Patents by Inventor Timothy David Farnham
Timothy David Farnham has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Patent number: 11567186Abstract: The present disclosure provides an error detector for determining an error vector between a radio trajectory and an image trajectory. The error detector includes: an input for monitoring a radio trajectory of an object from a radio signal and an image trajectory of an object from an image over an observation area; a correlation module arranged to correlate the radio trajectory with the image trajectory; an error module arranged to determine an error vector between the radio trajectory and the image trajectory; and an output arranged to transmit the error vector for use in determining an estimated trajectory of a target based on a target trajectory from a radio signal.Type: GrantFiled: March 19, 2019Date of Patent: January 31, 2023Assignee: Kabushiki Kaisha ToshibaInventors: Timothy David Farnham, Aftab Khan
-
Patent number: 11493644Abstract: A method of identifying item selection by a user, the method comprising: receiving signals at a receiver of a fixed terminal from a transmitter of a mobile terminal associated with the user, generating a signature at the receiver of the fixed terminal of the movement of the user based on changes in the signals received from the transmitter, matching the signature with prior stored movement information to determine the movement of the user, and identifying the item being selected by the user based on the determined movement of the user.Type: GrantFiled: March 15, 2019Date of Patent: November 8, 2022Assignee: Kabushiki Kaisha ToshibaInventors: Timothy David Farnham, Zubeir Bocus
-
Patent number: 11475255Abstract: A method of operating a network comprising an edge node and a server. The method comprises obtaining, by the edge node, a plurality of data samples, determining, by the edge node, a plurality of output labels by applying a first machine learning model using an input memory having a first input memory size to the plurality of data samples, calculating, by the edge node, an error term based on the confidence score of a first output label from the plurality of output labels, determining, by the edge node, based on the error term, whether to modify the first input memory size of the machine learning model and, if so, generating a second machine learning model based on the first machine learning model and a second input memory size.Type: GrantFiled: August 30, 2019Date of Patent: October 18, 2022Assignee: Kabushiki Kaisha ToshibaInventors: Aftab Khan, Timothy David Farnham
-
Publication number: 20220083916Abstract: A computer-implemented method for identifying and rectifying a machine learning drift in a federated learning deployment comprising a parameter server and a plurality of worker nodes, wherein a first worker node comprises: a first machine learning model trained using a first data source; and a second machine learning model trained using a second data source; wherein the first data source is generated by the first worker node and the second data source is generated by a second worker node; the method comprising calculating, by the first worker node, using a trusted data set, a first performance metric associated with the first machine learning model and a second performance metric associated with the second machine learning model and determining, by the first worker node, whether a potential drift has occurred in at least one of the first and the second machine learning models.Type: ApplicationFiled: September 11, 2020Publication date: March 17, 2022Applicant: Kabushiki Kaisha ToshibaInventors: Aftab KHAN, Pietro E. CARNELLI, Timothy David FARNHAM, Ioannis MAVROMATIS, Anthony PORTELLI
-
Patent number: 11240146Abstract: A computer implemented method of routing service requests to service instances in a service mesh. The method comprises monitoring one or more performance characteristics of each of a plurality of network links to service instances. The method further comprises making a prediction of the network latency of transmitting a service request via each of at least two of the plurality of network links and selecting one of the plurality of service instances and routing the service request thereto based on the predictions of the network latencies.Type: GrantFiled: October 30, 2019Date of Patent: February 1, 2022Assignee: Kabushiki Kaisha ToshibaInventor: Timothy David Farnham
-
Publication number: 20210135983Abstract: A computer implemented method of routing service requests to service instances in a service mesh. The method comprises monitoring one or more performance characteristics of each of a plurality of network links to service instances. The method further comprises making a prediction of the network latency of transmitting a service request via each of at least two of the plurality of network links and selecting one of the plurality of service instances and routing the service request thereto based on the predictions of the network latencies.Type: ApplicationFiled: October 30, 2019Publication date: May 6, 2021Applicant: Kabushiki Kaisha ToshibaInventor: Timothy David FARNHAM
-
Publication number: 20210064941Abstract: A method of operating a network comprising an edge node and a server. The method comprises obtaining, by the edge node, a plurality of data samples, determining, by the edge node, a plurality of output labels by applying a first machine learning model using an input memory having a first input memory size to the plurality of data samples, calculating, by the edge node, an error term based on the confidence score of a first output label from the plurality of output labels, determining, by the edge node, based on the error term, whether to modify the first input memory size of the machine learning model and, if so, generating a second machine learning model based on the first machine learning model and a second input memory size.Type: ApplicationFiled: August 30, 2019Publication date: March 4, 2021Applicant: Kabushiki Kaisha ToshibaInventors: Aftab Khan, Timothy David Farnham
