DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Status of Claims
This Office Action is in response to the application filed on 02/06/2024. Claim(s) 1 - 20 are presently pending and are examined in this first action on the merits (FAOM).
Information Disclosure Statement
The information disclosure statement(s) (IDS) submitted on 02/06/2024 has been considered by the Examiner.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1, 2, 4, 6-10, 12, 14-18 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Jakob Spiegelberg et. al. DE 102022202990 (“Spiegelberg”), in view of Qureshi et. al. Improved road segment based Geographical Routing Protocol for Vehicular AD Hoc Networks Kashif Naseer Qureshi et. al. Electronics 2020, 9, 1248 (“Qureshi”) and Arindam Banerjee et. al. US 20230169356 (“Banerjee”)
As per Claim 1, 9 and 17
Spiegelberg discloses,
A method for updating a machine learning model for vehicles (see at least [0006] a distributed machine learning method for a vehicle-related machine learning problem, which is executed by a central server, a corresponding method that is executed in a vehicle, a corresponding central server and a corresponding device in a vehicle, and [Abstract] a distributed machine learning model in a vehicle where the common model is sent with untrained model parameters to several vehicles, each of which has at least one computing unit with which the common model can be trained locally. The locally updated model parameters are received from the multiple vehicles and aggregated into globally updated model parameters. Based on the globally updated model parameters, a trained common model is provided)
aggregating a machine learning model of the selected vehicle and a machine learning model of the other vehicle of the pair of vehicle (see at least [0008] Aggregating the received locally updated model parameters into globally updated model parameters, and [0018] Means of aggregating received locally updated model parameters for the common model)
operating the selected vehicle based on the aggregated machine learning model (see at least [0018] Communication means (KOM) for sending the common model with untrained model parameters (MP U ) to several vehicles, each of which has at least one computing unit with which the common model can be trained locally, and for receiving locally updated model parameters (MP A ) from the several vehicles; - means for aggregating received locally updated model parameters for the common model; and - means are provided for providing a trained common model based on the aggregated model parameters, and [0019] Device for distributed machine learning in a vehicle, comprising - a communication unit (MF) for receiving a common model (ML) with untrained model parameters (MPu) from a central server and for sending locally updated model parameters (MP A ) to the central server and /or to at least one corresponding device of another vehicle)
Spiegelberg does not disclose,
obtaining an edge encounter score for each of a pair of vehicles calculated based on a movement momentum of each of the pair of vehicles and a direction from a location of each of the pair of vehicles to each of one or more edge servers;
selecting one of the pair of vehicles having a higher edge encounter score than other vehicle;
aggregating a machine learning model of the selected vehicle and a machine learning model of the other vehicle of the pair of vehicle
uploading the aggregated machine learning model to one of the one or more edge servers;
Qureshi teaches,
obtaining an edge encounter score for each of a pair of vehicles calculated based on a movement momentum of each of the pair of vehicles and a direction from a location of each of the pair of vehicles to each of one or more edge servers (see at least [Page 7 of 20] Direction difference-oriented score, Sd, when direction of both vehicles is same, direction-based score is 1. When it is 180 degrees, score is 0, and Speed difference score, Ss, is affected by link quality, and [Section 3.3] The progressive distance towards the destination is one of the significant routing metrics in the geographical routing protocol. The next routing metric is the direction towards the destination where the next forwarder node only selects the node that is moving towards the destination to avoid looping issues)
selecting one of the pair of vehicles having a higher edge encounter score than other vehicle (see at least [Section 3.3] Similar representation also follows in Equation (6) where the speed difference-based score is calculated using the function. The third metric is link quality when there is a tie in the direction-oriented score Sd between two or more nodes, and then ISR checks the speed-based link quality of the nodes and chooses the better node between them, as shown in Equation (6), for calculating speed difference-oriented score S5. The overall score S0 can be calculated as S0 = Sd + Ss for each candidate HN during selection. The node with the higher score is finally selected as the HN for the considered road segment. Here, the weighting factor for selecting the HN must be less than or equal to 1)
Spiegelberg disclose a distributed machine learning model in a vehicle where the common model is sent with untrained model parameters to several vehicles, each of which has at least one computing unit with which the common model can be trained locally and an aggregation approach and Qureshi teaches an improved protocol using the concept of a Head Node and a selection based on a scoring method using distance, direction of travel and speed of the vehicle.
As a result, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to provide the inventions as disclosed by Spiegelberg with the selection of vehicle as the head node as taught by Qureshi, with a reasonable expectation of success, to perform the information dissemination-centric routing in the urban VANETs environment (Page 5 of 20, Section 3).
