Geospatial Technology Associates, LLC v. United States ( 2023 )

No. 16-346C
(Originally filed: February 3, 2023)
(Reissued: February 21, 2023)1
*********************** Patent infringement;
Obviousness;
GEOSPATIAL TECHNOLOGY Motivation to combine;
ASSOCIATES, LLC, Indefiniteness; Improper
combination of method
Plaintiff, and apparatus claims;
Means-plus-function
v. structure
THE UNITED STATES,
Defendant.
***********************
Richard T. Matthews, Raleigh, NC, for plaintiff.
Jenna Munnelly, Trial Attorney, United States Department of Justice,
Washington, DC, with whom were Scott Stewart, Deputy Assistant Attorney
General, Gary L. Hausken, Director, for defendant.2
OPINION
BRUGGINK, Judge.
This is a patent and copyright case brought against the government for
infringement of plaintiff’s Patent No. 8,897,489 (the “‘489 patent”) and the
associated software protected by copyright. Plaintiff alleges that defendant
1
This opinion was originally issued under seal to afford the parties an
opportunity to propose redactions of protected information. They agreed that
none were necessary. The opinion thus appears in full.
2
Mr. Stewart was listed on the briefs at the time the motions were initially
filed and subsequently briefed. Brian M. Boyton is now the Principal Deputy
Assistant Attorney General at the Department of Justice on the government’s
briefs filed in this case.
1
improperly possessed and used the ‘489 invention in violation of its patent.
Defendant has responded, inter alia, by arguing that it did not infringe
because the ‘489 patent is invalid. Pending are the parties’ renewed motions
for summary judgment on patent validity. As explained below, plaintiff’s
motion is granted in part, and defendant’s is denied.
BACKGROUND
The ‘489 patent teaches the creation and use of a solid-object
detection and identification apparatus. The technology “eliminates the need
for an analyst to visually inspect all data and instead enables the analyst to
quickly focus on an accurate and ranked list of target detections” within each
image. ‘489 patent at 4. The invention functions by using a computer to
accept a series of images from its user, and then uses a statistical detection
algorithm to “filter each pixel in each of one or more images” to generate “a
statistical detection score” for each pixel in the image. Id. The apparatus then
examines the regions surrounding high-scoring pixels and uses an algorithm
to unmix the pixels, collate the scores, and determine an “object based score
for each [image’s] identified regions.” Id. at 4 (language added for clarity).
Once regions with high target scores are identified, the ‘489 invention cross-
references the scores with geographical data, further increasing the accuracy
of the target detection process. The apparatus then feeds a report on the
geographical location of the identified objects to the user who verifies if
identified pixels and regions hold objects of interest to the analysist. Id.
Five steps fully encompass this process (as described in the patent’s
second figure): 1) obtaining images; 2) applying a statistical filter to generate
a per pixel detection score; 3) “apply[ing] spatial process to per pixel . . .
scores to identify regions” and to determine the score “in each detection
plane;” 4) determining “corresponding geographical information for each
identified region;” 5) and providing the selected regions and object scores
“with corresponding geographical information.” Id. at 3. Steps two through
four can be repeated iteratively on each image to provide greater accuracy in
the final report issued to the user analyst.
The ‘489 invention was developed between 2008 and 2011 by Dr.
William Basener and was filed in a patent on January 28, 2011. Portions of
the invention’s software were subsequently copyrighted in July of 2017
under Registration No. TX 8-420-604. The invention’s intellectual property
rights were originally assigned by Dr. Basener to the Rochester Institute of
Technology (“RIT”), but were later reassigned on November 12, 2015 to
Geospatial Technology Associates, LLC. (“GTA”, the plaintiff here.
