DETAILED ACTION
Status of the Claims
Claims 1–7 are pending. Claim 1 is objected to. Claims 1–7 are rejected under 35 U.S.C. § 103.
Other Prior Art
Bush (US 10,218,628 B2) discloses a verification module that receives a schedule for transmission of data frames over a Time-sensitive Network together with a destination and a maximum tolerable latency for each data frame, and determines whether the received schedule satisfies those latencies (Abstract).
Sharma et al. (US 10,511,455 B1) disc loses a time-sensitive networking system having gate control circuits that control egress of data from respective queues, and a list execution circuit that configures those gate control circuits according to a current gate control list (Abstract).
Claim Objections
Claim 1 is objected to because the term “actual time for the transmission” (“actual transmission time”) is unclear. Claim 1 recites that an actual time for the transmission of the data ... is determined ... wherein time synchronization jitter and forwarding jitter are taken into account in the dynamic delays. This term is amenable to two different constructions, and it is unclear from the claim which is intended.
Under a first construction, the “actual transmission time” is a measured quantity – the time the network is observed to actually take in operation, for example as the difference between an egress timestamp and an ingress timestamp at the network devices. The specification supports this reading where it determines the network behavior “after measuring the network behavior of a network in practice” (Summary) and, in the debugging context, compares “the actual connection saturation with the calculated saturation” and “the calculated delay envelope curve with the actual behavior in the network” (section 7.3.1), where “actual” is expressly contrasted with “calculated.”
Under a second construction, the “actual transmission time” is a computed quantity – a time calculated from constituent components, namely the time synchronization jitter and the forwarding jitter recited in the claim. The specification supports this reading where the transmission time is “analyzed theoretically after modeling the network behavior” (Summary) using the best-case/worst case model of Section 5.1, which computes the delay from the sum of the forwarding jitters and the time-synchronization jitter j_timesync, and where the “actually determined time” is compared with the theoretically determined time “during the planning of a network that does not yet exist in reality” – a comparison that is possible only if the actual time is computed, since an unbuilt network cannot be measured.
These two constructions are in tension, and the claim does not resolve which is intended. In particular, it is unclear how a transmission time can be considered an actual time if it is determined by taking into account computed components such as the time synchronization jitter and the forwarding jitter. A time calculated from such components is a predicted or modeled time, whereas an actual transmission time, in its ordinary sense, is the time the network in fact takes, observed or measured. The claim recites both – it labels the time “actual” while reciting that it is “determined ... taking [ the two jitters] into account” – leaving one of ordinary skill in the art unable to determine with reasonable certainty whether the claimed “actual time” is a measured quantity or a quantity computed from the recited jitter components.
A further difficulty impacts the reading of the “actual time for the transmission” as a measured quantity. An actual, measured transmission time – the time the network in fact takes from the starting network device to the target network device – inherently already reflects both the time synchronization jitter and the forwarding jitter, because those jitters are physical contributors to the very delay being measured; they are constituents of the measured time, not inputs supplied to its determination. Under that reading, the recitation that time synchronization jitter and forwarding jitter are taken into account in determining the time is difficult to give effect: the two jitters cannot be separately “taken into account” in a measured time, since a measurement of the actual time already embeds their combined effect and does not separately incorporate either as a distinct, identifiable component. By contrast, a computed transmission time is built up from the two jitters as separate, additive inputs – for example as the sum of the forwarding jitters and the time synchronization jitter j_timesync described in the specification – so that each jitter is separately taken into account” in the manner the claim recites. The recited step of separately taking the two jitters into account is thus consistent with a computed time but is not readily reconcilable with a measured one, compounding the ambiguity identified above and leaving one of ordinary skill in the art unable to determine with reasonable certainty what is claimed.
