
`a signal exceeds the capacity (or dynamic range) of an electronic component, such
`
`as an amplifier, a digital-to-analog converter, or an analog-to-digital converter.
`
`
`
`11
`
`CSCO-1103
`Cisco v. TQ Delta, IPR2016-01007
`Page 11 of 49
`
`
`
`Declaration Of Robert Short, Ph.D.
`IPR2016-01020
`
`When the maximum dynamic range of a component is exceeded, the signal will
`
`
`
`
`become distorted or will “clip.”
`
`25. As a result of clipping, the portion of the signal exceeding the
`
`component’s dynamic range is truncated and the information in the cut off signal
`
`
`
`portion is lost.
`
`26. One way to reduce clipping is to use transceiver components that have
`
`larger dynamic ranges. Such components, however, can be expensive and may
`
`consume a relatively large amount of power. Increasing the dynamic ranges of the
`
`components, therefore, can be impractical.
`
`27.
`
`Instead of demanding ideal circuitry, multicarrier systems are
`
`designed to actually allow a certain amount of clipping. One design criterion is
`
`
`
`12
`
`CSCO-1103
`Cisco v. TQ Delta, IPR2016-01007
`Page 12 of 49
`
`
`
`Declaration Of Robert Short, Ph.D.
`IPR2016-01020
`
`specified as a “clipping rate.” One such multicarrier system is digital subscriber
`
`
`
`
`line (“DSL”). In DSL at the time of the invention (“ADSL-1995”) the maximum
`
`allowable clipping rate is one in every 107 (ten million) samples, which
`
`corresponds to a clipping probability of 10-7 (or one in ten million). Ex. 1017 at p.
`
`48, § 6.11.1. This exact clipping probability is also referenced in the ’243 patent.
`
`See Ex. 1001 at 2:6–8.
`
`28. DSL is subject technology in the ’243 patent (Ex. 1001 at 3:25–26),
`
`Shively (Ex. 1011 at 1:4–5), Stopler (Ex. 1012 at 12: 23–24), and Gerszberg (Ex.
`
`1013 at 1:19–26). A particular DSL standard in use at the time of the invention is
`
`defined in Exhibit 1017—ANSI standard T1.413-1995 (“ADSL-1995”). The
`
`ADSL-1995 standard is described in Shively (Ex. 1011 at 1:51–53 and 2:12–24).
`
`29.
`
`In ADSL-1995, the ideal sampling rate is approximately 2.2 million
`
`samples/second. Given this sampling rate and a clipping probability of 10-7, there
`
`would be a clipping error when processing a transmission signal about once every
`
`4.55 seconds on average. This clipping rate is deemed acceptable because, at this
`
`rate, error correction methods are capable of fixing the errors cause by clipping.
`
`30. A PAR “problem” exists when the actual clipping rate exceeds the
`
`maximum allowable rate. In the example above, if there is a clipping error once
`
`every 3 seconds (on average), then a PAR problem exists—because 3 seconds is
`
`less than 4.55 seconds. As the inventor of the ’243 patent recognized, “If the phase
`
`
`
`13
`
`CSCO-1103
`Cisco v. TQ Delta, IPR2016-01007
`Page 13 of 49
`
`
`
`Declaration Of Robert Short, Ph.D.
`IPR2016-01020
`
`of the modulated carriers is not random, then the PAR can increase greatly.” Ex.
`
`
`
`
`1001 at 2:15–16. “An increased PAR can result in a system with high power
`
`consumption and/or with high probability of clipping the transmission signal.” Id.
`
`at 2:25–27. Contrarily, if the clipping probability does not increase, then there is
`
`no PAR problem.
`
`31. One way to decrease the impact of PAR and reduce the probability of
`
`clipping is by reducing the overall signal power below the maximum overall signal
`
`power for which the system was designed, as depicted below.
`
`32. The system disclosed in Shively is an example of a system in which
`
`the overall signal power is reduced below the maximum overall signal power for
`
`
`
`
`
`14
`
`CSCO-1103
`Cisco v. TQ Delta, IPR2016-01007
`Page 14 of 49
`
`
`
`Declaration Of Robert Short, Ph.D.
`IPR2016-01020
`
`which the system was designed. Such power reduction is shown by Shively, and it
`
`
`
`
`results in a system with virtually no clipping at all.