2
Subsequently, the government licensed the use of the ‘489 invention
for a short image processing contract. Under this agreement, plaintiff used
the ‘489 invention to process and analyze images furnished by the
government. Because the analysis process used an iterative procedure that
used previous image results to accurately identify future images, at least one
copy of plaintiff’s invention remained on government computers after the
end of the image processing contract. Plaintiff alleges that the government’s
possession of this ‘498 invention copy and use of other detection programs
beyond the termination of the license, have infringed the ‘489 patent. In total,
the nine programs causing alleged infringement are: Full Spectrum Tool Kit
(“FSTK”), FTSK with Probabilistic Identification of Solid Materials
(“PRISM”), GeoReplay, GeoReplay with Prism, GeoReplay Full Spectrum
Exploitation (“GeoReplay-FX”), Lobo, HyperSEAL, GEOMATE, and
Object-Based Identification, Sorting, and Ranking (OBISR) algorithms. As
a result, plaintiff filed suit against the government on March 16, 2016, to seek
damages for the unlicensed use of plaintiff’s intellectual property.
On July 24, 2020, defendant filed a brief presenting three bases on
which it believed that the ‘489 patent was invalid: 35 U.S.C. §102 lack of
novelty, 35 U.S.C. §103 obviousness, and 35 U.S.C. §112 indefiniteness.3
Plaintiff responded to this brief with a motion for summary judgment
asserting the ‘489 patent’s validity; defendant replied with an opposing
cross-motion regarding invalidity. Both motions were fully briefed, and on
April 8, 2021, Judge Griggsby, previously assigned to this case, ruled against
plaintiff and held that all claims in the ‘489 patent were non-novel and invalid
under 35 U.S.C. §102 because they were fully anticipated by the Civil Air
Patrol’s Hyperspectral Sensor System (“Archer”). Geospatial Tech. Assocs.
v. United States, No. 16-346C, 2021 U.S. Claims LEXIS 1085 (Fed. Cl. Apr.
8, 2021). Because this decision invalidated the entire patent, Judge Griggsby
did not decide whether the ‘489 patent was obvious or indefinite. Plaintiff
then sought reconsideration of the holding of obviousness.
The case was thereafter transferred to the undersigned. On
reconsideration, plaintiff argued (amongst other unsuccessful arguments)
that the court previously had erred because, “for a court to find a feature of a
patent obvious or anticipated by some prior art, without that feature being
explicitly taught, that feature must be ‘necessarily present in the thing
recognized by persons of ordinary skill.’” Geospatial Tech. Assocs., LLC v.
3
The court invited defendant to present its views on the patent’s validity in
a brief. See ECF No. 196 (Order of July 14, 2020). Dispositive motions
were to be scheduled later, but plaintiff responded to the government’s brief
with a motion, and defendant followed course.
3
United States, 158 Fed. Cl. 113, 119 (2021) (quoting In re Robertson. 169
F.3d 743, 745 (Fed. Cir. 1999)). Plaintiff used this precedent to argue that
because the “unmixing” capabilities of the ‘489 patent were not found in the
ARCHER prior art, the ‘489 patent could not have been fully anticipated by
ARCHER. We agreed and, on November 29, 2021, held that “it is
insufficient that the prior art might be capable of performing a function of a
patented invention [unmixing] if only it were modified, tweaked, or . . . used
in a machine. For a feature to be inherent in the prior art, however, that
feature must be necessarily present in the steps, functions, or elements of the
prior art.” Id. at 6, language added for clarity.
Because the ARCHER system did not have “unmixing” as a step,
function, or element, the undersigned granted the motion in part and declared
that the ARCHER system did not anticipate claims 5, 14, 23, 29, 31, and 33
of the ‘489 patent. Id. Although this ruling had the effect of reviving the ‘489
patent, it did not adjudicate the obviousness and indefiniteness arguments for
patent invalidity. Because those issues are unresolved and have been fully
briefed, we will consider the validity issues.
Discussion
In its brief on patent invalidity, defendant argued that the ‘489 patent
was obvious under §103 because of existing prior art that taught target
detection and identification from images. Defendant also argued that the ‘489
patent was indefinite under §112 because the patent confused method and
apparatus claims and failed to disclose the “specific structure” of the patent’s
algorithm. ECF No. 201 at 13, 23-25. Plaintiff opposed these arguments and
sought summary judgment to declare that the ‘489 patent was both non-
obvious and definite.
