`571-272-7822
`
`
`Paper 15
`Entered: December 16, 2016
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`____________
`
`MICROSOFT CORPORATION, MICROSOFT MOBILE INC.,
`SAMSUNG ELECTRONICS AMERICA, INC., and SAMSUNG
`ELECTRONICS CO. LTD.,
`Petitioner,
`
`v.
`
`FASTVDO LLC,
`Patent Owner.
`____________
`
`Case IPR2016-01179
`Patent 5,850,482
`____________
`
`
`Before JEFFREY S. SMITH, PATRICK M. BOUCHER, and PETER P.
`CHEN, Administrative Patent Judges.
`
`SMITH, Administrative Patent Judge.
`
`DECISION
`Instituting Inter Partes Review
`37 C.F.R. § 42.108
`
`
`
`
`
`
`
`
`
`
`
`IPR2016-01179
`Patent 5,850,482
`
`
`I. INTRODUCTION
`
`Petitioner filed a Petition for inter partes review of claims 1–3, 5, 6,
`
`12–14, 16, 17, and 28 of U.S. Patent No. 5,850,482 (Ex. 1001, “the ’482
`
`patent”). Paper 3 (“Pet.”). Patent Owner filed a Preliminary Response.
`
`Paper 12 (“Prelim. Resp.”). Institution of an inter partes review is
`
`authorized by statute when “the information presented in the petition . . . and
`
`any response . . . shows that there is a reasonable likelihood that the
`
`petitioner would prevail with respect to at least 1 of the claims challenged in
`
`the petition.” 35 U.S.C. § 314(a); see 37 C.F.R. § 42.108.
`
`Upon consideration of the Petition and the Preliminary Response, we
`
`are persuaded Petitioner has demonstrated a reasonable likelihood that it
`
`would prevail in establishing the unpatentability of claims 1–3, 5, 6, 12–14,
`
`16, 17, and 28 of the ’482 patent. Accordingly, we institute an inter partes
`
`review.
`
`
`
`A. Related Matters
`
`The ’482 patent is the subject of the following related litigations:
`
`FastVDO LLC v. AT&T Mobility LLC et al., Case No. 3:16-cv-00385
`
`(S.D. Cal.), filed Feb. 11, 2016. Pet. 2; Paper 8 (Patent Owner’s Mandatory
`
`Notice).
`
`FastVDO LLC v. LG Electronics, Inc. et al., Case No. 3:16-cv-00386
`
`(S.D. Cal.), filed Feb. 11, 2016. Pet. 2; Paper 8 (Patent Owner’s Mandatory
`
`Notice).
`
`FastVDO LLC v. NEC Corp. et al., Case No. 3:16-cv-00389 (S.D.
`
`Cal.), filed Feb. 11, 2016 (terminated). Pet. 2; Paper 8 (Patent Owner’s
`
`Mandatory Notice).
`
`2
`
`
`
`IPR2016-01179
`Patent 5,850,482
`
`
`FastVDO LLC v. Nokia Corp. et al., Case No. 3:16-cv-00390 (S.D.
`
`Cal.), filed Feb. 11, 2016. Paper 8 (Patent Owner’s Mandatory Notice).
`
`FastVDO LLC v. ZTE Corp. et al., Case No. 3:16-cv-00394 (S.D.
`
`Cal.), filed Feb. 11, 2016. Pet. 2; Paper 8 (Patent Owner’s Mandatory
`
`Notice).
`
`FastVDO LLC v. Dell Inc. et al., Case No. 3:16-cv-00395 (S.D. Cal.),
`
`filed Feb. 11, 2016. Paper 8 (Patent Owner’s Mandatory Notice).
`
`FastVDO LLC v. Huawei Technologies Co., et al., Case No. 3:16-cv-
`
`00396 (S.D. Cal.), filed Feb. 11, 2016. Pet. 2; Paper 8 (Patent Owner’s
`
`Mandatory Notice).
`
`The ’482 patent is also the subject of IPR2016-01203. Paper 8 (Patent
`
`Owner’s Mandatory Notice).
