Free Remark - District Court of Delaware - Delaware

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191 Document 132-1 Filed 06/28/2007 Page 1 of of 19 Filed 01/18/2007 Page 1 19

UNITED STATES DISTRICT COURT FOR THE EASTERN DISTRICT OF TEXAS MARSHALL DIVISION REMBRANDT TECHNOLOGIES, LP Plaintiff, v. COMCAST CORPORATION; COMCAST CABLE COMMUNICATIONS, LLC; AND COMCAST OF PLANO, LP Defendants. PLAINTIFF REMBRANDT TECHNOLOGIES, LP'S REPLY BRIEF REGARDING CLAIM CONSTRUCTION Frank E. Scherkenbach Lawrence K. Kolodney Michael H. Bunis Thomas A. Brown FISH & RICHARDSON P.C. 225 Franklin Street Boston, MA 02110 Tel: 617-542-5070 Fax: 617-542-8906 Timothy Devlin FISH & RICHARDSON P.C. 919 N. Market Street, Suite 1100 Wilmington, DE 19899-1114 Tel: 302-652-5070 Fax: 302-652-0607 Alan D. Albright State Bar No. 00973650 FISH & RICHARDSON P.C. One Congress Plaza, 4th Floor 111 Congress Avenue Austin, TX 78701 Tel: 512-391-4930 Fax: 512-591-6837 Otis Carroll State Bar No. 03895700 Wesley Hill State Bar No. 24032294 IRELAND, CARROLL & KELLEY, P.C. 6101 S. Broadway, Suite 500 Tyler, Texas 75703 Tel: (903) 561-1600 Fax: (903) 581-1071 Email: [email protected] Civil Action No. 2:05-cv-00443-TJW

Counsel for Plaintiff REMBRANDT TECHNOLOGIES, LP

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191 Document 132-1 Filed 06/28/2007 Page 2 of of 19 Filed 01/18/2007 Page 2 19

TABLE OF CONTENTS Page I. II. INTRODUCTION .............................................................................................................1 DISCUSSION ....................................................................................................................2 A. U.S. Patent No. 5,719,858......................................................................................2 1. 2. 3. time-division multiplexed bus (claims 1, 7, 9, 11, 15, 20) ........................2 packet data (claims 1, 7, 9, 11, 15 and 20); synchronous data (claims 7, 9, 11).............................................................3 distributed packet manager (claims 1, 7); allocate access to the allotted bandwidth among said packet data sources [and corresponding limitations] (claims 1, 7, 15, 20) ...................................................................................4

B.

U.S. Patent No. 4,937,819......................................................................................5 1. 2. 3. 4. 5. application program[s] (claims 1, 14)........................................................5 time slot assigned to each of said application programs (claim 1); dividing a period of a clock . . . (claim 14) ...............................6 master network timing means . . . (claim 1)...............................................7 ranging means . . . (claim 1).......................................................................7 reservation request generator; reservation request processor (claim 2); priority bit (claim 11)................................................8

C.

U.S. Patent No. 5,852,631......................................................................................8 1. link layer (claims 1, 3, 4, 6, 8, 9, and 10) ..................................................8 a. b. c. 2. Rembrandt's Construction Is the Most Faithful to the Intrinsic Evidence ....................................................................8 Error Correction Is a Link Layer Feature ......................................9 The Extrinsic Evidence Supports Rembrandt's Construction.................................................................................10

Comcast's Indefiniteness Argument Violates the Local Rules and Is Contradicted by Its Own Expert..........................................10

D.

U.S. Patent No. 5,243,627....................................................................................11

PLAINTIFF REMBRANDT TECHNOLOGIES, LP'S REPLY BRIEF REGARDING CLAIM CONSTRUCTION - PAGE i

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191 Document 132-1 Filed 06/28/2007 Page 3 of of 19 Filed 01/18/2007 Page 3 19

1. 2. 3. 4.

trellis encoded channel symbol (claims 9, 19).........................................11 signal point (claims 9, 19)........................................................................12 distributed Viterbi decoder (claims 9, 19) ...............................................12 means for deinterleaving the interleaved signal points to recover said plurality of streams of trellis encoded channel symbols (claim 9) .......................................................................13

PLAINTIFF REMBRANDT TECHNOLOGIES, LP'S REPLY BRIEF REGARDING CLAIM CONSTRUCTION - PAGE ii

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191 Document 132-1 Filed 06/28/2007 Page 4 of of 19 Filed 01/18/2007 Page 4 19

TABLE OF AUTHORITIES Page(s) CAE Screenplates, Inc. v. Heinrich Fielder GmbH & Co. KG, 224 F.3d 1308 (Fed. Cir. 2000)..........................................................................1 CollegeNet, Inc. v. ApplyYourself, Inc., 418 F.3d 1225 (Fed. Cir. 2005)..........................................................................5 Curtiss-Wright Flow Control Crop. v. Velan, Inc., 438 F.3d 1374 (Fed. Cir. 2006)..........................................................................5 Electro Medical Sys., S.A. v. Cooper Life Sciences, Inc., 34 F.3d 1048 (Fed. Cir. 1994)............................................................................1 Gemstar-TV Guide Int'l, Inc. v. Int'l Trade Comm'n, 383 F.3d 1352 (Fed. Cir. 2004)..........................................................................6 Hormone Research Found. v. Genentech, Inc., 904 F.2d 1558 (Fed. Cir. 1990)..........................................................................5 Lampi Corp. v. Am. Power Prods., Inc., 228 F.3d 1365 (Fed. Cir. 2000)..........................................................................9 Phillips v. AWH Corp., 415 F.3d 1303 (Fed. Cir. 2005)................................................................1, 2, 11 Primos, Inc. v. Hunters Specialties, Inc., 451 F.3d 841 (Fed. Cir. 2006)..........................................................................12 SanDisk Corp. v. Memorex Prods., Inc., 415 F.3d 1278 (Fed. Cir. 2005)..........................................................................6 Sorensen v. Int'l Trade Comm'n, 427 F.3d 1375 (Fed. Cir. 2005)..........................................................................4 Teleflex, Inc. v. Ficosa N. Am. Corp., 299 F.3d 1313 (Fed. Cir. 2002)......................................................................1, 8 Vitronics Corp. v. Conceptronic, Inc., 90 F.3d 1576 (Fed. Cir. 1996)............................................................................1 STATUTES 35 U.S.C. § 112, ¶ 4.................................................................................................9 35 U.S.C. § 112, ¶ 6.............................................................................................7, 8

PLAINTIFF REMBRANDT TECHNOLOGIES, LP'S REPLY BRIEF REGARDING CLAIM CONSTRUCTION - PAGE iii

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191 Document 132-1 Filed 06/28/2007 Page 5 of of 19 Filed 01/18/2007 Page 5 19

I.