Claim Rejections – 35 U.S.C. § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim 1 is rejected under 35 U.S.C. § 103 as being unpatentable over the non-patent literature entitled “On Time Synchronization Issues in Time-sensitive Networks with Regulators and Nonideal Clocks” (“Boudec”) in further view of the non-patent literature entitled “Deterministic Networking (DetNet) Bounded Latency” (“Finn”).1
Regarding claim 1, Boudec teaches a method of operating a network, wherein multiple network devices, each having their own configuration, are connected to one another for data exchange and exchange data via these connections, wherein dynamic delays (jitters) are taken into account in the determining of the time for the transmission of the data. Boudec teaches a method of operating a network, in that Boudec models a network of interconnected nodes that forward data and computes the delay across the network while accounting for the delay-varying effects of the nodes’ nonideal clocks (col. 1, lines 33–34 and 41–42).
wherein the network is a time-sensitive network. Boudec teaches that the network is a time-sensitive network, in that Boudec is directed to time-sensitive networks (col. 1, lines 11–15).
wherein an actual time for the transmission of the data over the network devices from a starting network device to a target network device is determined while taking into account the dynamic delays. Boudec teaches determining an actual time for the transmission of the data over the network devices from a starting network device to a target network device while taking into account the dynamic delays, in that Boudec computes a per-hop and end-to-end delay bound for a flow from the source to the destination device by a computation that takes the delay-varying effects into account (col. 16,lines 35–40; col. 27, lines 30–38; col. 28, lines 28–33).
wherein time synchronization jitter ... [is] taken into account in the dynamic delays. Boudec teaches taking time synchronization jitter into account in the dynamic delays, in that Boudec accounts for the deviation of the nodes’ nonideal clocks as a delay penalty added to the per-hop forwarding (TAI) delay bound –– the time-error bound being on the order of 1 ms in a tightly synchronized network and 100 ms in a loosely synchronized one (col. 2, lines 39–42 and 45–46; col. 27, lines 30–38 (Prop. 8); col. 28, lines 28–33).
However, Boudec does not teach wherein ... forwarding jitter [is} taken into account in the dynamic delays. Nonetheless, Finn teaches taking forwarding jitter into account in the dynamic delays, in that Finn models the per-hop delay of a time-sensitive network as comprising an output delay, a link delay, a frame preemption delay, and a processing delay, and teaches that these delays are not constant but vary, such that their variation must be taken into account (pg. 6, lines 23–25 and 32–42; pg. 7, lines 7 and 16; pg. 8, lines 4–9). Finn further teaches summing these per-hop delays into an end-to-end delay bound (pg. 9, line 7; pg. 10, lines 41–47). This variation in the forwarding-path delay is the forwarding jitter taken into account in determining the transmission time.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Boudec so that the forwarding jitter taught by Finn is taken into account in the determination, as taught by Finn, by composing Finn’s end-to-end forwarding-delay bound –– the sum of the per-hop delays (pg. 6, lines 23–25; pg. 9, line 7; pg. 10, lines 41–47) –– with the clock-synchronization penalty that Boudec computes as a network-calculus contribution to be added to a per-hop forwarding-delay bound Boudec does not itself supply (col. 3, lines 35–36; col. 16, lines 35–40; col. 27, lines 30–38; col. 28, lines 28–33), because doing so yields the valid worst-case bound recited in claim 1 accounting for both the forwarding jitter and the time synchronization jitter, where omitting the clock-synchronization contribution can leave delay unbounded in non-synchronized networks and, in a loosely-synchronized network, can leave the omitted penalty (on the order of 125 ms) exceeding the required delay bound for flows with stringent delay requirements, rendering the computed bound unusable (col. 1, lines 33–34; col. 3, line 40; col. 4, lines 35–36; col. 28, lines 34–36).
Claim 2 is rejected under 35 U.S.C. § 103 as being unpatentable over Boudec and Finn, as applied to claim 1 above, in further view of Hummen (DE 10 2017 127 431 A1; “Hummen”).
Regarding claim 2, the combination of Boudec and Finn teaches the method according to claim 1, but does not teach wherein each network device inserts a timestamp in a data frame on its ingress port and on its egress port. Nonetheless, Hummen teaches taking an ingress time stamp and an egress time stamp on a network device ([0015]), and further teaches that, in addition to the time stamp collected upon receipt of a packet, a further time stamp is determined when the packet is transmitted, both time stamps in the same network device referring to that device’s local clock, so that each network device records both an ingress and an egress time stamp for the packet ([0007]). The ingress and egress time stamps are thereby inserted at the respective network device along the forwarding path ([0014]–[0015]).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Boudec and Finn so that each network device inserts an ingress and an egress timestamp in a data frame, as taught by Hummen, because doing so provides the per-node forwarding-delay values used in the computed determination of the transmission time.