`
`
`
`B. The ’482 Patent
`
`
`
`The ’482 patent relates generally to error resilient methods and
`
`apparatus for entropy coding, and the application of error resilient coding to
`
`image compression. Ex. 1001, 1:5–11. Entropy coding reduces the number
`
`of bits required to represent a data set by using variable length coding in a
`
`manner which exploits the statistical probabilities of various symbols in the
`
`data set. Id. at 4:36–39. For example, entropy coding assigns shorter code
`
`words to those symbols which occur frequently, while longer code words are
`
`assigned to those symbols which occur less frequently. Id. at 4:40–43.
`
`Error resilient entropy coding can utilize unequal error protection
`
`techniques, isolate effects of a bit error to a single code word, and constrain
`
`the resulting error to an interval. Id. at 6:33–47.
`
`3
`
`
`
`IPR2016-01179
`Patent 5,850,482
`
`
`
`The error resilient method and apparatus includes a code word
`
`generator that encodes data pursuant to split field coding, in which each code
`
`word includes a prefix field and an associated suffix field. Id. at Abstract.
`
`The prefix field includes information representing a characteristic of the
`
`suffix field, such as the length. Id. The suffix field includes information
`
`representing at least some of the original data. Id. If the prefix field is
`
`decoded without any errors, the method and apparatus can correctly
`
`determine the length of the suffix field and the range of values represented
`
`by the suffix field such that the suffix field is resilient to errors. Id. To
`
`increase the probability that the prefix field is correctly decoded, the prefix
`
`field is protected to a greater degree than the suffix field, such that the data
`
`can be more efficiently compressed. Id. Figure 1 of the ’482 patent is
`
`reproduced below.
`
`
`
`4
`
`
`
`IPR2016-01179
`Patent 5,850,482
`
`
`Figure 1 above shows a block diagram of error resilient data
`
`compression apparatus 10, including error resilient data encoder 16. Id. at
`
`8:48–51. Original data is initially transformed by data transformer 12. Id. at
`
`9:30–32. The original data can be transformed based upon one of a number
`
`of predetermined functions, such as a cosine function, a complex
`
`exponential function, or a wavelet transform. Id. at 9:41–52. The
`
`transformed data is then quantized by data quantizer 14 such that the
`
`quantized data has fewer unique data values or coefficients than the
`
`transformed data. Id. at 11:36–38. In one embodiment, transformed
`
`coefficients whose magnitudes fall below a certain level, called a clipping
`
`threshold, are designated as insignificant and set to zero. Id. at 11:55–61.
`
`Entropy encoder 16 shown in Figure 1 above includes code word
`
`generator 26 to generate code words that represent the quantized significant
`
`coefficients. Id. at 13:36–39. Each code word includes a first portion, or
`
`prefix field, and an associated second portion, or suffix field. Id. at 13:41–
`
`43. Code word generator 26 includes prefix generator 27 for generating the
`
`prefix field of each code word and suffix generator 28 for generating the
`
`associated suffix field of each code word. Id. at 13:44–48. Because each
`
`code word is formed of two fields, this method of coding is termed split field
`
`coding. Id. at 13:48–50.
`
`According to split field coding, the prefix field includes information
`
`representative of the associated suffix field, such as the number of characters
`
`which form the suffix field, or the range of coefficient values represented by
`
`the suffix field. Id. at 13:51–63. If the prefix field is decoded correctly, the
`
`length of the suffix field and the range of values represented by the suffix
`
`field can be determined. Id. at 15:61–66. Bit errors within the suffix field
`
`5
`
`
`
`IPR2016-01179
`Patent 5,850,482
`
`will not result in loss of code word synchronization, but instead will be
`
`isolated to that single code word. Id. at 16:1–4. Also, the resulting error
`
`will be within the range of coefficient values included in the prefix field. Id.
`
`at 16:4–9.
`
`The prefix fields are encoded at an appropriately high level of error
`
`protection in order to provide a high probability that the prefix fields will be
`
`decoded correctly. Id. at 16:15–18. The suffix field can be encoded with a
`
`lower level of error protection, which reduces storage requirements and
`
`transmission bandwidth while still providing error resiliency. Id. at 16:18–
`
`27. To provide error protection, entropy encoder 16 includes unequal error
`
`protection means 29 for providing appropriate levels of error protection to
`
`the encoded data. Id. at 17:1–4.