`
`D. A Note On Terminology
`
`33. There is some overlap and apparent inconsistency with certain
`
`terminology in the art. Particularly confusing is the use of “symbol.” Generally,
`
`“symbol” can have two meanings. First, “symbol” can refer to information
`
`transmitted on one carrier. Second, “symbol” can refer to all of the information
`
`transmitted in a “symbol period.” In the case of a multicarrier symbol, there are
`
`multiple carrier “symbols” but only one collective “symbol.” The individual
`
`carrier symbols are often referred to as “QAM symbols,” where “QAM”
`
`(Quatrature Amplitude Modulation) is a commonly-used type of modulation used
`
`to modulate a carrier symbol onto a carrier. A multicarrier “symbol” (i.e., the
`
`collection of multiple carrier symbols) in a DMT system is often referred to as a
`
`“DMT symbol,” where “DMT” (Discrete Multitone) is a type of multicarrier
`
`technology.
`
`34.
`
`In order
`
`to keep
`
`things
`
`clear
`
`and
`
`to
`
`avoid
`
`apparent
`
`inconsistency/overlap with the term “symbol,” this declaration employs the terms
`
`“carrier” and “symbol” as follows:
`
`•
`
`“carrier” means a carrier symbol (e.g., a QAM symbol); and
`
`
`
`15
`
`CSCO-1103
`Cisco v. TQ Delta, IPR2016-01007
`Page 15 of 49
`
`
`
`Declaration Of Robert Short, Ph.D.
`IPR2016-01020
`
`
`
`
`
`•
`
`“symbol” means a collective multicarrier symbol in a single symbol
`
`period (e.g., a DMT symbol).
`
`35. This declaration also uses appropriate editorializing to distinguish
`
`between a “carrier” and a “symbol” in the references of record and Petitioners’
`
`expert’s declaration.
`
`36. Further adding to potential confusion is that the terms “carrier,”
`
`“subcarrier,” “band,” “sub-band,” “bin,” “channel,” and “tone” are often used
`
`interchangeably. Ex. 1011 at 1:42–43 (“sub-bands or frequency bins”); id. at 1:48
`
`(“sub-band channels”); id. at 5:13–15 (“carrier”); id. at 10:40–41 (“subcarriers”);
`
`id. at 12:39 (“bin (channel)”); Ex. 1012 at 1:41 (“tones or bands”). For
`
`consistency, “carrier” is used as much as possible in this declaration.
`
`VI. CLAIM CONSTRUCTION————“SCRAMBLING…A PLURALITY OF
`CARRIER PHASES”
`
`37.
`
`In the context of the ’243 patent, “scrambl[e/ing]…a plurality of
`
`carrier phases”—or the variant “scramble…a plurality of phases”—should be
`
`construed to mean “adjusting the phases of a plurality of carriers in a single
`
`multicarrier symbol by pseudo-randomly varying amounts.” This construction is
`
`fully supported by the specification of the ’243 patent, and it clarifies that phase
`
`scrambling is performed amongst individual carrier phases in a single multicarrier
`
`
`
`16
`
`CSCO-1103
`Cisco v. TQ Delta, IPR2016-01007
`Page 16 of 49
`
`
`
`Declaration Of Robert Short, Ph.D.
`IPR2016-01020
`
`symbol. In other words, phase scrambling is not met if the phase adjustment only
`
`
`
`
`occurs over time from one symbol to the next.
`
`38. The ’243 patent is directed exclusively to multicarrier modulation
`
`systems. See Ex. 1001 at 1:26–29 (“This invention relates to communications
`
`systems using multicarrier modulation. More particularly, the invention relates to
`
`multicarrier communications systems that lower the peak-to-average power ratio
`
`(PAR) of transmitted signals.”); 3:32–37 (“Although described with respect to
`
`discrete multitone modulation, the principles of the invention apply also to other
`
`types of multicarrier modulation, such as, but not limited to, orthogonally
`
`multiplexed quadrature amplitude modulation (OQAM), discrete wavelet multitone
`
`(DWMT) modulation, and orthogonal frequency division multiplexing (OFDM).”).
`
`Furthermore, every
`
`independent claim
`
`is directed
`
`to a “multicarrier
`
`communications transceiver” (claims 1, 7, and 20) or a “multicarrier transmitter”
`
`(claim 13). The ’243 patent discloses several multicarrier techniques and uses
`
`“discrete multitone modulation” (“DMT”) as an example. Id. at 3:32–37.
