Ex Parte Gopalan et alDownload PDFPatent Trial and Appeal BoardMay 21, 201814176446 (P.T.A.B. May. 21, 2018) Copy Citation UNITED STA TES p A TENT AND TRADEMARK OFFICE APPLICATION NO. FILING DATE FIRST NAMED INVENTOR 14/176,446 02/10/2014 Prem K. Gopalan 87555 7590 05/23/2018 Riverbed Technology Inc. - PVFD c/o PARK, VAUGHAN, FLEMING & DOWLER LLP 2820 Fifth Street Davis, CA 95618 UNITED STATES DEPARTMENT OF COMMERCE United States Patent and Trademark Office Address: COMMISSIONER FOR PATENTS P.O. Box 1450 Alexandria, Virginia 22313-1450 www .uspto.gov ATTORNEY DOCKET NO. CONFIRMATION NO. RIV-0760-CONl 6994 EXAMINER PATEL, HITESHKUMARR ART UNIT PAPER NUMBER 2441 NOTIFICATION DATE DELIVERY MODE 05/23/2018 ELECTRONIC Please find below and/or attached an Office communication concerning this application or proceeding. The time period for reply, if any, is set in the attached communication. Notice of the Office communication was sent electronically on above-indicated "Notification Date" to the following e-mail address( es): j eannie@parklegal.com wendy@parklegal.com PTOL-90A (Rev. 04/07) UNITED STATES PATENT AND TRADEMARK OFFICE BEFORE THE PATENT TRIAL AND APPEAL BOARD Ex parte PREM K. GOP ALAN and BRYAN THOMAS EL VERSON 1 Appeal2017-007009 Application 14/17 6,446 Technology Center 2400 Before JUSTIN BUSCH, STACEY G. WHITE, and JOHN P. PINKERTON, Administrative Patent Judges. Opinion for the Board filed by Administrative Patent Judge PINKERTON. Opinion Dissenting-in-part filed by Administrative Patent Judge BUSCH. PINKERTON, Administrative Patent Judge. DECISION ON APPEAL Appellants appeal under 35 U.S.C. § 134(a) from the Examiner's rejection of claims 1-20, which are all the pending claims. We have jurisdiction under 35 U.S.C. § 6(b). We reverse. 1 The real party in interest identified by Appellants is Riverbed Technology, Inc. App. Br. 3. Appeal2017-007009 Application 14/176,446 CLAIMED SUBJECT MATTER Appellants' invention "relates to systems, methods, and techniques to detect anomalies in network traffic." Spec. 1. In particular, the system and method compare observed network traffic to a prediction interval, identify an outlier if the observed network traffic is outside the prediction interval, and identify an anomaly if the number of detected outliers within a certain time period exceeds a threshold. Id. at 2, 5. Claims 1, 8, and 15 are independent claims. Claim 1 is representative and reproduced below: 1. A computer-implemented method, comprising: identifying, by computer, an outlier for a network traffic metric if observed network traffic is outside of a prediction interval during a time period; and identifying an anomaly for the network traffic metric only when a count of outliers for the network traffic metric identified during the time period is greater than or equal to two, and exceeds an anomaly threshold for the network traffic metric. REJECTIONS Claims 1-5, 7-12, and 14--20 stand rejected under 35 U.S.C. § 103(a) as being unpatentable over Gopalan (US 2003/0208523 Al; Nov. 6, 2003) and Ho (US 6,597,777 Bl; July 22, 2003). Final Act. 4--11. Claims 6 and 13 stand rejected under 35 U.S.C. § 103(a) as being unpatentable over Gopalan, in view of Ho, and further in view of Yun (US 2009/0126023 Al; May 14, 2009). Final Act. 11-12. 2 Appeal2017-007009 Application 14/176,446 ANALYSIS CONSTRUCTION OF METHOD CLAIM 1 In the Final Office Action, the Examiner considers all of the limitations of claim 1, and finds they are taught or suggested, and rendered obvious, by the combined teachings of Gopalan and Ho. Final Act. 5---6. In the Answer, the Examiner takes a different approach. The Examiner finds that if, under the broadest reasonable interpretation of a method claim, one or more conditional method steps would not be invoked, the Examiner need not find the conditional method steps in the prior art to render the claimed method obvious. Ans. 4--5 (citing Ex parte Katz, No. 2010-006083, 2011 WL 514314, at *4--5 (BPAI 2011)). The Examiner concludes that the two steps of claim !