Node selection and / or communication resource allocation in hierarchical tree federated learning network
By acquiring and analyzing node topology information in the hierarchical tree federated learning network and optimizing communication resource allocation, the problem of insufficient resource allocation in traditional methods is solved, and the efficiency and effectiveness of federated learning are improved.
Patent Information
- Application Number
- PCT/CN2024/091318
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-05-09
- Filing Date
- 2024-05-07
- Publication Date
- 2024-11-14
AI Technical Summary
In traditional federated learning networks, the node topology information in the hierarchical tree structure is not fully considered when allocating communication resources, resulting in insufficient resource allocation, which affects the efficiency and effectiveness of federated learning.
By obtaining the topological information of each communication node in the hierarchical tree federated learning network, including the total data volume, number, number of layers and importance of nodes in the subtree, priority is given to allocating communication transmission resources and selecting nodes participating in federated learning. To achieve more accurate and efficient allocation of communication resources.
By considering node topology information, the allocation of communication resources is optimized, the efficiency and accuracy of federated learning are improved, so that the final generated model can better adapt to the actual situation of the communication network and improve resource utilization.
Smart Images

Figure CN2024091318_14112024_PF_FP_ABST
Abstract
Description
Node selection and / or communication resource allocation in hierarchical tree-structured federated learning networks Technical Field
[0001] The present disclosure relates generally to communication network technology, and more particularly to methods, electronic devices, and computer-readable storage media associated with node selection and / or communication resource allocation in a hierarchical tree-like federated learning network. Background Art
[0002] In traditional communications, the g Node-B (gNB) allocates communication resources based on, for example, the location of the user equipment and the measured values of each channel (e.g., signal-to-noise ratio (SNR), reference signal received power (RSRP), etc.).
[0003] Federated learning can be implemented in wireless communication networks. FIG1 shows a schematic structural diagram of an exemplary federated learning network 100 according to the prior art.
[0004] As shown in Figure 1, a federated learning network 100 may include multiple client nodes 101-1, 101-2, ..., 101-K and a server node 102. The client nodes may be user equipment (UE), and the server node may be a server or a gNB. As shown in Figure 1, multiple client nodes 101-1, 101-2, ..., 101-N communicate with server node 102 via a wireless channel.
[0005] Taking the FedAvg algorithm as an example, each client node can upload a locally learned model to the server node 102 , which aggregates the locally trained models received from each client node and then sends the aggregated model to each client node.
[0006] An exemplary process may be as follows:
[0007] 1) Each client node (101-1, 101-2, ..., or 101-K) accesses the server node 102 and obtains an initial global learning model via a downlink from the server node 102 to the corresponding client node;
[0008] 2) Each client node uses the data stored locally (such as D1, D2...D K As shown in the figure), the received initial global learning model is locally learned, and m local model iterations are completed to obtain the local model. Where m can be a preset number of iterations;
[0009] 3) Each client node transmits its learned local model to the server node 102 via an uplink from the corresponding client node Upload to server node 102;
[0010] 4) The server node 102 aggregates the local models received from each client node
[0011] in,
[0012] n k It is a natural number, which represents the number of samples used by the k-th client node when learning the local model, that is, the sample size.
[0013] 5) The server node 102 updates the global learning model W t+1 Send it to each client node again and run steps 2)-4) until the model converges.
[0014] Generally, when a server or gNB acts as a parent node and a user equipment (UE) acts as a child node, the link between the parent and child nodes can be an uplink / downlink. When both the parent and child nodes are UEs, the link between the parent and child nodes can be a sidelink.
[0015] In traditional communications, when allocating communication resources, whether uplink / downlink or sidelink, only directly connected user devices (such as child nodes directly connected to the root node) are considered, such as the data volume and location of the directly connected user devices themselves.
[0016] Summary of the Invention
[0017] This section provides a brief overview of the present disclosure in order to provide a basic understanding of some aspects of the present disclosure. However, it should be understood that this overview is not an exhaustive overview of the present disclosure. It is not intended to identify key or important parts of the present disclosure, nor is it intended to limit the scope of the present disclosure. Its purpose is simply to provide some concepts of the present disclosure in a simplified form as a prelude to the more detailed description that will be given later.
[0018] According to one aspect of the present disclosure, an electronic device is provided, comprising a processing circuit, wherein the processing circuit is configured to: obtain topology information of each of a plurality of communication nodes with the electronic device as a root node contained in a hierarchical tree-like federated learning network; and allocate communication transmission resources and / or select communication nodes participating in hierarchical tree-like federated learning based on the topology information of each of the plurality of communication nodes.
[0019] In some embodiments, the topology information of each communication node indicates at least one of the following: the total amount of data of the nodes in the subtree with the communication node as the root node; the total number of nodes in the subtree with the communication node as the root node; the number of layers of the subtree with the communication node as the root node; the number of important nodes in the subtree with the communication node as the root node; and the total importance of the nodes in the subtree with the communication node as the root node.
