# Oral argument — Samsung Electronics Co., Ltd. v. ITC > Briefs, arguments, decisions, and more. URL: https://www.frixlaw.com/law-library/documents/argument%3A8721 ## Record - **Collection:** Oral argument - **Document type:** Oral argument - **Published:** May 7, 2014 ## Text No, just before we begin, Mr. Verhoeven, Mr. Lee, I mean, I don't know if you have any comments on this. It seems to the panel, I think, that as you know, there are major, major portions of these briefs which are marked confidential. No one's tried to remove that designation, so I assume we're all constrained by that today in our argument in terms of what we say and what we cover. Am I right about that? Your Honor, we have attempted to meet and confer with all the parties that are involved with the confidentiality issues, including the third parties, the chip suppliers, and we have reached agreement that we can talk about, at a certain level, where we can talk, if I may, we can talk, we can talk, use the word soft fits. We can talk generally about the functionality, the specific identities of the function calls. These third parties have requested remain confidential in the software. We can use the word symbol, but we have not identified the specific code, and we can talk about soft fits, but they would prefer we not identify how many. I don't know if that helps, Your Honor. Do you have a comment on that? No, I think that's accurate. I think what we've agreed is, and with the third parties is, soft decision, hard decision, can be talked about. Symbol and fits can be discussed without discussing the specific number of fits that might comprise a symbol. That's part of the record. Thank you. Well, speaking only for myself, because we're all hearing this for the first time, I hope you realize that under these circumstances, I feel it is quite difficult. Maybe you understand and you had notice of what we can say and what we can't say and know exactly what it is, but I think you should appreciate that it puts the panel in a very precarious position to one, be hearing about this lifting at the 11th hour, and two, I'm not comfortable that I understand or appreciate exactly what the distinctions are. So you may be able to say what you want to say, but I think you need to recognize that it puts the panel, at least speaking for myself, it puts me, I feel, in a difficult position to making sure that I follow whatever rules of confidentiality you've created. Just leave it at that. Proceed. Thank you, Your Honor. May it please the Court. This bill represents, in our view, a classic situation where a court or tribunal has confused the inclusion of additional functionality in a comprising claim with the issue of non-infringement. In other words, here we have undisputedly comprising claims that identify certain numbers of bits that have to be transmitted a certain number of ways, and the question that the tribunal should have looked at is, for example, are there 60 bits extracted? Whether after there are 60 bits extracted, there's further processing where those 60 bits are combined with soft bits, is additional functionality that is expressly permitted under the court's doctrine of comprising. Well, your friends on the other side, I'm sure, will point out, as they did in the briefs, that this isn't a matter of applying the word comprising and additional functions. This goes to the heart of what is covered by the term bit. Bit is not the real witness testimony which they cite and cross-examination which decides, again, I don't know what I can say or can't say, but describes the parameters of the term bit. So, I mean, their view, which I appreciate at least, is that we're not talking about additional functions here. We're talking about the definitions of the claim term. We agree bit is a 1 or a 0, Your Honor. And what we're saying is, perhaps the best way to explain this is to go to the commission's own brief at pages 8 through 9. It's undisputed in this case that node B sends, via analog, sends 60 units of information, bits, whatever you want to call it, a 1 or a 0, 60 of them, over the air. It's undisputed that you can't add any bits while it's going over the air. Now, the ILJ found that the accused product here received that analog signal and processed it and stored it. And if you look at the commission's own brief, well, first let's look, if I could, Your Honor, so let's start with what the ALJ found. This is A314, and I'm quoting, the I and Q values. Now, this gets technical, but those are the values that the accused products take from the over-the-air analog signal, I and Q values. And the ALJ found the I and Q values are represented as symbols that are samples of the ALJ found that. That's undisputed. Now, if you look at the commission's brief at page 8 or 9, the commission candidly discusses what those symbols are. And I'll just quote selectively, but it's on page 8 or 9. You can see there the commission is talking about these very symbols, the I and Q components that are received over the each symbol can thus convey two bits of information, i.e., 00, 01, 10, and 11. And it continues, because of interference in the environment, the signals that an Apple product receives from the cell tower may be distorted. Dr. Min testified, and it's undisputed in the software, that what happens with these chips, these accused chips Now, you're not reading from the brief anymore. No, I'm not, Your Honor. This is Dr. Min at A11037, that these chips, they sample 30 times. That's what the software has them do. Sample 30 times. You get two bits per sample under the commission's own statements for the way that the Q and I sampling works. That's 60 bits. It's undisputed. Going back to the ID's description