[00:00:00] Speaker 00:
For argument 25-1063, once again, Amstead Rail v. Squires.

[00:00:13] Speaker 00:
So there are a few less issues on this one.

[00:00:15] Speaker 03:
Yes, that's correct, Your Honor. So with this appeal, I'm going to start and concentrate on the Board's decision with regard to claims 25 and 28.

[00:00:31] Speaker 03:
These claims require, respectively, software filters and digital filters.

[00:00:37] Speaker 03:
Amstead argued below that the only reference that actually discloses filtering only discloses analog filtering in connection with the preconditioning of signals for analog digital conversion. The board agreed with that characterization, yet the board found these claims obvious and by adopting a theory that the petitioner never advanced and was never argued by the parties, which is that the Barone's analog filtering, again, in relation to preconditioning of an analog signal for conversion to a digital signal, could be implemented in Armitage's firmware as a digital filter.

[00:01:20] Speaker 03:
So Amstead had no notice of this particular argument, had no opportunity to develop the record as to why a person with an ordinary skill would not use the disclosure of analog filtering in relation to preconditioning implement it in a digital environment, in the digital domain already. There's no reason to do that because the whole point of the disclosure in Barone is to prepare an analog signal for digital conversion.

[00:01:50] Speaker 02:
My understanding of the board's decision and the board's reading of the petition is that the petitioner was relying on Barone for the kind of broader conceptual teaching that it's good to filter out unwanted data.

[00:02:07] Speaker 02:
There's an advantage to parsing that out and then focusing on the desired data. And then when you do that in Armitage's sensing process, system with Armitage's firmware, well, you would do that kind of filtering in a software form, and that Mr. Ricca, Dr. Ricca, the expert said, all these different forms of filtering, mechanical, electronic, software, are all super well known and have been super well known for decades.

[00:02:52] Speaker 02:
So therefore, it's just once you decide you want to do filtering in Armitage, you just plug and play in Armitage's environment, which would lead you to a digital software filter.

[00:03:09] Speaker 03:
I guess the point is that you would not be looking to Barone to do that.

[00:03:14] Speaker 03:
Barone doesn't discuss digital filtering at all, filtering in the digital domain at all, and the only filtering that's disclosed, at least decided, is, again, signal preconditioning.

[00:03:24] Speaker 02:
My understanding is that The point is that we're not looking at Barone's teaching of an analog filter. We're just picking up Barone as evidence that, yes, it was very, very well known in the art that when you get a bunch of data, it's good to filter some of that data because you can excise the unwanted data. And then you can do that in other environments where you're...

[00:03:57] Speaker 02:
you know, sensing train safety data, and that would be Armitage. And so once you apply that concept to Armitage, where would that lead you? That would funnel you right into Armitage's firmware, which would lead you to a software form of technology.

[00:04:16] Speaker 03:
So...

[00:04:18] Speaker 03:
Again, that really wasn't, certainly wasn't how I read the petition to be arguing. The petitioner continually, and the petitioner's expert, continually stated, Barone discloses a digital filter. They said that in relation to claim one. They said that in relation to claim 25 and 28 was a reference. But in any case, that's the point, is that the argument that was raised in the petition was, is that Verone discloses a digital filter that you could implement in Armitage's firmware.

[00:04:54] Speaker 03:
And it doesn't. There is no disclosure of a digital filter in Armitage, and in fact, one of the pieces of evidence actually states, in relation to signal preconditioning prior to analog-digital conversion, that those types of filters always must be in the analog domain, because, and the reason why is that you end up, if you encode into the digital domain without conducting that filtering, you get a signal that is not, that has a lot of artifacts in it.

[00:05:27] Speaker 03:
That's, you know...

[00:05:29] Speaker 03:
It must be in the analog domain. And so it's not clear to us if there was any explanation for why a person of ordinary skill would look at Barone, the reference that was selected, to arrive at generic filtering in Armitage's firmware, which Armitage doesn't disclose filtering at all.

[00:05:55] Speaker 02:
That's a different issue.

[00:05:56] Speaker 03:
That's a different issue.

[00:05:58] Speaker 03:
But, you know, after Amstead brought this up in the patent owner response, the petitioner doubled down, saying that Barone is not limited to analog and that it teaches the filtering of digital signals.

