ChanBond, LLC v. Atlantic Broadband Group, LLC

Opinion

IN THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF DELAWARE

CHANBOND, LLC, : : Plaintiff, : : v. : Civil Action No. 15-842-RGA : Consolidated ATLANTIC BROADBAND GROUP, LLC, : et al., : : Defendants. :

MEMORANDUM ORDER

In connection with the proposed Chanbond v. Cox pretrial order, there was a dispute about the admissibility of portions of the deposition of Anthony Wechselberger. (D.I. 500 at 33- 36). Mr. Wechselberger was deposed on April 28, 2017, in connection with an instituted IPR concerning the ‘822 patent, one of the three asserted patents-in-suit. (See id; D.I. 438). He offered opinions during the deposition; Plaintiff asserts some of them are consistent with its expert’s opinions and inconsistent with Cox’s expert’s opinions. Both experts have some citations to this deposition testimony in their expert reports. It seems extremely likely Mr. Wechselberger will be unavailable for the trial in this case. In the proposed pretrial order, the parties argue about the applicability of Federal Rule of Evidence 804(b)(1), and Cox also argues that the testimony should be excluded under Rule 403. Plaintiff does not address Rule 403. (D.I. 500 at 33-36). We discussed the issue at the pretrial conference. (D.I. 509 at 23-31, 41-44). We agreed that there would be additional submissions, which were duly filed. (D.I. 508, 510, 514). The submissions include the proposed testimony,1 with various annotations by the parties. (D.I. 508- 4; see D.I. 510-4; D.I. 514-2). In the additional submissions, both sides make arguments concerning Rule 804(b)(1) and Rule 703, but only Cox addresses Rule 403 concerns. I believe that Plaintiff is only asserting one claim of the ‘822 patent—claim 14. (D.I. 505 at 1; D.I. 506 at 1).2

I quote the five proposed excerpts. Q. So the question is the second embodiment that you mentioned is at column 5, lines 22 through 25; correct? A. Yes. Q. In this embodiment the sets of additional channels are predefined; right? A. That's what Tiedemann says, yes. Q. These additional channels are still dynamically allocated; correct? A. Yes. Q. So a person of ordinary skill would know that a channel can be dynamically allocated even though the set of possible channels is pre-assigned before communication starts; is that right? A. You'll have to run that question by me again. It took a bit to come out. I lost it. Q. A person of ordinary skill would know that channels can be dynamically allocated even though the set of possible channels is pre-assigned before communication starts? A. I would agree with that.

Q. If you turn to Exhibit 1001, the '822 patent, at column 9, line 3 through 5, you can read it to yourself, but on line 5 there is a mention of parallel-to-serial conversion. Do you see that? A. So here he's talking about Figure 2. Q. Please feel free to review whatever other portions of the '822 patent you need to. A. Figure 2 is called a local RF receiver baseband out for use in sending baseband information to a wideband network and receiving digital and nondigital information. Back to column 9. Okay. Okay. And your question again, if there was one? Q. Yes. At line 5 -- sorry. Column 9, line 5 there is a mention of parallel-to-serial conversion. Do you see that? A. Yes. Q. What does "parallel" mean in this context? A. So the output -- looking at Figure 2, the output of demodulator 220, which is referenced right above the particular citation you focused on, says the demodulator -- I'm reading line 1. It says the demodulator 220 strips the RF carrier signal from the digital baseband signal as known in the art. Following the demodulation the IP signals, plural, are combined by a digital combiner 212, which is shown in the Figure. So into combiner 212 -- into the digital combiner 212 are -- you could see at least two inputs coming from the demodulator 220. Those are parallel inputs, which he describes as being converted into a single serial stream. Q. And just to -- for clarity of the record, we're talking about Figure 2; correct? A. Yes. The lead-in to the top of column 9, unless

1 Plaintiff refers to the highlighted portions of the Wechselberger testimony as “representative.” (D.I. 508 at 2 n.3). I am not sure what is meant by this word, but my interpretation (in light of what I requested at the pretrial conference) is that the highlighted testimony is essentially the universe of what Plaintiff wants to offer, although it might offer less.

