Patents by Inventor Hengjun Zhu
Hengjun Zhu 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).
-
Publication number: 20240134939Abstract: A deep learning-based method for fusing multi-source urban energy data and a storage medium are provided to perform data fusion on multi-source urban energy data found in big data and perform multi-scale and multimodal information fusion by using a cross-modal transformer, thereby implementing cross-modal mutual fusion of multi-source heterogeneous types of data to obtain a fused feature for prediction of a quantity of energy that will be used in the future and a quantity of energy that needs to be produced. The present disclosure proposes a multi-scale cooperative multimodal transformer architecture to enhance an effect of representation learned from an unaligned multimodal sequence. Not only there is a higher degree of correlation in multi-source urban energy data fusion, but also a system becomes more lightweight.Type: ApplicationFiled: December 4, 2022Publication date: April 25, 2024Applicant: STATE GRID JIANGSU ELECTRIC POWER CO., LTD NANJING POWER SUPPLY COMPANYInventors: Zhengyi ZHU, Honghua XU, Weiya ZHANG, Long LIANG, Jinjie MA, Hengjun ZHOU, Wendi WANG, Xin QIAN, Linqing YANG
-
Publication number: 20240134323Abstract: According to a transformer-architecture-based method and system for IoT for and smart control of urban integrated energy, urban integrated energy IoT information acquired by a terminal from different areas is output to a server after being standardized into a sequence signal, where energy demand prediction is performed through learning of a deep learning network configured on the server, a smart control network is configured and built by using the deep learning network, real-time prediction processing is performed on the urban integrated energy IoT information acquired in real time, and a result is uploaded to a workstation for reviewing. Based on the transformer architecture, the deep learning network is combined with a fast Fourier transform algorithm and inverse transform on a sequence, and an FFT-Attention mechanism is proposed. Compared with a conventional transformer architecture, frequency domain information of a sequence is given more emphasis.Type: ApplicationFiled: December 4, 2022Publication date: April 25, 2024Applicant: STATE GRID JIANGSU ELECTRIC POWER CO., LTD NANJING POWER SUPPLY COMPANYInventors: Honghua XU, Weiya ZHANG, dongxu ZHOU, Zhengyi ZHU, Xing LUO, Hui WU, Long LIANG, Jinjie MA, Hengjun ZHOU
-
Patent number: 11792046Abstract: A method for generating forwarding information includes generating, by a controller, first forwarding information, wherein the first forwarding information comprises an identifier of a service chain comprising a first service processing entity (SPE) attached to a first service forwarding entity (SFE); generating, by the controller, second forwarding information, wherein the second forwarding information comprises the identifier of the service chain comprising a second SPE attached to a second SFE; sending, by the controller, the first forwarding information to the first SFE; and sending, by the controller, the second forwarding information to the second SFE. Hence, the controller generates forwarding information for the SFE such that a service chaining technology can be applied to a service environment in which a large quantity of services and a large quantity of service processing function units exist.Type: GrantFiled: November 25, 2020Date of Patent: October 17, 2023Assignee: HUAWEI TECHNOLOGIES CO., LTD.Inventors: Yuanlong Jiang, Lehong Niu, Hengjun Zhu
-
Patent number: 11777857Abstract: This application relates to the field of communications technologies, and discloses a service forwarding method and a network device that performs such method. The method includes: forwarding, by a first network device, a data packet of a first service to a second network device in a period (T1); and if the data volume of the forwarded first service reaches a threshold, forwarding, by the first network device, a data packet of a second service to the second network device. The first service is a low-latency service, and the second service is a non-low-latency service. In addition, the period (T1) is determined based on a delay allowed by a device for forwarding the data packet of the first service, and the threshold is a value determined based on a maximum transmission rate of the first service.Type: GrantFiled: September 15, 2021Date of Patent: October 3, 2023Assignee: HUAWEI TECHNOLOGIES CO., LTD.Inventors: Yuanlong Jiang, Hengjun Zhu, Hongguang Guan
-
