`United States Patent No. 5,659,891
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`______________________
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`______________________
`ARRIS GROUP, INC., ARUBA NETWORKS, INC.,
`HEWLETT PACKARD ENTERPRISE COMPANY, and HP, INC.,
`Petitioner,
`v.
`MOBILE TELECOMMUNICATIONS TECHNOLOGIES, LLC
`Patent Owner.
`______________________
`Case IPR2016-00768
`Patent 5,659,8911
`______________________
`Before MEREDITH C. PETRAVICK, SCOTT A. DANIELS, and
`MIRIAM L. QUINN, Administrative Patent Judges.
`
`
`REBUTTAL DECLARATION OF DR. APOSTOLOS K. KAKAES IN
`SUPPORT OF PETITIONERS’ REPLY
`TO PATENT OWNER’S RESPONSE
`
`1.
`
`I, Dr. Apostolos K. Kakaes have previously been asked to testify as an
`
`expert witness in this action. As part of my work in this action, I have been asked
`
`by Petitioners to respond to certain assertions offered by Mobile
`
`Telecommunications Technologies, LLC (“Patent Owner”) in IPR2016-00766 and
`
`
`1 Case IPR2016-00766 has been joined with the instant proceeding.
`
`
`
`1
`
`IPR 2016-00768
`Petitioners, Ex. 1018 Page 00001
`
`
`
`IPR2016-00768
`United States Patent No. 5,659,891
`IPR2016-00768. I hereby declare, under penalty of perjury under the laws of the
`
`United States of America, as follows:2
`
`I.
`
`INTRODUCTION
`
`2.
`
`I am the same Apostolos K. Kakaes who provided Declarations in
`
`these matters executed on August 7, 2015, submitted as Exhibit 1003 (IPR2016-
`
`00768) and ARRIS1003 (IPR2016-00766).3
`
`3. My experience and qualifications are provided in this prior
`
`Declaration (¶¶1-8, 10, 12-13) and curriculum vitae (Ex.1003, pp.19-23).
`
`4.
`
`In this Rebuttal Declaration, I respond to certain assertions in Patent
`
`Owner’s Response (“POR”) (Paper 28) and the Declaration of Dr. Jay P. Kesan
`
`(Ex.2011) submitted on January 9, 2017.
`
`5.
`
`In reaching the conclusions described in this Rebuttal Declaration, I
`
`have relied on the documents and materials cited herein as well as those cited
`
`within and identified in my prior Declaration (see Ex.1003¶9).
`
`6. My opinions are also based upon my education, training, research,
`
`knowledge, and personal and professional experience.
`
`
`2 Throughout this Rebuttal Declaration, all emphasis and annotations are added
`
`unless noted.
`
`3 All references made herein are to Exhibit 1003 in IPR2016-00768.
`
`
`
`2
`
`IPR 2016-00768
`Petitioners, Ex. 1018 Page 00002
`
`
`
`IPR2016-00768
`United States Patent No. 5,659,891
`I hereby declare that all statements made herein of my own
`
`7.
`
`knowledge are true and that all statements made on information and belief are
`
`believed to be true; and further that these statements were made with the
`
`knowledge that willful false statements and the like so made are punishable by fine
`
`or imprisonment, or both, under 18 U.S.C. § 1001. If called to testify as to the truth
`
`of the matters stated herein, I could and would testify competently.
`
`II. OPINIONS
`A. Level of Ordinary Skill in the Art
`In my August 7, 2015 Declaration, I opined that a person of ordinary
`8.
`
`skill in the art (“POSITA”) as of June 7, 1995 (the priority date of the ’891 patent)
`
`would have at least a B.S. degree in electrical engineering, computer science,
`
`computer engineering, or equivalent education. Ex.1003¶10. This person would
`
`also need to have at least two years of experience in the design and configuration
`
`of wireless paging systems, or other two-way wireless communications systems
`
`and be familiar with the operation and functionality of multicarrier transmissions.
`
`Ex.1003¶10. I have reviewed Dr. Kesan’s opinion regarding the level of skill of a
`
`POSITA with respect to the ‘891 patent (requiring “a bachelor’s degree in
`
`electrical engineering or its equivalent and about four years working in the field of
`
`wireless telecommunications networks and…knowledge regarding frequency,
`
`amplitude, and masks as used in telecommunications, or equivalent education and
`
`
`
`3
`
`IPR 2016-00768
`Petitioners, Ex. 1018 Page 00003
`
`
`
`IPR2016-00768
`United States Patent No. 5,659,891
`work experience”). Ex.2011¶9. Under either my definition or Dr. Kesan’s
`
`definition, I met or exceeded the level of skill required as of June 7, 1995, and my
`
`opinions are the same.