INTRODUCTION Comcast's claim construction Response (Docket No. 121) includes a number of persistent

errors, even from its opening words. For example, in its introduction, Comcast suggests that interpreting the claims more broadly than preferred embodiments ­ in other words, as they are written ­ is somehow improper, because the patents are being applied to later-developed cable technology in this case. (Response at 1.) The law is clear, however, that patents are frequently construed more broadly than the preferred embodiments. Electro Med. Sys., S.A. v. Cooper Life Sciences, Inc., 34 F.3d 1048, 1054 (Fed. Cir. 1994) (holding that "particular embodiments appearing in a specification will not be read into the claims when the claim language is broader than such embodiments"). This is far different from construing the claims in a manner that is inconsistent with the specification, or, even worse, in a manner that excludes the preferred embodiments from the claims. Such constructions ­ including many proposed by Comcast ­ are plainly improper. See Vitronics Corp. v. Conceptronic, Inc., 90 F.3d 1576, 1583 (Fed. Cir. 1996) (a construction that would exclude a preferred embodiment is "rarely, if ever, correct and would require highly persuasive evidentiary support"). Comcast makes other fundamental errors as well. While its Response includes a number of purported "incontrovertible rules" of claim construction (several of which Comcast proceeds to violate), the list conveniently omits several well-known principles: Claims should not be limited to the preferred embodiment. Teleflex, Inc. v. Ficosa N. Am. Corp., 299 F.3d 1313, 1328 (Fed. Cir. 2002) (cautioning against limiting the claimed invention to a preferred embodiment in the specification). Different language in claims generally leads to different meanings. CAE Screenplates, Inc. v. Heinrich Fielder GmbH & Co. KG, 224 F.3d 1308, 1317 (Fed. Cir. 2000) ("In the absence of any evidence to the contrary, we must presume that the use of these different terms in the claims connotes different meanings.") Extrinsic evidence should be rejected where it is inconsistent with the patent claims or specification. Phillips v. AWH Corp., 415 F.3d 1303, 1318 (Fed. Cir. 2005) (en banc) (stating that a court should discount extrinsic evidence that is at odds with the intrinsic evidence).
PLAINTIFF REMBRANDT TECHNOLOGIES, LP'S REPLY BRIEF REGARDING CLAIM CONSTRUCTION - PAGE 1

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191 Document 132-1 Filed 06/28/2007 Page 6 of of 19 Filed 01/18/2007 Page 6 19

Expert testimony on the meaning of claim terms should be viewed with scrutiny. Phillips, 415 F.3d at 1318-1319 (noting the unreliability of expert testimony as compared to intrinsic evidence). Comcast's Response violates these rules over and over again and further shows why Rembrandt's constructions are the right ones.1 II. DISCUSSION A. U.S. Patent No. 5,719,858 1. time-division multiplexed bus (claims 1, 7, 9, 11, 15, 20)

Comcast makes three separate arguments regarding the TDM bus limitation, none of which has any merit. Comcast first argues that Rembrandt does not further define the term "bus," but Comcast itself states that "it is understood that a bus is a group of (one or more) conductors shared by several users (i.e., transmitting and receiving sources)." (Response at 3.)2 There is simply no need to define this easily understood term using more words, particularly when one of those words ­ conductor ­ may be confusing and incorrect. Comcast may attempt to argue, for example, that the term "conductor" does not include a fiber optic connection, but fiber optics were well known in 1995 to communicate data. Second, Comcast asserts that a TDM bus "requires that the transmission sequence be defined . . . ." (Response at 3.) This statement, like Comcast's construction, directly contradicts the '858 patent specification. In the preferred embodiment there is no defined transmission sequence for the data sources which use the TDM bus. This is shown best in Figure 6 of the '858 patent, in which Module 6 follows Module 2, skipping over Modules 3, 4 and 5. Later on this order could change, depending on when each Module has data to send. (Ex. 25 at 62:2164:11; see also 58:16-59:6.)3
1

It should be noted that, due to space constraints, Rembrandt does not delineate every error in Comcast's brief, nor does Rembrandt even address all claim limitations. No such omission is intended to suggest agreement with Comcast's position; Rembrandt rests on its opening brief and will be prepared to address such issues, if the Court wishes, at the upcoming hearing. 2 Unless otherwise indicated, emphasis in this Reply has been added. 3 Exhibits 1-24 refer to exhibits attached to Rembrandt's Opening Claim Construction Brief (Docket No. 119). Exhibits 25-28 are attached to this Reply Brief. Exhibits 7 and 8 to Rembrandt's Opening Claim Construction Brief include rough transcripts of the depositions of Drs. Curtis Siller and Harry Bims. Exhibits 25 and 26 to this Reply include the final
PLAINTIFF REMBRANDT TECHNOLOGIES, LP'S REPLY BRIEF REGARDING CLAIM CONSTRUCTION - Page 2

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191 Document 132-1 Filed 06/28/2007 Page 7 of of 19 Filed 01/18/2007 Page 7 19

Third, Comcast tries to avoid this problem by arguing that in the '858 patent, over repeated frames, consistent portions of bandwidth are allocated collectively to the packet sources and synchronous sources. (Response at 4.) In other words, according to Comcast, because there is some repetition to how different groups of sources access the bus, the '858 patent specification is consistent with Comcast's construction. Even if Comcast's characterization of the '858 patent were true, the argument fails to cure the problems in Comcast's construction. That construction requires that "each user," not a group of users, access the bus in a defined, repeated sequence. In light of this proposed language, it is inexplicable how Comcast asserts that its construction "does not even relate to what data source is allowed to transmit at a particular time." (Id. at 4.) This statement is belied by its own proposed construction. Comcast's construction would read out the preferred embodiment of the '858 patent, and should be rejected. 2. packet data (claims 1, 7, 9, 11, 15 and 20); synchronous data (claims 7, 9, 11)

Comcast attempts to draw the line between packet and synchronous data according to the format of the data, i.e. whether it includes header information or not. This forces Comcast to (1) ignore an express definition within the '858 patent, and (2) ignore its own expert's Declaration. As set forth in Rembrandt's opening brief and conceded by Comcast, the '858 patent expressly defines packet data as "variable-bit-rate" data. (Ex. 1 at 1:9-10; Response at 5.) This is the construction proposed by Rembrandt, and there is nothing in the specification to the contrary. Comcast wrongly suggests that Rembrandt seeks to "limit" the term beyond this definition. (Response at 5.) Rembrandt simply proposes that the proper construction is the express definition set out in the specification. Comcast's arguments regarding "synchronous data" are belied by its own expert. Dr. Siller states in his Declaration that "synchronous" refers to "a mode of transmission in which the sending and receiving equipment are operating continuously at the same rate and are maintained versions of each of these transcripts, including pages cited in Rembrandt's Opening Claim Construction Brief.
PLAINTIFF REMBRANDT TECHNOLOGIES, LP'S REPLY BRIEF REGARDING CLAIM CONSTRUCTION - Page 3

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191 Document 132-1 Filed 06/28/2007 Page 8 of of 19 Filed 01/18/2007 Page 8 19

in a desired phase relationship." (Siller Decl. at ¶ 23.) This description makes no reference to the format of the data. Indeed, Comcast admits that synchronous data "can be sent in packets." (Response at 5.) There is no basis in the patent or otherwise to limit synchronous data to data "not transmitted in packets," as Comcast proposes. Comcast's reference to the prosecution history wrongly cites the Examiner's description of a prior art reference, not the '858 patent invention. (Comcast Ex. 4 at 2-3.) Moreover, the Examiner's statements are not a basis for contradicting an express definition within the specification. Sorensen v. Int'l Trade Comm'n, 427 F.3d 1375, 1379 (Fed. Cir. 2005) (noting that "the statements of an examiner will not necessarily limit a claim"). 3. distributed packet manager (claims 1, 7); allocate access to the allotted bandwidth among said packet data sources [and corresponding limitations] (claims 1, 7, 15, 20)