of the facts, the I and Q values are representative symbols that are samples of the node B signal, which are stored in a buffer. The issue here that was presented for non-infringement is the following claim language. A physical channel D mapper for extracting a 60-bit rate-matched block from a signal received from node B. That is what the commission said there wasn't enough evidence on, but it's undisputed. The undisputed evidence is that you get this analog signal, and there's I and Q signals extracted out, which are two bits for each sample, and you sample it 30 times. Is there a difference between extracting and estimating or determining bits? What you're doing when you're extracting, Your Honor, is you're taking the data that's in that analog signal and capturing it. That's what a person of ordinary skill can understand in this context of this patent. Now, what the Q systems do is once they get those 60 bits of information, the 60 items, whatever you want to call them, they're ones or zeros, it then creates an error correction, additional error correction, to layer on top of what this patent already is, which is dealing with error correction and making sure the absolute grant never gets corrupted. That's the whole purpose of this patent, by the way. What this additional functionality does is it layers on top, belt and suspenders, to try and ensure even further that there's no error in the absolute grant. So, let me, I'm a little, this is obviously very confusing, at least to me, but it seems to me that what you're saying now is a factual assertion about how the system works, and I believe your colleague on the other side is going to have a different view of the facts of how the system works. So, am I right in that this is, you're now veering into a kind of substantial evidence question about whether the commission properly determined whether the accused devices extract rather than determine? There's two issues, and unfortunately the way these opinions are written, both the ID and the final determination, it's very hard to ascertain whether they're construing claims or not. What happened below, your honor, if I could go to commission brief at 22, this is the commission's brief, quote, the ALJ explained that in order to extract a binary digit from a sample of a received signal, a receiver would have to make a final decision about whether it had received a zero or a one. That's their admission of what the ALJ did. Okay, that's legal error. There is no limitation in this patent about making a final decision. To suggest, in fact, if you look at the claims here, what happens is if you look at these bits, and the following steps are steps designed to correct for errors in those bits. The notion that those extracted bits... Well, I guess that's what I'm asking, though, and I don't, again, maybe I'm not understanding your answer the way I should be, but even taking that, the patent talks about extracting and the error correction. If the commission determined, and at least it seems that your opponent's view of the fact that the chips don't extract at all, they just get the information and they estimate something, and stop me if I'm veering into something I shouldn't be talking about. That seems to me that they've made a factual distinction between how these chips operate and how the patent operates. If that's the case, then we have to look at whether there's evidence to support that or not. That's one argument they're going to definitely make. I will state to you on a factual level, they cannot dispute that 60 bits are sent and 60 bits are extracted. What they're going to try and do is say that it goes into a chip and it comes out as soft bits, but that's not looking at what actually happens, and I just read to you from their brief, the undisputed evidence, and from the ID, the actual factual findings in the ID, the extraction occurs at the very level where the sampling of the I's and Q symbols happens, and the storing it in the buffer. And so what the ALJ focused on was, hey, that's not final because the argument they raised is soft bits is needed in order to make a final decision as to whether it's a one or a zero, so therefore it's not a bit. But what they ignore, your honors, is what comes into that chip, which is 60 bits, which is sampled and goes into that chip before a software program manipulates it and adds additional confidence bits. And so as a factual matter, your honor, it really boils down to can you add additional functionality after you've extracted it because it's a comprising claim? The answer should be yes, you can. And I should note, your honors, that there's also an important claim construction here of this very phrase, and that claim construction was that this 60-bit rate match block, we had a dispute because below, Apple said that block, those bits all have to be contiguous. There can't be any bits between them. And as the commission notes in its opinion at A96, the ALJ rejected that argument and said that block can have other bits interspersed between them or around them or whatever. You just have to have the 60 bits. That's a correct construction, but the application of that was flawed because the ALJ thought those bits had to be final bits, as the commission said, or as the commission in the brief said. Now, what happens at the commission level is the commission says nothing about final. The commission concludes, it looks only at the only evidence about soft bits in the whole case, which is the cross-examination of Dr. Min. Their experts submitted nothing, and he said, oh, there's no extraction going on. It's determining. It's not extracting. Well, that is contrary, your honors, to the positions of both And so there's an