-"identifying ... an outlier" and "identifying an anomaly"-need not be found in the prior art because they are both conditional limitations. Id. at 5 (citing Cybersettle, Inc. v. Nat 'l Arbitration Forum, Inc., 243 F. App'x 603, 607 (Fed. Cir. 2007) (unpublished) and Applera Corp. v. Illumina, Inc., 375 F. App'x 12, 21 (Fed. Cir. 2010) (unpublished)). Appellants argue the Examiner essentially concludes "that the claim elements in the independent claims should not be given patentable weight merely because the claim elements include the words 'if and 'when."' Reply Br. 6. Appellants contend Cybersettle and Applera do not directly address how conditional claim language is to be interpreted under the broadest reasonable interpretation standard, but Ex parte Schulhauser, Appeal 2013-007847, 2016 WL 6277792 (PTAB Apr. 28, 2016) (precedential) is directly on point. Reply Br. 6-7. Appellants argue 3 Appeal2017-007009 Application 14/176,446 Schulhauser applies only to method claims and, therefore, all the limitations of the computer-readable storage medium claims (claims 8-14) and the apparatus claims (claims 15-20) should be given patentable weight because those claims still require the structure for performing the recited conditional functions. Id. at 7 (citing Schulhauser). Appellants also argue Schulhauser does not apply to method claims 1-7 because failing to give patentable weight to the steps of claim 1, as the Examiner proposes, would result in the claimed method "not requir[ ing] any step at all, which would render the entire claim meaningless." Id. at 7-8 (citing Ex parte Hehenberger, Appeal No. 2015-007421, 2017 WL 588626 (PTAB Jan. 27, 2017) (concurring opinion)). Initially, unlike Appellants, we do not understand the Examiner's conclusion with respect to whether the conditional limitations of the claims are to be given patentable weight to apply to all the independent claims. Instead, we interpret the Examiner's conclusion in this regard to apply only to method claim 1. Regardless, we agree with Appellants that Schulhauser applies only to method claims and, like the "system" claim in Schulhauser, the conditional functional limitations of claim 8 (directed to "a computer- readable storage medium") and claim 15 (directed to "an apparatus") should be given patentable weight because the broadest reasonable interpretation of these claims "requires structure for performing the function should the condition occur." See Schulhauser, 2016 WL 6277792, at *7. Regarding method claim 1, we are persuaded by Appellants' arguments that the Examiner errs by failing to give patentable weight to the limitations of claim 1. Reply Br. 8. First, we agree with Appellants' argument that the Examiner's construction would result in the claimed 4 Appeal2017-007009 Application 14/176,446 method "not requir[ing] any step at all, which would render the entire claim meaningless." Id. at 7-8 (citing Ex parte Hehenberger, Appeal No. 2015- 007421 (Jan. 27, 2017) (concurring opinion)). We also agree with Appellants that the Examiner's construction is overly broad and unreasonable because a person of ordinary skill in the art, considering the claim in the context of the Specification as a whole, would interpret the claim to be limited to the method described in which the recited conditions occur. Id. at 8. Second, we find Schulhauser is not controlling because it is factually distinguishable. The method claim in Schulhauser was directed to monitoring cardiac conditions and recited, inter alia, (1) collecting electrocardiac signal data, (2) comparing the data with a threshold electrocardiac criteria for indicating a strong likelihood of a cardiac event, (3) "triggering an alarm state if the ... data is not within the threshold," and (4) "determining the current activity level of the subject ... if the electrocardiac signal data is within the threshold." Schulhauser, 2016 WL 6277792, at * 1-2 (emphases added). The Board found the "triggering" and "determining" steps are "mutually exclusive" and that the claim "as written covers at least two methods"-"one in which the prerequisite condition for the triggering step is met and one in which the prerequisite condition for the determining step is met." Schulhauser, 2016 WL 6277792, at *3--4. Thus, the Board concluded that the broadest reasonable interpretation of the method claim encompasses a method where only the steps of "collecting," "comparing," and "triggering an alarm state if the electrocardiac signal data is not within the threshold" are performed. Id. at *4. The Board further found that the Examiner did not need to present evidence of the obviousness 5 Appeal2017-007009 Application 14/176,446 of the remaining method steps that are not required to be performed under the broadest reasonable interpretation of the claim. Id. Unlike the method claim in Schulhauser, the steps of method claim 1 are not mutually exclusive and the claim, as written, covers only one method. The steps of "identifying ... an outlier" and "identifying an anomaly" are not dependent on the happening of a condition precedent that would result in the claim operating in two separate methods as in Schulhauser. Instead, these steps work