`
`B. Claim Construction
`Patent Owner’s assertions and Dr. Kesan’s opinions
`1.
`regarding “the band edge of the mask” are incorrect (claims
`1, 3, 5)
`
`9.
`
`I understand Patent Owner construes “the band edge of the mask” to
`
`mean “the band edge that is nearest to the center frequency of each outer most
`
`carrier at the highest power level of each outer most carrier.” POR27; see also
`
`POR16-43; Ex.2011¶¶47-85, 90. I disagree. I have reviewed the ‘891 patent and,
`
`in my opinion, a POSITA would not have understood the ‘891 specification as
`
`describing “the band edge that is nearest to the center frequency of each outer most
`
`carrier at the highest power level of each outer most carrier.” Instead, a POSITA
`
`would have understood that the band edge of the mask defines the channel, as
`
`required by the claims, which expressly recite: “the band edge of the mask defining
`
`said channel” and a “single mask-defined bandlimited channel.” Ex.1001, claims
`
`1, 3, 5. In my opinion, a POSITA would have understood the ‘891 claims thus
`
`require “the band edge of the mask defining the channel” to be located where the
`
`entire channel has been included. If “the band edge of the mask” is located so as to
`
`include only a portion of the channel, it will no longer “defin[e] the channel.”
`
`
`
`4
`
`IPR 2016-00768
`Petitioners, Ex. 1018 Page 00004
`
`
`
`IPR2016-00768
`United States Patent No. 5,659,891
`Ex.1001, claims 1, 3, 5; see also Ex.2012, 37:11-25, 38:1-15, 46:24-47:10;
`
`Ex.2013, 180:5-181:4.
`
`10.
`
`In addition, the ‘891 specification states “FCC masks typically require
`
`the power spectral density of a signal to be attenuated at least 70 dB at the band
`
`edge.” Ex.1001, 1:57-61; see also Ex.1012, 48; Inst.9. The ‘891 specification
`
`further states “FIG. 4…depict[s] an exemplary FCC emissions mask that requires
`
`the power spectral density to be attenuated at least 70 dB within 10 kHz from
`
`center frequency.” Ex.1001, 3:16-18; see also id., 4:47-49. Thus, a POSITA
`
`would have understood the ‘891 specification describes “the band edge” in Figure
`
`4’s mask is at ±10 kHz from the center frequency. See also Ex.2012, 68:10-69:3,
`
`78:1-6. A POSITA would have also understood that Figure 4 depicts the band
`
`edge of the mask at ±10 kHz, which defines the channel and includes the entire
`
`channel. Fig. 4; see also Ex.2012, 68:21-69:3.
`
`
`
`5
`
`IPR 2016-00768
`Petitioners, Ex. 1018 Page 00005
`
`
`
`IPR2016-00768
`United States Patent No. 5,659,891
`
`Ex.1001, Fig.4.
`
`
`
`11.
`
`I understand Patent Owner asserts that I “confirmed that the term
`
`‘band edge’ is not clear from the plain language of the claim.” POR17-18. I
`
`disagree. As I explained during my deposition and above (¶¶9-10), a POSITA
`
`would have understood the ‘891 claims require “the band edge of the mask
`
`defining the channel” to be located where the entire channel has been included, and
`
`the ‘891 specification further describes “the band edge” in Figure 4’s mask is at
`
`±10 kHz from the center frequency. Ex.2012, 37:11-25, 38:1-15, 46:24-47:10,
`
`68:10-69:3, 78:1-6; Ex.2013, 180:5-181:4. Moreover, in purported support of its
`
`assertion, Patent Owner cites my testimony regarding Petrovic (Ex.2012, 35:22-
`
`36:2), where I did not need to determine the outer bounds of the claim term as
`
`
`
`6
`
`IPR 2016-00768
`Petitioners, Ex. 1018 Page 00006
`
`
`
`IPR2016-00768
`United States Patent No. 5,659,891
`Petrovic discloses the limitation regardless of the outer bounds. Based on a
`
`POSITA’s understanding of the phrase “the band edge of the mask defining the
`
`channel,” as used in the ‘891 claims and specification, Petrovic discloses Elements
`
`1.C, 3.C, and 5.D as discussed below (¶¶42-51).
`
`12.