extraction of more than that. And then they carried on with what I would call a box-drawing exercise, ignoring the actual 60 bits that were extracted and drawing a box around all the soft bits together with the extracted bits and saying it doesn't meet the rest of the claims. But if you focus only on the bits that were And so, go ahead. I just want to, I'm not going to quote it because I don't know if it's confidential or not, but on page 22 and 23 of your brief, you quote Dr. Min's testimony. And it seems to go, 22 and 23 of your blueprint, and it seems to go directly to the point that I was asking about, and sorry, I'm asking this for probably the third time, but is there a difference between What he was saying, your honor, is what's extracted is these 60 bits. You can't extract more than that out of the analog signal. That's what the extracting is. But the chip then takes that data and generates the bit plus additional soft bits. And so he was, but what my friend across the aisle was attempting to do in cross-examination, you'll see if you read it in the entire context, was trying to get him to admit that more than the 60 bits were extracted. And Dr. Min was trying to draw a distinction between the bit that was extracted from the air and the additional bits, the soft bits, which he said were determined. Now, the way the chip works is after you get that data, the next thing out of the chip will be the whole set. And so he was saying at the same time in that sense, he's talking about what comes out of the chip. Can I just ask you, is this testimony confidential? I'm not sure exactly what testimony. I won't read. But I'm having the same struggle that Judge Post identified at first. I don't want to veer into this, but I want to ask you about this because this is important to me. It seems to me that, again, you're arguing a factual question about how this operates. And your view is it extracts a bit and then adds the additional confidence stuff, right? Yes. And your opponent's view is it doesn't do that. It does it all at once and creates a soft bit. I can't, right? I think that's their view. Am I rejoining to that? Well, let me just finish. I think that's their view. And it seems to me that this testimony at least somewhat is substantial evidence supporting their view. Okay. Two points on that, Your Honor. It's physically impossible for their system to create the soft bits if they didn't extract the bit from the system. I'd like you to ask counsel that question. If you have nothing coming in that you're extracting, you can't process it. It's physically impossible. So if the system works, you have to extract the bits. Okay? That has to happen. But can't you get the signal that's coming through the air and take it and it's not going to come in as a pure one or a zero or a negative one or a one. However, it's doing it through the radio wave. It's going to come in as some other version. Clearly, you're measuring that. But can't you just take the measurement and based upon all the other information in the signal, estimate a soft bit rather than extract that particular one or zero and then add confidence rates? No. The extraction is occurring at the I and Q symbol level where the analog symbol is sampled at predetermined places at predetermined times that correspond to the ones and zeros sent over by the node B. That's where the 60 bits get extracted. And you can't even create the soft bits unless you have that information factually. I know I'm running out of time. So if I could just say one more thing on this. One of the reasons you're looking at the evidence part of this case and scratching your head is because of the way the evidence came in in this case. And we had no notice that this was even a non-infringement theory. The initial contentions did not identify this. The first time it was raised was in a summary judgment motion after our expert had submitted his expert report with no notice that they were even going to make this argument. We moved to strike and ALJ – well, their expert then submitted an expert report. We moved to strike. The ALJ granted it. I did the trial, Your Honors. And at the trial, before Dr. Manning even testified, I tried to make sure I understood what the rules of the road were. And I asked the ALJ, I trust that this won't open the door for them to manufacture this non-infringement. I think I'm going to try to cut you off here. These are matters that I know. I know everything you're going to say because I read it in your brief already. Thank you, Your Honors. It's pretty well covered. We're ahead of time. We're two minutes of rebuttal and we'll add four minutes to the other side to keep it even. So you can each add two minutes to your time if that's what you want to do. May it please the Court, Clark Cheney for the International Trade Commission. I think Samsung's counsel said it best that we look at the evidence in the record here and scratch our heads. And it was Samsung's burden to produce an evidentiary record that showed infringement. And Samsung did not meet that burden. And therefore, the commission found that there was no violation of Section 337. I'd like to briefly review what that substantial evidence and what the conflicting evidence created by Samsung was. First, Samsung's expert, Dr. Min, in his expert report called soft bits a binary digit. And that's at A16832. In his direct testimony, he continued this theory, pointing to portions in the source code that counted soft bits as equating to the term bit in the claim. That's at A10772, A10805, and A10810. Then under cross-examination, Samsung's expert, Dr. Min, was forced to admit that soft bits are not bits under the parties' agreed meaning of the claim