`
`
`
`C. Illustrative Claim
`
`Claims 1, 12, and 28 of the challenged claims of the ’482 patent are
`
`independent. Claim 1 is illustrative of the claimed subject matter:
`
`An error resilient method of encoding data
`1.
`comprising the steps of:
`
`generating a plurality of code words representative of
`respective portions of the data, wherein each code word
`comprises a first portion and an associated second portion, and
`wherein said code word generating step comprises the steps of:
`
`generating the first portion of each code word,
`wherein said first portion generating step comprises the
`step of including information within the first portion that
`is representative of a predetermined characteristic of the
`associated second portion; and
`
`generating the second portion of each code word,
`wherein said second portion generating step comprises the
`step of including information within the second portion
`
`6
`
`
`
`IPR2016-01179
`Patent 5,850,482
`
`
`that is representative of the respective portion of the data;
`and
`
`providing error protection to at least one of the first
`portions of the plurality of code words while maintaining any
`error protection provided to the respective second portion
`associated with the at least one first portion at a lower level than
`the error protection provided to the respective first portion.
`
`Ex. 1001, 18:8–29.
`
`
`
`D. References
`
`Petitioner relies on the following references. Pet. 4.
`
`
`Wei
`Kato
`
`
`US 5,243,629
`US 5,392,037
`
`Sept. 7, 1993
`Feb. 21, 1995
`
`Ex. 1004
`Ex. 1002
`
`E. Asserted Grounds of Unpatentability
`
`
`
`Petitioner contends that claims 1–3, 5, 6, 12–14, 16, 17, and 28 of the
`
`’482 patent are unpatentable based on the following specific grounds:
`
`Reference(s)
`
`Kato
`
`Kato and Wei
`
`
`
`Basis
`
`Challenged Claims
`
`§ 103
`
`1–3, 5, 12–14, 16, and 28
`
`§ 103
`
`6 and 17
`
`II. ANALYSIS
`
`A. Claim Construction
`
`The term of the ’482 patent expired April 17, 2016. For claims of an
`
`expired patent, the Board’s claim interpretation is similar to that of a district
`
`court. See In re Rambus Inc., 694 F.3d 42, 46 (Fed. Cir. 2012). “[T]he
`
`words of a claim ‘are generally given their ordinary and customary meaning’
`
`. . . that the term would have to a person of ordinary skill in the art in
`
`question at the time of the invention.” Phillips v. AWH Corp., 415 F.3d
`
`7
`
`
`
`IPR2016-01179
`Patent 5,850,482
`
`1303, 1312–13 (Fed. Cir. 2005) (en banc) (citations omitted). “[T]he person
`
`of ordinary skill in the art is deemed to read the claim term not only in the
`
`context of the particular claim in which [it] appears, but in the context of the
`
`entire patent, including the specification.” Phillips, 415 F.3d at 1313. For
`
`example, a “claim construction that excludes [a] preferred embodiment
`
`[described in the specification] ‘is rarely, if ever, correct and would require
`
`highly persuasive evidentiary support.’” Adams Respiratory Therapeutics,
`
`Inc. v. Perrigo Co., 616 F.3d 1283, 1290 (Fed. Cir. 2010) (citation omitted).
`
`But “a claim construction must not import limitations from the specification
`
`into the claims.” Douglas Dynamics, LLC v. Buyers Products Co., 717 F.3d
`
`1336, 1342 (Fed. Cir. 2013) (citation omitted). Therefore, “it is improper to
`
`read limitations from a preferred embodiment described in the specification–
`
`even if it is the only embodiment–into the claims absent a clear indication in
`
`the intrinsic record that the patentee intended the claims to be so limited.”
`
`Dealertrack, Inc. v. Huber, 674 F.3d 1315, 1327 (Fed. Cir. 2012) (citation
`
`omitted).
`
`1. “code word”
`
`Each of independent claims 1, 12, and 28 recites “a plurality of code
`
`words.” Petitioner contends one of ordinary skill in the art would have
`
`understood “code word” as used in the ’482 patent to mean “‘code from a
`
`code book representing a symbol.’” Pet. 19. Patent Owner does not propose
`
`construction of the phrase “code word” in its Preliminary Response.