`
`I understand Patent Owner and Dr. Kesan assert that the ‘891
`
`specification refers to a “nearest band edge” that must be chosen out of multiple
`
`band edges. POR16-17, 20-28, 43, 49-50; Ex.2011¶¶65, 127, 90, 103, 70. I
`
`disagree. I understand Patent Owner and Dr. Kesan base this argument on a
`
`passage in the ‘891 specification (Ex.1001, 4:30-34), which describes Figure 3B
`
`and states the “frequency difference between the center frequency of each carrier
`
`and the nearest band edge of the mask is greater than half the frequency difference
`
`between the center frequencies of the two carriers.” POR16-17, 20-21;
`
`Ex.2011¶127; see also Ex.1001, Fig.3B. But a POSITA would have understood
`
`this passage as using the phrase “nearest band edge” to distinguish the left band
`
`edge from the right band edge of the mask, and not selecting a “nearest band edge”
`
`to be a frequency corresponding to a frequency along for example, the “diagonal”
`
`or “slanted” lines of the Figure 3B mask.
`
`
`
`7
`
`IPR 2016-00768
`Petitioners, Ex. 1018 Page 00007
`
`
`
`IPR2016-00768
`United States Patent No. 5,659,891
`
`Ex.1001, Fig.3B.
`
`
`
`13. For example, as shown in Figure 3B above, a POSITA would have
`
`understood the “nearest band edge” to the left-most carrier 32a is the vertical line
`
`depicting the band edge of the mask on the left side of Figure 3B, as opposed to the
`
`vertical line depicting the band edge of the mask on the right side, which is located
`
`farther away. See also Inst.11. In particular, the ‘891 patent addresses the
`
`relationship between an outer most carrier (i.e., the left-most carrier) and the band
`
`edge on its respective side (i.e., the band edge on the left). The preceding passage
`
`of the ‘891 specification (Ex.1001, 4:17-20) confirms this understanding and states
`
`“the frequency spacings between adjacent carriers…can be smaller than the
`
`
`
`8
`
`IPR 2016-00768
`Petitioners, Ex. 1018 Page 00008
`
`
`
`IPR2016-00768
`United States Patent No. 5,659,891
`frequency spacings between the band edges of the mask and the nearest respective
`
`carrier.” Thus, a POSITA would have understood the ‘891 specification uses
`
`“nearest” to distinguish, i.e., the left band edge from the right band edge, and the
`
`left-most carrier from the right-most carrier. In addition, I have reviewed the ‘891
`
`patent, and it does not describe anywhere that a mask has multiple band edges,
`
`where a “nearest band edge” is chosen out of such multiple band edges on a given
`
`side based on any consideration, including “the highest power level of each outer
`
`most carrier.” Ex.1001.
`
`14. Dr. Kesan asserts “[a] PHOSITA would…understand that the term
`
`‘band edge of a mask’ means all points along the edge of the mask that limits the
`
`frequency band.” Ex.2011¶48; see also Ex.2011¶¶47-51; POR27-29. I disagree.
`
`A POSITA would have understood the ‘891 patent does not describe “the band
`
`edge of a mask” as “all points along the edge of the mask that limits the frequency
`
`band.” The claims require calculating a “frequency difference” between the center
`
`frequency of the outer most carrier and “the band edge of the mask,” thus a
`
`POSITA would have understood “the band edge of the mask” is a single
`
`frequency—or a single point—not “all points” along the mask as Dr. Kesan opines.
`
`Ex.1001, claims 1, 3, 5; see also Ex.2012, 63:16-64:16. Furthermore, a POSITA
`
`would have understood the ‘891 patent describes “the band edge” with respect to
`
`Figure 4’s mask as located at one point on the left side of the mask and one point
`
`
`
`9
`
`IPR 2016-00768
`Petitioners, Ex. 1018 Page 00009
`
`
`
`IPR2016-00768
`United States Patent No. 5,659,891
`on the right side of the mask—not at multiple points along the mask. See, e.g.,
`
`Ex.1001, 1:57-61, 3:16-18, 4:47-49, Fig. 4.
`
`15. Dr. Kesan asserts other “current” FCC masks provided in Appendix C
`
`of his declaration “have multiple band edges.” Ex.2011¶51; see also Ex.2011¶¶49-
`
`50; POR28-29. I disagree. As an initial matter, Appendix C of Dr. Kesan’s
`
`Declaration has a copyright date of 2015. Thus, I understand the “current masks”
`
`described in Appendix C are irrelevant to the understanding of a POSITA as of
`
`June 7, 1995. In any event, Figures 1-3 of Appendix C, which Dr. Kesan cites, do
`
`not even describe a “band edge of a mask,” much less describe that there are
`
`“multiple band edges of a mask.” Dr. Kesan also incorrectly states that “[o]n
`
`June 7, 1995, the applicants filed an information disclosure statement (IDS) with
`
`an exemplary mask.” Ex.2011¶51. The information disclosure statement (IDS)
`
`Dr. Kesan cites was actually filed on August 9, 1995—not June 7, 1995. Ex.1012,
`
`79-84. In addition, the mask provided in the August 9, 1995 IDS does not even use
`
`the term “band edge of the mask,” much less describe that there are “multiple band
`
`edges of the mask.” As explained above, a POSITA would have understood the
`
`‘891 specification describes “the band edge of the mask” as one point on the left
`
`side of the mask and one point on the right side of the mask—not as multiple
`
`points along the mask.