Banerjee teaches,
aggregating a machine learning model of the selected vehicle and a machine learning model of the other vehicle of the pair of vehicle (see at least [0235] Leader nodes may aggregate the results from other nodes of the same cluster, calibrate the result and communicate the results outside of the cluster e.g., to the second layer 152. The learnt result, the global belief 805, and data from the first layer 151 may be aggregated and sent up the fog hierarchy to the second layer 152 and third layer 153, comprising neighborhood and regional fog nodes respectively, for further analysis and distribution, and [0253] The global belief may be aggregated at the leader node level that may be communicated to other same-layer clusters or higher level of fog layer for further accumulation of actionable insights).
uploading the aggregated machine learning model to one of the one or more edge servers (see at least [0011] The improved model may then be shared with the entire network)
Spiegelberg disclose a distributed machine learning model in a vehicle where the common model is sent with untrained model parameters to several vehicles, each of which has at least one computing unit with which the common model can be trained locally and an aggregation approach and Banerjee teaches a leader node that coordinates the decentralized learning process and it is responsible for aggregating the local beliefs of the other nodes in the first set of nodes , and together with its own local belief, update an existing predictive model.
As a result, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to provide the inventions as disclosed by Spiegelberg with the coordination of decentralized learning and aggregation of model as taught by Banerjee, with a reasonable expectation of success, to explore the development of leader nodes in hierarchical fog architecture for the benefit of faster communication and context-aware task management [0262].
As per Claim 2, 10 and 18
Spiegelberg does not disclose,
wherein the edge encounter score for each of a pair of vehicles is calculated further based on a distance from the location of each of the pair of vehicles to each of the one or more edge servers.
Qureshi teaches,
wherein the edge encounter score for each of a pair of vehicles is calculated further based on a distance from the location of each of the pair of vehicles to each of the one or more edge server (see at least [Abstract] It divides the forwarding area into a number of road segments and selects a head node on each segment by focusing on traffic-aware information including the location, direction, and link quality-centric score for every vehicle on each road segment.
Spiegelberg disclose a distributed machine learning model in a vehicle where the common model is sent with untrained model parameters to several vehicles, each of which has at least one computing unit with which the common model can be trained locally and an aggregation approach and Qureshi teaches an improved protocol using the concept of a Head Node and a selection based on a scoring method using distance, direction of travel and speed of the vehicle.
As a result, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to provide the inventions as disclosed by Spiegelberg with the selection of vehicle as the head node as taught by Qureshi, with a reasonable expectation of success, to perform the information dissemination-centric routing in the urban VANETs environment (Page 5 of 20, Section 3).
As per Claim 4, 12 and 20
Spiegelberg does not disclose,
wherein the movement momentum of each of the pair of vehicles is calculated based on a weighted sum of a previous movement momentum and a current motion of corresponding vehicle.
Qureshi teaches,
wherein the movement momentum of each of the pair of vehicles is calculated based on a weighted sum of a previous movement momentum and a current motion of corresponding vehicle (see at least [Section 3.4] the score of each node is calculated using Equations (5) and (6) by considering the direction- and speed-based weighting score of nodes. Moreover, the overall score is calculated by adding the direction-based score and the speed-centric link quality score for each node for the final HN selection).
Spiegelberg disclose a distributed machine learning model in a vehicle where the common model is sent with untrained model parameters to several vehicles, each of which has at least one computing unit with which the common model can be trained locally and an aggregation approach and Qureshi teaches an improved protocol using the concept of a Head Node and a selection based on a scoring method using distance, direction of travel and speed of the vehicle.
As a result, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to provide the inventions as disclosed by Spiegelberg with the selection of vehicle as the head node as taught by Qureshi, with a reasonable expectation of success, to perform the information dissemination-centric routing in the urban VANETs environment (Page 5 of 20, Section 3).
As per Claim 6 and 14
Spiegelberg discloses,
further comprising: transmitting, by the selected vehicle, the aggregated machine learning model to the other vehicle (see at least [0006] a central server is used to coordinate the participating devices, to combine the model parameter updates from the participating devices, and to communicate these model parameter updates back to the participating devices. However, decentralized approaches to distributed learning are also known, in which updates to the model parameters are directly transmitted to the other participating devices, and [0017] the local training of the common model can take place in an iterative process, whereby locally updated model parameters are transferred to the central server during the iterative process, and globally updated model parameters are transferred from the central server to the other vehicles).
As per Claim 7 and 15
Spiegelberg does not disclose,
wherein each of the pair of vehicles calculates corresponding edge encounter score and transmits corresponding edge encounter sore to the other vehicle.
Qureshi teaches,
wherein each of the pair of vehicles calculates corresponding edge encounter score and transmits corresponding edge encounter sore to the other vehicle (see at least [Page 6 of 20, Section 3.2] The head node (HN) is an appropriate forwarding vehicle node for information dissemination in a traffic environment. The HN selection process starts for every segment after the completion of segment formation operation on the nearby road map. Each vehicular node shares the information of its own position, direction, and link quality to its neighbor nodes within the segment. The protocol checks the position and direction of each neighboring vehicle for selecting the HN).