together, and both must be used, to accomplish the goal of reducing false positives. As explained in the Specification: "A single outlier is not necessarily enough [to] cause an anomaly. Using a threshold number of outliers to identify an anomaly can provide the advantage of reducing the number of false positives." Spec. 13:22-25. In other words, outliers are first identified "if observed network traffic is outside of a prediction interval during a time period," and anomalies are then identified based on the count of outliers-"only when a count of outliers for the network traffic metric identified during the time period is greater than or equal to two, and exceeds an anomaly threshold for the network traffic metric." Claims App'x 17. Although each of these steps is conditional, they are integrated into one method or path and do not cause the claim to diverge into two methods or paths, as in Schulhauser. Thus, we conclude that the broadest reasonable interpretation of claim 1 requires the performance of both steps of "identifying ... an outlier" and "identifying an anomaly." Third, neither Schulhauser, nor other precedent, requires that all conditional claim limitations are not to be give patentable weight. In fact, Schulhauser gave patentable weight to the conditional "triggering" step-- 6 Appeal2017-007009 Application 14/176,446 "triggering an alarm state if the electrocardiac signal data is not within the threshold electrocardiac criteria." Schulhauser, 2016 WL 6277792, at *4. In Ex parte Fleming, Appeal 2014-002849, 2014 WL 7146104, at *2 (PTAB Dec. 12, 2014) (expanded panel decision on rehearing), which is cited in Schulhauser, at * 10, fn 4, the Board stated "we did not and do not hold that every conditional step recitation in a method claim is not entitled to patentable weight." Fourth, the other cases cited in Schulhauser, and by the Examiner, are also distinguishable. Ex parte Urbanet, Appeal 2011-002606, 2012 WL 4460637 (BPAI Sept. 19, 2012), cited in Schulhauser, at *10, fn 4, is factually distinguishable because the Board found the claim at issue alternatively covers at least two methods because of a conditional "wherein" clause. See Urbanet, at *2. For the reasons explained above, that is not the situation here because claim 1 recites one method. In Katz, as cited by the Examiner, the Board held "[a]s under the broadest scenario, the steps dependent on the 'if' conditional would not be invoked, the Examiner was not required to find these limitations in the prior art in order to render the claims obvious." Katz, at *4. Katz is not controlling because, as discussed above, the broadest reasonable interpretation of method claim 1 here requires the performance of both steps of "identifying ... an outlier" and "identifying an anomaly." Regarding the cases cited by the Examiner, we agree with Appellants that Cybersettle and Applera do not address "the issue at hand," namely how conditional claim language is interpreted under the broadest reasonable interpretation standard. Reply 6-7. In particular, Cybersettle concerns the issue of infringement of a method claim. See Cybersettle, 243 F. App'x at 607. And, Applera concerns whether a 7 Appeal2017-007009 Application 14/176,446 sequence of steps recited in the method claim was required to be repeated multiple times. See Applera, 375 F. App'x at 21. Because the Examiner erred in construing claim 1, we do not sustain the Examiner's rejection of claims 1-7 on the ground that the claim limitations are not to be given patentable weight. Thus, we proceed to consider the Examiner's substantive rejections of claims 1-20 under 35 U.S.C. § 103 in view of Appellants' arguments. REJECTION OF CLAIMS 1-20 UNDER 35 U.S.C. § 103 The Examiner rejects claims 1-5, 7-12, and 14--20 as obvious in view of Gopalan and Ho and claims 6 and 13 as obvious in view of Gopalan, Ho, and Yun. Final Act. 4--12. Of particular relevance to Appellants' arguments, the Examiner finds Ho teaches the identifying an anomaly step recited in claims 1, 8, and 15. Final Act. 5---6 (citing Ho 3:25-35); Ans. 3--4 (citing Ho 5:11-14, 11:1-20, Figs. 8-9). Appellants argue Ho fails to teach or suggest the identifying an anomaly step because the Examiner's rejection depends on an improper implicit construction of at least one of "outlier" and "only when," as recited in the independent claims. App. Br. 13-15. In particular, Appellants assert Ho triggers an alert upon the first occurrence of an outlier. Id. at 14. Appellants further explain "that each occurrence of the anomaly in Ho corresponds to the term 'outlier' in the claims of the instant