terms. He makes that admission at A11046, A11053, and A11054. In fact, he goes on to admit that what he previously called a binary digit was, in fact, a multi-bit variable. That was at A11049 and A11053. Then on redirect, he comes up with a new theory, that in fact, it's not the soft bits that are extracted. It's the first bit of the multi-bit variable that's extracted. In cross-examination, he says that there's no such thing as a line of code that shows the first bit in a soft bit being extracted and other bits appended to it. That's A11301. He says that all bits are determined at the same time. That's A11303. We have testimony from Dr. Min that contradicts itself about what a bit is, whether it's extracted or not. Samsung's attorneys continue the contradiction. First, they claim that the extracted bits that read on the claim terms are soft bits. That's at A05253. Could you just address your friend's argument, though, that in order to do all this processing in the first place, you have to sample and get the 60 extracted bits? Certainly. There are two places in the commission's brief that address that argument. First, at pages 47 and 48 of the commission's brief, the commission addresses the argument that, hey, the phone works on the system, so it must practice the claims. There's substantial evidence in the record that a phone can work on the system without practicing the claims. That evidence is provided by Apple's expert and also conceded to by Samsung's expert. Apple's expert testimony can be found at A12278, and Dr. Min's concession on that point can be found at A11031. There's substantial evidence that the system can work without practicing the claim. Now, I'd also like to address what Samsung pointed to in terms of the soft bits. What is it that's actually creating that soft bit? Maybe I didn't understand your answer, but that didn't answer the question I thought I was asking, which is your friend talked about the sampling and doing, I think, two samples of 30 times and sampling the signal, so you sample at 60 places. Why isn't that extracted? Actually, Samsung never relied on sampling to prove the extracting limitation. What Samsung relied upon was the output of a component that I'm not sure I can name, but I can direct your honors to the place in the brief where we discuss it. It's at pages 43 and 44 of the commission's brief. This is the part of the circuit that outputs what Samsung was relying upon and equating to a soft bit. If you'll look at the top of page 44, you'll see some references in the record. Dr. Min admitted that these soft bits that they're relying upon were not extracted from the received signal, but in fact are determined within the accused devices. There's also a passage in one of the pretrial submissions from Samsung's attorneys that makes a similar point. That's at A14032. Samsung's attorneys say that soft bits can only be generated by first extracting the transmitted and received bits. There, Samsung is saying that soft bits are not the extracted bits. They're generated later. Samsung is pointing to something that's not extracted from the received signal to prove its infringement case. I hope that's more responsive to your question. If I may, I just would add that the commission certainly agrees with the court's view of this case that this case does not turn on whether the claims allow additional functionality. Both the ALJ and the commission expressly said that the comprising language would allow for additional steps, but what the commission found was that there was no proof that those steps actually occur. Dr. Min testified that all of the bits in a soft bit are determined at the same time. You started by saying that the commission agrees with the court's view. What do you mean the court's view? I'm sorry. During my colleague's presentation, I believe you expressed your view that the case… That was just a question. I didn't mean to characterize it more strongly than was appropriate. Thank you. This case does turn on substantial evidence, and I hope that I've walked through sufficient evidence for the court to find support for the commission's determination. This court in the expansion case, another ITC case cited in our standard of review section, states that even if some of the record evidence would contradict a different conclusion than the one the commission came to, this court doesn't reweigh the evidence to come to its own conclusion. That's essentially what Samsung is asking the court to do here. The commission's decision on this infringement point turned heavily on witness credibility. That was noted by the ALJ at A00315 and by the commission at A00094-96. I'd like to turn briefly, if I may, to the three procedural errors alleged by Samsung in its brief. All of them are reviewed for an abuse of discretion standard, and under the Winbon case, this court is very deferential to the commission's decisions in managing its caseload. Samsung argues that it should have been allowed to file a rebuttal expert witness report. Yeah, but this seems quite an unusual process that went down, right? Because it was after discovery had closed and after the expert reports were all submitted, that's when the new theory of non-infringement was presented, right? Apple was obligated to provide its non-infringement theory by January 20th, 2012, and it provided it for the first time on February 29th. So it was after it had submitted its non-infringement contentions, but it was also after the first time that all the parties had had an opportunity to review the source code. And the ALJ did provide a substantial remedy for Apple's late non-infringement contention. The ALJ struck scores of paragraphs from Apple's expert report and refused