-
Publication number: 20200301004Abstract: The present disclosure provides an error detector for determining an error vector between a radio trajectory and an image trajectory. The error detector includes: an input for monitoring a radio trajectory of an object from a radio signal and an image trajectory of an object from an image over an observation area; a correlation module arranged to correlate the radio trajectory with the image trajectory; an error module arranged to determine an error vector between the radio trajectory and the image trajectory; and an output arranged to transmit the error vector for use in determining an estimated trajectory of a target based on a target trajectory from a radio signal.Type: ApplicationFiled: March 19, 2019Publication date: September 24, 2020Applicant: Kabushiki Kaisha ToshibaInventors: Timothy David FARNHAM, Aftab Khan
-
Publication number: 20200292715Abstract: A method of identifying item selection by a user, the method comprising: receiving signals at a receiver of a fixed terminal from a transmitter of a mobile terminal associated with the user, generating a signature at the receiver of the fixed terminal of the movement of the user based on changes in the signals received from the transmitter, matching the signature with prior stored movement information to determine the movement of the user, and identifying the item being selected by the user based on the determined movement of the user.Type: ApplicationFiled: March 15, 2019Publication date: September 17, 2020Applicant: Kabushiki Kaisha ToshibaInventors: Timothy David Farnham, Zubeir Bocus
-
Patent number: 10567918Abstract: A radio-location method of locating a target device that is contained within a region of space, comprising obtaining a first set of detected signals by detecting, at a plurality of points in space, a beacon signal from the target device, or by detecting at the target device, beacon signals sent from one or more radio transmitters in the region of space; determining a current estimate of the target device's location; obtaining successive sets of detected signals, for each successive set of detected signals: determining one or more candidate locations for the target device; comparing the candidate locations with a radio map of the region of space, identifying one or more feasible candidate locations that are consistent with the radio map; determining a new estimate of the target device's location based on the one or more feasible candidate locations; and setting the new estimate as the current estimate of the target device's location.Type: GrantFiled: November 20, 2017Date of Patent: February 18, 2020Assignee: Kabushiki Kaisha ToshibaInventor: Timothy David Farnham
-
Patent number: 10517100Abstract: Radio resource allocation is carried out on the basis of a radio environment map. The radio environment map is constructed based on received reports of signal quality and/or strength. Using history and triangulation, estimates of station positions can be determined, and expectations can be determined for interference between stations and between stations and access points. Resource requests can then be fulfilled on the basis of separate treatment of requests which have little potential for causing interference, and those which have potential to cause interference.Type: GrantFiled: December 8, 2014Date of Patent: December 24, 2019Assignee: Kabushiki Kaisha ToshibaInventor: Timothy David Farnham
-
Patent number: 10404535Abstract: A method for managing the configuration of a wireless connection used to transmit sensor readings from a sensor to a data collection facility, the method comprising receiving a sensor reading from the sensor, comparing the value of the sensor reading with a forecast value for the sensor reading, determining the value of a normality parameter for the sensor, wherein the value of the normality parameter defines the extent to which the value of the sensor reading differs from the forecast value; and determining whether to retain a current configuration of the wireless connection between the sensor and data collection facility or to change the configuration, wherein the determination to retain the current configuration or change configuration is based at least in part on the value of the normality parameter.Type: GrantFiled: December 11, 2015Date of Patent: September 3, 2019Assignee: KABUSHIKI KAISHA TOSHIBAInventor: Timothy David Farnham
-
Patent number: 10405324Abstract: A device and method for wirelessly communicating over a transmission channel with one or more wireless devices. The device comprises a wireless network interface and a controller. The controller is configured to receive a wireless signal via the wireless network interface, if strength of the received signal is greater than a first threshold, determine that the channel is busy, and, if the strength of the received signal is not greater than the first threshold, but is greater than a second threshold that is lower than the first threshold, determine whether the received signal is intended for the device and if so, determine that the channel is busy.Type: GrantFiled: January 26, 2017Date of Patent: September 3, 2019Assignee: Kabushiki Kaisha ToshibaInventors: Zhenzhe Zhong, Fengming Cao, Parag Gopal Kulkarni, Timothy David Farnham
-