application," such that Ho triggers an alert in response to detecting a single outlier. Id. Appellants contend Ho, therefore, does not reduce false positives, whereas the claims reduce false positives because the claims identify an anomaly "only when" the count of outliers detected is greater than one during a time period. Id. 8 Appeal2017-007009 Application 14/176,446 Ho is directed to detecting network anomalies using traffic data to determine traffic intensity. Ho 3:10-23, 5:11-14, Abstract. For each service class, Ho determines "historical data, predicted baseline, and upper and lower time-dependent traffic intensity thresholds." Id. at 3:23-27. Ho measures the real-time traffic intensity and compares that measurement to the upper and lower thresholds to determine whether the measured traffic intensity is within (or outside of) the determined upper and lower thresholds for longer than a time period. Id. at 3:27-32. Ho further teaches detecting and acting upon "an anomalous condition" in response to determining whether the currently detected traffic intensity is outside the determined thresholds and, if so, for how long and by what degree. Id. at 7:5-32. More specifically, Ho describes that "deviations (in both magnitude and duration, as defined by a set of fault criteria)" outside the thresholds indicate service anomalies. Id. at 9:10-13 (emphasis added). One embodiment of Ho teaches identifying an anomaly when the traffic intensity persistently falls outside the thresholds. Id. at 9: 14--43; see also id. at 10:20-23 ("Detector processor 604, upon determining that the current traffic intensity in any service class falls above the upper threshold or below the lower threshold for a duration of 15 or more minutes, generates an alarm." (Emphasis added)), 11: 12-15 ("The present invention, however, detects the fact that the traffic intensity persistently exceeds the upper threshold within the first 15 minutes of the onset of the server failure" (emphasis added)). We agree with Appellants that Ho's determination of whether the real- time traffic intensity falls within the determined thresholds is not equivalent to Appellants' recited detecting an anomaly because Ho's thresholds are analogous to Appellants' recited prediction interval, which is used to 9 Appeal2017-007009 Application 14/176,446 determine whether a particular measurement is an "outlier." See App. Br. 14 (arguing that "each occurrence of the anomaly in Ho corresponds to the term 'outlier' in the claims of the instant application"). We disagree, however, with Appellants' assertion that Ho, therefore, "triggers an alert as soon as it identifies a single outlier." See id. Ho's determination that traffic intensity is outside its determined thresholds is similar to Appellants' recited identifying step. However, contrary to Appellants' alleged distinction that Ho triggers an alert upon identification of a single outlier, id., Ho discloses generating an alarm when the detected traffic intensity is persistently outside of Ho's determined thresholds. See Ho 9:14--43, 10:20-23, 11:12-15. Appellants' claims recite a "count of outliers," which is compared to an anomaly threshold, and Appellants explain that this disputed aspects of the claim reduces false positives. App. Br. 14. Ho suggests reducing false positives by detecting an anomaly only when the traffic intensity is persistently outside the determined thresholds. Thus, both Appellants' claims and Ho's disclosure address the statistical significance of detected outliers. Nevertheless, the Examiner has not sufficiently explained how Ho's anomaly detection would have taught or suggested the claimed count of outliers or "identifying an anomaly ... only when a count of outliers ... is greater than or equal to two, and exceeds an anomaly threshold." Nor has the Examiner provided a rationale for modifying Ho's disclosed method of ensuring the statistical significance of its outliers. Therefore, constrained by this record, we are persuaded the Examiner erred in rejecting independent claims 1, 8, and 15. For similar reasons, we are persuaded the Examiner erred in rejecting claims 2-7, 9-12, 14, and 16- 10 Appeal2017-007009 Application 14/176,446 20, which ultimately depend from and incorporate the limitations of one of claims 1, 8, and 15. Similarly, because the Examiner does not rely on Yun to cure the identified deficiency in the rejection of claims 6 and 13, we also are persuaded the Examiner erred in rejecting claims 6 and 13 for the same reasons. DECISION We reverse the Examiner's decision rejecting claims 1-20 under 35 U.S.C. § 103. REVERSED 11 