I. Obviousness
The aim of the U.S. patent system is to encourage innovation and
promote the spread of new technology throughout the public by creating a
disclosure that will “give the public the benefit of an invention after the
patent shall expire.” Beidler v. United States 253 U.S. 447, 453 (1920). In
line with this goal, it is Congress’s desire only to allow patents for true
innovations, and not for the use of previously known elements in an obvious
combination. See generally KSR Intern. Co. v. Teleflex Inc., 550 U.S. 398,
416 (2007).
To achieve this purpose, Congress included §103 as a qualification for
earning a valid U.S. Patent: “A patent for a claimed invention may not be
4
obtained . . . if the differences between the claimed invention and the prior
art are . . . obvious . . . to a person having ordinary skill in the art.” This
means that, when previous inventions (prior art) can be combined in obvious
ways to result fully in the now-claimed invention, any patents granted for
that invention are invalid and not able to serve as the basis for a patent
infringement lawsuit. In this case, defendant argues that the ‘489 patent runs
afoul of §103 because the claimed invention is:
(1) a combination of prior art elements according to known methods
to yield predictable results, (2) a simple substitution of one known
element for another to yield predictable results, (3) an application of
known techniques to known devices ready for improvement to yield
predictable results, (4) and/or obvious to try.
ECF No. 201 at 14-15.
In support of these assertions, defendant commissioned Alan Stocker,
an engineering consultant, as an expert to review the ‘489 patent and write a
report (the “Stocker Report”) evaluating the ‘489 patent’s validity. In it, Mr.
Stocker opines that two primary pieces of prior art existed before the ‘489
patent’s filing date that together disclose all of the claims of the ‘489 patent,
rendering it obvious under §103. The first of the prior art references
identified by Mr. Stocker is a type of “automated algorithm” used “for
detecting gas plumes and identifying their chemical constitutes via a detailed
analysis of aggregated pixel spectra.” ECF No. 201-1, at 12-13. The second
piece of prior art is an image identification algorithm named “GeoID”
developed by the Air Force national Air and Space Intelligence Center to
“process spectral imagery from . . . sensors as well as several other
instruments.” Id. Taking these two together, Mr. Stocker avers that all of the
features and limitations of the ‘489 patent are found. Assuming that to be
true, which plaintiff disputes, the only remaining question is whether a
person of ordinary skill in the art (POSITA) would have been motived to
combine the two algorithms at the time of the invention at issue.
Defendant points to a particular line in Mr. Stocker’s report as
indicative of just this motivation in the field at the time: “there was an
‘emphasis’ in the H.S.I. [(Hyper-spectral imaging)] community towards
utilizing known methods and algorithms of target identification used for gas
problems to address solid-matter problems.” ECF NO. 201 at 15. Thus, any
POSITA would have thought it simple and obvious to “combine target
detection [(gas plume detection)] and target identification [(the GeoID
algorithm)] to reduce false alarm rate[s]” and thereby develop the ‘489
5
invention. Id. If so, the ‘489 patent’s invention is obvious under § 103 and
ineligible for intellectual property protection.
Plaintiff disagrees on all counts, emphasizing that an invention which
combines known elements is novel if there is no known motivation to
combine them at the time of the invention. Plaintiff points out that “the fact
[that] a POSITA might or even would be able to combine certain references
does not mean they would have had a reason or motivation to do so.” ECF
No. 204 at 19 (citing TQ Delta, LLC v. Cisco Sys., 942 F.3d 1352 (Fed. Cir.
2019); InTouch Techs., Inc. v. VGo Communs., Inc., 751 F.3d 1327 (Fed. Cir.
2014)) (emphasis omitted). Plaintiff finds the line in Mr. Stocker’s report
regarding the use of gas plume identification algorithms for solid matter
problems lacking the crucial details of how that would have motivated a
skilled artisan to combine the two pieces of cited prior art in a manner that
teaches the claim limitations of the ‘489 patent. Plaintiff also doubts the
veracity of Mr. Stocker’s statement, presenting 2011 emails from Mr.
Stocker in which he expressed doubt regarding combining these two types of
algorithms to meet the need for solid matter identification. ECF No. 204-6.
