Case 1:20-cv-00765-DAE Document 54 Filed 03/29/23 Page 1 of 12
`
`IN THE UNITED STATES DISTRICT COURT
`FOR THE WESTERN DISTRICT OF TEXAS
`AUSTIN DIVISION
`
`BANDSPEED, LLC,
`
`Plaintiff,
`
`v.
`
`REALTEK SEMICONDUCTOR
`CORPORATION,
`
`Defendant.
`
`
`
`§
`§
`§
`§
`§
`§
`§
`§
`§
`§
`§
`
`CASE NO. 1:20-cv-00765-LY
`
`
`
`BANDSPEED’S REPLY CLAIM CONSTRUCTION BRIEF
`
`
`
`
`
`
`
`
`
`
`
`

`

`Case 1:20-cv-00765-DAE Document 54 Filed 03/29/23 Page 2 of 12
`
`
`
`I.
`
`CONTESTED CLAIM TERM NOS. 1-11
`
`A.
`
`Contested Term No. 1: “selection kernel”
`
`Realtek argues that “[c]laims 1, 6, and 11 of the ’643 patent recite a selection kernel, a term
`
`that has no meaning in the art and does not connote structure by itself.” Doc. 41 at 2. Realtek
`
`provides no expert opinion to support this statement. Bandspeed’s expert testified that a POSITA
`
`would understand the meaning of the term and cited to, as an example, the basic selection kernel
`
`described in an earlier Bluetooth specification. See Doc. 40-1 at 8-9; Doc. 40-13 at 129. Realtek’s
`
`argument is further belied by its brief, where Realtek describes what a selection kernel exists in
`
`the prior art Bluetooth Standard. Doc. 41 at 8-10. Further, the PTAB determined “selection kernel”
`
`was not indefinite and adopted a definition consistent with Bandspeed’s construction. Doc. 40-14
`
`at 11. Realtek cites no evidence to support its contention that “[d]ifferent competitors could
`
`implement the selection of channels each in different ways, even ways not possible at the time of
`
`the invention and clearly not contemplated by the patentee, yet each would satisfy Plaintiff’s
`
`proposed construction as long as they achieve the stated result.” Realtek’s unsupported argument
`
`should be rejected. Finally, Realtek argues Bandspeed’s proposal lacks “any corresponding
`
`structure.” This is not a MPF claim so no corresponding structure must be identified.
`
`Realtek cannot meet its burden to “show by clear and convincing evidence that a skilled
`
`artisan could not discern the boundaries of the claim based on the claim language, the specification,
`
`and the prosecution history, as well as her knowledge of the relevant art area.”1 Consistent with
`
`Bandspeed’s proposal, a POSITA would understand a “selection kernel” to be “a device or
`
`mechanism of a participant that selects communications channels to form the hopping sequence.”
`
`B.
`
`Contested Term No. 2: “the hopping sequence”
`
`
`1 Haemonetics Corp. v. Baxter Healthcare Corp., 607 F.3d 776, 783 (Fed. Cir. 2010).
`
`
`
`1
`
`