[00:06:16] Speaker 03:
It actually cited additional evidence that was not considered by the board because it was not part of the petition.

[00:06:23] Speaker 02:
What if we conclude that Armitage... There's substantial evidence that Armitage discloses filtering unwanted data.

[00:06:36] Speaker 03:
Well, I mean, I would challenge that. Amstead's view of what Armitage discloses is more of a sampling technique. At the time of data collection, it only collects data from particular axes of the accelerometer, for example. And that's, as our expert discussed below, that is more akin to a sampling technique as opposed to filtering.

[00:07:03] Speaker 03:
And so that would be our argument for why our research does not disclose filtering.

[00:07:11] Speaker 03:
And again, that's at the point of data collection.

[00:07:21] Speaker 03:
So I'd like to also briefly touch on the implicit construction of one or more filters.

[00:07:31] Speaker 03:
In Amstead's view, the claim tax specification in prosecution history indicate that the filtering, which is recited in the claim, occurs after the accelerations are sensed.

[00:07:51] Speaker 03:
And that's based on reading of the claim, antecedent basis, various disclosures in the specification.

[00:08:01] Speaker 01:
And on that argument, do we have to agree with you that claim 26 is plagued by a drafting error?

[00:08:09] Speaker 03:
I don't think so. I'll just refer to the line from, I don't have the case site in front of me, but if the specification is clear, the claim differentiation can't wag the dog.

[00:08:25] Speaker 01:
Don't mechanical filters have to be within the scope of the independent claim if they're in claim 26 and 26 is not an error?

[00:08:39] Speaker 03:
That would be the application of the claim differentiation doctrine, yes.

[00:08:43] Speaker 01:
So what you're saying is no, because claim differentiation doesn't mean that?

[00:08:49] Speaker 03:
I'm saying that the clear disclosure and the specification would trump that. And we go through the history of the claims and the prosecution, that the specification has been very clear, was very clear, on the distinction between mechanical filtering, and the filtering which is the core of the invention.

[00:09:09] Speaker 01:
So in your view, claim 26 is invalid?

[00:09:15] Speaker 03:
Very likely, yes.

[00:09:17] Speaker 02:
Did you cancel it?

[00:09:19] Speaker 03:
We did not. We did not.

[00:09:44] Speaker 04:
Good morning, Your Honors. It may please the Court. Shallow in for the Director of the USPTO.

[00:09:49] Speaker 04:
I'll begin by addressing the issue of Claims 25 and 28.

[00:09:53] Speaker 04:
The Board's reading of the petition was correct. The Board did find that Barone's signal conditioning is analog, but then it read the petition to propose implementing the filtering of Barone in the firmware that is indisputably found in Armitage.

[00:10:10] Speaker 04:
And you can look at the petition at Appendix... 265 to 268, and then again at Appendix 276 to 277, the Board's reading of the petition is the correct one.

[00:10:24] Speaker 04:
The petition makes clear that it's arguing that Barone discloses both analog and digital filtering. And When the petition discusses the dependent claims with the digital filtering limitations, it refers back to that general idea discussed in reference with claim one, which is that a personal skill in the art would have been motivated to combine filtering, as taught in Barone, with the Armitage reference, to the extent Armitage can be read to not contain a filtering disclosure at all.

[00:10:56] Speaker 04:
So the evidence cited in the petition is the same evidence that appears in the Board's decision at Appendix 37 to 38. Excuse me.

[00:11:10] Speaker 04:
The board relies on the testimony of petitioner's expert and then cites back to teachings in the prior art about that a person of skill in the art would have known that there are times when you can implement filtering in analog and times you can implement filtering in digital domains.

[00:11:31] Speaker 04:
Yes, the board found that Barone's signal conditioning is analog, but the petition was based on the broader concept of combining Barone's filtering.

[00:11:41] Speaker 04:
I'll turn next to, unless the court has questions on this issue, I'll turn next to Claim 26, and to start, you're absolutely correct. There's no universe in which Claim 1 and Claim 26 can both exist. Claim 26 further limits Claim 1 and says wherein the one or more filters aren't. a mechanical filter, and that just disposes of their claim construction argument. To the extent Council was suggesting that the specification controls, well, there we have the preferred embodiment, which teaches mechanical filtering.