Patent number: 11757754Abstract: A method and network node for obtaining a target transmission path, where the method includes obtaining, by a first network node in a network domain, topology information of a plurality of network nodes on each path between an ingress node and an egress node that are in the network domain, obtaining, by the first network node, a transmission delay of each path according to the topology information, where the transmission delay of each path includes a sum of physical link delays between all network nodes on each path and node residence times of all the network nodes on each path, and determining, by the first network node, the target transmission path according to the transmission delay of each path.Type: GrantFiled: March 3, 2021Date of Patent: September 12, 2023Assignee: HUAWEI TECHNOLOGIES CO., LTD.Inventors: Yuanlong Jiang, Hengjun Zhu, Min Zha
-
Publication number: 20220006744Abstract: This application relates to the field of communications technologies, and discloses a service forwarding method and a network device that performs such method. The method includes: forwarding, by a first network device, a data packet of a first service to a second network device in a period (T1); and if the data volume of the forwarded first service reaches a threshold, forwarding, by the first network device, a data packet of a second service to the second network device. The first service is a low-latency service, and the second service is a non-low-latency service. In addition, the period (T1) is determined based on a delay allowed by a device for forwarding the data packet of the first service, and the threshold is a value determined based on a maximum transmission rate of the first service.Type: ApplicationFiled: September 15, 2021Publication date: January 6, 2022Inventors: Yuanlong JIANG, Hengjun ZHU, Hongguang GUAN
-
Publication number: 20210367896Abstract: A service packet processing method, apparatus, and system, the method including receiving, by a first network device, a first packet, where the first packet includes a first service path identifier, and the first service path identifier indicates a first service path for transmitting the first packet, generating, by the first network device, a second packet based on the first packet, where the second packet includes a second service path identifier indicating a second service path, and the second service path identifier is different from the first service path identifier, and sending, by the first network device, the second packet via the second service path.Type: ApplicationFiled: August 5, 2021Publication date: November 25, 2021Inventors: Yong Huang, Hengjun Zhu, Lehong Niu, Jinchun Xu, Xuhui Bai
-
Patent number: 11140085Abstract: This application relates to the field of communications technologies, and discloses a service forwarding method and a network device that performs such method. The method includes: forwarding, by a first network device, a data packet of a first service to a second network device in a period (T1); and if the data volume of the forwarded first service reaches a threshold, forwarding, by the first network device, a data packet of a second service to the second network device. The first service is a low-latency service, and the second service is a non-low-latency service. In addition, the period (T1) is determined based on a delay allowed by a device for forwarding the data packet of the first service, and the threshold is a value determined based on a maximum transmission rate of the first service.Type: GrantFiled: October 9, 2019Date of Patent: October 5, 2021Assignee: Huawei Technologies Co., Ltd.Inventors: Yuanlong Jiang, Hengjun Zhu, Hongguang Guan
-
Patent number: 11102133Abstract: An embodiment method includes: a first service atom receives a second packet sent by a central switching device, where a first service packet is encapsulated in the second packet and the second packet further includes a first service path identifier. The first service atom performs first service processing according to information in the first service packet, to obtain a first processing result. The first service atom queries a first path switching entry according to the first processing result and the first service path identifier. The first service atom sends a third packet to the central switching device, where a source device identifier of the third packet is a device identifier of the first service atom, a second service packet is encapsulated in the third packet, and the third packet includes the second service path identifier.Type: GrantFiled: March 26, 2018Date of Patent: August 24, 2021Assignee: HUAWEI TECHNOLOGIES CO., LTD.Inventors: Yong Huang, Hengjun Zhu, Lehong Niu, Jinchun Xu, Xuhui Bai
-
Publication number: 20210194789Abstract: A method and network node for obtaining a target transmission path, where the method includes obtaining, by a first network node in a network domain, topology information of a plurality of network nodes on each path between an ingress node and an egress node that are in the network domain, obtaining, by the first network node, a transmission delay of each path according to the topology information, where the transmission delay of each path includes a sum of physical link delays between all network nodes on each path and node residence times of all the network nodes on each path, and determining, by the first network node, the target transmission path according to the transmission delay of each path.Type: ApplicationFiled: March 3, 2021Publication date: June 24, 2021Inventors: Yuanlong Jiang, Hengjun Zhu, Min Zha