`
`The language of claims 1–3, 5, 6, 12–14, 16, 17, and 28 at issue does
`
`not contain the word “codebook.” In support of the argument that the
`
`claimed “code word” is limited to code “from a codebook,” Petitioner relies
`
`on the following language from the Specification of the ’482 patent: “‘The
`
`8
`
`
`
`IPR2016-01179
`Patent 5,850,482
`
`assignment of code words for entropy coding is typically governed by means
`
`of a codebook which must be known to both the encoder and decoder.’” Pet.
`
`19 (emphasis omitted) (citing Ex. 1001, 4:51–54). But it is improper to read
`
`limitations from the Specification into the claims absent a clear disavowal.
`
`See Dealertrack, 674 F.3d at 1327; Douglas Dynamics, 717 F.3d at 1342.
`
`Here, the Specification merely states that the code word is typically
`
`governed by a codebook. The Specification does not state that the code
`
`word must be generated from a codebook. We decline to include the
`
`“codebook” limitation in our construction of the term “code word.”
`
`The ’482 patent explains that entropy coding assigns shorter code
`
`words to symbols which occur frequently, and longer code words to symbols
`
`which occur less frequently. Ex. 1001, 4:36–43; 14:27–31 (italics added).
`
`An example of entropy coding is Huffman coding. Id. at 4:25–26, 43–46.
`
`The ’482 patent also explains that Huffman coding represents data symbols
`
`using code words that each have a length consisting of a number of bits. Id.
`
`at 4:43–46 (italics added).
`
`Based on the record before us, in light of these explanations, we
`
`construe “code word” as used in the ’482 patent to encompass “a number of
`
`bits assigned to a symbol and used to represent the symbol.”
`
`2. “first portion of each code word”
`
`Each of independent claims 1 and 28 recites a “first portion of each
`
`code word.” Ex. 1001, Claim 12 recites “a plurality of code words . . .
`
`which have respective first and second portions.”
`
`Petitioner contends one of ordinary skill in the art would have
`
`understood “first portion of each code word” as used in the ’482 patent to
`
`mean, “‘prefix field of a code word generated in a manner such that a bit
`
`9
`
`
`
`IPR2016-01179
`Patent 5,850,482
`
`error in the field could result in a potential loss of code word
`
`synchronization.’” Pet. 20 (citation omitted). Patent Owner does not
`
`propose construction of the phrase “first portion of each code word” in its
`
`Preliminary Response.
`
`The language of independent claims 1, 12, and 28 does not include the
`
`requirement that the first portion is generated in a manner such that a bit
`
`error could result in loss of code word synchronization. The ’482 patent
`
`discloses that each code word includes at least a first portion, termed a prefix
`
`field, and an associated second portion, termed a suffix field. Ex. 1001,
`
`13:41–50. The ’482 patent also provides that “Each code word has . . . a
`
`first or prefix field which is susceptible to bit errors. . . . [T]he code words
`
`can be generated such that a bit error in the prefix field of a code word could
`
`result in a potential loss of code word synchronization . . . . ” Id. at 6:52–59
`
`(italics added). The ’482 patent states that the first portion is susceptible to
`
`errors, but does not state that the first portion must be generated such that an
`
`error in the prefix field could result in loss of synchronization. Rather, the
`
`’482 patent uses permissive language, stating that the first portion can be
`
`generated such that a bit error could result in loss of synchronization.
`
`Therefore, we decline to import the merely permissive language into the
`
`claim.
`
`For purposes of this Decision, we construe a “first portion of each
`
`code word” as used in the ’482 patent to encompass “a first or prefix field
`
`which is susceptible to bit errors.”
`
`3. “second portion of each code word”
`
`Each of independent claims 1 and 28 recites a “second portion of each
`
`code word.” Claim 12 recites “a plurality of code words . . . which have
`
`10
`
`
`
`IPR2016-01179
`Patent 5,850,482
`
`respective first and second portions.” Petitioner contends one of ordinary
`
`skill in the art would have understood “second portion of each code word” as
`
`used in the ’482 patent to mean a “‘suffix field of a code word generated in a
`
`manner such that a bit error in the field results in a miscoded value that falls
`
`in a predetermined range about the correct value.’” Pet. 21. Patent Owner
`
`does not propose construction of the phrase “second portion of each code
`
`word” in its Preliminary Response.