[0020] In some embodiments, the processing circuit is further configured to prioritize allocating communication transmission resources to a communication node and / or selecting the communication node to participate in hierarchical tree-based federated learning when a communication node meets at least one of the following conditions: the total data volume of the nodes in the subtree with the communication node as the root node exceeds a first threshold; the total number of nodes in the subtree with the communication node as the root node exceeds a second threshold; the number of layers of the subtree with the communication node as the root node exceeds a third threshold; the number of important nodes in the subtree with the communication node as the root node exceeds a fourth threshold, wherein each important node has a corresponding importance, and when the importance of a node exceeds a fifth threshold, the node is determined to be an important node; and the total importance of the nodes in the subtree with the communication node as the root node exceeds a sixth threshold.
[0021] In some embodiments, the operation of obtaining topological information of each of the multiple communication nodes contained in the hierarchical tree-like federated learning network with the electronic device as the root node further includes at least one of the following: obtaining the data amount of each node in the subtree with the communication node as the root node and determining the total data amount of the nodes in the subtree with the communication node as the root node; obtaining the total number of nodes in the subtree with the communication node as the root node; obtaining the number of layers of each node in the subtree with the communication node as the root node and determining the number of layers of the subtree with the communication node as the root node; obtaining the importance of each node in the subtree with the communication node as the root node and determining the number of important nodes in the subtree with the communication node as the root node; and obtaining the importance of each node in the subtree with the communication node as the root node and determining the total importance of the nodes in the subtree with the communication node as the root node.
[0022] In some embodiments, the amount of data for each node is based on the number of samples used by the node to participate in hierarchical tree-based federated learning.
[0023] In some embodiments, the importance of a node is based on a positional relationship between the node and the region of interest.
[0024] According to another aspect of the present disclosure, a method performed by an electronic device is provided, including: obtaining topology information of each of a plurality of communication nodes with the electronic device as a root node contained in a hierarchical tree-like federated learning network; and allocating communication transmission resources and / or selecting communication nodes participating in hierarchical tree-like federated learning based on the topology information of each of the plurality of communication nodes.
[0025] The topology information of each communication node indicates at least one of the following: the total amount of data of the nodes in the subtree with the communication node as the root node; the total number of nodes in the subtree with the communication node as the root node; the number of layers of the subtree with the communication node as the root node; the number of important nodes in the subtree with the communication node as the root node; and the total importance of the nodes in the subtree with the communication node as the root node.
[0026] When a communication node meets at least one of the following conditions, communication transmission resources are allocated to the communication node and / or the communication node is selected to participate in hierarchical tree-based federated learning on a priority basis: the total data volume of the nodes in the subtree with the communication node as the root node exceeds a first threshold; the total number of nodes in the subtree with the communication node as the root node exceeds a second threshold; the number of layers of the subtree with the communication node as the root node exceeds a third threshold; the number of important nodes in the subtree with the communication node as the root node exceeds a fourth threshold, wherein each important node has a corresponding importance, and when the importance of a node exceeds a fifth threshold, the node is determined to be an important node; and the total importance of the nodes in the subtree with the communication node as the root node exceeds a sixth threshold.
[0027] According to another aspect of the present disclosure, a computer storage medium is provided, storing computer executable instructions, which, when executed by a processor, perform the method described above. BRIEF DESCRIPTION OF THE DRAWINGS
[0028] The present disclosure may be better understood by referring to the detailed description given below in conjunction with the accompanying drawings, wherein the same or similar reference numerals are used throughout the drawings to represent the same or similar elements. All drawings, together with the following detailed description, are incorporated into and form a part of this specification and are used to further illustrate the embodiments of the present disclosure and to explain the principles and advantages of the present disclosure. Among them:
[0029] FIG1 shows a schematic structural diagram of an exemplary federated learning network according to the prior art.
[0030] FIG2 shows a schematic structural diagram of an exemplary hierarchical tree-like federated learning network according to an embodiment of the present disclosure.
[0031] FIG3 shows a flowchart of an exemplary method for node selection and / or communication resource allocation in a hierarchical tree-like federated learning network according to an embodiment of the present disclosure.
[0032] FIG4 is a block diagram illustrating a first example of a schematic configuration of a base station to which the technology of the present disclosure can be applied.
[0033] FIG5 is a block diagram illustrating a second example of a schematic configuration of a base station to which the technology of the present disclosure can be applied.
[0034] FIG. 6 is a block diagram illustrating an example of a schematic configuration of a smartphone to which the technology of the present disclosure can be applied.
[0035] FIG. 7 is a block diagram showing an example of a schematic configuration of a car navigation device to which the technology of the present disclosure can be applied.
[0036] The features and aspects of the present disclosure will be clearly understood by reading the following detailed description with reference to the accompanying drawings. DETAILED DESCRIPTION
[0037] Various exemplary embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings. For the sake of clarity and conciseness, not all implementations of the embodiments are described in this specification. However, it should be noted that when implementing the embodiments of the present disclosure, many implementation-specific settings can be made according to specific needs in order to achieve the developer's specific goals. In addition, it should be understood that although the development work may be complex and laborious, for those skilled in the art who benefit from the content of this disclosure, such development and disclosure are merely routine tasks.