Comcast would improperly require the "distributed packet manager" to perform every function for placing packet data on the bus. This once again reads out a preferred embodiment, because the '858 patent itself describes that the central "network access module" or "NAM" is what reserves a portion of bus bandwidth for the packet sources. (Ex. 1 at 5:11-13.) The '858 patent does not demand the complete elimination of all central control, as Comcast proposes. (Comcast's construction of "network access module" is also incorrect for this same reason.) Comcast also seeks to read into these limitations a requirement of preventing packet collisions, but even Comcast's brief shows that such a construction would limit the patent to the preferred embodiment, namely "a specific solution . . . whereby only one packet source `captures' permission to use the MAPC at a time." (Response at 8.) This "specific solution" is the preferred method of access, using "packet request" and "packet hold" signals, described at columns 7 to 10. Comcast cites no support that justifies limiting the patent to this embodiment. This is particularly true in light of known methods of time division multiplexing that allow for packet collisions. (Ex. 25 at 92:22-93:14.) Comcast's attempts to avoid the doctrine of claim differentiation have no merit. Its cases are inapplicable, because they relate to situations where claim differentiation would lead to a
PLAINTIFF REMBRANDT TECHNOLOGIES, LP'S REPLY BRIEF REGARDING CLAIM CONSTRUCTION - Page 4

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191 Document 132-1 Filed 06/28/2007 Page 9 of of 19 Filed 01/18/2007 Page 9 19

construction that was incompatible with the specification and would render a claim term "nearly meaningless." See Curtiss-Wright Flow Control Crop. v. Velan, Inc., 438 F.3d 1374, 1379 (Fed. Cir. 2006); Hormone Research Found. v. Genentech, Inc., 904 F.2d 1558, 1567 n.15 (Fed. Cir. 1990). That is not the case here. Moreover, the '858 patent includes dependent claims that recite the "specific solution" for allocating access which Comcast seeks to import into the independent claims. (See, e.g., claim 2, reciting the "packet request" and "packet hold" signals.) Comcast's efforts to effectively read these dependent limitations into broader independent claims should be rejected, as set forth in Comcast's own cited case law. Curtiss-Wright, 438 F.3d at 1380 (observing that "reading an additional limitation from a dependent claim into an independent claim would not only make that additional limitation superfluous, it might render the dependent claim invalid"). B. U.S. Patent No. 4,937,819 1. application program[s] (claims 1, 14)

Having failed initially to cite any relevant evidence in support of its construction for "application program," Comcast now cites to two new extrinsic sources. One of the sources (the IEEE dictionary) is five years late, and the other source (Newton's Telecom) actually supports Rembrandt's construction: "APPLICATION[:] A software program that carries out some useful task." (Comcast Ex. 13.) Comcast's arguments also mischaracterize the intrinsic record. Contrary to Comcast's assertion, the prosecution history broadened the phrase "host application" to "application program," so that the program in question no longer needed to be running at a central host. This is clear from the following Office Action, after the claim was broadened, in which the Examiner refers to using a "remote unit to execute [an] application program." (Ex. 27 at 4.) Also contrary to Comcast's suggestion, there is no inconsistency between Rembrandt's construction and the patent specification, because there is no requirement that every single program have an associated channel. The patent claims include the phrase "comprising," and so can accommodate additional elements such as extra programs that do not require any data
PLAINTIFF REMBRANDT TECHNOLOGIES, LP'S REPLY BRIEF REGARDING CLAIM CONSTRUCTION - Page 5

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191 Document 132-1 Filed 06/28/2007 Page 1010 of 19 Filed 01/18/2007 Page of 19

transmission. CollegeNet, Inc. v. ApplyYourself, Inc., 418 F.3d 1225, 1235 (Fed. Cir. 2005) ("The transitional term `comprising' . . . is inclusive or open-ended and does not exclude additional, unrecited elements or method steps."). Finally, Comcast's selective citation of inventor testimony ignores the general statement by the inventor that his invention is "transparent to the application." (Ex. 23 at 58:6-7.) 2. time slot assigned to each of said application programs (claim 1); dividing a period of a clock . . . (claim 14)

The primary dispute regarding these elements is whether time slots can be assigned dynamically. Comcast's efforts to read out this capability are contrary to the plain teachings of the patent. Although one disclosed method of assigning time slots is to do so at initialization, the patent discloses another method by which remote units request additional time slots during transmission, and the time slots are assigned at that time. Comcast concedes all of the patent disclosure regarding this second method (Response at 12), but somehow seeks to avoid calling this procedure an "assignment" of time slots. This strained interpretation makes no sense. In fact, the '819 patent itself refers to this dynamic allocation as an "assignment" of time slots. Figure 8 shows a flow chart of the preferred dynamic allocation method, in which a remote unit makes a "reservation request" for additional time slots, and they are granted by the master unit. (Ex. 2 at Fig. 8.) The notation on the right hand side of Figure 8 refers to the grant as "Assignment of Transmission Period for Reservation Requests." Comcast's efforts to construe the claim so as to exclude this disclosure should be rejected. Comcast also wishes to read in further limitations regarding whether information is "packetized," but this overstates the arguments made during prosecution. A disclaimer in prosecution must be "clear and unmistakable." SanDisk Corp. v. Memorex Prods., Inc., 415 F.3d 1278, 1286 (Fed. Cir. 2005). Where as here the patentee merely differentiated prior art on a number of exemplary bases, the argument does not rise to the level of a disclaimer. Gemstar-TV Guide Int'l, Inc. v. Int'l Trade Comm'n, 383 F.3d 1352, 1375 (Fed. Cir. 2004) ("Gemstar's
PLAINTIFF REMBRANDT TECHNOLOGIES, LP'S REPLY BRIEF REGARDING CLAIM CONSTRUCTION - Page 6

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191 Document 132-1 Filed 06/28/2007 Page 1111 of 19 Filed 01/18/2007 Page of 19

statements in the prosecution history do not indicate a disavowal or disclaimer of claim scope . . . but merely provide an example to illustrate differences between the invention and the prior art."). 3. master network timing means . . . (claim 1)

Comcast mistakenly asserts that the "master network timing means" limitation recites a function, but in fact it merely recites a "wherein" clause that defines some features of how bandwidth is partitioned in the '819 patent. Even if the "wherein" clause is interpreted as functional language, the patent specification does not associate it with the master network timing means. Comcast itself concedes that a user can set up these partitions, not the master network timing means. (Response at 15.) Accordingly, this limitation does not meet the requirements for the application of 35 U.S.C. § 112, ¶ 6. Comcast belatedly raises an indefiniteness argument premised on the fact that its purported "function" is not associated with a purported "algorithm." (Response at 14, n. 11.) Citing Rembrandt's expert Dr. V. Thomas Rhyne, Comcast falsely asserts that the "parties agree that the network timing and control processor 12 performs the claimed functions." (Id. at 14.) Rembrandt does not agree, and Dr. Rhyne's declaration says no such thing. Moreover, Comcast's indefiniteness argument was never raised in its Invalidity Contentions or previously in claim construction disclosures, and should be rejected. If anything, Comcast's admission that its purported "function" is not associated with a purported "algorithm" only confirms that § 112, ¶ 6 should not apply. As set forth in Rembrandt's opening Brief, the patent specification discloses a clock that keeps network timing, and this is the only construction needed to capture the meaning of a "master network timing means." 4. ranging means . . . (claim 1)

In its Response, Comcast virtually ignores the essential problem with its proposed construction. Specifically, the function Comcast seeks to associate with the "ranging means" is in fact associated with the master unit, not the ranging means. This function can be performed by other components of the master unit. Comcast's only response is a conclusory footnote, in
PLAINTIFF REMBRANDT TECHNOLOGIES, LP'S REPLY BRIEF REGARDING CLAIM CONSTRUCTION - Page 7