`

`Case 1:20-cv-00765-DAE Document 54 Filed 03/29/23 Page 3 of 12
`
`
`
`Realtek claims that the alleged lack of antecedent basis “is an indication of indefiniteness,”
`
`but fails to show that a POSITA would not understand the term read in light of the specification.2
`
`Further, Realtek provides no explanation for its failure to disclose its contention in its invalidity
`
`contentions. Doc. 40 at 8-9. Realtek cannot prove indefiniteness by clear and convincing evidence.
`
`Although the sole basis for Realtek’s claim of indefiniteness of “the hopping sequence” is
`
`the lack of antecedent basis, Doc. 35-1 at 5, Realtek relies on two new arguments, both unrelated
`
`to a lack of antecedent basis, that it omitted from its invalidity contentions. The Court may reject
`
`the arguments on this ground alone. Doc. 40 at 8-9. In addition, both arguments fail on the merits.
`
`First, Realtek mistakenly asserts that claim 2 requires the claimed device to use
`
`simultaneously “a communication channel from both the first and second set of communications
`
`channel[s] . . . .” Doc. 41 at 6. Realtek omits the language from claim 2 showing each hopping
`
`sequence is “based on the frequency hopping protocol” that includes the first and second sets of
`
`communications channels: “at each hop in the hopping sequence based on the frequency hopping
`
`protocol, only one communications channel of the [first/second] set . . . is used.”3 Claim 2 further
`
`discloses a sequence in which the first set of communications channels is used “for a first period
`
`of time” and then the second set of communications channels is used “for a second period of time
`
`that is after the first period of time . . . .”4 Thus, the claim does not require “simultaneous
`
`communications” over a channel from both the first and second sets and would be readily
`
`understood by a POSITA. Doc. 40-1 at 13-15.
`
`Second, Realtek incorrectly claims that the specification’s definition of “hopping
`
`
`2 MPEP § 2173.05(e); See also In re Downing, 754 Fed. App’x 988, 996 (Fed. Cir. 2018);
`Energizer Holdings, Inc. v. Int'l Trade Comm’n, 435 F.3d 1366, 1370-71 (Fed. Cir. 2006).
`3 Doc. 40-6 at 27:25-26, 30-31.
`4 Id. at 27:9-14 (emphasis added).
`
`
`
`2
`
`

`

`Case 1:20-cv-00765-DAE Document 54 Filed 03/29/23 Page 4 of 12
`
`
`
`sequence” is ambiguous. Doc. 41 at 6. A POSITA would readily ascertain the meaning of the term
`
`“the hopping sequence” as defined by the specification. Doc. 40-1 at 13-15. Indeed, Realtek
`
`agreed to the same definition of “hopping sequence” for the ’418 and ’614 patents. Doc. 35-1 at 3.
`
`C.
`
`Contested Term No. 3: “[first/second/third] performance [criterion/criteria]”
`
`Realtek argues the jury should not be permitted to apply their “ordinary understanding” of
`
`the term “performance” because it has a broad scope Doc. 41 at 7. Claim terms are to be construed
`
`based on their plain and ordinary meaning with “only two exceptions to [the] general rule,” which
`
`are when the patentee acts as his own lexicographer or disavows the full scope of the claim term
`
`either in the specification or during prosecution.5 Neither of those conditions are present. The jury
`
`should be permitted to apply the plain meaning of the word. Realtek cites a dictionary definition
`
`for “performance” of “[m]anner of functioning.” Doc. 41 at 7. But, as used in the asserted claims,
`
`this “manner of functioning” itself makes sense. The claim language is clear that it is “performance
`
`of a plurality of communications channels” such that even if it were read as “[manner of
`
`functioning] of a plurality of communications channels,” it would be acceptable. The term
`
`“performance” is a common word with no special meaning. No construction is necessary.
`
`D.
`
`Contested Term No. 4: “a [first]/[second] time”
`
`Realtek provides no justification that the claims are limited only to taking “a fraction of a
`
`second to test the performance of a communications channel.” In fact, Realtek states that the first
`
`and second time element in the ’418 patent refer to “the time required to test a communications
`
`channel,” with that duration of time not limited, and particularly not limited to an “instant” or
`
`fraction of time. To the contrary, the specification identifies determining channel performance at
`
`various times other than “an instant of time” including testing performance “according to a
`
`
`5 Thorner v. Sony Computer Entm’t Am. LLC, 669 F.3d 1362, 1365 (Fed. Cir. 2012).
`
`
`
`3
`
`