`
`39. A multicarrier signal includes the combination of a plurality of
`
`carriers, where each carrier is transmitted at a different frequency and has its own
`
`phase. In the embodiment of the ’243 patent, each of the plurality of carriers
`
`corresponds to a different QAM symbol. See, e.g., id. at 4:13–14 (“The modulator
`
`46 modulates each carrier signal with a different QAM symbol 58.”). Each carrier
`
`
`
`17
`
`CSCO-1103
`Cisco v. TQ Delta, IPR2016-01007
`Page 17 of 49
`
`
`
`Declaration Of Robert Short, Ph.D.
`IPR2016-01020
`
`(or QAM symbol) has its own phase or phase characteristic. See, e.g., id. at 4:7–9
`
`
`
`
`(“The QAM symbols 58 represent the amplitude and the phase characteristic of
`
`each carrier signal.”). The combination of these carriers (or QAM symbols) is
`
`referred to as a DMT symbol (which is an exemplary type of multicarrier symbol).
`
`See, e.g., id. at 9:8–9 (“…a set of QAM symbols 58 produces a DMT symbol
`
`70….”).
`
`40. The term “phase characteristic” in the ’243 patent is interchangeable
`
`with “phase.” See, id., at 1:40–42 (“The DMT transmitter typically modulates the
`
`phase characteristic, or phase, and amplitude of the carrier signals….”).
`
`41. The ’243 patent repeatedly discloses a “phase scrambler” that
`
`scrambles the phases or phase characteristics of carriers within a single DMT
`
`symbol. See, id., at 6:52–8:13.
`
`42. There is no disclosure in the ’243 patent of scrambling carrier phases
`
`amongst different symbols. As the ’243 patent explains, it is the adjustment of a
`
`plurality of carrier phases within a single DMT symbol that reduces the PAR of the
`
`transmission signal. Ex. 1001 at 4:35–38 and 6:48–53. If the carrier phases were
`
`only adjusted from one DMT symbol to the next, PAR would not be reduced.
`
`
`
`18
`
`CSCO-1103
`Cisco v. TQ Delta, IPR2016-01007
`Page 18 of 49
`
`
`
`Declaration Of Robert Short, Ph.D.
`IPR2016-01020
`
`VII. OVERVIEW OF ASSERTED REFERENCES—SHIVELY AND
`STOPLER
`
`
`
`
`A.
`
`Shively
`
`43. Shively discloses a concept that is intended to increase the useable
`
`bandwidth in a multicarrier communications system. Ex. 1011 at 1:5–20. Shively
`
`teaches a theoretical way to transmit data over a transmission medium having high
`
`signal attenuation at frequencies corresponding to a significant number of carriers.
`
`44. To appreciate Shively’s teachings, it necessary to understand such
`
`impaired transmission mediums. Shively specifically describes “long loop”
`
`systems, where the length of cable between transmitting and receiving DSL
`
`modems is at least 18,000 feet (about 3.4 miles):
`
`Referring to FIGS. 1 and 2, a transmitting modem 31 is connected to a
`receiving modem 32 by a cable 33 having four twisted pairs of
`conductors. In long loop systems where cable 3 is of length of the
`order 18,000 feet or more, high signal attenuation at higher
`frequencies (greater than 500 kHz) is usually observed. This
`characteristic of cable 33 is represented graphically by curve A in
`FIG. 1.
`
`Id. at 9:63–10:2 (emphasis added). See also id. at 11:11–12 (“Such noisy and/or
`
`highly attenuated sub-bands can occur for example in long-run twisted pair
`
`conductors.”).
`
`45. FIG. 1 of Shively, which is annotated with color below, is illustrative:
`
`
`
`19
`
`CSCO-1103
`Cisco v. TQ Delta, IPR2016-01007
`Page 19 of 49
`
`
`
`Declaration Of Robert Short, Ph.D.