`
`
`
`10
`
`IPR 2016-00768
`Petitioners, Ex. 1018 Page 00010
`
`
`
`IPR2016-00768
`United States Patent No. 5,659,891
`16. Dr. Kesan asserts “the band edge nearest in frequency to the outer
`
`most carrier would be chosen for the asymmetric condition in order to minimize
`
`the frequency distance, Dm, of Drawing 9” and that “minimizing Dm[,] necessarily
`
`minimizes the frequency distance between carriers, Dc,” which “allows more
`
`carriers to be placed in the channel, which increases the message capacity of the
`
`channel, which is the stated goal of the ’891 patent.” Ex.2011¶67; see also
`
`Ex.2011¶40 (Drawing 9 – reproduced below); Ex.2011¶¶52-69; POR33-34, 29-32,
`
`42-43, 51-53. I disagree. The claims do not require minimizing the frequency
`
`difference between the band edge and the outermost carrier nor do they require
`
`minimizing the frequency difference between adjacent carriers. Ex.1001, claims 1-
`
`5.
`
`“Drawing 9” of Dr. Kesan’s Declaration (Ex.2011¶40).
`
`17.
`
`Instead, the ‘891 specification explains that using transmitter co-
`
`location, “carrier spacings far closer than would ordinarily be allowed (e.g., 5 to 10
`
`
`
`
`
`11
`
`IPR 2016-00768
`Petitioners, Ex. 1018 Page 00011
`
`
`
`IPR2016-00768
`United States Patent No. 5,659,891
`kHz) are achievable” (Ex.1001, 4:12-15), however, the ‘891 specification does not
`
`require minimizing those carrier spacings. For example, this same passage of the
`
`‘891 specification describes a range of achievable spacings (e.g., from 5 to 10 kHz)
`
`and does not require that the spacing between carriers is minimized to, e.g., 5 kHz.
`
`The ‘891 specification also explains that the spacing between carriers is considered
`
`among other parameters, explaining that “operating parameters, including the peak
`
`frequency[] deviation[,] bit rate, and carrier frequencies” and “other parameters
`
`can be adjusted so that the carriers generated and transmitted according to the
`
`present invention will remain within the FCC emission limits while providing
`
`optimal transmission performance.” Ex.1001, 4:35-46, Figs. 5A, 6A, 7A; see also
`
`Ex.2013, 191:3-192:3, 192:22-194:15, 211:17-212:8. Figure 3B in the ‘891 patent
`
`further shows that Dm is not always minimized to minimize Dc. For example, in
`
`the below diagram annotated by Patent Owner (see POR21), “(Dm)a” has not been
`
`minimized to minimize “(Dc)a”.
`
`
`
`12
`
`IPR 2016-00768
`Petitioners, Ex. 1018 Page 00012
`
`
`
`IPR2016-00768
`United States Patent No. 5,659,891
`
`POR21.
`
`
`
`18.
`
`I understand Patent Owner also asserts that I “agreed that minimizing
`
`Dm necessarily minimizes Dc.” POR33, 43. I disagree. During my deposition, I
`
`testified that decreasing Dm does not necessarily mean decreasing Dc to satisfy the
`
`claim and that conversely, increasing Dc does not necessarily mean increasing Dm.
`
`Ex.2013, 185:7-17, 186:12-18. Figure 3B of the ‘891 specification confirms this is
`
`true—for example—if Dm in Figure 3B above is decreased, Dc need not
`
`necessarily be decreased to meet the claim requirement of Dm > Dc, and likewise,
`
`if Dc is increased, Dm need not necessarily be increased to meet the claim
`
`requirement of Dm > Dc.
`
`
`
`13
`
`IPR 2016-00768
`Petitioners, Ex. 1018 Page 00013
`
`
`
`IPR2016-00768
`United States Patent No. 5,659,891
`I understand Patent Owner asserts “the nearest band edge” is
`
`19.
`
`determined according to “the highest power level of each outer most carrier.”