`

`Case 1:20-cv-00765-DAE Document 54 Filed 03/29/23 Page 5 of 12
`
`
`
`specified schedule, such as the expiration of a specified time period.” Doc. 40-3 at 6:47-57; id. at
`
`8:24-29. The word “instant” is not claimed. It should not be added.
`
`E.
`
`Contested Term Nos. 5 and 6: “channel index” and “apply[ing] an index . . .”
`
`Although Realtek contends that the ’520 patent provides no explanation of these terms, it
`
`ignores the claim language and acknowledges sections of the specification providing explanation
`
`more than adequate to allow a POSITA to understand the terms. For example, although Realtek
`
`claims “channel index does not appear in the specification,” Doc. 41 at 8, it ignores the claim
`
`language explaining that the device “load[s] channel indices of the subset of communications
`
`channels into [a] register,”6 revealing that channel indices are indicators or identifiers of channels.
`
`Realtek also cites Fig. 5A of the ’520 Patent, showing that the “channel indices” “load[ed] into the
`
`register,” as described in the claim language,7 consist of channel indicators such as addresses.8
`
`Thus, Realtek’s argument is reduced to the dubious claim that a POSITA would be unable to
`
`determine that “channel indices” refers to more than one “channel index.”
`
`Similarly, although Realtek claims the specification “sheds no light on” the application of
`
`an index to the indicator of the identified communications channel, it cites the portion of the
`
`specification that explains the process of applying an index to the channel indicator of a candidate
`
`channel to determine a replacement channel for use9 and acknowledges that this index “is applied
`
`to the output of the selection kernel” and is different from the channel index (i.e. channel
`
`
`
`6 Doc. 40-10 at 28:6-7.
`7 Id. at 28:6-7 (“load channel indices of the subset of communications channels into the register”).
`8 Doc. 41 at 10 (“an address is shown” in Fig. 5A); id. at Fig. 5A; Doc. 40-1 at 15 (“[T]he patent
`describes the use of a channel index for each channel, which, based on the claim language and the
`specification, a POSITA would understand is an indicator of a channel such as an address.”). See
`also Ex. 18, RANDOM HOUSE WEBSTER’S UNABRIDGED DICTIONARY at 971 (“Computers a. a value
`that identifies and is used to locate a particular element within a data array or table”).
`9 Compare Doc. 40 at 14 (citing Doc. 40-10 at 20:30-59) with Doc. 41 at 9-10.
`
`
`
`4
`
`

`

`Case 1:20-cv-00765-DAE Document 54 Filed 03/29/23 Page 6 of 12
`
`
`
`indicator).10 Realtek also ignores a related portion of the specification identifying the output of the
`
`selection kernel as a channel address,11 which a POSITA would understand is a channel index.12
`
`Thus, these terms are not indefinite and do not require construction.13
`
`F.
`
`Contested Term Nos. 7 and 8: the “distinct channels” terms
`
`Realtek argues that “[n]othing in the specification suggests that the same channel would
`
`ever be included more than once in the same set of channels,” which suggests that the specification
`
`at least discloses sets of channels having distinct channels as claimed. Doc. 41 at 11. Realtek’s
`
`argument is not that it is not possible for a set of channels in a communications system to include
`
`the same channel more than once in a same set of channels, but rather that Realtek believes the
`
`specification does not suggest this possibility for the claimed invention. But this does not make
`
`the claim language’s use of “distinct channels” in claiming “the number of distinct channels” a
`
`nullity or indefinite. Rather, it clarifies the scope of the claimed invention for a POSITA. The
`
`meaning of “distinct” and “distinct channels” should be given their plain and ordinary meaning.
`
`G.
`
`Contested Term No. 9: “rescanning [the default channels]”
`
`Realtek incorrectly asserts “rescanning [the default channels]” is indefinite because neither
`
`the claims nor the specification implies “a prior, initial” determination of channel performance.
`
`Doc. 41 at 12. The claims and the specification refer to an initial determination of performance of
`
`
`10 Doc. 41 at 8 (“[T]his is a different usage of the term ‘index’ (i.e., not a ‘channel index’) as it is
`referring to something that is applied to the output of the selection kernel . . . .”). See also Ex. 18,
`RANDOM HOUSE WEBSTER’S UNABRIDGED DICTIONARY at 971 (“Computers b. a reference table
`that contains the keys or references needed to address data items”).
`11 See supra notes 7-Error! Bookmark not defined..
`12 Doc. 40-10 at 20:4-5 (“The output of the selection kernel is a set of addresses for each slot in
`the register.”); Doc. 40-1 at 15 (“[T]he patent describes the use of a channel index for each channel,
`which, based on the claim language and the specification, a POSITA would understand is an
`indicator of a channel such as an address.”).
`13 Alternatively, “channel index” may be construed as “an indicator of a channel” and “apply[ing]
`an index . . .” may be construed as “indexing the channel indicator of the candidate
`communications channel for use to determine a replacement channel for use.” Doc. 40 at 15.
`
`
`
`5
`
`