to let Apple's expert testify about that non-infringement theory. That substantial relief, which I'm sure my colleague will agree was prejudicial to his side of the case, is not even mentioned in Samsung's brief. The ALJ provided a substantial remedy here. Now, Samsung seems to believe that it was allowed to put Dr. Min on the sand and say a multi-bit variable is a single binary digit and not have that question. But under both the Administrative Procedure Act, the Commission's rules, and Supreme Court precedent, cross-examination of a witness is a matter of right. So once Dr. Min gets up there and says the soft bit is a bit, Apple is entitled to test the truthfulness of that statement. And in fact, on cross-examination, Dr. Min admitted that it was not a single digit, it was a multi-digit variable, and it did not equate to the term bit as the parties had agreed in the patent in their claim construction stage of the proceedings. I'd like to leave my colleague plenty of time to address whatever his points are, unless the Court has further questions for the Commission. Thank you, Your Honor. Yes. May it please the Court, my name is Bill Lee, and together with Rochelle Neal, we represent Apple, and we represented Apple below. Let me address two points. One is the substantial evidence issue, which I do think is the core of the issue. And I think one of the most important facts is that what's before the Court now is the fourth theory of infringement, and a theory that was not litigated before the ALJ or the Commission. And I think it will answer, Judge Hughes, your question about determining and extracting. If we start with the first theory, and I'm quoting from Samsung's pre-hearing brief. The quote was, when viewed in isolation, each individual zero or one in a single soft bit, the soft bits we're talking about, is meaningless. That's A3648. So coming into the hearing, the theory being offered was that a soft bit equals a bit under the agreed-upon claim interpretation. And, Your Honor, by the filing of the pre-hearing briefs, these theories were out there. There was an earlier non-imprisonment report, but then much of the source code was in the hands of third parties. Both parties got it very late in the discovery period. Both parties supplemented. That's why the timing worked out the way it did. But what happened is this. On cross-examination, he admitted that a soft bit is not a bit, as defined by the Court. And that's at A1057. So theory one was set aside. And theory two, which Judge Hughes, you asked about, was, well, you're extracting one bit, and then a number of confidence bits after that point in time. And so you're really extracting the 60 bits. But then again, on cross-examination, Dr. Min admitted that wasn't true, that they were determined simultaneously. And they're determined because you're determining these ratios, not zeros and ones. But the soft decision coding process is more complicated. It requires more computing power. It requires more computations. It's something you can do in the modern day because processes are more powerful, and you can get a better result. On cross-examination, again, he admits that, no, they're all determined together. Then on redirect, there's an attempt to offer a new doctrine of equivalence theory. It's excluded, as never having been offered before. But equally important, when Dr. Min testified on direct for nine hours, my very first question to him on cross was, you have no equivalence opinion, correct? Yes. So I'm not going to ask you about that. And he says, yes. That was opinion number three, and that's where things stood when it went to the Commission. And the Commission decided it as a substantial evidence question. In fact, while the briefing here is focused on extracting, it actually is the claim term bit and rate match block, which are the focus of the Commission's decision, neither of which are contested here. This question of what extracting means is a little bit of a red herring. In some sense, it doesn't matter if it is extracting in the plain meaning, extracting by the definition that they now propose, determining, because at the end of the day, it's bit and rate match block which makes the determination. And now on appeal, we have what I call theory number four, which is sampling. And it's the argument that's been made to you before, but Dr. Min testified to directly the opposite. And I know that we apologize on the confidentiality. We learned of the rules of the road only this morning ourselves. But at A11041 to A11042, you will see the examination of Dr. Min where he's asked about these I's and Q's and what happened. And there's nothing here about one bit or two bits or 60 bits. And in fact, you find at A11042 that the sampling is of this collection of bits without giving the precise number. The transcript does give the precise number. And it says it is this collection of bits that are sampled that are then used to determine this ratio. And it's the ratio that makes this more robust and powerful. So in some sense, it doesn't matter which of the three or four theories that we focus upon. Dr. Min contradicts each one of the four on cross-examination. All of them depend upon claim interpretation's bit and rate match block, which are not in dispute. And all of them are inconsistent with the evidence. Now, there's one important point I think that I would add to what the Commissioner has said, and it's this. It wasn't just Dr. Min's cross-examination. Intel witnesses testified. Qualcomm witnesses testified. The Intel source code was in evidence. The Qualcomm source code was in evidence. And while portions of Dr. Stark's opinions were stricken, not all of his opinions were stricken, and he