`
`POR27; POR23-27; POR34-40. I disagree. I have reviewed the ‘891 patent, and
`
`as discussed below, a POSITA would have understood that it does not describe
`
`selecting a “nearest band edge” according to “the highest power level of each outer
`
`most carrier.” Ex.1001. The purpose of the mask is to limit the power level of the
`
`modulated carrier to prevent interference in adjacent channels. Therefore, it is
`
`circular and illogical to define a property of the mask (namely, the band edge of
`
`the mask) according to the “highest” (or otherwise) power level of the very carrier
`
`that the mask is designed to limit.
`
`20. Dr. Kesan asserts “[a] PHOSITA would understand that using the
`
`mask of Figure 4…in Figure 5A of the ‘891 Patent would not produce the desired
`
`result if the carriers are operating at full power, even though the mask of Figure 4
`
`includes edges at -10 kHz and 10 kHz” and that “[a] PHOSITA would also
`
`understand that using the mask of Figure 4…in Figure 5A of the ’891 Patent could
`
`produce the desired result if the carriers are not operating at full power.”
`
`Ex.2011¶¶75, 77 (Drawings 10, 10b); see also Ex.2011¶¶70-80; POR34-37, 51. I
`
`disagree. The FCC expressly specifies in 47 C.F.R. § 22.106 with respect to the
`
`emissions mask that “the power of any emission shall be attenuated below the
`
`unmodulated carrier power (P) in accordance with the following schedule.”
`
`
`
`14
`
`IPR 2016-00768
`Petitioners, Ex. 1018 Page 00014
`
`
`
`IPR2016-00768
`United States Patent No. 5,659,891
`Ex.1012, 82 (47 C.F.R. § 22.106 (1994)); Ex.1001, 5:10-15. Thus, Dr. Kesan’s
`
`analysis is fundamentally incorrect because it fails to consider at all the
`
`unmodulated carrier power of the signal. A POSITA would have understood that
`
`the power level of the carriers cannot simply be lowered to fit under a mask, as the
`
`mask itself is defined relative to the total power of the unmodulated carrier signal.
`
`Thus, Dr. Kesan’s concepts of “operating at full power” and “not operating at full
`
`power” are completely nonsensical.
`
`21.
`
`In addition, based on the FCC requirements described above, a
`
`POSITA would have understood the “0 dB” in Figure 4 indicates a power level as
`
`a frame of reference relative to the total power of an unmodulated signal.
`
`Ex.1012, 82 (47 C.F.R. § 22.106 (1994)); Ex.1001, 5:10-15. A POSITA would
`
`have also understood the “0 dB” in Figure 5A indicates a power level as a frame of
`
`reference relative to the maximum power level of the modulated signal, as Figure
`
`5A depicts the power spectral density of a modulated signal and the maximum
`
`power level of the modulated signal is shown to be at “0 dB,” which defines the
`
`frame of reference. Thus, placement of the mask as shown in Dr. Kesan’s
`
`“Drawing 10” (reproduced below) is incorrect because it effectively treats Figure
`
`4’s “0 dB” frame of reference (defined relative to the total power of the
`
`unmodulated carrier) and Figure 5A’s “0 dB” frame of reference (defined relative
`
`to the maximum power level of the modulated signal at a given frequency) as the
`
`
`
`15
`
`IPR 2016-00768
`Petitioners, Ex. 1018 Page 00015
`
`
`
`IPR2016-00768
`United States Patent No. 5,659,891
`same frame of reference, while, in fact, they are different. See also Ex.2012,
`
`133:10-135:8.
`
`Ex.2011¶75.
`
`
`
`22. Furthermore, based on the FCC specifications in 47 C.F.R. § 22.106, a
`
`POSITA would have understood that the correct placement of the Figure 4 mask
`
`over the carriers in Figure 5A is based on the total power of the unmodulated
`
`carrier. Ex.1012, 82 (47 C.F.R. § 22.106 (1994)); Ex.1001, 5:10-15. Again,
`
`because the mask is defined relative to the total power of the unmodulated carrier,
`
`the power level of the modulated sub-carriers cannot simply be lowered to fit under
`
`a mask as Dr. Kesan has incorrectly done in “Drawing 10b” (reproduced below).
`
`
`
`16
`
`IPR 2016-00768
`Petitioners, Ex. 1018 Page 00016
`
`
`
`IPR2016-00768
`United States Patent No. 5,659,891
`
`Ex.2011¶77.