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191 Document 132-1 Filed 06/28/2007 Page 1212 of 19 Filed 01/18/2007 Page of 19

which it notes that the ranging means is within the master unit. (Response at 15.) This statement does nothing to correlate any function with the "ranging means" itself, as would be required for the application of § 112, ¶ 6. Comcast's expert conceded this very point in deposition. (Ex. 25 at 181:23-182:14.) 5. reservation request generator and processor (claim 2); priority bit (claim 11)

These limitations are further examples of Comcast seeking to restrict the claims to a single preferred embodiment described in the specification. Such restrictions are improper, and should be rejected. Teleflex, 299 F.3d at 1328 (cautioning against limiting the claimed invention to a preferred embodiment in the specification). C. U.S. Patent No. 5,852,631

Only two substantial disputes among the parties remain with respect to the '631 patent: "link layer" and two means-plus-function terms.4 1. link layer (claims 1, 3, 4, 6, 8, 9, and 10) a. Rembrandt's Construction Is the Most Faithful to the Intrinsic Evidence

The '631 patent refers to "error correction" forty-eight times in the context of a link layer protocol. (Ex. 3 at 1:53; 2:42, 46, 47, & 49; 4:28; 6:59, 62, 64 & 65; 7:29, 39, 49 & 63; 11: 26, 32, 35, 43, 60, 64 & 67; 12: 1, 3, 5, 10, 11, 24, 28, 30, 43, 51, 55 & 58; 13: 2, 4, 21, 22, 25, 27, 31, 37, 38 & 39; 14: 3, 43, 48 & 67). In not one of these instances does the '631 patent limit the kind of error correction that is to be accomplished by the link layer. Comcast nonetheless seeks to restrict the scope of the '631 patent to link layer protocols utilizing frame retransmission, on the basis of an out-of-context snippet from a single sentence in the '631 patent. When read in full, however, this sentence provides no support for Comcast's position:

4

With respect to the term "negotiated physical layer modulation," Rembrandt does not dispute that a negotiation requires an exchange of information, nor that the negotiation must take place at run time. These concepts are already inherent in Rembrandt's proposed construction (requiring that the modems "agree on a common supported physical layer modulation").

PLAINTIFF REMBRANDT TECHNOLOGIES, LP'S REPLY BRIEF REGARDING CLAIM CONSTRUCTION - Page 8

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191 Document 132-1 Filed 06/28/2007 Page 1313 of 19 Filed 01/18/2007 Page of 19

The data link layer is the second lowest layer of the OSI seven layer model and is provided to perform error checking functions as well as retransmitting frames that are not received correctly. (Ex. 3 at 1:11-14.) This sentence does not purport to define the link layer, but merely incorporates ­ and partially characterizes ­ a pre-existing concept of "link layer" defined in the OSI seven layer model. The seminal Zimmerman article introducing this model broadly defines the link layer as encompassing those "specific techniques to be used in order to transmit data between systems despite a relatively high error rate . . . ." (Ex. 16 at 429.) Nothing in the Zimmerman article limits link layer error correction to a particular type, such as frame retransmission. To the contrary, Zimmerman explicitly recognizes that the types of error control procedures that may be found in the link layer may change over time. "It must also be recognized that new physical communications media . . . will require quite different data-link control procedures [from those already in standard use]." (Id. at 429.) The patent claims themselves support this understanding. Claim 1 recites a "method for establishing a link layer connection." (Ex. 3 at 14:25.) Claim 3 depends from claim 1 and further requires that "said link layer connection is an error-correcting protocol." (Ex. 3 at 14:4243.) Yet Comcast would have this Court rule that claim 1 is already limited to error-correcting protocols that operate through frame retransmission ­ in other words, that independent claim 1 is narrower than dependent claim 3. Such a construction is impermissible. See 35 U.S.C. § 112, ¶ 4 ("a claim in dependent form shall . . . specify a further limitation of the subject matter claimed"); Lampi Corp. v. Am. Power Prods., Inc., 228 F.3d 1365, 1376 (Fed. Cir. 2000) (construing an independent claim to ensure that a dependent claim was not of broader scope). In short, no one disputes that, where used, frame retransmission is part of the link layer. But nothing in Zimmerman or the '631 patent suggests that frame retransmission is a necessary feature of the patent. b. Error Correction Is a Link Layer Feature

Comcast maintains that because the V.34 standard includes a protocol for establishing a physical layer connection, and also includes trellis encoding, that trellis encoding must be part of
PLAINTIFF REMBRANDT TECHNOLOGIES, LP'S REPLY BRIEF REGARDING CLAIM CONSTRUCTION - Page 9

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191 Document 132-1 Filed 06/28/2007 Page 1414 of 19 Filed 01/18/2007 Page of 19

the physical layer. (Response at 22.) This argument is may easily be dismissed because Comcast's own expert testified that protocols like the V.34 standard include functionality in multiple layers: "There are many examples of communications protocol standards that encompass multiple layers in the OSI protocol model." (Ex. 26 at 93:17-19.) Comcast's expert also explained that "it's generally understood as to which functions go within which layer of the OSI protocol model." (Id. at 94:7-9.) In view of the repeated statements in the '631 patent and the Zimmerman article that error correction is a function of the link layer, this testimony confirms that the V.34 protocol spans multiple layers of the OSI model. While some of the V.34 protocol's functionality relates to the physical layer, the trellis-encoding functionality, which pertains to error correction, exists within the link layer. c. The Extrinsic Evidence Supports Rembrandt's Construction

Comcast's citations to treatises and dictionary definitions to buttress its link layer argument are unavailing. While as noted above frame retransmission was one known method of implementing link layer error correction, this was clearly not the only known method. Comcast acknowledges that the Lim article (Ex. 20), published only shortly after the filing date of the '631 patent, utilizes a form of error correction in the link layer that does not use frame retransmission. (Response at 23; Bims Decl. ¶ 22.) Even a patent issued to Comcast's own expert clearly states that error correction in the link layer can be achieved through Reed-Solomon coding, as opposed to frame retransmission. (See Ex. 28 at 8:6-9 ("In one embodiment, data link layer 408 is implemented . . . with Reed-Solomon decoding . . . .")) 2. Comcast's Indefiniteness Argument Violates the Local Rules and Is Contradicted by Its Own Expert

In Comcast's opening claim construction brief, Comcast argues for the first time that two means-plus-function terms in the '631 patent5 are invalid as indefinite. Comcast's argument
5

"means for establishing said link layer connection based upon the negotiated physical layer connection" and "means for presetting said link layer parameters based on the negotiated physical layer modulation."