Claims 3-5 and 7 are rejected under 35 U.S.C. § 103 as being unpatentable over Boudec and Finn, as applied to claim 1 above, in further view of Finzi et al. (US 11,367,024 B2; “Finzi”).
Regarding claim 3, the combination of Boudec and Finn teaches the method according to claim 1, but does not teach wherein a theoretical time for the transmission of the data over the network device from the starting network device to the target network device is determined.2 Nonetheless, Finzi teaches that, during the offline generation of the network schedule and prior to operation of the network (col. 5, lines 24–29; col. 13, lines 12–21), a real-time requirement is determined for each rate-constrained flow, the requirement being expressly an end-to-end delay bound –– a determined worst-case end-to-end delay-time for the transmission of the data over the network from the sender node to the receiver node (col. 5, line 65 – col. 6, line 2). That determined end-to-end delay-time computed during planning and serving as the target the transmission must satisfy, is the theoretically determined time recited.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to further modify the system of Boudec and Finn so that a theoretical time for the transmission of the data from the starting network device to the target network device is determined, as taught by Finzi, because doing so provides, during the planning of the network and before it is built up in practice, a determined target transmission time against which a configuration’s computed transmission time can be evaluated.
Regarding claim 4, the combination of Boudec, Finn, and Finzi teaches the method according to claim 3, and Finzi further teaches wherein the actually determined time is compared with the theoretically determined time. The actually determined time is the worst-case end-to-end transmission time computed for a configuration, as determined at claim 1 by Boudec and Finn; the theoretically determined time is the determined target end-to-end delay bound, as determined at claim 3 by Finzi. Finzi compares the two: during the offline generation of the schedule, Finzi computes, for each rate-constrained flow, the worst-case end-to-end delay along the flow’s path from sender node to receiver node –– itself computed during synthesis via Finzi’s Network Calculus framework (col. 3, lines 1-12) –– and compares that computed worst-case end-to-end delay against the determined end-to-end delay bound required of the flow (col. 10, lines 15-25; col. 13, lines 36-45). The computed worst-case end-to-end delay is the actually determined time and the required end-to-end delay bound is the theoretically determined time, so that Finzi compares the actually determined time with the theoretically determined time, reading on the recited comparison.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to further modify the system of Boudec, Finn, and Finzi so that the actually determined time is compared with the theoretically determined time, as taught by Finzi, by comparing the worst-case end-to-end transmission time computed for the configuration by Boudec and Finn against the theoretically determined target end-to-end delay bound determined by Finzi, because doing so reveals, during the planning of the network and before the configuration is committed, whether the configuration’s computed transmission time in fact satisfies its determined target.
Regarding claim 5, the combination of Boudec, Finn, and Finzi teaches the method according to claim 4, and Finzi further teaches wherein when the comparison exceeds a predeterminable threshold value, then the configuration of at least one network device is changed for debugging purposes between the starting network device and the target network device. As to the triggering condition –– then, when the comparison exceeds a predeterminable threshold value, a responsive action is taken –– Finzi checks by a feedback loop whether the computed worst-case end-to-end delay of a rate-constrained flow exceeds the determined end-to-end delay bound required of that flow and, when it does, reschedules the problematic time-triggered messages, the computed delay exceeding the required bound being the comparison exceeding the presettable threshold value (col. 10, lines 15-25; col. 13, lines 36-45). As to the responsive action –– the configuration of at least one network device is changed ... between the starting network device and the target network device –– Finzi computes modified transmission times for each output port of the selected flow’s path (col. 6, lines 29–32; column 13, lines 36-45), which path runs through the nodes and star couplers between the sender and receiver nodes (fig. 1), thereby changing the configuration of at least one network device between the starting and target network devices. The recitation that the configuration is changed “for debugging purposes” is a statement of intended use given no patentable weight, as it imposes no further structure or step beyond the change of configuration already taught and the prior art need not share that purpose to meet the limitation.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to further modify the system of Boudec, Finn, and Finzi so that, when the comparison exceeds a presettable threshold value, the configuration of at least one network device between the starting and target network devices is changed, as taught by Finzi, because doing so corrects a configuration whose computed transmission time fails its determined target, bringing the network into conformance with that target before the configuration is committed.