2 The PTAB declined to institute the IPR on claim 14. The IPR resulted in the invalidation of claim 1 of the ‘822 patent, but absent some circumstance I am now unaware of, the jury will not be permitted to hear that. In the IPR, “channel” was construed as “a path for transmitting electric signals.” No other term required construction. (Although I construed various terms, “channel” was not one of them. (D.I. 86)). I missed a transition, begins at the top of line 8 in a description of Figure 2, taking us all the way through the various components; and Figure 2 is still the subject by the time you get to the top of column 9. Q. So going back to my earlier question, what does the '822 patent mean by parallel- to-serial conversion here? A. It means simply there are parallel streams of digital information or parallel -- since he's talking about an IP signal there, that stands for Internet Protocol, so we would assume those signals are digital. It says "digital signals are combined." So you have a plurality or at least more than one IP digital signal arriving at the combiner 212 at its input, which is the right-hand side of that box. So you have simultaneous streams of IP digital information entering the combiner. Those will then be converted into a single bitstream, which he then denotes as a high speed serial digital output. . . . Q. Yes. Can you explain your opinion that Tiedemann's multiplexer 127 performs parallel-to-serial conversion. A. I have a citation at paragraph 224 of my report which is actually the same one you just called to my attention a few moments ago at column 8. The citation says, quote, multiplexer, or MUX, M-U-X, 127 combines the decoded data transmitted on the primary channel with the decoded data transmitted on the additional channels. The resembled data packet is provided to data sink 130. So in conjunction with the Figure I show in paragraph 221 of my report, which was taken from Tiedemann Figure 3, you can see at the top of page 137 of my report the primary decoder 122 outputting the information on the primary channel, and the output of the additional channels is shown in the box 126 called additional decoder, and those are both going into the multiplexer 127. So here you see the parallel paths of information, just like we talked about in the '822 patent, with the single now combined information going to the data sink 130. Q. Can you think of a multiplexer that would not perform parallel-to-serial conversion? A. Well, the term "multiplexer" generally would be understood by a person of ordinary skill to take in multiple inputs and combine them into some different number of outputs. It doesn't have to be one output. You could have eight in, two out. You could have eight in, one out. So multiplexing comes in all different kinds of flavors. And it's not typical, in my experience, to usually refer to that as parallel-to-serial conversion, although certainly in the context of the claim here it's understood what's going on.

Q. If you look at Exhibit 1009 Tiedemann at column 8, lines 23 and 24, I'm just going to read it because it's very short. It says, "The reassembled data packet is provided to data sink 130." Did I read that right? A. Yes. Q. Do you read that Tiedemann's multiplexer 127 reassembles the data packet before providing -- before providing it to data sink 130? A. Well, one has to take him literally at what he says, which is what you read answers your question. He says the reassembled, past tense, data packet is provided to data sink. So whatever else the multiplexer did, it also reassembled. Q. So the answer is "yes"? A. Yes. The answer is yes. Q. So a person of ordinary skill in the art would know that the combiner in the claims of the '822 patent does not simply merge the data together but also properly resequences the data in the original order? A. Now you're referencing Claim 1C and 19D? Q. I am referencing the combiner limitation of Claims 1 and 19 of the '822 patent. A. So I'm looking at my report at page 138, paragraph -- the box leading into paragraph 224. So there we have that limitation. Q. That is correct. A. And what was the question again about that? Q. The question is a person of ordinary skill in the art would know that the combiner limitation of the '822 patent does not simply merge the data together but also resequences the data in the right order. A. Well, the claim only cites what it claims and it doesn't say anything either way about your question, so I don't think it's proper to read into the claim anything other than the claim language. It doesn't say anything about resequencing. . . . Q. So I think you can agree that the claims of the '822 patent do not exclude a combiner that in addition to combining data also resequences the data in the correct order? A. The claim doesn't say that data sequence has any function. So, you know, it doesn't exclude it and it doesn't include it. It's silent on the issue. . . . Q. You don't know a reason why the '822 patent would exclude a combiner that resequences the data in the right order? A. I can't think of a reason why it would exclude it. I just said I don't recall if it addressed that issue. Q. So as best you understand, a person of ordinary skill in the art would know that the combiner in the '822 patent claims can resequence the data in the right order? A. That issue is nonexistent in the claim, so there is no reason for a person of ordinary skill reading the claim to even consider it.