Plaintiff further cites to its own validity expert, Dr. Jones, who opined that
the prior art teaches away from the combination of methods found in the ‘489
patent. See ECF No. 202-2 (Jones Report) at ¶¶ 170-71, 167-68, 423-26. At
best, per plaintiff, Mr. Stocker’s statements raise a disputed question of fact.
Plaintiff also raises the more basic point that defendant has not
presented any evidence, and very little argument, as to how the prior art
compares to the patent claims, i.e., a claim-by-claim analysis of whether the
pre-existing algorithms disclosed the limitations contained in the asserted
patent. It finds any such analysis lacking from Mr. Stocker and asserts that
the claim chart appended to the government’s motion is grossly inadequate,
or, as counsel described it, “a ‘choose-your-own adventure’ endeavor.” ECF
No. 204 at 22.
Defendant replies that it has presented undisputed evidence on both
points. On the question of motivation, it admits that the precise term was not
used by Mr. Stocker but argues that his statements at deposition and in his
report are sufficient to show the motivation present at the time of the
invention. The government quotes the following from Mr. Stocker’s
deposition testimony: “[the] hyperspectral community emphasis shifted
towards solid-material targets with reflective-band spectral signatures,
analytical methods and algorithms previously optimized for the gas problem
in the thermal infrared region were tailored and extended to apply to the
solids realm using reflective-band imagery.” ECF No. 201, Ex. A ¶¶ 37-38.
This “shift” or “emphasis,” according to the government, demonstrates that,
6
as of the patent’s priority date, other researchers in the field were working on
developing methods to use gas signatures for solid object identification. If
those in the field were already trying to use known identification methods for
solid objects, then the motivation was already present and further motive
need not be proved, argues defendant. That may ultimately be proven, but
the facts are in dispute.
A showing of motivation to combine is critical to the obviousness
analysis because “inventions in most, if not all, instances rely on building
blocks long since uncovered, and claimed discoveries almost of necessity
will be combination of what, in some sense, is already known.” KSR Int'l
Co. v. Teleflex Inc., 550 U.S. 398, 418-19 (2007). Patents are not often issued
in truly novel areas of endeavor. Science and industry are largely advancing
incrementally, building upon what came before. Thus, courts must consider
carefully whether a motivation to combine previously known in elements in
the way claimed by the patent in suite was present in the field of art before
declaring a patent invalid. Because “obviousness determinations cannot rely
on hindsight,” courts must carefully consider motivation to combine
arguments to prevent ex post reasoning. Belcher Pharm., LLC v. Hospira,
Inc., 450 F. Supp. 3d 512, 538 (D. Del. 2020) (citing KSR, 550 U.S. at 421).
The government’s Stocker Report states that, at the time of the ‘489
patent’s filing, other non-party researchers were moving towards using
existing gas analysis programs for solid object identification. Plaintiff
presents evidence that casts at least some doubt on this asserted inertia in the
form of Mr. Stocker’s own prior statements and Dr. Jones’ opinion that prior
art taught away from combining in the manner that the invention did, which
casts doubt on a finding of the necessary motivation to combine.
“Obviousness is a question of law based on underlying findings of fact,
which include the motivation to combine multiple prior art references and
any objective indicia of non-obviousness.” S. Ala. Med. Sci. Found. v.
Gnosis S.p.A., 808 F.3d 823, 826 (Fed. Cir. 2015). On the present record,
motivation is disputed, and very little evidence or argument regarding the
secondary indicia has been presented. For those reasons alone, summary
judgment is inappropriate. We note, further, however, that, as plaintiff points
out, no claim-by-claim analysis was presented beyond the government’s
chart. No citations to or argument from it were made in defendant’s briefing,
however. The issue of whether the cited prior art meets the limitations of the
‘489 claims is also preserved for trial as the record at this point is immature
on the issue.
7
II. Indefiniteness
Defendant also challenges the patent’s validity on definiteness
grounds. Because the ultimate purpose of a patent is to encourage both
innovation and disclosure of new technology to the public—a delicate
balance to be sure—the patent must adequately give notice of what it claims.