`
`
`
`23. A POSITA would have understood the correct placement of Figure
`
`4’s mask over Figure 5A is determined by first calculating the total power of the
`
`unmodulated carrier signal, relative to the “0 dB” reference point of Figure 5A. A
`
`POSITA would have understood the total power of the unmodulated carrier is
`
`equal to the total area under the curve formed by the power spectral density of the
`
`modulated sub-carriers as depicted in Figure 5A. To estimate the total power in
`
`Figure 5A, I first estimated the power level of the signal at 1-kHz intervals from 0
`
`to +10 kHz. At each of the ten 1-kHz intervals, I assessed a best approximation of
`
`the power level (A1BA, A2BA, A3BA, A4BA, A5BA, A6BA, A7BA, A8BA, A9BA,
`
`A10BA) and an approximation of the maximum power level (A1MAX, A2MAX,
`
`A3MAX, A4MAX, A5MAX, A6MAX, A7MAX, A8MAX, A9MAX, A10MAX) relative to the
`
`
`
`17
`
`IPR 2016-00768
`Petitioners, Ex. 1018 Page 00017
`
`
`
`IPR2016-00768
`United States Patent No. 5,659,891
`unknown power level X of the power level at which the power spectral density is
`
`maximal, as shown in Table A below.
`
`Decibels
`from 0 dB
`Frame of
`Reference
`
`-33 dB
`-12 dB
`-1.5 dB
`0 dB
`-1.5 dB
`-1.5 dB
`-2 dB
`-12 dB
`-33 dB
`-50 dB
`5.89 dB
`5.89 dB
`
`Best Approximation
`of Power
`
`A1BA
`A2BA
`A3BA
`A4BA
`A5BA
`A6BA
`A7BA
`A8BA
`A9BA
`A10BA
`P1BA (0 to +10 kHz)
`P2BA
`(-10 kHz to 0 kHz)
`PBA
`(-10 kHz to 10 kHz)
`
`
`8.9 dB
`
`7.79X
`
`TABLE A
`Power Level
`Maximum
`Relative to
`Approximation of
`Power
`Power
`Spectral
`Density (X)
`X/1995
`X/16
`X/1.4
`X
`X/1.4
`X/1.4
`X/1.6
`X/16
`X/1995
`X/100,000
`3.89X
`3.89X
`
`A1MAX
`A2MAX
`A3MAX
`A4MAX
`A5MAX
`A6MAX
`A7MAX
`A8MAX
`A9MAX
`A10MAX
`P1MAX (0 to +10 kHz)
`P2MAX
`(-10 kHz to 0 kHz)
`TotalMAX
`(-10 kHz to 10 kHz)
`
`Decibels from
`0 dB Frame of
`Reference
`
`Power Level
`Relative to
`Power Spectral
`Density (X)
`
`-25 dB
`-6 dB
`0
`0
`0
`0
`-2 dB
`-6 dB
`-24 dB
`-40 dB
`7.1 dB
`7.1 dB
`
`X/316
`X/4
`X
`X
`X
`X
`X/1.6
`X/4
`X/251
`X/10,000
`5.13X
`5.13X
`
`10.1 dB
`
`10.3X
`
`24. For each of the best approximation and maximum approximation, I
`
`added the power level estimates at each interval to obtain an estimate from the 0 to
`
`+10 kHz range, which I determined to be P1BA (0 to +10 kHz) = 3.89 times the power
`
`spectral density (X) and P1MAX (0 to +10 kHz) = 5.13 times the power spectral density (X).
`
`I multiplied this value by two to account for the two-sided power spectral density
`
`in the -10 to +10 kHz range and obtained a best approximation of the total power
`
`of the carriers as PBA (-10 to +10 kHz) = 7.79 times the power spectral density (X) or 8.9 dB
`
`above X and an approximation of the maximum value of the total power of the
`
`carriers as PMAX (-10 to +10 kHz) = 10.3 times the power spectral density (X) or 10.1 dB
`
`
`
`18
`
`IPR 2016-00768
`Petitioners, Ex. 1018 Page 00018
`
`
`
`IPR2016-00768
`United States Patent No. 5,659,891
`above X. A POSITA would have understood that the total power of the modulated
`
`carriers shown in Figure 5A is the same as the total power of the unmodulated
`
`carrier. Based on this amount of power, a POSITA would have understood the
`
`correct placement of the mask in Figure 4 over the carriers in Figure 5A is at
`
`around 8.9-10.1 dB above the 0 dB reference point as shown below in blue. This is
`
`consistent with the ‘891 specification, which states “the carriers [in Figure 5A]
`
`remained within the FCC mask.” Ex.1001, 4:56-63.
`
`Ex.1001, Fig.5A (annotated).
`
`
`
`25.