PLAINTIFF REMBRANDT TECHNOLOGIES, LP'S REPLY BRIEF REGARDING CLAIM CONSTRUCTION - Page 10

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191 Document 132-1 Filed 06/28/2007 Page 1515 of 19 Filed 01/18/2007 Page of 19

comes too late. P.R. 3-3(d) requires that Comcast disclose any indefiniteness argument in its Preliminary Invalidity Contentions, but no such disclosure was made. Nor did Comcast contend that these terms were indefinite when the parties filed their Amended Joint Claim Construction Statement. (See Docket No. 112 at A-15 ­ A-17.) Nor did Comcast's expert opine that these terms were indefinite. (Ex. 26 at 241:15-242:3; 242:18-24; see also id. at 214:7-13.) Because Comcast can cite no evidence supporting its indefiniteness argument, and because Comcast provided no notice to Rembrandt of its position, the Court should not entertain Comcast's late contentions now and should adopt Rembrandt's proposed construction. D. U.S. Patent No. 5,243,627

Comcast's constructions of the '627 patent exemplify one of the basic methodological defects described at the outset of this brief. When conducting claim construction, the court should "rely heavily on the written description for guidance as to the meaning of the claim terms." Phillips, 415 F.3d at 1317. Comcast instead relies extensively on extrinsic evidence and unsubstantiated claims of its expert, while treating the patent specification ­ the "single best guide" for claim construction, id. at 1315 ­ as an afterthought. Because Comcast's constructions depart from (and at least in one case contradict) the explicit teaching of the '627 patent specification, they are incorrect as a matter of law. 1. trellis encoded channel symbol (claims 9, 19)

Comcast's construction of this term requires trellis encoded channel symbols to be generated using the outputs of a "single state transition" of the trellis encoder. Comcast asserts that its construction is consistent with the preferred embodiment because the patent is purportedly "clear" that pairs of subset identifiers are "generated in parallel" and thus generated during a single state transition. (Response at 29.) The reality is that Comcast points to nothing in the specification that supports this notion. It relies instead exclusively on its expert. (Response at 29; Bims Decl. ¶ 53.) The patent specification, in fact, teaches exactly the opposite of what Comcast contends. It repeatedly states that subset identifiers are generated sequentially rather than in parallel.
PLAINTIFF REMBRANDT TECHNOLOGIES, LP'S REPLY BRIEF REGARDING CLAIM CONSTRUCTION - Page 11

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191 Document 132-1 Filed 06/28/2007 Page 1616 of 19 Filed 01/18/2007 Page of 19

"[T]he words identified on lead 109 are used by trellis encoder 119 to sequentially identify on lead 121 N subsets . . . ." (Ex. 4 at 3:53-55.) "A succession of N outputs from the trellis encoder identifies a particular one of the 2N-dimensional subsets . . . ." (Id. at 4:19-21.) Since Comcast's proposed construction would exclude the channel symbols specifically described in the '627 patent, that construction is presumptively incorrect. Primos, Inc. v. Hunters Specialties, Inc., 451 F.3d 841, 848 (Fed. Cir. 2006) (noting that "we also should not normally interpret a claim term to exclude a preferred embodiment"). 2. signal point (claims 9, 19)

Comcast claims that the term "signal point" is "overwhelmingly" used to refer to a "mapped" point in a signal constellation. (Response at 31.) Yet Comcast does not identify a single description of this "mapping," either in the '627 patent or in its own extrinsic evidence. Comcast's citations to the '627 patent ("[t]his constellation is comprises of 32 signal points, which are divided into four subsets, A through D, each comprised of eight signal points") and to the extrinsic evidence '000 patent ("[e]ach signal point in the [signal] constellation has an associated bit code") mention nothing about mapping. (Response at 30-31.) Comcast's inclusion of "mapped" in its construction is unsubstantiated, making its construction incorrect. Comcast's sole argument against Rembrandt's construction is that signal points are not transmitted. The patent specification, however, directly contradicts this argument. "Those two signal points are thereupon communicated over the channel by QAM encoder 124 and modulator 128 as described above." (Ex. 4 at 4:1-3.) Moreover, Comcast's expert also concedes that signal points are transmitted. (Bims Decl. ¶ 35 ("Each signal point corresponds to certain waveform characteristics (e.g., amplitude, frequency, etc.), and as such can be transmitted over the medium (e.g., a wire).")) 3. distributed Viterbi decoder (claims 9, 19)

By requiring "two or more Viterbi decoders" in its construction, Comcast completely fails to address the explicit disclosure of the '627 patent cited in Rembrandt's opening brief.

PLAINTIFF REMBRANDT TECHNOLOGIES, LP'S REPLY BRIEF REGARDING CLAIM CONSTRUCTION - Page 12

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191 Document 132-1 Filed 06/28/2007 Page 1717 of 19 Filed 01/18/2007 Page of 19

[M]ultiple trellis encoders and decoders can be realized using a single program routine which, through the mechanism of indirect addressing of multiple arrays within memory, serves to provide the function of each of the multiple devices. (Ex. 4 at 9:61-66.) The specification makes clear that the distributed Viterbi decoder can be implemented as a single Viterbi decoder that emulates the function of multiple physical devices. Such a distributed Viterbi decoder, contrary to Comcast's assertion, is distinguishable over a prior art Viterbi decoder because it has multiple, rather than a single, Viterbi decoding processes. 4. means for deinterleaving the interleaved signal points to recover said plurality of streams of trellis encoded channel symbols (claim 9)

Comcast asserts that the function of this limitation is "expressly limited to reversing the signal point interleaving" that is performed by interleaver 341 in the transmitter. (Response at 33.) This notion is contradicted by the claim language itself, which states that the claimed deinterleaving must "recover [a] plurality of streams of trellis encoded channel symbols" ­ a function that is clearly performed by switching circuit 431. (See Ex. 4 at Fig. 4.) Comcast is also incorrect that switching circuit 431 "by itself" cannot deinterleave the interleaved signal points. (Response at 33-34.) The claims describe interleaving signal points in two different ways: one way ensures that "signal points of each channel symbol are nonadjacent" and a second way causes "signal points of adjacent symbols in any one of said channel symbol streams [to be] non-adjacent." The corresponding structure identified by Comcast ­ essentially the deinterleaver 441 ­ is involved in reversing only the first of those processes, and is therefore incomplete. By contrast, Rembrandt's construction correctly identifies both of the structures that perform deinterleaving in the '627 patent.

PLAINTIFF REMBRANDT TECHNOLOGIES, LP'S REPLY BRIEF REGARDING CLAIM CONSTRUCTION - Page 13

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191 Document 132-1 Filed 06/28/2007 Page 1818 of 19 Filed 01/18/2007 Page of 19

Dated: January 17, 2007

Respectfully submitted, FISH & RICHARDSON P.C.

By: /s/ Thomas A. Brown Otis Carroll State Bar No. 03895700 Wesley Hill State Bar No. 24032294 IRELAND, CARROLL & KELLEY, P.C. 6101 S. Broadway, Suite 500 Tyler, TX 75703 Tel: (903) 561-1600 Fax: (903) 581-1071 Email: [email protected] Frank E. Scherkenbach Lawrence K. Kolodney Michael H. Bunis Thomas A. Brown FISH & RICHARDSON P.C. 225 Franklin Street Boston, MA 02110 Tel: 617-542-5070 Fax: 617-542-8906 Timothy Devlin FISH & RICHARDSON P.C. 919 N. Market Street, Suite 1100 Wilmington, DE 19899-1114 Tel: 302-652-5070 Fax: 302-652-0607 Alan D. Albright State Bar No. 00973650 FISH & RICHARDSON P.C. One Congress Plaza, 4th Floor 111 Congress Avenue Austin, TX 78701 Tel: 512-391-4930 Fax; 512-591-6837 Counsel for Plaintiff REMBRANDT TECHNOLOGIES, LP

PLAINTIFF REMBRANDT TECHNOLOGIES, LP'S REPLY BRIEF REGARDING CLAIM CONSTRUCTION - Page 14

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191 Document 132-1 Filed 06/28/2007 Page 1919 of 19 Filed 01/18/2007 Page of 19