`

`Case 1:20-cv-00765-DAE Document 54 Filed 03/29/23 Page 7 of 12
`
`
`
`the default channels to identify a set of “good” channels. Claims 5, 10, and 15 depend from
`
`independent claims 1, 6, and 11, respectively. Each independent claim refers to initially loading a
`
`default set of channels into a default channel register and a good set of channels into a good channel
`
`register.14 As shown by the specification, the system initially identifies the “good” channels by
`
`determining the performance of the default channels and classifying each as “bad” or “good”15:
`
`
`
`Doc. 40-7, Fig. 6A. The register of “good” channels includes the default channels initially
`
`classified as “good” based upon performance.16 When performance of the “good” channels falls
`
`below a specific threshold, Doc. 40-7, Fig. 6B, step 666, the default channels are rescanned and
`
`channel performance is redetermined, id., Fig. 6C, step 694. Realtek’s claim that an initial set of
`
`
`14 Doc. 40-7 at 26:59-61, 27:20-22, 28:13-16.
`15 Id., Fig. 6A, 23:12-17 (In block 632 * * * testing of each channel is typically performed for a
`specified number of times, such as 10 to get a fair representation of the typical performance of
`each channel and thereby avoid relying on one or a few tests that may not accurately reflect typical
`channel performance.”); see also id. at 6:31-33 (“An initial set of channels is selected based on
`one or more selection criteria at the start-up of the communications network.”).
`16 Doc. 40-7 at Fig. 5A, 20:15-18 (“Register with good channels 550 includes default channels
`552a – 552n that are channel numbers for the good channels, such as may be provided to the
`participant using good channel packet 400.”), 20:52-54 (“Assume for this example that default
`channels 522b and 522n-1 are classified as bad and that the remaining default channels are
`classified as good.”) (emphasis added).
`
`
`
`6
`
`

`

`Case 1:20-cv-00765-DAE Document 54 Filed 03/29/23 Page 8 of 12
`
`
`
`good channels can be loaded from the default channels without any need for an initial
`
`determination of channel performance is contrary to the language of the claims and specification.
`
`H.
`
`Contested Term Nos. 10 and 11: the “bad” and “good” terms
`
`Realtek argues the specification does not dictate which of the disclosed test methods must
`
`be used or a specific threshold or value that must be used in a system, and as such lay jurors would
`
`have no way of knowing what a “good” or “bad” channel is. Furthermore, Realtek states
`
`incorrectly and without expert support that “there is no way to determine which method to use to
`
`set the threshold in any particular situation.” Doc. 41 at 13-14. A POSITA, however, would
`
`understand the use of the terms “good” and “bad” in selecting the test method, values or thresholds
`
`for a particular application. For example, a threshold may be set higher or lower depending upon
`
`overall application requirements and associated system tradeoffs. Doc. 40-1 at 20-22.
`
`Realtek argues the “written description does not even limit the threshold to a number value.
`
`The threshold could be any criteria, including non-numerical ones.” Doc. 41 at 14 citing ’769
`
`Patent at 7:26-28. Realtek’s citation, however, is from the wrong part of the specification because
`
`it relates to channel selection rather than channel classification. Doc. 40-9 at 7:13-28.17 Claim 1
`
`requires “based at least on results of the monitoring, classifying one or more communications
`
`channels of the plurality of communication channels as good and classifying one or more
`
`communications channels of the plurality of communication channels as bad.” Doc. 40-9 at cls.
`
`1(emphasis added), 11 & 20.18 Channel selection is not relevant to the analysis. Thus, in the
`
`
`17 “After classifying the performance of the channels, a set of channels is selected based on the
`selection criteria […] channels may be selected based on comparing performance results to one or
`more criteria without first classifying the channels. Also, other criteria may be used, for example,
`that particular channels are, or are not, to be selected regardless of the test results.” Doc. 40-9 at
`7:13-28.
`18 Claims 2, 12 and 21 require, “based at least on results of the monitoring and the one or more
`votes from the [other/slave] wireless communications device, classifying one or more
`
`
`
`
`7
`
`