Q. That single datastream that comes out of multiplexer 127 would include data addressed to the first addressable device and data addressed to the second addressable device? A. If there are more than one addressable devices and you had MUX 127 put everything down into a single stream, then that single stream would now contain a multiplex of information destined for two addressable devices. Q. That would still be within the scope of the claims of the '822 patent; correct? A. Yeah, appears to be. . . . Q. So you agree that a person of ordinary skill in the art reading the '822 patent would understand that the combiner limitation of Claims 1 and 19 can be met by a combiner that combines the at least two channels into a single datastream for all traffic, regardless of what specific addressable device it is intended for? A. Well, putting a mild correction on the very end of that, that single stream would contain the information destined for all the addressable devices, and there is nothing in the combiner limitation that says you could not do that, because it says -- because it combines it into a digital stream. So, and if there is at least one addressable device, there could be more addressable devices so that a digital stream would indeed contain data that was destined for more than one addressable device. . . . Q. So you agree that the digital stream of the '822 patent, Claims 1 and 19, can contain data bound for different addressable devices? A. Yes. . . . Q. Now assume, as we did before, that two addressable devices are connected to Tiedemann's mobile station in this combination with Gorsuch. Is that scenario realistic? A. Yes, I believe so. Q. In that case the channel-in-use information would still identify the channels assigned to the mobile station and not to any specific addressable device; correct? A. The question again? Q. In the case where two addressable devices are connected to the Tiedemann's mobile station in combination with Gorsuch, the channel-in-use information would identify the channels assigned to the mobile station and not to any specific addressable device; is that correct? A. I think I agree, because Gorsuch provides the targeting for the addressable devices in conjunction with the traffic management processes that are embodied in its ISDN modems, whereas Tiedemann, contrastly, identifies the additional -- the primary and additional channels through the Walsh Code assignments. . . . Q. In your proposed combination of Tiedemann and Gorsuch, the channel-in- use information does not identify channels carrying traffic addressed to any particular addressable device; is that right? A. It doesn't identify traffic addressed to a particular addressable device. I would agree with that. Q. So a person of ordinary skill in the art would know that the language of the claims of the '822 patent can be met where the channel-in-use information does not identify any channels with information addressed to a particular addressable device? A. Yeah. That requirement is not found in the first claim element. It just says it receives channel-in-use information which identifies the channel in the modulated RF signal that includes information addressed to at least one addressable device, so it's not particular as to which device. Q. A person of ordinary skill in the art would know that the language of the claim in the input limitation can be met where the channel-in-use information identifies the channels with information addressed to any of the addressable devices but not a specific addressable device. A. I agree with that.

Q. Now on to the additional channels. The base station sends the channel assignment message to the mobile station, Tiedemann, in advance of sending data to the mobile station; correct? A. In advance of...? Q. Sending data on those channels to the mobile station. A. Correct. Q. So in Tiedemann at the time the mobile station receives the channel assignment message, no additional channels are in use yet? A. Channel assignment -- the answer to that is maybe, because there may have been a previous channel assignment message which established, say, one additional channel. Maybe that's not enough bandwidth; so, you know, a new channel assignment message can come out allocating additional channels. So at the time the additional channels are allocated, there may have been a predecessor additional channel in effect. Q. But it's also possible that there were no additional channels prior to that channel assignment message; is that right? A. Sure. Q. In your opinion, even in that scenario, Tiedemann would meet the claim language; correct? A. Which claim language? Q. The language of channel-in-use information. Let me spell out it would be the limitation of an input configured to receive a modulated RF signal containing multiple channels and to receive channel-in-use information. A. So what's the question? Q. The question is even in the scenario where there were no additional channels prior to the channel assignment message, Tiedemann would need the claim language of an input configured to receive channel-in-use information? A. Yes.

I now address the arguments of the parties. First, Rule 804(b)(1). “In order for former testimony to be admissible as an exception to the hearsay rule: (1) the declarant must be unavailable; (2) testimony must be taken at a hearing, deposition, or civil action or proceeding; and (3) the party against whom the testimony is now offered must have had an opportunity and similar motive to develop the testimony by direct, cross, or redirect examination.” Kirk v. Raymark Indus., Inc., 61 F.3d 147, 164 (3d Cir. 1995). The point in dispute is the third. In the IPR, the party was Cisco, which is indemnifying Cox in this case. There is no dispute before me that Cisco counts as Cox’s predecessor-in- interest. The argument is about whether Cisco had “similar motive to develop the testimony.” I think the answer is no. At the IPR, the issue in dispute was whether various claims of the ‘822 patent, not including claim 14, were invalid as obvious under the prior art combination of Tiedemann, Gilhousen, and Gorsuch. That is not the issue, though, for which Plaintiff offers it now. Rather, Plaintiff suggests that the testimony shows infringement. I think Plaintiff is advancing two arguments about why the testimony supports infringement. One is extremely convoluted, but I would describe it as, if Tiedemann meets an