-
Publication number: 20210105153Abstract: A method for generating forwarding information includes generating, by a controller, first forwarding information, wherein the first forwarding information comprises an identifier of a service chain comprising a first service processing entity (SPE) attached to a first service forwarding entity (SFE); generating, by the controller, second forwarding information, wherein the second forwarding information comprises the identifier of the service chain comprising a second SPE attached to a second SFE; sending, by the controller, the first forwarding information to the first SFE; and sending, by the controller, the second forwarding information to the second SFE. Hence, the controller generates forwarding information for the SFE such that a service chaining technology can be applied to a service environment in which a large quantity of services and a large quantity of service processing function units exist.Type: ApplicationFiled: November 25, 2020Publication date: April 8, 2021Inventors: Yuanlong Jiang, Lehong Niu, Hengjun Zhu
-
Patent number: 10965581Abstract: A method and network node for obtaining a target transmission path, where the method includes obtaining, by a first network node in a network domain, topology information of a plurality of network nodes on each path between an ingress node and an egress node that are in the network domain, obtaining, by the first network node, a transmission delay of each path according to the topology information, where the transmission delay of each path includes a sum of physical link delays between all network nodes on each path and node residence times of all the network nodes on each path, and determining, by the first network node, the target transmission path according to the transmission delay of each path.Type: GrantFiled: May 17, 2018Date of Patent: March 30, 2021Assignee: HUAWEI TECHNOLOGIES CO., LTD.Inventors: Yuanlong Jiang, Hengjun Zhu, Min Zha
-
Patent number: 10893019Abstract: The present invention discloses an address management method and apparatus, to help dynamically adjust an address resource and improve address resource utilization. The method includes: establishing a communication connection to a network device based on a specified routing protocol; receiving an address pool status message sent by the network device; determining, based on the address pool status message, whether an address pool of the network device is in an address insufficient state or an address abundant state; and sending a first routing message to the network device by using the communication connection if the address pool of the network device is in the address insufficient state; or sending a second routing message to the network device by using the communication connection if the address pool of the network device is in the address abundant state.Type: GrantFiled: March 8, 2019Date of Patent: January 12, 2021Assignee: Huawei Technologies Co., Ltd.Inventors: Weiping Xu, Hengjun Zhu, Shucheng Liu
-
Patent number: 10892884Abstract: A method for determining a clock synchronization path, and a device, where the method includes determining a first clock synchronization path from a clock injection node of the first network to the first network element based on a request of the first network element and the clock synchronization topology of the first network. A clock synchronization topology is automatically updated based on clock synchronization capability information of a network element, and a clock synchronization path is determined to reduce costs of deploying a clock synchronization path.Type: GrantFiled: September 18, 2018Date of Patent: January 12, 2021Assignee: HUAWEI TECHNOLOGIES CO., LTD.Inventors: Jinchun Xu, Yuanlong Jiang, Hengjun Zhu, Xian Liu
-
Patent number: 10868699Abstract: A method for generating forwarding information includes determining, by a controller, a first service forwarding entity (SFE) and a second SFE forwarding a service stream, where the first SFE and the second SFE are on a service function chain corresponding to the service stream, generating, by the controller, first forwarding information for the first SFE and second forwarding information for the second SFE, and sending, by the controller, the first forwarding information to the first SFE and sending the second forwarding information to the second SFE. Hence, the controller generates forwarding information for the SFE such that a service chaining technology can be applied to a service environment in which a large quantity of services and a large quantity of service processing function units exist.Type: GrantFiled: October 30, 2018Date of Patent: December 15, 2020Assignee: HUAWEI TECHNOLOGIES CO., LTD.Inventors: Yuanlong Jiang, Lehong Niu, Hengjun Zhu
-