Publication number: 20190158982Abstract: A radio-location method of locating a target device that is contained within a region of space, comprising obtaining a first set of detected signals by detecting, at a plurality of points in space, a beacon signal from the target device, or by detecting at the target device, beacon signals sent from one or more radio transmitters in the region of space; determining a current estimate of the target device's location; obtaining successive sets of detected signals, for each successive set of detected signals: determining one or more candidate locations for the target device; comparing the candidate locations with a radio map of the region of space, identifying one or more feasible candidate locations that are consistent with the radio map; determining a new estimate of the target device's location based on the one or more feasible candidate locations; and setting the new estimate as the current estimate of the target device's location.Type: ApplicationFiled: November 20, 2017Publication date: May 23, 2019Applicant: Kabushiki Kaisha ToshibaInventor: Timothy David FARNHAM
-
Publication number: 20180034694Abstract: A method for managing the configuration of a wireless connection used to transmit sensor readings from a sensor to a data collection facility, the method comprising receiving a sensor reading from the sensor, comparing the value of the sensor reading with a forecast value for the sensor reading, determining the value of a normality parameter for the sensor, wherein the value of the normality parameter defines the extent to which the value of the sensor reading differs from the forecast value; and determining whether to retain a current configuration of the wireless connection between the sensor and data collection facility or to change the configuration, wherein the determination to retain the current configuration or change configuration is based at least in part on the value of the normality parameter.Type: ApplicationFiled: December 11, 2015Publication date: February 1, 2018Applicant: KABUSHIKI KAISHA TOSHIBAInventor: Timothy David FARNHAM
-
Publication number: 20170265201Abstract: Radio resource allocation is carried out on the basis of a radio environment map. The radio environment map is constructed based on received reports of signal quality and/or strength. Using history and triangulation, estimates of station positions can be determined, and expectations can be determined for interference between stations and between stations and access points. Resource requests can then be fulfilled on the basis of separate treatment of requests which have little potential for causing interference, and those which have potential to cause interference.Type: ApplicationFiled: December 8, 2014Publication date: September 14, 2017Applicant: KABUSHIKI KAISHA TOSHIBAInventor: Timothy David FARNHAM
-
Publication number: 20170223718Abstract: A device and method for wirelessly communicating over a transmission channel with one or more wireless devices. The device comprises a wireless network interface and a controller. The controller is configured to receive a wireless signal via the wireless network interface, if strength of the received signal is greater than a first threshold, determine that the channel is busy, and, if the strength of the received signal is not greater than the first threshold, but is greater than a second threshold that is lower than the first threshold, determine whether the received signal is intended for the device and if so, determine that the channel is busy.Type: ApplicationFiled: January 26, 2017Publication date: August 3, 2017Applicant: Kabushiki Kaisha ToshibaInventors: Zhenzhe ZHONG, Fengming CAO, Parag Gopal KULKARNI, Timothy David FARNHAM
-
Patent number: 9680580Abstract: A method in a wireless system, including: receiving, at a wireless device, over first and second networks, from each of plural radios of the first and second networks, respective first and second sets of reference signals received from first radios of the first and second networks, and plural further reference signals received from remaining radios of plural radios of the first and second networks, the first radios of the first and second networks located at a common location; calculating first and second sets of normalized reference signals from ratios for each further reference signal of the first and second sets of reference signals with the first and second reference signals of the first and second sets of reference signals; comparing predictions of signal loss based on the first and second sets of normalized reference signals; and determining an estimate of shadow fading or signal reflectors using the comparison result.Type: GrantFiled: October 14, 2013Date of Patent: June 13, 2017Assignee: Kabushiki Kaisha ToshibaInventor: Timothy David Farnham
-
Patent number: 9411038Abstract: Determining a location estimate of a wireless communications device can depend on the dynamic and/or static state of the device. Compensation processing is applied to a location estimate (or to information from which such an estimate is derived, in use) if required, given the dynamic and/or static state of the device.Type: GrantFiled: January 30, 2013Date of Patent: August 9, 2016Assignee: Kabushiki Kaisha ToshibaInventor: Timothy David Farnham
-
Publication number: 20140119217Abstract: A method in a wireless system, including: receiving, at a wireless device, over first and second networks, from each of plural radios of the first and second networks, respective first and second sets of reference signals received from first radios of the first and second networks, and plural further reference signals received from remaining radios of plural radios of the first and second networks, the first radios of the first and second networks located at a common location; calculating first and second sets of normalized reference signals from ratios for each further reference signal of the first and second sets of reference signals with the first and second reference signals of the first and second sets of reference signals; comparing predictions of signal loss based on the first and second sets of normalized reference signals; and determining an estimate of shadow fading or signal reflectors using the comparison result.Type: ApplicationFiled: October 14, 2013Publication date: May 1, 2014Applicant: Kabushiki Kaisha ToshibaInventor: Timothy David FARNHAM