CERTIFICATE OF SERVICE The undersigned hereby certifies that a true and correct copy of the above and foregoing document has been served on seventeenth day of January, 2007 to all counsel of record who are deemed to have consented to electronic service via the Court's CM/ECF system per Local Rule CV-5(a)(3). Any other counsel of record will be served by first class mail. Brian L. Ferrall, Esq. Leo L. Lam, Esq. Eric MacMichael, Esq. Keker & Van Nest, L.L.P. 710 Sansome Street San Francisco, CA 94111-1704 Jennifer Haltom Doan, Esq. John Peyton Perkins, III, Esq. Haltom & Doan, LLP 6500 N. Summerhill Road Crown Executive Center, Suite 1A P. O. Box 6227 Texarkana, TX 75505-6227 Attorneys for Defendants Comcast Corporation, Comcast Cable Communications, LLC and Comcast of Plano, LP

Attorneys for Defendants Comcast Corporation, Comcast Cable Communications, LLC and Comcast of Plano, LP

/s/ Thomas A. Brown Thomas A. Brown
21534251.doc

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191-2 Document 132-2 Filed 06/28/2007 Page 1 1 of 23 Filed 01/18/2007 Page of 23

Exhibit 25

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191-2 Document 132-2 Filed 06/28/2007 Page 2 2 of 23 Filed 01/18/2007 Page of 23
Page 1

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

UNITED STATES DISTRICT COURT FOR THE EASTERN DISTRICT OF TEXAS MARSHALL DIVISION ) ) Plaintiff, ) ) vs. ) ) COMCAST CORPORATION; COMCAST ) CABLE COMMUNICATIONS, LLC; ) AND COMCAST OF PLANO, LP ) ) Defendants. ) ______________________________) REMBRANDT TECHNOLOGIES, LP

No. 2:05-cv-00443-TJW

VIDEOTAPED DEPOSITION OF CURTIS A. SILLER, Ph.D., San Francisco, California Wednesday, December 20, 2006

Reported by: Rebecca Romano, CSR No. 12546 Job No. 201462

Esquire Deposition Services (215) 988-9191

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191-2 Document 132-2 Filed 06/28/2007 Page 3 3 of 23 Filed 01/18/2007 Page of 23
Page 6

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

San Francisco, California, Wednesday, December 20, 2006 9:27 A.M. - 4:22 P.M.

(Siller's Exhibit Nos. 1 through 4 were marked for identification.) THE VIDEOGRAPHER: Good morning. Here begins

Videotape No. 1 in the deposition of Curtis Siller in the matter of Rembrandt v. Comcast in the United States District Court for the Eastern District of Texas Marshall Division, Case No. 205-CV-00443-DJW. Today's date is December 20th, 2006. is 9:27 A.M. This deposition is being taken at 710 Sansome Street, San Francisco, California. The videographer is The time

Marty Marjoub, here on behalf of Esquire Deposition Services, 505 Sansome, Suite 502, San Francisco, California. Will all counsel present please identify yourselves and state whom you represent. MR. DEVLIN: Tim Devlin of Fish & Richardson

for plaintiff Rembrandt. MR. WERDEGAR: Matthew Werdegar of Keker & Van Also present is Audrey -Walton-Hadlock.

Nest for defendant Comcast. MS. WALTON-HADLOCK: MR. WERDEGAR:

-- Walton-Hadlock.

Esquire Deposition Services (215) 988-9191

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191-2 Document 132-2 Filed 06/28/2007 Page 4 4 of 23 Filed 01/18/2007 Page of 23
Page 54

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

over.

I think I -The part on the left is the packet channel. A Q A Q Understood you to have said that, yes. Let me just say it all again clearly. Okay. We'll start from the start. We're looking at that dotted line down the

middle of figure 5. right? A Q Correct.

And it divides it into two regions,

The region on the left is the channel for

packet sources. MR. WERDEGAR: Q A Q (By Mr. Devlin) Yes. And the region on the right are the channels Objection. Form.

Is that fair?

for the synchronous sources. MR. WERDEGAR: THE DEPONENT: Objection. Form.

Not labeled as such, but I

would infer that to be the case. Q (By Mr. Devlin) There are time slots shown on

the left-hand side of the figure. A Q Right. What are those? MR. WERDEGAR: Objection. Form.

Esquire Deposition Services (215) 988-9191

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191-2 Document 132-2 Filed 06/28/2007 Page 5 5 of 23 Filed 01/18/2007 Page of 23
Page 55

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 A Q Q A Q A Q

THE DEPONENT:

Those are individual time

slots, fixed in size, that make up a portion of the part of the overall frame that is set aside for packet communications. Q (By Mr. Devlin) On the right-hand side of the

figure it doesn't actually say the word "time slot." Do you see that? Yes, I see that. Based on your understanding and reading of the

entire patent, is it your understanding that those time slots exist on the right-hand side of the figure also? A Q That would be my understanding. They are just not shown in figure 5. MR. WERDEGAR: (By Mr. Devlin) That's fair. There's a designation that says, "Start of Objection. Form.

Is that fair?

packet 50." Do you see that? I do. And that starts -- a thick bold line that

starts on frame 1, ends at that dotted line, and then continues onto the left-hand side of frames 2 and 3, continues into frame 4, again, on the left-hand side, and then ends.

Esquire Deposition Services (215) 988-9191

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191-2 Document 132-2 Filed 06/28/2007 Page 6 6 of 23 Filed 01/18/2007 Page of 23
Page 57

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

Q

Are you able to tell from figure 5 which of

the packet data sources would transmit the next packet after the one that's shown? MR. WERDEGAR: THE DEPONENT: Q (By Mr. Devlin) Objection. Form.

Not from figure 5. If you knew from the text of

the patent -- this is a hypothetical. If the text of the patent told you which of the packet data sources was transmitting that packet -A Q A Q Uh-huh. -- in figure 5 -Yes. -- could you then tell me what the next packet

source would be to transmit a packet? MR. WERDEGAR: THE DEPONENT: Q A (By Mr. Devlin) Objection. I could not. Why is that? Form.

Because the very next -- next one may not have

any information to send. Q A Q And if it doesn't, what would happen? It would go to the next one. So there's no particular order in which the

packet data sources have to transmit data in the '858 patent invention. MR. WERDEGAR: Objection. Form.

Esquire Deposition Services (215) 988-9191

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191-2 Document 132-2 Filed 06/28/2007 Page 7 7 of 23 Filed 01/18/2007 Page of 23
Page 58

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Q Q A Q A Q

THE DEPONENT:

I believe this is description

of a sequential process in which there is a sequential arbitration to invite the individual packet modules to transmit. Q (By Mr. Devlin) Let me -- let me see if I can

get an answer.

This is a foundational question.

There's -What is --- no actual -What is a foundational question? I just want to get a more basic bit of

information -A Q A Q Oh. -- from you. Okay. In the '858 patent, there's no actual sequence

by which the packet sources have to transmit their information. MR. WERDEGAR: (By Mr. Devlin) MR. WERDEGAR: THE DEPONENT: (By Mr. Devlin) Objection. Is that fair? Objection. Form.

I think that's fair. And what you were saying

before is that there is some sequence of arbitration, right?

Esquire Deposition Services (215) 988-9191

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191-2 Document 132-2 Filed 06/28/2007 Page 8 8 of 23 Filed 01/18/2007 Page of 23
Page 59

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 it.