`IPR2016-01020
`
`
`
`
`
`
`
`46. FIG. 1 of Shively shows carriers at increasing frequencies along the x-
`
`axis. Each carrier is delineated by vertical lines. Power level is indicated along the
`
`y-axis. Green line (A) represents an attenuation/noise floor, which increases as a
`
`function of frequency. Id. at 2:1–12. Shively explains that attenuation at higher
`
`frequencies is a problem across long cables. Id. at 9:65–10:2 (“In long loop
`
`systems where cable 3 is of length of the order 18,000 feet or more, high signal
`
`attenuation at higher frequencies (greater than 500 kHz) is usually observed.”).
`
`Green line (A) is a characteristic of a communications channel, and it does not
`
`illustrate a transmitted signal. Id. at 10:61–11:12.
`
`
`
`20
`
`CSCO-1103
`Cisco v. TQ Delta, IPR2016-01007
`Page 20 of 49
`
`
`
`Declaration Of Robert Short, Ph.D.
`IPR2016-01020
`
`
`
`
`
`47. Blue
`
`line
`
`(B)
`
`is
`
`the minimum power margin above
`
`the
`
`attenuation/noise floor (green line (A)) that is required to transmit a single bit. Id.
`
`at 2:8–10. Red line (C) illustrates a “spectral density mask,” which is a type of
`
`power limit imposed by system design. Id. at 2:10–12 (“Curve C represents the
`
`limits imposed by a power spectral density mask imposed by an external
`
`communications standard.”). Power transmitted in a given carrier cannot exceed
`
`red line (C).
`
`48. As FIG. 1 illustrates, blue line (B) is below red line (C) for the carriers
`
`shaded in purple. In these purple-shaded carriers, there is sufficient headroom to
`
`transmit a signal representing at least one bit without exceeding a power limit. For
`
`the carriers shaded in orange, however, blue line (B) exceeds red line (C). Because
`
`the noise and attenuation for these orange-shaded carriers is too high (A), a bit
`
`cannot be reliably transmitted without exceeding the imposed spectral density
`
`mask. In other words, the minimum required power margin (B) is greater than the
`
`spectral density mask (C). Id. at 10:65–11:3.
`
`49. Shively proposes a way to transmit data on some of the orange-shaded
`
`carriers. Specifically, replicated data is sent across multiple orange-shaded
`
`(impaired) carriers using power or energy levels at or below the spectral density
`
`mask—i.e., red line (C). Because red line (C) is lower than blue line (B), the
`
`power level on a given orange-shaded carrier is too low to reliably transmit a bit.
`
`
`
`21
`
`CSCO-1103
`Cisco v. TQ Delta, IPR2016-01007
`Page 21 of 49
`
`
`
`Declaration Of Robert Short, Ph.D.
`IPR2016-01020
`
`Shively makes use of this available power by “spreading” a single bit of data
`
`
`
`
`across multiple impaired carriers.
`
`50. Shively’s concept combines (adds) these otherwise too-low-power
`
`signals (that were sent on the impaired carriers) at the receiver to recover the
`
`information. “According to the invention, digital modulator 14 replicates
`
`(‘spreads’) a k-bit symbol over multiple adjacent bands with correspondingly less
`
`energy in each band. At the receiving end, detector 49 coherently recombines
`
`(‘despreads’) the redundant symbols in the noisy/attenuated sub-bands. In
`
`recombining the symbols, the symbols are simply arithmetically added. Because
`
`the noise is incoherent while the signal is coherent, the noise tends to be averaged
`
`out while the signal is reinforced by the addition process.” Id. at 11:16–24.
`
`51. Although not explicitly depicted in FIG. 1, one having ordinary skill
`
`in the art understood that there are carriers in addition to those shaded in purple
`
`and orange that are completely unusable under any circumstance—even with
`
`Shively’s concept. According to this reality, FIG. 1 can be expanded to look like
`
`this:
`
`
`
`22
`
`CSCO-1103
`Cisco v. TQ Delta, IPR2016-01007
`Page 22 of 49
`
`
`
`Declaration Of Robert Short, Ph.D.
`IPR2016-01020
`
`
`
`
`
`
`
`52. The pink-shaded carriers are at frequencies higher than the orange-
`
`shaded carriers. The pink-shaded carriers are completely unusable because the
`
`noise/attenuation floor (green line (A)) is greater than the imposed spectral density
`
`mask power limit (red line (C)).