`
`In addition, the correct placement of Figure 4’s mask over the carriers
`
`in Figure 5A (discussed above) further demonstrates Patent Owner’s proposed
`
`construction is wrong. Using Patent Owner’s construction, the alleged “nearest
`
`
`
`19
`
`IPR 2016-00768
`Petitioners, Ex. 1018 Page 00019
`
`
`
`IPR2016-00768
`United States Patent No. 5,659,891
`band edge,” looking at “the highest power level of each outer most carrier,” would
`
`be located at approximately ±7.5 kHz (shown below). But a band edge of
`
`approximately ±7.5 kHz in the above scenario does not even meet the claim
`
`limitations because the “frequency difference between the center frequency of the
`
`outer most of said carriers and the band edge” (e.g., 7500 Hz – 4590 Hz = 2910
`
`Hz) is not “more than half the frequency difference between the center frequencies
`
`of each adjacent carrier” (e.g., 9180 Hz ÷ 2 = 4590 Hz). A band edge of
`
`approximately ±7.5 kHz in the above scenario is further inconsistent with the ‘891
`
`specification, which specifies “the band edge” of Figure 4’s mask is ±10 kHz from
`
`the center frequency. Ex.1001, 1:57-61, 3:16-18, 4:47-48, Fig.4.
`
`Ex.1001, Fig.5A (annotated)
`
`
`
`
`
`20
`
`IPR 2016-00768
`Petitioners, Ex. 1018 Page 00020
`
`
`
`IPR2016-00768
`United States Patent No. 5,659,891
`26. Dr. Kesan asserts “Drawing 11,” in which Dr. Kesan placed the
`
`August 9, 1995 IDS mask (Ex.1012, 84) over the carriers in Figure 5A also
`
`“confirms that [the] band edge has to be chosen where the limit of the mask is most
`
`likely to be first exceeded by the frequency spread of the carrier due to
`
`modulation.” Ex.2011¶84; see also Ex.2011¶¶81-85; POR37-40, 51. I disagree.
`
`As an initial matter, the information disclosure statement (IDS) cited by Dr. Kesan
`
`was filed on August 9, 1995—after the filing date of the ‘891 patent. Ex.1012, 79-
`
`84. In any event, Dr. Kesan has incorrectly placed the August 9, 1995 IDS mask
`
`over the carriers in Figure 5A and has failed to take into account that the FCC
`
`mask set forth limits the attenuation relative to the total power of the unmodulated
`
`carrier signal. Ex.1012, 82 (47 C.F.R. § 22.106 (1994) (“the power of any
`
`emission shall be attenuated below the unmodulated carrier power (P) in
`
`accordance with the following schedule”)); Ex.1001, 5:11-15.
`
`
`
`21
`
`
`
`IPR 2016-00768
`Petitioners, Ex. 1018 Page 00021
`
`
`
`IPR2016-00768
`United States Patent No. 5,659,891
`Ex.2011¶83 (Drawing 11).
`
`27. Accordingly, Dr. Kesan’s “Drawing 11” is incorrect. To correctly
`
`determine how the August 9, 1995 IDS mask would be placed over the carriers in
`
`Figure 5A, it is necessary to determine the total power of the unmodulated carriers
`
`(P), which as I detailed above is around 8.9 – 10.1 dB above the 0 dB reference
`
`point of the power spectral density. See ¶¶20-24. Based on this amount of power,
`
`a POSITA would have understood the correct placement of the August 9, 1995
`
`IDS mask over the carriers in Figure 5A would be at around 8.9-10.1 dB above the
`
`0 dB reference point as shown below in blue.
`
`Ex.1001, Fig.5A (annotated).
`
`
`
`28. Accordingly, in my opinion, Dr. Kesan’s “Drawing 10,” “Drawing
`
`10b” and “Drawing 11” are incorrect. As specified in 47 C.F.R. § 22.106, which
`
`
`
`22
`
`IPR 2016-00768
`Petitioners, Ex. 1018 Page 00022
`
`
`
`IPR2016-00768
`United States Patent No. 5,659,891
`Dr. Kesan acknowledges but does not apply himself (Ex.2011¶82), the mask itself
`
`is defined relative to the total power of the unmodulated carrier signal. Ex.1012,
`
`82; Ex.1001, 5:10-15. Thus, Dr. Kesan’s analysis is critically flawed, as Dr. Kesan
`
`assumes the power level of the modulated carriers is adjusted to fit the mask
`
`(Ex.2011¶¶77-80), when in fact, the total power of the unmodulated carrier signal
`
`is first determined and the mask is then applied relative to the total power of that
`
`unmodulated carrier signal and a determination is thereafter made as to whether or
`
`not the modulated signal fits within the mask.