A Q

Yes. But that arbitration doesn't mean that the

packet data's -- data sources will actually transmit in a given sequence. MR. WERDEGAR: THE DEPONENT: Q figure 6. Could you describe for me what's happening in figure 6 of the '858 patent? MR. WERDEGAR: THE DEPONENT: Objection. Yeah. Form. (By Mr. Devlin) Objection. Form.

That's right. In fact, let's turn to

I will try to describe

Time is shown as elapsing across the top. Q A Q A (By Mr. Devlin) From left to right?

Yes, from left to right. Thank you. And there is a suggestion that among the

packet data there have been -- has been a closing flag indicating that one of the packet sources has ceased communicating. Q Can I ask you where you are referring to that

closing flag? A figure 6. It's in the lower left-hand part of

Esquire Deposition Services (215) 988-9191

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191-2 Document 132-2 Filed 06/28/2007 Page 9 9 of 23 Filed 01/18/2007 Page of 23
Page 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

further on the right for P6, between what's designated as T19 and T23, I suppose it would be. Between T19 in

the right-hand side of the figure, the only packet data source that's transmitting is P6. Fair to say? A Again, that would -MR. WERDEGAR: THE DEPONENT: interpretation. Q (By Mr. Devlin) Okay. So just to summarize Objection. Form.

That would be my

this figure, figure 6, we are moving from the left to the right, and first, packet source 2 transmits information, then there's a closing flag, and then the next packet source to transmit information is packet 6. MR. WERDEGAR: Q (By Mr. Devlin) Objection. Form.

Is that -- is that a fair

interpretation of the figure? A Yes. If we overlook the fact that the

preceding module, which is not identified, has transmitted its closing flag. Q Will -- will module 6 always transmit

information after module 2? MR. WERDEGAR: Q (By Mr. Devlin) Objection. Form.

Based on your understanding

of the '858 patent?

Esquire Deposition Services (215) 988-9191

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191-2 Document 132-2 Filed 06/28/2007 Page 10 of 23 Filed 01/18/2007 Page 10 of 23
Page 63

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

A Q

I don't think that's a necessary condition. Why is that? MR. WERDEGAR: THE DEPONENT: Objection. Form.

Because module 2 and the

preceding modules may have had nothing to transmit at all. Q (By Mr. Devlin) So -- so I'm saying, next

time around, module 2 may transmit again sometime later in time, correct? A Q It may or may not. It may never transmit information again. Is that fair? A Q That's true. If it did, would it necessarily follow that

module 6 would be the next module to transmit information? MR. WERDEGAR: THE DEPONENT: Objection. Form.

This level of detail is getting

outside my area of expertise in terms of how the actual arbitration process works for the packet modules on the bus. Q (By Mr. Devlin) Is it your understanding that

the next time module 2 transmitted information, it could be followed by module 3? MR. WERDEGAR: Objection. Form.

Esquire Deposition Services (215) 988-9191

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191-2 Document 132-2 Filed 06/28/2007 Page 11 of 23 Filed 01/18/2007 Page 11 of 23
Page 64

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

Q A Q A Q

(By Mr. Devlin) Yeah. Or module 4? Possibly.

Is that fair?

Just depends on who has information to send

after module 2 and the nature of the arbitration technique that's used. MR. WERDEGAR: Q (By Mr. Devlin) MR. WERDEGAR: THE DEPONENT: Q (By Mr. Devlin) Objection. Is that fair? Objection. Form.

That's right. In figure 6, there's

something called a P-R-E-Q, all capital letters next to one another. Do you see that? A Q P-H-O-L-D. A Q Okay. I sometimes refer to that as PREQ and PHOLD. Is that how you refer to those? A I don't really refer to them because, as I Yes. And then there is something called a

indicated, the details of the arbitration technique for the packet sources to get on is outside my area of expertise.

Esquire Deposition Services (215) 988-9191

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191-2 Document 132-2 Filed 06/28/2007 Page 12 of 23 Filed 01/18/2007 Page 12 of 23
Page 66

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

Q line 18? A Q

Do you see a paragraph beginning around

Yes, I do. Reference to something called asynchronous

transfer mode, ATM? A Q A Q A That's right. Do you have an understanding of what ATM is? Yes, I do. What is it? It's a packetization technique for carrying a

variety of traffic, and it's most noted by the fact that the packets are fixed in size. Q A Why are they fixed in size? It was a compromise between the

telecommunication and the data industry that the particular size of the packets that are used for ATM, if carrying voice, are well-suited to the PSTN, the public switch telephone network. Whereas, people working in the computer industry are more inclined to want to send information in a packetized form, so they compromised on the size that met the needs of telephony and a fact that packetization is used for data communications. Q A What do you mean by "telephony"? Telephone calls.

Esquire Deposition Services (215) 988-9191

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191-2 Document 132-2 Filed 06/28/2007 Page 13 of 23 Filed 01/18/2007 Page 13 of 23
Page 67

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 cell?

Q

There is a sentence that begins at around line

22, "In the case of ATM." Do you see that? A Q Yes. Yeah. "In the case of ATM." I will just read that sentence.

And it says, "In the case of ATM cells, the oxidants which form the cells need to be aligned within DS0 channels." Do you see that? A Q I do. Do you have an understanding of what an ATM

A Q about? A Q

Yes, I do. That's that 48 bits you were just talking

Well, it's actually 53. Oh, 53 bytes. Thank you.

And they are bytes.

Is there anything else that makes up part of an ATM cell? A header. Well, it has the two ingredients. It has the information field. It has the

Part of the

AT -- information field can be given over to what they call the ATM adaptation layer. Q A What is in the header? The header would contain addressing

Esquire Deposition Services (215) 988-9191

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191-2 Document 132-2 Filed 06/28/2007 Page 14 of 23 Filed 01/18/2007 Page 14 of 23
Page 68

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

information. Q A

Yeah, addressing information.

Anything else? It may indicate the fact -- what type of AAL, There are five, AAL

ATM adaptation layer is being used. 1 through AAL 5. Q Okay.

Are ATM cell -- when you are using an

asynchronous transfer mode network -A Q Uh-huh. -- are ATM cells used for both the packetized

data and for telephony? MR. WERDEGAR: THE DEPONENT: one more time. Q (By Mr. Devlin) Yeah. Let me try to ask a Objection. Form.

Would you repeat the question

simpler one. Are ATM cells used for telephony applications in ATM networks? MR. WERDEGAR: THE DEPONENT: Q (By Mr. Devlin) Objection. Form.

Yes, they are. That sentence continues to

say, "The octecs which form the cells need to be aligned within DS0 channels." Do you see that? A Q I do. Do you understand what I mean by octecs?

Esquire Deposition Services (215) 988-9191

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191-2 Document 132-2 Filed 06/28/2007 Page 15 of 23 Filed 01/18/2007 Page 15 of 23
Page 92

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Q

MR. WERDEGAR: THE DEPONENT: it's an example. Q (By Mr. Devlin)

Objection.

Form.

It says on the second line that

Are you aware of other ways

to do time-division multiplexing that don't involve collision avoidance? MR. WERDEGAR: THE DEPONENT: techniques. Q A (By Mr. Devlin) Can you tell me an example? For Objection. Yes. Form.

I think there are other

Well, there are pulling strategies.

example, a lower, I think is one of them, in which you have multiple sources that try to access a shared medium, but they are only given permission if they receive a token which they pick up. So that would be an example of one. That does not use collision avoidance? MR. WERDEGAR: THE DEPONENT: Objection. Form.