`
`The ’482 patent discloses that each code word includes at least a first
`
`portion, termed a prefix field, and an associated second portion, termed a
`
`suffix field. Ex. 1001, 13:41–50. The ’482 patent also provides that “Each
`
`code word has . . . an associated second or suffix field which is resilient to
`
`bit errors. [The] code words can be generated such that . . . a bit error in the
`
`suffix field of a code word will . . . fall within a predetermined interval about
`
`the correct value.” Id. at 6:52–65. The ’482 patent does not state that the
`
`second portion must be generated in this manner. We do not import this
`
`limitation into the claims.
`
`For purposes of this Decision, we construe a “second portion of each
`
`code word” as used in the ’482 patent to encompass at least “a second or
`
`suffix field associated with the first field which is resilient to bit errors.”
`
`We further determine that none of the other terms require express
`
`construction at this stage.
`
`
`
`B. Asserted Obviousness Over Kato: Claims 1–3, 5, and 28
`
`Petitioner, relying on the declaration of Robert L. Stevenson, Ph. D.,
`
`challenges claims 1–3, 5, and 28 as obvious over Kato. Pet. 38–52, 56–59
`
`(citing Ex. 1005 ¶¶ 79–115, 131–135). The first embodiment of Kato
`
`11
`
`
`
`IPR2016-01179
`Patent 5,850,482
`
`entropy encodes input data into a first portion and a second portion. Ex.
`
`1002, 10:44–11:4. The fourth embodiment of Kato entropy encodes input
`
`data into a first portion and a second portion, then provides unequal error
`
`protection to the first and second portions. Id. at 24:2–59, 31:52–67.
`
`1. Kato (Ex. 1002) first and fourth embodiments
`
`a. first embodiment
`
`
`
`Kato provides a method of efficient encoding that can reduce the total
`
`number of bits of recorded or transmitted data. Ex. 1002, 1:7–9. Figure 1(a)
`
`of Kato is reproduced below.
`
`
`
`Figure 1(a) of Kato above shows an efficient encoding apparatus that
`
`uses an efficient coding method according to a “first embodiment.” The
`
`efficient encoding apparatus includes prediction circuit 102, subtraction
`
`circuit 103, classification circuit 104, data conversion circuit 105, remainder
`
`calculation circuit 106, main encoding circuit 107, sub encoding circuit 109,
`
`sub encoding circuit 110, and multiplexing circuit 111. Ex. 1002, 9:46–51.
`
`12
`
`
`
`IPR2016-01179
`Patent 5,850,482
`
`
`Input terminal 101 is subject to input data Di to be encoded. Id. at
`
`9:41–42. Prediction circuit 102 generates an estimate Pi of the input data.
`
`Id. at 9:52–53. The estimate Pi is subtracted from the input data Di by
`
`subtraction circuit 103 to generate estimation error Si. Id. at 9:53–55.
`
`Classification circuit 104 receives the estimation error Si and outputs a
`
`category index Ji. Id. at 9:56–57. Data conversion circuit 105 generates
`
`divisor data OUi in response to the category index Ji. Id. at 9:57–59.
`
`Remainder calculation circuit 106 divides data Di by divisor data OUi to
`
`generate remainder data Ei. Id. at 9:59–62. Main encoding circuit 107
`
`encodes the category number Ji and remainder data Ei into coded data Ci,
`
`which is transmitted through output terminal 108. Id. at 9:62–65. Main
`
`encoding circuit 107 includes sub encoding circuit 109 which encodes
`
`category index Ji into a Huffman code CJi. Id. at 9:65–10:1. Sub encoding
`
`circuit 110 in main encoding circuit 107 determines a bit number Mi of
`
`remainder data Ei, and outputs coded data CEi. Id. at 10:1–2. Multiplexing
`
`circuit 111 adds coded data CEi to the end of the coded data CJi to produce
`
`multiplexed coded data Ci. Id. at 10:2–4.
`
`Table 1 of Kato is reproduced below.