See Nautilus, Inc. v. Biosig Instruments, Inc., 572 U.S. 898, 909-10 (2014)
(clarifying that the indefiniteness requirement is aimed at ensuring adequate
notice to the public of what is claimed and what is not); see generally Pfaff
v. Wells Elecs., Inc., 525 U.S. 55, 63-64 (1998) (discussing the “carefully
crafted bargain” of the patent system that encourages innovations and
disclosure of technology). This notice requirement is found in 35 U.S.C.
§112, which states that a patent “shall contain a written description of the
invention, and of the manner and process of making and using [the
invention], in such full, clear, concise, and exact terms as to enable a
[POSITA] to make and use the [invention].” Should an invention fail to meet
this standard of disclosure, the patent is indefinite and invalid.
Defendant presents two reasons why it believes the ‘489 patent does
not adequately disclose the claimed invention to the public. First, the
government argues that the ‘489 patent claims disclose both a method and an
apparatus. Claims “reciting both an apparatus and a method of using that
apparatus render a patent claim indefinite.” I.P.X.L. Holdings, L.L.C. v.
Amazon Inc, 430 F.3d 1377, 1384 (Fed. Cir. 2005). The second argument is
that the ‘489 patent is indefinite because the specifications fails to adequately
disclose the algorithm claimed by the ‘489 invention. “[I]f there is no
structure in the specification corresponding [to the invention’s details] . . .
the claim will be found as indefinite.”4 Biomedino, L.L.C. v. Waters
Technologies Corp., 490 F.3d 946, 950 (Fed. Cir. 2007), (clarification
added). Defendant argues that “when general purpose computers are
disclosed, the specification must disclose a specific algorithm to transform
the general-purpose microprocessor to a special purpose computer
programed to perform the disclosed algorithm.” ECF No. 201 at 22.
A. Method and Apparatus
4
Defendant also argues that claims 5 and 29 are indefinite due to a failure to
“provide reasonable certainty of the scope of the invention.” ECF No. 201
at 21-22. However, this question is encompassed by the broader means-plus-
function argument addressed herein.
8
Claims 5 and 29 of the ‘489 patent describe methods of using a
computer containing the ‘489 algorithm for “identification, sorting, and
ranking detections of one or more targets.” ‘489 Patent at 7, 9. Conversely,
claims 14, 23, 31, and 33 describe the actual physical set-up of a computer
programed to perform the ‘489 algorithm, including “a non-transitory
computer medium having stored thereon instructions for identification,
sorting and ranking detection of one or more targets . . . .” Id. at 8-10.
The government contends these claims are indefinite by noting that
the ‘489 patent claims mention an apparatus (alternatively referred to as a
“medium”), but then detail a series of steps (“unmixing the target detection
processing apparatus,” or “comparing [the image] with the target detection
processing apparatus”) that must be performed for the invention to function.
See claims 5, 14, 23, 29, 31, 33 of U.S. Patent No. 8,897,489 B2 (issued Nov.
25, 2014). The government argues the inclusion of these steps comprises a
process patent claim which, when combined with the apparatus in the claim’s
preamble, makes the patent so indefinite that a skilled artisan could not
reproduce the invention.
Plaintiff makes two responses to this method-apparatus combination
allegation. First, it argues that two of the claims (claim 5 and claim 29) are
method claims that avoid the issue of combination indefiniteness because
they claim only methods and not apparatuses. Plaintiff argues that these two
claims are not methods that claim an apparatus, but rather methods that
describe the functioning of the apparatus as a limitation of the method itself.
Plaintiff argues that this distinction prevents claims 5 and 29 from being
indefinite. Id. We find this line of reasoning persuasive and note that it is
consistent with existing precedent that “a method claim may include an
apparatus element that invokes 35 U.S.C. § 112.” Spherix Inc. v. Vtech
Telcoms. Ltd, 2015 U.S. Dist. LEXIS 171949 (N.D. Tex. Mar. 19,
2015), citing the Federal Circuit Court of Appeal’s reasoning in both On
Demand Mach. Corp. v. Ingram Indus., Inc., 442 F.3d 1331, 1336, 1340-41
(Fed. Cir. 2006) and J & M Corp. v. Harley-Davidson, Inc., 269 F.3d 1360,
1364 & n. 1, 1367 (Fed. Cir. 2001).