Spiegelberg disclose a distributed machine learning model in a vehicle where the common model is sent with untrained model parameters to several vehicles, each of which has at least one computing unit with which the common model can be trained locally and an aggregation approach and Qureshi teaches an improved protocol using the concept of a Head Node and a selection based on a scoring method using distance, direction of travel and speed of the vehicle.
As a result, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to provide the inventions as disclosed by Spiegelberg with the selection of vehicle as the head node as taught by Qureshi, with a reasonable expectation of success, to perform the information dissemination-centric routing in the urban VANETs environment (Page 5 of 20, Section 3).
As per Claim 8 and 16
Spielberg discloses,
wherein each of the pair of vehicles is an autonomous driving vehicle (see at least [0002] The importance of machine learning (ML) is constantly increasing in many technical fields, including its use in vehicles, for example for driver assistance systems in partially automated driving or for safety systems in fully automated driving)
Claims 3, 11 and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Spiegelberg in Qureshi and Banerjee” as applied to Claim 1 above and further in view of Danxing Wang et. al. CN 112596892 A (“Wang”)
As per Claim 3, 11 and 19
Spiegelberg does not disclose,
wherein the edge encounter score for each of a pair of vehicles is calculated further based on utilization status of each of the one or more edge servers
Wang teaches,
wherein the edge encounter score for each of a pair of vehicles is calculated further based on utilization status of each of the one or more edge servers (see at least The present invention provides a data interaction method for a multi-node edge computing device, which is characterized in that it includes the following steps:
Step S1: Obtain real-time data processing status information of a number of node servers to determine the busy/idle status of the node server, and select a number of connectable nodes from the number of node servers according to the busy/idle status of the node server;
Further, in the step S1, real-time data processing status information of a number of node servers is obtained to determine the busy/idle status of the node server, and based on the busy/idle status, select from the number of node servers Obtaining several connectable node servers specifically include:
Step S101: Obtain the real-time operating load value and the maximum operating load value of each of the several node servers, and use the ratio between the real-time operating load value and the maximum operating load value as the actual load ratio of the node server;
Step S102, judging whether the node server is in an idle state or a busy state according to the actual load ratio).
Spiegelberg disclose a distributed machine learning model in a vehicle where the common model is sent with untrained model parameters to several vehicles, each of which has at least one computing unit with which the common model can be trained locally and an aggregation approach and Wang teaches a method to use busy/idle state of the edge servers to find a number of connectable node servers are filtered from a number of the node servers.
As a result, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to provide the inventions as disclosed by Spiegelberg with the utilization status (busy/idle state) of the edge node as taught by Wang, with a reasonable expectation of success, to improve the operating efficiency of the multi-node server system.
Claims 5 and 13 are rejected under 35 U.S.C. 103 as being unpatentable over Spiegelberg in Qureshi and Banerjee” as applied to Claim 1 above and further in view of Seungchul Ko et. al. US 20190227567 (“Ko”).
As per Claim 5 and 13
Spiegelberg does not disclose,
identifying that each of the one or more edge servers is within a predetermined distance of one of the pair of vehicles.
Ko teaches,
identifying that each of the one or more edge servers is within a predetermined distance of one of the pair of vehicles (see at least [0085] the central server 1000 may determine edge servers 2000 located within a threshold distance from the traveling path based on the position information of the plurality of edge servers 2000, [0086] For example, referring to FIG. 6B, the central server 1000 may determine a first edge server 2000 a, a second edge server 2000 b, and a third edge server 2000 c located within the threshold distance from the traveling path among first to eighth edge servers 2000 a to 2000 h , [0088] central server 1000 may divide the traveling path into a plurality of sections based on a point closest to or within a range of each of the at least one edge server 2000 among points on the traveling path, and [0089] the central server 1000 may divide the traveling path into a plurality of sections based on a point spaced apart by a reference distance in a moving direction from a point closest to each of the at least one edge server 2000)
Spiegelberg disclose a distributed machine learning model in a vehicle where the common model is sent with untrained model parameters to several vehicles, each of which has at least one computing unit with which the common model can be trained locally and an aggregation approach and Ko teaches a method to determine the edge server located within a threshold distance.
As a result, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to provide the inventions as disclosed by Spiegelberg with the selection of edge server based on distance as taught by Wang, with a reasonable expectation of success, to determine at least one edge server among the plurality of edge servers, based on positions of the regions respectively corresponding to the plurality of edge servers [0010].
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Applicants should take note of the prior art in the PTO-892.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ASHUTOSH PANDE whose telephone number is (571)272-6269. The examiner can normally be reached Monday -Friday 9:00am -5:00 PM EST.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Fadey Jabr can be reached at 5712721516. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/A.P./Examiner, Art Unit 3668
/Thomas Ingram/Primary Examiner, Art Unit 3668
Prosecution Insights