element, Tiedemann is the same as an accused product in that regard, so therefore the accused product also meets the limitation.3 The other is more straightforward. Some testimony opines on what a POSA would understand about various subjects, at least some of which appear to be relevant to issues that I expect will arise at trial. Regardless, I do not think the proper motive comparison is between Cisco’s raising of invalidity defenses and Cox’s raising of the same invalidity defenses (see D.I. 514); rather, I think the proper motive comparison is between Cisco’s raising of invalidity defenses and Cox’s raising of non-infringement arguments. For invalidity purposes, Cisco/Cox wanted to read the prior art and the claims as broadly as possible; for infringement purposes, Cisco/Cox does not necessarily care about the prior art, but surely wants to read the claims narrowly. Those goals are not the same or similar motivation; they are

different motivations. Thus, I do not think the Wechselberger testimony is admissible as former testimony under Rule 804(b)(1). Second, Rule 703. “An expert may base an opinion on facts or data in the case that the expert has been made aware of or personally observed. If experts in the particular field would reasonably rely on those kinds of facts or data in forming an opinion on the subject, they need not be admissible for the opinion to be admitted. But if the facts or data would otherwise be inadmissible, the proponent of the opinion may disclose them to the jury only if their probative

3 I would tend to think this argument, which is essentially “reverse practicing the prior art,” is probably a non-starter for the same reason that “practicing the prior art” is not a permissible argument. value in helping the jury evaluate the opinion substantially outweighs their prejudicial effect.” Fed. R. Evid. 703. There are a number of problems with Plaintiff’s argument on its Rule 703 theory. First, it is less than clear that Dr. Nettles is relying upon anything Mr. Wechselberger said. In the papers

submitted to me, in particular, the Reply Report, there are three places that have been cited to me where Dr. Nettles quotes Mr. Wechselberger. The first two are introduced with, “My opinions are consistent with the testimony of Mr. Wechselberger . . .” (D.I. 514-2, ¶¶ 94, 108). The third place has a different introduction, which is immediately followed by, “To be clear, I am relying on the testimony of Mr. Wechselberger only to the extent it shows that Defendants’ position is inconsistent with that of their own expert.” (Id., ¶ 294). Dr. Nettles does not rely upon the Wechselberger testimony to form any of his opinions. He formed them independently. Rather, he relies upon the Wechselberger testimony as a substitute for cross-examining Cox’s expert (Dr. Cromarty) with someone else’s opinions. Plaintiff has not addressed the requirement that it needs to show that the “probative value in helping the jury evaluate [Dr. Nettles’] opinion

substantially outweighs their prejudicial effect.” Thus, without Plaintiff even attempting the Rule 703 predicate for admission, its argument fails. The testimony will not be admitted under Rule 703 either. As I see it, the only point of stating anything about the Wechselberger testimony is essentially because Mr. Wechselberger stated it to advance Cox’s interests and it is contrary to what Cox now sees as its interests. It makes Cox look dishonest. But Plaintiff does not rely upon the Rule 801(d) bases for admitting statements of a party-opponent, presumably because those grounds have usually been rejected as a basis for allowing such evidence. Given that Plaintiff’s expert does not appear to be relying upon the Wechselberger testimony for any of his opinions, and certainly is not mainly relying upon it for anything, I will grant Cox’s request under Rule 403 to exclude not only the testimony but any reference to the testimony. It has at most little probative value, which is substantially outweighed by the risk of

unfair prejudice, confusion, and waste of time, as the introduction of such testimony would open the door to arguments about Tiedemann (which is referred to frequently in the proposed testimony, but which the jury would not otherwise hear about), explanation of what an IPR is, explanation as why claim 14 was not part of the decision, and possibly the explanation of the relationship between Cisco and Cox. See Fed. R. Evid. 403. Both sides are prohibited from referring to Mr. Wechselberger or his testimony. IT IS SO ORDERED this 19th day of April 2021.

/s/ Richard G. Andrews United States District Judge