`
`53. Shively discloses two different modes of operation for ADSL-1995:
`
`(1) a normal mode and (2) a “power-boost” mode. The normal mode is referenced
`
`by Shively’s statement that: “The other limit is on the aggregate power, also
`
`defined by an external communication standard, e.g., ANSI Standard T1.413-1995
`
`[(ADSL-1995)] limits the total power for all sub-bands to 100 mWatts.” Ex. 1011
`
`at 2:12–15. When referring to the cited standard, one having ordinary skill in the
`
`
`
`23
`
`CSCO-1103
`Cisco v. TQ Delta, IPR2016-01007
`Page 23 of 49
`
`
`
`Declaration Of Robert Short, Ph.D.
`IPR2016-01020
`
`art would have understood that this aggregate power limit corresponds to the
`
`
`
`
`normal mode. The normal mode has an aggregate power limit of 20.4 dBm, which
`
`is about 109.6 mW (approximately 100 mW). Ex. 1017 at p. 65. § 6.13.3 (“The
`
`normal aggregate power level shall not exceed…20.4 dBm if all sub-carriers are
`
`used[].”).
`
`54. The power-boost mode of ADSL-1995 is also described by Shively:
`
`The power spectral density mask may be dictated by the standard used
`in a particular country implementing the standard (such as A.N.S.I.
`standard T1.413-1995 [(ADSL-1995)])…For example, the power
`limit for frequencies or tones between 0 and 200 kilohertz must be
`less than -40 dBm/Hz (a power level referenced to one milliwatt over
`1 Hz bandwidth). Above 200 kHz (to frequencies in the megahertz of
`spectrum), the constraint may be -34 dBm/Hz.
`
`Ex. 1011 at 1:51–65. When referring to ADSL-1995, one having ordinary skill in
`
`the art would have understood that this spectral density mask scheme—lower
`
`power (-40 dBm/Hz) at lower frequencies and higher power (-34 dBm/Hz) at
`
`higher frequencies—describes the ADSL-1995 power-boost mode.
`
`55. The power-boost mode is illustrated below in a figure excerpted from
`
`the ADSL-1995 standard, Ex. 1017 at p. 66:
`
`
`
`24
`
`CSCO-1103
`Cisco v. TQ Delta, IPR2016-01007
`Page 24 of 49
`
`
`
`Declaration Of Robert Short, Ph.D.
`IPR2016-01020
`
`
`
`
`
`
`
`Note, in the figure above, “PSD” stands for power spectral density, which
`
`corresponds to the spectral density mask power limit. See Ex. 1017 at p. 61, § 6.8.
`
`56. While
`
`the normal mode has an aggregate power
`
`limit of
`
`approximately 110 mW, the aggregate limit of the power-boost mode is
`
`approximately 344 mW. See Ex. 1017 at p. 66 (“a power boost…total power = the
`
`sum of the powers (-4 + 10log(ncdown1)) and (2 + 10log(ncdown2)), where
`
`ncdown1 and ncdown2 are the number of subcarriers used in the sub-bands i = 0 to
`
`50, and i = 51 to 255, respectively.”). Petitioners’ expert, however, misunderstood
`
`Shively by imagining that a 100 mW aggregate power limit would be used in the
`
`power-boost mode. In particular, during cross-examination regarding the bases for
`
`
`
`25
`
`CSCO-1103
`Cisco v. TQ Delta, IPR2016-01007
`Page 25 of 49
`
`
`
`Declaration Of Robert Short, Ph.D.
`IPR2016-01020
`
`his opinions regarding Shively, Dr. Tellado testified that he interpreted Shively as
`
`
`
`
`disclosing a single mode in which the spectral density mask is -40 dBm/Hz for
`
`carriers in the frequency band up to 200 kHz and -34 dBm/Hz for carriers in the
`
`frequency band above 200 kHz and an aggregate power limit of approximately 100
`
`mW. Ex 2002 at 43:7–25. But this is wrong because it is inconsistent with the
`
`ADSL-1995 standard to which Shively’s disclosure is directed.
`
`57.
`
`In fact, having such a low aggregate power limit (less than 1/3 of the
`
`actual limit) would defeat the purpose of having a power-boost mode. One having
`
`ordinary skill in the art would have readily paired the correct aggregate power limit
`
`and spectral density mask for a particular mode when consulting the ADSL-1995
`
`standard to which Shively’s disclosure is directed. To the extent Shively is
`
`incorrectly inte