Both claims 5 and 29 describe the process of the ‘489 invention’s
operation, and by their very nature must mention aspects of the apparatus’s
components. This does not mean that claims 5 and 29 are claiming the
apparatus components, but rather that 5 and 29 are using the apparatus as
limitation for the method being described. We do not find that this use of
apparatus as a limitation makes claims 5 and 29 vague to the point that a
POSITA would be unable to reproduce the method due to confusion as to
whether those claims are for a method or for an apparatus.
9
Plaintiff’s second response to the government’s method and apparatus
combination argument is that the remaining apparatus claims (14, 23, 31, and
33) do not include a method claim that renders them indefinite. Instead,
plaintiff argues that the alleged method of these claims is actually a limitation
of the apparatus because the apparatus, not a user, performs the action.
We again agree with plaintiff. The Federal Circuit’s decision in
Mastermine Software instructs that an apparatus claim can have functional
language so long as “the functional language does not appear in isolation, but
rather, is specifically tied to the physical structure” of the apparatus
Mastermine Software Inc. v. Microsoft Corp. 874 F3.d 1307, 1316 (Fed. Cir.
2017). So long as the function is performed by the structure of the apparatus
and is not a series of steps performed by the user, the functional language
merely defines the limitations and capacity of what the apparatus can do and
is not indefinite because it combines an apparatus and a method. Id. Such is
the case here.
There is no aspect of any part of claims 14, 23, 31, or 33 that requires
the apparatus user to perform any action or step. Claims 14 and 31 detail a
“computer readable medium” which “store[s] thereon instructions for
identification, sorting and ranking detections of one or more targets.” This
language specifies a computer containing instructions for identification,
sorting, and ranking, not a step wherein the computer’s user identifies, sorts,
and ranks targets. Thus, claims 14 and 31 meet the Mastermine standard.
Similarly, claims 23 and 33 specify an “apparatus comprising . . . processors
which are configured to execute programed instructions stored in the
memory.” The programmed instructions include “determining a target
detection score,” “identifying a region,” and unmixing the pixel with the
highest determined . . . score.” None of these steps are performed by the
apparatus’s user, and all are stipulated as limitations of what the apparatus
can perform. Therefore, claims 23 and 33 also satisfy the Mastermine
definiteness rule.
All functions performed in these four claims are performed by the
apparatus, therefore the description of the apparatus’s function is a limitation
of the apparatus being claimed by the ‘489 patent. See also UltimatePointer,
L.L.C. v. Nintendo Co., where the Federal Circuit held that “if an apparatus
claim is ‘clearly limited to a[n apparatus] possessing the recited structure and
capable of performing the recited functions,’ then the claim is not invalid as
indefinite.” 816 F.3d 816, 825 (Fed. Cir. 2016) (citing Microprocessor
Enhancement Corp. v. Texas Instruments Inc., 520 F.3d 1367, 1375 (Fed.
Cir. 2008)).
10
Accordingly, we do not find that the ‘489 patent is indefinite due to
an invalid combination of method and apparatus claims. The method claims
and the apparatus claims of the ‘489 patent are sufficiently distinct as to not
be confused for each other by a person of ordinary skill.
This decision on the '489 claims, however, only pertains to whether
the ‘489 patent is indefinite for combination-induced indefiniteness. As
discussed below, there are other questions raised by defendant that are
unresolvable at this point.
B. Necessary Structure
The government also argues that “the ‘489 patent relies upon a means-
plus-function terminology that is crippled by the fact there is no disclosure
of a corresponding algorithm for the ‘determining,’ ‘identifying,’ and
‘processing’ function . . . . Rather, the specification vaguely references
hardware and software, but provides no identification of a specific
software/algorithm.” ECF No. 204 at 24, (emphasis added). Instead of
teaching the algorithm, the patent merely describes what the ‘489 formula
can or should accomplish, says defendant.
The patent’s specifications, in column 5, disclose the following:
. . . target detection processing apparatus identifies the highest
scoring pixel in the image collection (e.g. call this highest
scoring pixel X) and selects a local region around this pixel.