`
`13
`
`
`
`IPR2016-01179
`Patent 5,850,482
`
`
`
`
`Table 1 above shows category indexes Ji, corresponding estimation
`
`error ranges (SNi – SXi), divisor data OUi, and word lengths Mi of
`
`remainder data Ei. Id. at 7:56–59. The estimation error Si is classified in
`
`accordance with the value thereof by referring to Table 1. Id. at 7:1–5. A
`
`decision is made regarding which range, or category, the estimation error is
`
`in, and the category index Ji denoting this range is determined. Id. at 7:5–7.
`
`The bit number Mi of remainder data Ei is determined in accordance with
`
`category index Ji by referring to Table 1. Id. at 10:64–67.
`
`b. fourth embodiment
`
`Kato also provides, in what it styles as its “fourth embodiment,” a
`
`method of efficient encoding which encodes input data into variable-length
`
`code words each having a first portion and a second portion, wherein the
`
`first portion includes a bit pattern that can determine a code length of the
`
`related word, and wherein the second portion is equal to a part of the related
`
`word except the first portion. Ex. 1002, 4:31–42. The first portions are
`
`collected into a group and arranged into a data store region. Id. at 4:42–44.
`
`14
`
`
`
`IPR2016-01179
`Patent 5,850,482
`
`The second portions are also collected into a group and arranged into the
`
`data store region. Id. at 44–46. Figure 6(a) is reproduced below.
`
`
`
`Figure 6(a) above shows a transmitter using an efficient coding
`
`method according to the so-called fourth embodiment. Ex. 1002, 23:47–49.
`
`The transmitter includes encoding circuit 602, ECC (error correction code)
`
`encoder 603, and modulation circuit 604. Id. at 23:52–54. Encoding circuit
`
`602 encodes input data Di into a variable-length code, and places the
`
`variable-length code in a data store region within a transmission format. Id.
`
`at 23:54–57. ECC encoder 603 adds an error correction code to the data
`
`output from encoding circuit 602. Id. at 23:57–59. Modulation circuit 604
`
`modulates the data from ECC encoder 603 and feeds the signal to output
`
`terminal 605. Id. at 23:59–62.
`
`15
`
`
`
`IPR2016-01179
`Patent 5,850,482
`
`
`Encoding circuit 602 includes read only memory (ROM) 606, shift
`
`register 607, control circuit 608, addition circuit 609, subtraction circuit 610,
`
`selection circuit 611, selection circuit 612, register 613, register 614,
`
`selection circuit 615, counter 616, random access memory (RAM) 617, and
`
`side information adding circuit 618. Id. at 23:62–24:21. ROM 606 receives
`
`input data Di via input terminal 601 and encodes it into data Vi, and
`
`generates word length data L1 and L2. Id. at 24:2–5. Shift register 607
`
`converts data Vi from bit-parallel format to bit-series format, which is stored
`
`in RAM 617. Id. at 24:7–10.
`
`A first portion Pi of code word Ci has length L1. Id. at 25:20–21.
`
`When writing Pi to RAM 617, control circuit 608 controls register 613,
`
`selection circuit 615, and counter 616 to write the L1 bits into RAM 617. Id.
`
`at 25:45–58. A second portion Ri of code word Ci has length L2. Id. at
`
`25:21–23. When the second portion Ri is written into RAM 617, control
`
`circuit 608 controls register 614, selection circuit 615, and counter 616 to
`
`write the L2 bites into RAM 617. Id. at 25:59–26:4.
`
`Encoding circuit 602 encodes input data Di into a variable length code
`
`word Ci and locates the code word Ci in a data store region within a data
`
`transmission format. Id. at 24:40–45. Figure 7 of Kato is reproduced below.
`
`
`
`16
`
`
`
`IPR2016-01179
`Patent 5,850,482
`
`
`Figure 7 above shows an example of conditions of the arrangement of
`
`variable-length code word in the data store region within the data
`
`transmission format. Id. at 24:46–48. Each variable length code word Ci is
`
`divided into a first portion Pi and a second portion Ri. Id. at 24:48–50. The
`
`first portion Pi includes a bit pattern from which the code length L
`
`(L=L1+L2) of the word Ci can be detected. Id. at 24:50–53. The first
`
`portion Pi has L1 former bits of word Ci, and the second portion Ri is equal
`
`to the remaining portion having L2 remaining bits of word Ci. Id. at 24:53–
`
`58. First portions Pi and second portions Ri of respective code words Ci are
`
`arranged in the data store region. Id. at 24:58–63.