UNITED STATES PATENT AND TRADEMARK OFFICE BEFORE THE PATENT TRIAL AND APPEAL BOARD Ex parte PREM K. GOP ALAN and BRYAN THOMAS ELVERSON Appeal2017-007009 Application 14/17 6,446 Technology Center 2400 Before JUSTIN BUSCH, STACEY G. WHITE, and JOHN P. PINKERTON, Administrative Patent Judges. BUSCH, Administrative Patent Judge, DISSENTING-IN-PART. I agree with, and join, the majority decision with respect to its treatment of claims 8-20. I write separately, however, because I respectfully disagree with the majority's claim construction of claim 1 and, therefore, the majority's ultimate conclusion that claims 1-7 would not have been obvious. As discussed in the majority opinion, the Examiner concludes that the two steps of claim 1-"identifying ... an outlier" and "identifying an anomaly"-need not be found in the prior art because they are both conditional limitations. Ans. 5 (citing Cybersettle, Inc. v. Nat 'l Arbitration Forum, Inc., 243 F. App'x 603, 607 (Fed. Cir. 2007) (unpublished); Applera Corp. v. Illumina, Inc., 375 F. App'x 12, 21 (Fed. Cir. 2010) (unpublished)). Appellants contend Cybersettle and Applera do not directly address how conditional claim language is to be interpreted under the broadest reasonable interpretation standard, arguing that Ex parte Schulhauser, Appeal 2013- Appeal2017-007009 Application 14/176,446 007847, 2016 WL 6277792 (PTAB Apr. 28, 2016) (precedential) "is directly on point." Reply Br. 6-7. I agree with Appellants that Schulhauser is the most relevant case regarding interpreting conditional claim language under the broadest reasonable interpretation standard. See Schulhauser, 2016 WL 6277792, at *3-5 (holding that in a method claim, a step reciting a condition precedent does not need to be performed if the condition precedent is not met); see also Cybersettle, Inc. v. Nat'! Arbitration Forum, Inc., 243 F. App'x 603, 607 (Fed. Cir. 2007) (unpublished) ("It is of course true that method steps may be contingent. If the condition for performing a contingent step is not satisfied, the performance recited by the step need not be carried out in order for the claimed method to be performed."). For the reasons below, I disagree with the majority's conclusion that Schulhauser is not controlling in this case. See Maj. Op. 5---6. Claim 1 's first step recites, in part, "identifying ... an outlier ... if observed network traffic is outside of a prediction interval." I agree with the majority that both of claim l's steps are conditional limitations, and that claim 1 's first and second steps are not mutually exclusive alternatives to each other. Maj. Op. 6 (explaining that each of claim l's steps are conditional and that the two steps work together to reduce false positives). Put another way, claim 1 identifies an outlier if the recited condition of observed network traffic occurs and claim 1 further identifies an anomaly when the count of identified outliers is greater than one. 2 Thus, claim 1 's 2 For this opinion, I assume the broadest reasonable interpretation, consistent with Appellants' Specification, of claim l's recited "count of outliers" is simply the number of times the condition precedent in the first step occurs. 2 Appeal2017-007009 Application 14/176,446 second identifying step will not occur until and unless claim 1 's first identifying step is performed at least twice. It is undisputed that Schulhauser 's particular method recited a pair of conditional limitations, wherein if one condition was evaluated as true, the other condition would be evaluated as false within the same iteration. Another way to describe Schulhauser 's claim is that evaluation of the first pair of conditional limitations led to executing one of two mutually exclusive alternative "branches" (one of which included further branching), wherein one "branch" was executed if a condition was evaluated as true and the other "branch" was executed if the same condition was evaluated as false. The majority determines Schulhauser is not controlling in this case and the cases cited therein are distinguishable from this case, the Examiner's construction is overly broad and unreasonable, and our precedent does not require giving all conditional limitations no patentable weight. My view diverges from the majority's view because I believe Schulhauser, as suggested by the cases cited therein, is controlling in this case and, therefore, Schulhauser governs the proper construction and analysis of whether or not the prior art must perform the conditional limitations in order to render the claim obvious. Therefore, the condition precedent of the second step would be met when the first condition precedent occurs at least twice within a specified time interval (and therefore the first identifying step is performed at least twice). I do not consider whether the broadest reasonable interpretation of method claim 1, which uses the transitional phrase "comprises," encompasses incrementing the count of outliers through other outlier detection steps not explicitly recited. 