Next, end members (e1. e2, . . . . en) are chosen by target
detection processing apparatus 12 from this local region after
the top statistical detection scores for the given target are
masked out. The convex hull of these end members is a
geometric model of the background for the detected pixel and
is determined by target detection processing apparatus. The
pixel is then “unmixed' by target detection processing
apparatus by finding abundances a1, a2,..., ai, at, that give the
best approximation of the pixel spectra as a linear combination
of the background end members and the target.
11
The background portion of the pixel is
and the residual portion of the pixel is r X-b. The estimated
target fill percentage of the pixel is a, and the unmixing error
is given by the target detection processing apparatus 12 com
paring the target spectra to the residual spectra using a metric,
such as spectra angle or percentage.
‘489 Patent at 7.
Defendant cites Mr. Stocker, its validity expert, for the proposition
that these mathematical equations are only descriptions of the algorithm’s
functions, not the algorithm itself. It thus asserts that Mr. Stocker and
POSITAs like him would be unable to determine precisely what is claimed
in the limitations by terms such as “determining,” “identifying,” and
“processing,” with the result that these claims are indefinite. Neither the
claims nor the specifications define the terms, and the patent would be
indefinite for lack of antecedent basis.
Plaintiff responds that “a lack of antecedent basis renders a claim
invalid only when the claim term would not ‘reasonably be understood by
persons of ordinary skill when read in light of the specification.’” ECF No.
204 at 27 (quoting Energizer Holdings v. ITC, 435 F.3d 1366, 1370 (Fed.
Cir. 2006)). The law, according to plaintiff only requires enough specificity
to prevent undue experimentation. Typhoon Touch Techs., Inc. v. Dell, Inc.,
659 F.3d 1376, 1385-86 (Fed. Cir. 2011) (discussing the “enablement”
requirement of § 112). Plaintiff alleges that any POSITA would reasonably
understand the meaning of the ‘489 claims considering the specification’s
description of the algorithm and cites its expert, Dr. Creed Jones, a professor
of electrical and computer engineering, for support. In his report, Dr. Jones
states that the ‘489 patent has an “adequate corresponding structure”
supporting its claims, thus allowing a person of ordinary skill to reproduce
the algorithm without undue experimentation. ECF No. 201-2 at 123-139.
Although plaintiff is correct that an exact copy of every part of the
algorithm is not required to create a valid patent, sufficient “structure” of the
invention, whether algorithm or otherwise, must be disclosed so that the
patent terms’ meaning are ascertainable and the public is on notice of how to
12
reproduce the invention. See Typhoon Touch Techs., Inc. v. Dell, Inc., 659
F.3d 1376, 1385-86 (Fed. Cir. 2011). The Federal Circuit made clear,
however, that, even when a computer executes the function, the law does not
always require the source code or a detailed algorithm. Id. (citing Aristocrat
Techs. Australia Pty. Ltd. v. Int’l Game Tech., 521 F.3d 1328, 1338 (Fed.
Cir. 2008); Finisar Corp v. DirecTV Group, Inc., 523 F.3d 1323, 1340 (Fed.
Cir. 2008)). “A description of the function in words may ‘disclose, at least
to the satisfaction of [a POSITA], enough of an algorithm to provide the
necessary structure under §112.’” Id. at 1386 (quoting Finisar, 523 F.3d at
1340). The parties rely on conflicting expert reports, creating a disputed
material fact. Nat'l Westminster Bank, PLC v. United States, 69 Fed. Cl. 128,
147 (2005). Both parties’ motions for summary judgment on the issue of
antecedent basis for means-plus-function indefiniteness are thus denied. The
question of whether the ‘489 patent is indefinite must go to trial. We have
considered all other presented arguments regarding patent validity and found
none of them persuasive.
Conclusion
For the abovementioned reasons, plaintiff’s renewed motion is
granted in part and denied in part and defendant’s motion is denied. The
patent is not indefinite due to a combined method and apparatus claims. The
other issues of obviousness and indefiniteness, including the imbedded
questions of fact and secondary indicia, must proceed to trial. A pretrial
schedule has already been set by order on January 6, 2023.
s/ Eric G. Bruggink
ERIC G. BRUGGINK
Senior Judge
13