[00:12:13] Speaker 04:
So it could not be more clear.

[00:12:15] Speaker 04:
that the independent claims, when they're talking about one or more filters, they're talking about mechanical as well as visual.

[00:12:22] Speaker 00:
Well, it seems like your friend's only answer to that is kind of, and I think he may be characterizing some of our cases correctly, like we not rested on claim construction that frequently, sometimes saying it's kind of like the tail wagging the dog. So do you have a response to that?

[00:12:40] Speaker 04:
I understand your question to mean not resting on claim differentiation. Okay. And that's the correct characterization of the case law from this court. But again, in this case, you have the specification at column 5 and at figure 4. This is at appendix 113 and appendix 100, talking about the preferred embodiment that talks about mechanical filters that are used for filtering accelerations that are considered noise.

[00:13:11] Speaker 04:
So the claim construction argument, whether you look at claim differentiation or just the specification or both, the intrinsic evidence supports that one or more filters configured to filter at least a portion of accelerations is broader than just digital filters or mechanical filters.

[00:13:28] Speaker 02:
What about dependent claim 3?

[00:13:31] Speaker 04:
So that's an interesting one, Your Honor.

[00:13:33] Speaker 02:
I didn't see your green brief mention dependent claim 3.

[00:13:38] Speaker 04:
I can answer that question now, Your Honor.

[00:13:39] Speaker 02:
I think dependent claims... Are you going to walk back with what the Board said about dependent claim 3?

[00:13:45] Speaker 02:
I'm not sure which... The Board seemed to rely on dependent claim 3 along with dependent claims 23 and 26 as evidence for why claim 1 should be understood to include mechanical filters upstream of a sensor. But of course, when you look at claim 3, it it calls for additional filtering to be added to Claim 1, and that additional filtering is, in fact, mechanical filters upstream.

[00:14:16] Speaker 04:
Correct, Your Honor. And I think it's a fair reading of Claim 3 that it's... If we're going to look at the dependent claims, claim 26 maps on more clearly. The argument for claim constriction, claim differentiation, I'm sorry, is stronger with claim 26. With claim three, and I'll pull it up so that I'm not just speaking from memory.

[00:14:38] Speaker 04:
With Claim 3, it's limiting Claim 1 further, and it's talking about a wireless sensor node and then saying that there's a mechanical filter in the wireless sensor node. Claim 1 isn't talking about a wireless sensor node in that way. So I think if you were to look at the support from the dependent claims, Claim 26 and Claim 23 are stronger evidence of... what the one or more filters term means in Claim 1 than Claim 3.

[00:15:10] Speaker 04:
But even with Claim 3, it's sort of introducing further limitations before it's discussing filtering or mechanical filtering. So that would be my response to that.

[00:15:21] Speaker 04:
And then just briefly, Your Honor, as to if you agree, you were asking the Cousin Council if you agree that Armitage disposes filtering. The petition ground was Armitage in combination with Barone. This was sort of a belts and suspender approach from Petitioner. They talk about Armitage disclosing filtering and then saying to the extent Armitage is not disclosed filtering, you have Barone, and there's a motivation to combine the two.

[00:15:52] Speaker 04:
Okay, unless the Court has further questions, I yield my time. Thank you, Your Honor.

[00:16:05] Speaker 03:
I will just very briefly address the Claim 3 issue and the issue of the preferred embodiment that my colleague just mentioned. So one thing to note is that Claim 3 was originally filed in the PCT application as Claim 49. Claim 26 was added much later. So when we say that the original description supports our reading, one of the things we're referring to is Claim 3. That is part of the original specification as it was filed.

[00:16:38] Speaker 03:
Additionally, it's just simply not true that the preferred embodiment must include a mechanical filter. The wireless sensor node in Figure 5 is expressly disclosed as not having a mechanical filter, and that embodiment is no less preferred.

[00:16:55] Speaker 03:
And a logical reading of the claims is that What is covered by Claim 1 would be the WSN of Figure 5, and what would be covered by Claim 3 would be the WSN of Claim 4, which expressly states that it has an optional mechanical filter in it. So the idea that mechanical filter must be included for being an impaired embodiment just simply isn't supported by the specification.

[00:17:25] Speaker 00:
Thank you. We thank both sides. The case is dismissed.