was allowed to say that he disagreed with Dr. Min, and he, in fact, did. The second point, I think, goes to this question of extracting versus determining. And let me say these two things. The first is this was not a new theory, that there was a difference between the extracting described in this specification and the plain meaning, and the determination of these complicated ratios. And, in fact, in A4458, you will see in our pre-hearing brief, we made this distinction between the two. But the more important point is probably this. Could you just answer me? How could you actually determine a soft bit unless you extract something from the signal? I know that seems very simplistic, but... Well, Your Honor, I think it actually goes back to what Dr. Min actually says accurately, which is that you are determining a ratio based upon a collection of bits. That's the key. And that's why, actually, we agree with what they said at the outset. No one bit by itself has a meaning. You need the X number of bits in order to determine the ratio. You are determining 60 ratios, but those ratios are not bits because the bits are 0s or 1s. That's not dependent upon how you determine whether extracting is determining. If the claim term were extracting, if it were determining, if it were there, I think it's at note 3, page 41 of their brief, it doesn't matter because a bit's a bit. It's a 0 or 1. These aren't bits. These are ratios. That's the power of soft decision computing. And they're not 60-bit rate match plots. They are something different. And when you focus on the number of bits, as the court knows, its magnitude is greater. And the reason its magnitude is greater is that's where you get the power of the computing. Now, the last point I'll make is this. On the discretionary rulings in terms of how the evidence came in, those are rulings committed to the discretion of the ALJ. There actually was not anything that was, once you recognize the source code came in late, and that was the reason for some of the procedural maneuvering by both parties. And both parties had portions of expert reports stricken that were a little late. What happened was an expert gets on the stand. He gives an opinion. There's cross-examination on his opinion. He testifies inconsistently with that opinion. He offers a new opinion on recross. He testifies inconsistently with that. That would be enough to be substantial evidence. But, again, there's the intel witnesses. And, in fact, we asked them. We showed Dr. Min the intel witness testimony about whether they extract. And the witness said no. And he said, do you have a reason to disagree? He said no. So there's an intel testimony, the Qualcomm testimony, the source code that was in play, plus Dr. Stark. All of that is substantial evidence, and it supports the decision of the commission. Thank you. Dr. Min did testify on direct examination that the – and this is at A11037 through A11038. Here's an excerpt of a question. What happens at line 152 is the Qualcomm check actually counts 30 symbol pairs, correct? Answer, correct. And it counts 60 different symbols, correct? That's how the Qualcomm terminology works. And this is just one example. It's in the record. Dr. Min put it in the record. That's where the extraction occurs. You've heard from the other side about his cross-examination. But his cross-examination was about what comes out of the integrated circuit chip. And that is after the extraction is done and after further processing has been done. And that's when you get the soft bits. That's after the extraction. The extraction occurs before the output. And all of this back and forth that they're saying about inconsistencies is about what the soft bit is. There was no cross-examination about the I and Q symbols and whether those are sampled 30 times and whether or not there's 60 bits that gets put in a buffer. The ALJ in his ID found, and I gave you the site earlier, that that's what happens in the accused products. They sample the I and Q values, which are two bits each, and they put them in a buffer. And it's undisputed they do it 30 times. That's 60 bits that's extracted. By pointing to the output of the chip and pointing to alleged inconsistencies and how the word soft bit, which doesn't appear in the claims, is characterized, that does not raise a substantial issue as to whether or not these I and Q symbols are sampled and put into a buffer before the output of the chip. That's the position Dr. Min has had. Can you point to me in your blue brief where you make this argument? I don't know off the top of my head, Your Honor, but this goes together with all of these arguments. What we were pointing out, Your Honor, is that when the soft bit is output, and this is what we focused on because it's the focus of their attack, when the soft bit is output from that chip, the extracted bit is represented by the first of the group. And that's how you trace it through the rest. And what we were pointing out to you, Your Honors, is that if you trace the correct bit and we know from the claim construction that rate match block is not limited to 60 bits. It is 60 bits, but it doesn't limit it to only 60 bits. It's got bits all around it. So you trace those specific 60 bits and you follow them through the claim language and ask are they manipulated in the way that the claims say you're going to find they are. And it's only by them drawing a box around all of the soft bits together with the extracted bit that's being processed that they get that. And it's your honor. Time is up. Okay. Thank you. We thank all counsel. The case is submitted. --- Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/argument%3A8721. Public record. Not legal advice.