Publication number: 20200044974Abstract: This application relates to the field of communications technologies, and discloses a service forwarding method and a network device that performs such method. The method includes: forwarding, by a first network device, a data packet of a first service to a second network device in a period (T1); and if the data volume of the forwarded first service reaches a threshold, forwarding, by the first network device, a data packet of a second service to the second network device. The first service is a low-latency service, and the second service is a non-low-latency service. In addition, the period (T1) is determined based on a delay allowed by a device for forwarding the data packet of the first service, and the threshold is a value determined based on a maximum transmission rate of the first service.Type: ApplicationFiled: October 9, 2019Publication date: February 6, 2020Inventors: Yuanlong JIANG, Hengjun ZHU, Hongguang GUAN
-
Patent number: 10554764Abstract: Embodiments of the present invention provide a method, device and system for carrying out a service in a network. The control apparatus comprises: an orchestrator unit, configured to receive a service request for requesting to carry out a service, wherein the service request comprises service deployment information for requesting to deploy a first service, service flow identification information for identifying a first packet flow and service flow steering information; and a control unit, configured to send the service deployment information to a computing apparatus, and send the service flow identification information and the service flow steering information to an identifying and steering apparatus. Technical solutions of the present invention provide a new service mode, which can meet developing needs of OTT applications.Type: GrantFiled: October 22, 2015Date of Patent: February 4, 2020Assignee: Huawei Technologies Co., Ltd.Inventors: Yong Huang, Hengjun Zhu, Lehong Niu, Xiaoyu Gong, Min Zha
-
Publication number: 20190238500Abstract: The present invention discloses an address management method and apparatus, to help dynamically adjust an address resource and improve address resource utilization. The method includes: establishing a communication connection to a network device based on a specified routing protocol; receiving an address pool status message sent by the network device; determining, based on the address pool status message, whether an address pool of the network device is in an address insufficient state or an address abundant state; and sending a first routing message to the network device by using the communication connection if the address pool of the network device is in the address insufficient state,; or sending a second routing message to the network device by using the communication connection if the address pool of the network device is in the address abundant state.Type: ApplicationFiled: March 8, 2019Publication date: August 1, 2019Inventors: Weiping Xu, Hengjun Zhu, Shucheng Liu
-
Patent number: 10263901Abstract: The present invention discloses a service packet processing method, apparatus, and system. The method includes a first service atom receiving a second packet sent by a central switching device, where a first service packet is encapsulated in the second packet and the second packet further includes a first service path identifier. The first service atom performs first service processing according to information in the first service packet, to obtain a first processing result. The first service atom queries a first path switching entry according to the first processing result and the first service path identifier. The first service atom sends a third packet to the central switching device, where a source device identifier of the third packet is a device identifier of the first service atom, a second service packet is encapsulated in the third packet, and the third packet includes the second service path identifier.Type: GrantFiled: August 21, 2015Date of Patent: April 16, 2019Assignee: Huawei Technologies Co., LtdInventors: Yong Huang, Hengjun Zhu, Lehong Niu, Jinchun Xu, Xuhui Bai
-
Publication number: 20190074995Abstract: A method for generating forwarding information includes determining, by a controller, a first service forwarding entity (SFE) and a second SFE forwarding a service stream, where the first SFE and the second SFE are on a service function chain corresponding to the service stream, generating, by the controller, first forwarding information for the first SFE and second forwarding information for the second SFE, and sending, by the controller, the first forwarding information to the first SFE and sending the second forwarding information to the second SFE. Hence, the controller generates forwarding information for the SFE such that a service chaining technology can be applied to a service environment in which a large quantity of services and a large quantity of service processing function units exist.Type: ApplicationFiled: October 30, 2018Publication date: March 7, 2019Inventors: Yuanlong Jiang, Lehong Niu, Hengjun Zhu