Regarding claim 7, the combination of Boudec, Finn, and Finzi teaches the method according to claim 5, and Finzi further teaches wherein the configuration of the at least one network device is changed until the comparison no longer exceeds the predeterminable threshold value. Finzi reschedules the problematic time-triggered messages when the computed worst-case end-to-end delay exceeds the determined required bound, and repeats that modify-and-recompute step –– recomputing the worst-case end-to-end delay for the modified configuration and again comparing it against the required bound –– until a configuration is found whose computed delay no longer exceeds the required bound, i.e., until the real-time requirements of the flows are fulfilled (col.10, lines 40-41; col. 6, lines 29-32; col. 13, lines 30-45). The configuration is thereby changed until the comparison no longer exceeds the presettable threshold value.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to further modify the system of Boudec, Finn, and Finzi so that the configuration of the at least one network device is changed until the comparison no longer exceeds the presettable threshold value, as taught by Finzi, by repeating Finzi’s modify-and-recompute step until the recomputed worst-case end-to-end delay no longer exceeds the determined required bound, because doing so drives the network to a configuration that in fact satisfies the target rather than stopping after a single adjustment that may leave the target unmet, repetition of a known-result driven step until a known goal is reached being within the ordinary skill in the art. Xu et al. (US 7,165,252 B1) likewise teaches repeating a recompute-and-compare step until a determined time satisfies its timing requirement, confirming the technique was known in the art.
Claim 6 is rejected under 35 U.S.C. § 103 as being unpatentable over Boudec, Finn, and Finzi, as applied to claim 5 above, in further view of the non-patent literature entitled “Simulation-based Evaluation of a Synchronous Transaction Model for Time-sensitive Software-Defined Networks” (“Haugg”).
Regarding claim 6, the combination of Boudec, Finn, and Finzi teaches the method according to claim 5, but does not teach wherein the configuration of the starting network device and/or the target network device is also changed. Nonetheless, Haugg teaches that, in a fully scheduled network, the endpoints as well as the switches implement the common TDMA schedule (pg. 2), and that this schedule is reconfigured by a central SDN controller that pushes the synchronized reconfiguration to the scheduled devices (pgs. 2–3). Because the endpoints ––which include the starting and target network devices –– are among the devices implementing the schedule that is reconfigured, changing that schedule changes the configuration of the starting network device and/or the target network device.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Boudec, Finn, and Finzi so that the configuration of the starting network device and/or the target network device is also changed, as taught by Haugg, so that the endpoints implementing the common schedule are reconfigured together with the intermediate devices, because doing so brings the starting and target network devices –– which themselves contribute to the end-to-end transmission time –– into compliance rather than limiting the adjustment to the intermediate devices.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Andrew Georgandellis whose telephone number is 571–270–3991. The examiner can normally be reached on Monday through Friday, 7:30–5:00 PM EST. If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Tonia Dollinger, can be reached on 571–272–4170. The fax phone number for the organization where this application or proceeding is assigned is 571–273–8300.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/ANDREW C GEORGANDELLIS/Primary Examiner, Art Unit 2459
1 For purposes of examination, and as discussed in the Claim Objections above, the term “actual transmission time” is amenable to two constructions – a measured quantity or a quantity computed from the recited time synchronization jitter and forwarding jitter. Until the ambiguity is resolved by amendment or clarification, the Examiner construes the term under the computed construction – i.e., a time computed from the recited time synchronization jitter and forwarding jitter rather than measured – and the following rejection proceeds under that construction.
2 As set forth in the Specification ([0033]), the theoretical time is the determined target time toward which the actually determined time is driven –– the time, determined during the planning of the network, that the data transmission is required to satisfy, and against which the actually determined time is compared until a satisfactory time has been achieved.
Prosecution Insights