`
`First portions Pi are sequentially arranged from the starting edge of
`
`the data store region toward the ending edge, and second portions Ri are
`
`sequentially arranged from the ending edge toward the starting edge. Id. at
`
`24:63–25:8. First portions Pi are separated from second portions Ri. Id. at
`
`31:51–55. This separating arrangement is adopted because an error can
`
`propagate through the first portions Pi, but cannot propagate through the
`
`second portions Ri. Id. at 31:55–58. ECC encoder 603 of Figure 6(a) adds
`
`error correction codes of higher ability to first portions Pi to increase the
`
`average number of code words which can be decoded correctly in the
`
`presence of a transmission error. Id. at 31:59–67.
`
`2. Analysis of Claims 1–3 and 5
`
`Petitioner contends, “An error resilient method of encoding data” as
`
`recited in the preamble of claim 1, is disclosed by the “fourth embodiment”
`
`of Kato, which describes that ECC encoder 603 increases the average
`
`number of code words which can be decoded correctly in the presence of a
`
`transmission error. Pet. 41 (citing Ex. 1002, 31:62–65).
`
`17
`
`
`
`IPR2016-01179
`Patent 5,850,482
`
`
`Petitioner contends “generating a plurality of code words
`
`representative of respective portions of the data” is disclosed by the “first
`
`embodiment” of Kato, which describes generating code words Ci having two
`
`portions, CJi and CEi. Pet. 41–42 (citing Ex. 1002, 10:45–49 and 10:65–
`
`11:2).
`
`Petitioner contends “wherein each code word comprises a first portion
`
`and an associated second portion” is disclosed by the first embodiment of
`
`Kato, which discloses that code word Ci comprises a first portion CJi and an
`
`associated second portion CEi. Pet 42–43 (citing Ex. 1002, 10:1–4).
`
`Petitioner also contends this limitation is disclosed by the fourth
`
`embodiment of Kato, which describes a code word Ci comprising a first
`
`portion Pi and an associated second portion Ri. Pet. 43 (citing Ex. 1002,
`
`23:46–57 and 24:40–45).
`
`Petitioner contends “generating the first portion of each code word,
`
`wherein said first portion generating step comprises the step of including
`
`information within the first portion that is representative of a predetermined
`
`characteristic of the associated second portion” is disclosed by the first
`
`embodiment of Kato, which describes generating a first portion of a code
`
`word CJi from a category index Ji, and an associated second portion CEi
`
`generated from remainder data Ei. Pet 43 (citing Ex. 1002, Table 1, 6:55–
`
`7:59). Petitioner contends the length Mi of remainder data Ei is determined
`
`by Ji, as shown in Table 1 of Kato, which represents a predetermined
`
`characteristic of CEi, namely, the length of remainder data Ei represented by
`
`CEi. Pet. 43–44. Petitioner also contends this limitation is disclosed by the
`
`fourth embodiment of Kato, which describes a code word Ci generated to
`
`have first portion Pi that provides information about a second portion Ri.
`
`18
`
`
`
`IPR2016-01179
`Patent 5,850,482
`
`Pet. 44. According to Petitioner, Kato describes that Pi includes a bit pattern
`
`from which the length L (L=L1+L2) of the code word Ci can be determined,
`
`which in turn determines the length of Ri, where the first portion Pi has L1
`
`bits and the second portion Ri has L2 bits. Pet. 44–45 (citing Ex. 1002,
`
`24:50–58).
`
`Petitioner contends “generating the second portion of each code word,
`
`wherein said second portion generating step comprises the step of including
`
`information within the second portion that is representative of the respective
`
`portion of the data” is disclosed by the first embodiment of Kato, which
`
`describes a code word Ci having a second portion CEi generated from
`
`remainder data Ei. Pet. 45–46 (citing Ex. 1002, Table 1; 7:1–59 and 9:64–
`
`10:4).