`
`2.
`
`Patent Owner’s assertions and Dr. Kesan’s opinions
`regarding “transmitting carriers from the same location”
`are incorrect (claims 1, 3; see also claim 5)
`I understand Patent Owner construes “transmitting carriers from the
`
`29.
`
`same location” as “transmitting carriers from the same location at the same time.”
`
`POR44-46. I disagree. In my opinion, a POSITA would not have understood
`
`“transmitting carriers from the same location” to require transmission of those
`
`carriers “at the same time” in view of the ’891 patent’s claims and specification.
`
`Ex.1001, 2:26-36, 2:37-46, 2:47-59, 4:8-11 (describing only transmitting multiple
`
`carriers from the same location).
`
`30.
`
`I understand Patent Owner and Dr. Kesan assert that Figure 1 of the
`
`’891 patent requires two carriers transmitting “at the same time.” POR59;
`
`Ex.2011¶37. I disagree. The ’891 patent’s description of Figure 1 merely states
`
`
`
`23
`
`IPR 2016-00768
`Petitioners, Ex. 1018 Page 00023
`
`
`
`IPR2016-00768
`United States Patent No. 5,659,891
`that it is “a co-located multicarrier transmitter system in a linear amplifier
`
`configuration….” Ex.1001, 3:44-45. It does not mention that Figure 1 has any
`
`temporal requirement. According to the ’891 patent, Figure 1 reflects “a co-
`
`located multicarrier transmitter system” that comprises two data sources, two
`
`modulators, a summation circuit, a linear RF amplifier, and an antenna. Ex.1001,
`
`3:44-48. The ’891 patent claims and specification do not specify or limit the
`
`modulation scheme employable by Figure 1’s two modulators. Rather, the ’891
`
`patent merely states that “[e]ach modulator converts the incoming digital
`
`information into a representative modulated signal or carrier.” Ex.1001, 3:51-53;
`
`see also id., 3:66-4:1 (same regarding Figure 2). Based on the ’891’s disclosures, a
`
`POSITA would have understood that a multicarrier transmitter system, like the one
`
`depicted in Figure 1, could be used with a number of different modulation
`
`schemes, including a well-known modulation scheme called ON/OFF keying
`
`(“OOK”) (including permutation modulation and multicarrier on-off keying). In
`
`the OOK modulation scheme, at any given instant of time, each carrier is either
`
`keyed “on” (and, therefore, transmitting power) or keyed “off” (and, therefore, not
`
`transmitting power4). Therefore, a POSITA would have understood that during
`
`
`4 As explained below (¶39), a carrier that is keyed “off” is nevertheless operational
`
`and conveying information.
`
`
`
`24
`
`IPR 2016-00768
`Petitioners, Ex. 1018 Page 00024
`
`
`
`IPR2016-00768
`United States Patent No. 5,659,891
`operation of Figure 1’s multicarrier transmitter system under the OOK modulation
`
`scheme there would be time instants where the summation circuit would receive a
`
`non-zero carrier signal (i.e., non-zero power) from both the first and second
`
`modulators and combine them into a single output signal, but that there would also
`
`be time instants where the summation circuit receives a non-zero carrier signal
`
`from only the first modulator or only the second modulator. In the latter scenario,
`
`the summation circuit still “combines” the received modulated signals from each of
`
`the two carriers, but for all practical purposes there is nothing to “combine”
`
`because the summation circuit received only one non-zero signal (from one of the
`
`first or second modulator). Thus, the “combined” output signal would consist of
`
`the signal contribution of a single modulated carrier signal. And, the antenna
`
`would transmit that single modulated carrier signal. Furthermore, and in the
`
`interest of completeness, at yet other times, when both modulators transmit a
`
`logical “0,” they will be in the OFF position, i.e., generate no signal and thus
`
`contribute no power. The antenna, in turn, would also not transmit any power at
`
`such instances. Accordingly, a POSITA would have understood that Figure 1 of
`
`the ’891 patent does not require two carriers transmitting at the same time instant.
`
`31. Similarly, in another well-known modulation scheme called
`
`frequency-shift keying (“FSK”), there will be instants of time where the
`
`summation circuit receives a non-zero signal from only the first or second
`
`
`
`25
`
`IPR 2016-00768
`Petitioners, Ex. 1018 Page 00025
`
`
`
`IPR2016-00768
`United States Patent No. 5,659,891
`modulator, for example, when the waveform of one of the modulators is at 0 (i.e.,
`
`no power is transmitted on that signal) and the othe