Well, it depends on what you Certainly using the token

mean by collision avoidance.

avoids the incident of a collision. Q (By Mr. Devlin) Are there any examples of

time-division multiplexing, that you're aware of, where collisions can take place? MR. WERDEGAR: Objection. Form.

Esquire Deposition Services (215) 988-9191

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191-2 Document 132-2 Filed 06/28/2007 Page 16 of 23 Filed 01/18/2007 Page 16 of 23
Page 93

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Q Q place? Q A Q

THE DEPONENT:

Where collisions can take

(By Mr. Devlin) Yes.

Uh-huh.

Can you give me an example of one of those

types of time-division multiplexing techniques? MR. WERDEGAR: THE DEPONENT: Objection. Form.

There are techniques where one

or more sources -- excuse me -- two or more sources can try to access a bus, and they do this simultaneously, and a collision, in fact, does take place. They gather

a number of techniques that detect that a collision has occurred, and then they back off a random amount of time and reattempt to access the bus. Q (By Mr. Devlin) Is there a specific example

of that, that you have in mind? MR. WERDEGAR: THE DEPONENT: (By Mr. Devlin) Objection. No. No. Okay. Form.

Is it like Ethernet works? MR. WERDEGAR: (By Mr. Devlin) Objection. Form.

My question, just in case it Do you know

wasn't clear because I said it fast, was: how Ethernet works, Dr. Siller? A No.

I don't know the details of how Ethernet

Esquire Deposition Services (215) 988-9191

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191-2 Document 132-2 Filed 06/28/2007 Page 17 of 23 Filed 01/18/2007 Page 17 of 23
Page 161

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

confirming, when you said "network administrator," you meant an actual person whose job it is to be a network -A Q Right. -- administrator. That's what you meant? A Q period? MR. WERDEGAR: THE DEPONENT: Objection. Form. And Yeah. Okay. What's a network clock framing

Well, there are subframes.

if there are subframes, there are presumably frames. Q A (By Mr. Devlin) Uh-huh.

And since these are delineated in time between

the acknowledged here and the claims terms, I think that explains it. Q Does this suggest to you that the user could

do the division of the period into frames and subframes and time slots? MR. WERDEGAR: THE DEPONENT: opinion on that. Q (By Mr. Devlin) Okay. In any case, Objection. Form.

I'm not sure I have an expert

there's -- there's nothing in that phrase that we just read or anything else that you can think of that states

Esquire Deposition Services (215) 988-9191

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191-2 Document 132-2 Filed 06/28/2007 Page 18 of 23 Filed 01/18/2007 Page 18 of 23
Page 162

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

that the network timing means actually does that division, right? A Q A Right. Okay. In fact, I believe that to be the structure

that count- -- is a counterpart to the patent claim terms you had me read. Q When you say the structure -- now, let's talk

about that. I'm looking at the chart again. A Q That's sorry. And now I'm carrying over -- unfortunately, I'm sorry.

this thing goes from A-23 to A-24, and on A-24 you see the structure. A Q A Q Oh, fine. Right? Yeah. Okay. And so the structure that you designate

is the network timing and control processor 12, right? A Q there. A Q Yeah. Can I ask you this: What exactly is the Is it just the Correct. Okay. And then you provide some citations

structure that you are citing here?

Esquire Deposition Services (215) 988-9191

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191-2 Document 132-2 Filed 06/28/2007 Page 19 of 23 Filed 01/18/2007 Page 19 of 23
Page 181

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

Q A

Uh-huh. -- in Figures No. 1 and 3. So it's not

specific what within the master unit does that. Q Right. It doesn't necessarily say that it's

the ranging means within the master unit that does that. It's -- it could be something else in the master unit. Is that fair? MR. WERDEGAR: THE DEPONENT: interpretation. Q (By Mr. Devlin) Is that a fair statement, Objection. Form. Leading.

That's a possible

that something else in the master unit could perform that function; that is, transmitting the transmission time to each of the respective remote units? MR. WERDEGAR: THE DEPONENT: Objection. Form. Leading.

I think that some of those

capabilities are captured in the structure that we enumerate over here. Q (By Mr. Devlin) I'm -- I'm with you there.

But what I'm asking you is -- so the -- let's just start at the start again. A Q Okay. This text doesn't specify that this

transmission from the master unit to each of the respective remote units is performed by the ranging

Esquire Deposition Services (215) 988-9191

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191-2 Document 132-2 Filed 06/28/2007 Page 20 of 23 Filed 01/18/2007 Page 20 of 23
Page 182

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

means, right? A Q means. A Q It does. Okay. But there's other things in the master That's correct. Okay. The master unit includes the ranging

unit also, right? A Q That's right. Okay. And it could be something else in the

master unit that performs that transmission function. MR. WERDEGAR: Q (By Mr. Devlin) MR. WERDEGAR: THE DEPONENT: Q (By Mr. Devlin) Objection. Leading.

Is that fair? Objection. Form.

Yes, that's fair. Okay. Let's talk now about

the function that you lay out -- oh, let me back up for a second. So apart from this claim language that we have just been discussing, is there anywhere else in the patent -- so we have been talking about the basis for your opinion that this claim term is, in fact, subject to 112, 6, right? A Q That's right. And you just gave me this claim language as

one of your bases, and we talked about it, right?

Esquire Deposition Services (215) 988-9191

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191-2 Document 132-2 Filed 06/28/2007 Page 21 of 23 Filed 01/18/2007 Page 21 of 23
Page 213

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

Q A aspect. Q

(By Mr. Devlin)

Sure.

And you haven't --

So I haven't formed an opinion on every

Okay.

You do agree with me, though, that the

description in the patent indicated that time slots could be assigned dynamically? MR. WERDEGAR: Beyond the scope. THE DEPONENT: description. Q (By Mr. Devlin) Let me just make sure we are I have read that in the Objection. Form. Leading.

on the same page. A Q Okay. What we read in the description at column 2,

lines 18 through 26 -- do you see that? A Q Yeah. -- indicates to you that time slots can be

assigned dynamically. Is the fair? MR. WERDEGAR: Objection. Form. Leading.

Beyond the scope of his expert designation. THE DEPONENT: Well, the master unit is going

to make a decision as to whether or not the requesting remote unit should use additional access slots. doesn't say that they are allocated to them. It

Esquire Deposition Services (215) 988-9191

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191-2 Document 132-2 Filed 06/28/2007 Page 22 of 23 Filed 01/18/2007 Page 22 of 23
Page 219

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

communication, for example, and there are several people using a network and you want to establish some users as taking precedence over other users so they have a different relative priority. Q (By Mr. Devlin) What about priority in terms

of one type of data versus another; do you ever have that type of priority? A Q A Yeah. That -- that could occur as well.

And could you give me an example of that? I don't think of a specific example; I think You have two

that maybe your words were sufficient. data communication flows.

There's a decision made

somehow that one has a higher relative priority than another, and that's that. Q Would it be possible for one type of

application -- data associated with one type of application to be granted a higher priority than data associated with another application? Would that be possible? MR. WERDEGAR: THE DEPONENT: possible. Q (By Mr. Devlin) Were you dealing with any Objection. Yes. Form.

I think that would be

issues of priority in the late 1980s or early 1990s when this patent was in the patent office and when it was

Esquire Deposition Services (215) 988-9191

Case 1:07-cv-00398-GMS Case 2:05-cv-00443-TJW-CE Document 191-2 Document 132-2 Filed 06/28/2007 Page 23 of 23 Filed 01/18/2007 Page 23 of 23
Page 223

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16