3 Appeal2017-007009 Application 14/176,446 In particular, contrary to the majority, I do not believe Schulhauser 's analysis and conclusions are limited to situations involving mutually exclusive alternative branches because, other than the requirement that one could practice the recited method without performing the conditional step when the condition precedent does not occur, I see no caveats for applying Schulhauser to conditional statements. Schulhauser found "[ t ]he Board previously has construed similar method steps in this same manner," citing cases indicating Schulhauser 's holding is not limited to conditional limitations creating mutually exclusive alternative branches. Schulhauser, 2016 WL 6277792, at *4 n.4 (citing Ex parte Fleming, Appeal 2014-002849, 2014 WL 7146104 (PTAB Dec. 12, 2014) (expanded panel decision on rehearing); Ex parte Urbanet, Appeal 2011-002606, 2012 WL 4460637 (PTAB Sept. 19, 2012); Ex parte Katz, Appeal 2010-006083, 2011WL514314 (BPAI Jan. 27, 2011)). In Fleming, 2014 WL 7146104 at *1-3, the Board concluded the claimed method was satisfied when the recited condition precedent was not met and, therefore, the conditional limitation ("generating an alert ... if' a position condition is met), which was not part of a set of mutually exclusive alternatives, was not performed. Similarly, in Urbanet, 2012 WL 4460637 at *2, the Board concluded the conditional limitation need not be performed to satisfy the claim, stating the claimed method covered a method "without the conditional steps and prerequisite condition in the wherein clause." Finally, in Katz, 2011 WL 514314 at *4, the Board concluded "the 'if condition' as employed in the method claim 1 is not a limitation against which prior art must be found because the step of deducting ... is conditioned on" a particular scenario and, because "under the broadest scenario, the steps dependent on 4 Appeal2017-007009 Application 14/176,446 the 'if' conditional would not be invoked, the Examiner was not required to find these limitations in the prior art in order to render the claims obvious." None of the cases Schulhauser cited in support of its conclusion involved claims with mutually exclusive alternative conditions. Accordingly, it follows that Schulhauser 's analysis and conclusions are not limited to only mutually exclusive alternative conditions. To the majority's point that the Examiner's construction is unreasonable because a person of ordinary skill in the art would have understood the method to be limited to a method in which the recited conditions occur, I see no distinction between the claims here and the claims in Schulhauser and, therefore, would follow Schulhauser in this case. Finally, I believe the majority's point that not requiring any steps would render the claim meaningless goes to the definiteness of the claim. In order to give patentable weight to every word in a claim, including the recited conditions, following Schulhauser results in a determination that a conditional limitation does not need to be performed if the condition precedent is not satisfied. Applying Schulhauser to conditional limitations only in claims that also recite non-conditional limitations would produce an odd result here (and in other cases }-namely, the conditional limitations in claim 1 would need to be performed, but those same limitations would not need to be performed in, for example, claim 4, which depends from claim 1 and recites additional, non-conditional, limitations. 3 3 The majority references Appellants' citation of Ex parte Hehenberger, Appeal No. 2015-007421, 2017 WL 588626 (PTAB Jan. 27, 2017) (Delmendo, A.P.J., concurring) in support of Appellants' position that the Examiner's construction would render the claim meaningless. Maj. Op. 4 5 Appeal2017-007009 Application 14/176,446 For the above reasons, I agree with the Examiner's construction of claim 1. Accordingly, for the reasons stated by the Examiner, I disagree with the majority that the Examiner erred in rejecting claims 1-7 as obvious under 35 U.S.C. § 103. (citing Reply Br. 7-8). I note we are not bound by a concurrence to a non- precedential Board decision. 6 Copy with citationCopy as parenthetical citation