`

`Case 1:20-cv-00765-DAE Document 54 Filed 03/29/23 Page 9 of 12
`
`
`
`asserted claims, “bad” and “good” are used in relation to classified or classifying channels (as
`
`opposed to selecting channels) as good or bad based on results of monitoring or vote count. Doc.
`
`40-1 at 20. Realtek’s argument fails.
`
`II.
`
`INSTRUCTIONS TERMS NOS. 12-15
`
`I. BEAUREGARD CLAIMS DO NOT MIX APPARATUS AND METHOD ELEMENTS
`
`Realtek’s argument that claims 1, 3, and 5 of the ’608 Patent and claim 11 of the ’643
`
`Patent are indefinite because they improperly mix apparatus and method elements—even if they
`
`are Beauregard claims19—is legally incorrect. Realtek’s reliance on IPXL fails to support
`
`indefiniteness because the IPXL claims were not Beauregard claims.20 Indeed, post-IPXL courts
`
`have rejected indefiniteness of Beauregard claims in the face of IPXL / mixed-method apparatus
`
`challenges.21 Infringement of Beauregard claims occurs at least at the time the claimed devices
`
`containing the required software code are sold.22 These claims are not indefinite.
`
`J.
`
`UNDER §112, ¶6, THE SPECIFICATIONS CONTAIN THE REQUIRED ALGORITHMS
`
`Realtek asserts that each of the “instructions” terms require MPF treatment and specific
`
`algorithms. Bandspeed’s Opening Brief explained in detail why Realtek is incorrect. Doc. 40 at
`
`21-23. As a threshold issue, Realtek individually challenges only ’418 Patent, cl. 88 (“the
`
`
`communications channels of the plurality of communication channels as good and classifying one
`or more communications channels of the plurality of communication channels as bad.” Doc. 40-9
`at cls. 2, 12, and 21.
`19 Realtek Br. at 18-19. Realtek’s arguments about ’643 cl. 1 and MPF treatment should be
`considered withdrawn as the parties agreed this claim was not subject to MPF construction and
`does not contain the term at issue. Compare Modified JCCS (Doc. 39-1) that JCCS (Doc. 35-1 at
`8-9). Realtek makes no argument why MPF treatment even applies as the term does not use the
`words “means” (or even “instructions.”) Realtek Br. at 19.
`20 IPXL Holdings L.L.C v. Amazon.com, Inc., 430 F.3d 1377, 1384 (Fed. Cir. 2005).
`21 See e.g., Apple, Inc. v. Samsung Elecs. Co., 877 F. Supp. 2d 838, 896 (N.D. Cal.), rev'd on other
`grounds, 695 F.3d 1370 (Fed. Cir. 2012)(See ‘172 Patent, cl. 27); Uniloc 2017 LLC v. Google
`LLC, No. 2:18-CV-00493-JRG-RSP, 2020 WL 339802, at *30–31 (E.D. Tex. Jan. 20, 2020),
`supplemented, No. 2:18-CV0-00499-JRG-RSP, 2020 WL 1324910 (E.D. Tex. Mar. 20, 2020).
`22 Id; See also, VirnetX Inc. v. Apple Inc., 792 F. App'x 796, 807–08 (Fed. Cir. 2019).
`
`
`
`8
`
`