`
`Petitioner contends “providing error protection to at least one of the
`
`first portions of the plurality of code words while maintaining any error
`
`protection provided to the respective second portion associated with the at
`
`least one first portion at a lower level than the error protection provided to
`
`the respective first portion” is disclosed by the fourth embodiment of Kato,
`
`which describes providing greater error protection to first portions Pi than to
`
`second portions Ri. Pet. 47 (citing Ex. 1002, 31:51–67).
`
`Petitioner contends a person of ordinary skill in the art would apply
`
`the unequal error protection provided to the first and second portions of a
`
`code word as taught by the fourth embodiment of Kato, to the first and
`
`second portions CJi and CEi of the code word of the first embodiment of
`
`Kato, because an error can propagate through the first portions but not the
`
`second portions. Pet. 40 (citing Ex. 1002, 31:55–59; 32:14–20; Ex. 1005 ¶¶
`
`83–87). Petitioner contends that adding the error correction codes of higher
`
`19
`
`
`
`IPR2016-01179
`Patent 5,850,482
`
`ability to the first portions increases the average number of code words
`
`which can be decoded into correct data in the presence of transmission errors
`
`as taught by Kato. Pet. 40–41 (citing Ex. 1002, 31:63–65; Ex. 1005 ¶ 88);
`
`Pet. 48 (citing Ex. 1005 ¶¶ 107, 108).
`
`Patent Owner contends the fourth embodiment of Kato does not teach
`
`providing unequal levels of error protection to first and second portions of a
`
`code word, such as CJi and CEi of the first embodiment of Kato. Prelim.
`
`Resp. 16–17. Patent Owner also contends the declaration of Dr. Stevenson
`
`does not provide a factual basis to conclude that incorporating unequal error
`
`protection from the fourth embodiment of Kato into the first and second
`
`portions of a code word taught by the first embodiment of Kato could be
`
`achieved without undue experimentation and yield predictable results. Id. at
`
`18.
`
`Patent Owner contends that the fourth embodiment of Kato discloses
`
`an encoding circuit that encodes input data into a variable length code, and
`
`places the code in a data store region with a data transmission format having
`
`a first portion Pi and a second portion Ri separated by an open area. Id. at
`
`19–20. Patent Owner further contends the code word of the first
`
`embodiment of Kato encodes input data into first portion CJi and second
`
`portion CEi, and multiplexes them into Ci. Id. at 21–22. According to
`
`Patent Owner, Petitioner has not explained why the fourth embodiment of
`
`Kato, where the first portion and second portion of a code word are
`
`separated by an open area, is compatible with an arrangement of coded data
`
`Ci, where CJi and CEi are sequentially arranged as taught by the first
`
`embodiment of Kato, nor how the sequentially arranged first and second
`
`portions can receive different levels of error protection. Id. at 23–24.
`
`20
`
`
`
`IPR2016-01179
`Patent 5,850,482
`
`
`Petitioner contends Kato’s fourth embodiment discusses the benefits
`
`of providing unequal error protection to first and second portions of a code
`
`word, due to the relative importance of the first portion, which contains
`
`information about the second portion, without which it would be difficult to
`
`decode the second portion of the code word and subsequent code words.
`
`Pet. 39. Petitioner contends it would have been obvious to apply unequal
`
`error protection, as taught by the fourth embodiment, to the first and second
`
`portions CJi and CEi of the first embodiment, because the fourth
`
`embodiment teaches an error can propagate through the first portion, but not
`
`the second portion. Pet. 39–41 (citing Ex. 1005 ¶¶ 87, 88).
`
`Dr. Stevenson testifies that “Kato’s fourth embodiment discusses the
`
`benefits of providing unequal error protection to first and second portions of
`
`a code word, particularly due to the relative importance of the two code
`
`word portions.” Ex. 1005 ¶ 83. Dr. Stevenson continues that
`
`one of ordinary skill in the art would have been motivated to
`provide higher level of error protection, as described in Kato’s
`fourth embodiment, to the first portions CJi from Kato’s first
`embodiment (where, like first portions Pi in Kato’s fourth
`embodiment “an error can propagate through,”) than to the
`associated second portions CEi (where, like second portions Ri
`in Kato’s fourth embodiment, “an error can not propagate
`through.”).
`
`Id. at ¶ 87. Dr. Stevenson further testifies that adding unequal error
`
`protection from the fourth embodiment to the encoding apparatus of the first
`
`embodi