`

`Case 1:20-cv-00765-DAE Document 54 Filed 03/29/23 Page 10 of 12
`
`
`
`“instructions that generate”) and ’608 Patent, cl. 5 (the “instructions causing”) terms—thus
`
`waiving other arguments. For this reason, even under MPF rules the Court should find each of the
`
`remaining terms definite and adopt Bandspeed’s constructions.23
`
`Realtek incorrectly posits the ’418 Patent specification fails to explain how the indication
`
`or selection of the channels [in the respective channel sets] occurs. Doc. 41 at 20. To the contrary,
`
`the specification notes, for example, the invention’s use of data that indicate channels 0-78 as the
`
`set of (candidate) channels in the Bluetooth or IEEE 801.15.1 specifications. Doc. 40-3 at 7:66-
`
`8:6.24 The specification has further structure with good/bad channel set selection, iterative good
`
`channels identification, and special packets with good channels information. Doc. 40 at 27-28.
`
`Realtek further challenges the ’608 Patent as not indicating how separate communications
`
`frequencies are identified. Doc. 41 at 20. But Realtek ignores the significance of the agreed nature
`
`of the registers: “an area of memory for storing a set of two or more communications channels.”
`
`Doc. 40 at 31. Realtek also fails to challenge the specification’s algorithmic support, which
`
`describes cyclical loading and random selection for loading the channel data into registers. Doc.
`
`40 at 31-32. Further, Realtek ignores that the ’608 patent describes that data identifying channels,
`
`e.g. channels 0-78, as identifying the separate communications channels, and that this data is stored
`
`in the register memory. Doc. 40 at 31-32.
`
`Finally, Realtek provides no expert support for its assertions about lack of algorithms.
`
`Thus, the Court should reject indefiniteness and adopt Bandspeed’s constructions.
`
`
`
`23 See also, Doc. 40 at 24-36.
`24 Intelligent Automation Design, LLC v. Zimmer Biomet CMF & Thoracic, LLC, 799 F. App’x
`847, 851 (Fed. Cir. 2020)(internal citations omitted)(“[T]he specification must disclose an
`algorithm for performing the claimed function on the microprocessor in any understandable terms
`including as a mathematical formula, in prose, or as a flow chart, or in any other manner that
`provides sufficient structure.”).
`
`
`
`9
`
`

`

`Case 1:20-cv-00765-DAE Document 54 Filed 03/29/23 Page 11 of 12
`
`
`
`Dated: March 29, 2023
`
`
`
`
`
`
`
`Respectfully submitted,
`
`/s/ Adam G. Price
`Adam G. Price
`Texas State Bar 24027750
`Christopher V. Goodpastor
`Texas State Bar 00791991
`Gabriel Gervey
`Texas State Bar 24072112
`Michael French (Pro Hac Vice)
`Texas State Bar No. 24116392
`DiNovo Price LLP
`7000 N. MoPac Expressway
`Suite 350
`Austin, Texas 78731
`Telephone: (512) 539-2626
`Facsimile: (512) 539-2627
`Email: aprice@dinovoprice.com
`cgoodpastor@dinovoprice.com
`ggervey@dinovoprice.com
`mfrench@dinovoprice.com
`
`ATTORNEYS FOR PLAINTIFF
`BANDSPEED, LLC
`
`
`
`
`
`
`
`
`
`
`10
`
`

`

`Case 1:20-cv-00765-DAE Document 54 Filed 03/29/23 Page 12 of 12
`
`
`
`CERTIFICATE OF SERVICE
`
`I certify that on March 29, 2023, I electronically field the foregoing with the Clerk of Court
`
`using the CM/ECF system, which will send notification of such filing to all counsel of record.
`
`
`
`
`
`
`
`
`
`/s/ Adam G. Price
` Adam G. Price
`
`
`
`
`
`11
`
`