[0038] In addition, it should be noted that in order to avoid obscuring the present disclosure with unnecessary details, only the processing steps and / or device structures closely related to the technical solutions of the present disclosure are shown in the accompanying drawings. The following description of the exemplary embodiments is merely illustrative and is not intended to limit the present disclosure and its applications.
[0039] This specification includes references to "one embodiment" or "an embodiment." The appearance of the phrase "in one embodiment" or "in an embodiment" does not necessarily refer to the same embodiment. The particular features, structures, or characteristics may be combined in any suitable manner consistent with the present disclosure.
[0040] The term "comprising" is open-ended. As used in the appended claims, the term does not exclude additional structures or steps. Consider a claim that states: "An apparatus comprising one or more processor units..." Such a claim does not exclude the apparatus from including additional components (e.g., a network interface unit, graphics circuitry, etc.).
[0041] Various units, circuits, or other components may be described or stated as being “configured to” perform one or more tasks. In such contexts, “configured to” is used to imply a structure (e.g., circuitry) that performs the one or more tasks during operation by indicating that the unit / circuit / component includes that structure. Thus, the unit / circuit / component may be configured to perform the task even when the specified unit / circuit / component is not currently operational (e.g., not turned on). Units / circuits / components used with the “configured to” language include hardware—e.g., circuitry, memory storing program instructions executable to implement the operation, etc. Additionally, “configured to” may include a general-purpose structure (e.g., a general-purpose circuit) manipulated by software and / or firmware (e.g., an FPGA or a general-purpose processor executing software) to operate in a manner capable of performing the one or more tasks to be solved. “Configured to” may also include adjusting a manufacturing process (e.g., a semiconductor fabrication facility) to manufacture a device (e.g., an integrated circuit) suitable for implementing or performing the one or more tasks.
[0042] In this document, the terms "first," "second," etc. serve as labels for the nouns that precede them and do not imply any type of ordering (e.g., spatial, temporal, logical, etc.).
[0043] The term "based on" is used to describe one or more factors that influence a determination. This term does not exclude additional factors that may influence the determination. That is, the determination may be based solely on these factors, or at least in part on these factors. Consider the phrase "determine A based on B." In this case, B is a factor that influences the determination of A, and the phrase does not exclude that A may also be determined based on C. In other cases, A may be determined based solely on B.
[0044] The present disclosure considers implementing hierarchical federated learning in wireless communication networks. When allocating communication transmission resources and / or selecting nodes for federated learning, topological information of the nodes below the root node (e.g., a gNB or server) is considered. Specifically, information related to the structure of the subtree rooted at each of the subtrees and / or the nodes within the subtrees, such as the total data volume of the nodes in the subtree, the total number of nodes in the subtree, the number of layers in the subtree, the number of important nodes in the subtree, and the overall importance of the nodes in the subtree, is considered. Allocating communication resources and / or selecting nodes for federated learning based on this topological information of the subtrees can achieve accurate and effective communication resource allocation and / or an efficient federated learning process.
[0045] FIG2 shows a schematic structural diagram of an exemplary hierarchical tree-like federated learning network 200 according to an embodiment of the present disclosure.
[0046] As shown in Figure 2, a hierarchical tree-like federated learning network 200 may include a root node 10 and several subordinate nodes. Root node 10 may be a server or a gNB. Subordinate nodes to root node 10 include internal nodes 12-14 and 24 and leaf nodes 11, 21-23, and 31. A leaf node is a node that is no longer connected to any child nodes. Internal nodes are all nodes other than the root node and leaf nodes.
[0047] Internal nodes can be edge servers, vehicle mounted relays (VMRs), unmanned aerial vehicles (UAVs), road side units (RSUs), etc. They can also be terminal devices such as mobile phones and pads. Leaf nodes can usually be terminal devices.
[0048] Still taking the FedAvg algorithm as an example, it is assumed that the root node 10 of the hierarchical tree-like federated learning network 200 includes three layers.
[0049] The model training process can be as follows:
[0050] 1) First, the root node (located at layer 0) sends the global initial model layer by layer to all lower-level nodes, namely, each internal node and leaf node;
[0051] 2) Each non-root node in each layer receives the model from its parent node (in the upper layer) and performs model training and model update locally based on the local sample data and the received model;
[0052] Layer 3:
[0053] 3) After node 31 in the third layer completes k3 times of local model training and updating, the local model obtained as the result Report to its (upper layer) parent node 24, where k3 can be a pre-set natural number;
[0054] Layer 2:
[0055] 4) After node 24 in the second layer completes k3 times of local model training and updating, the local model generated as the result The model reported by the corresponding node 31 Perform local aggregation to obtain a local aggregation model When the number of local aggregations is not an integer multiple of m2, node 24 will Send it to node 31, and when the number of local aggregations is equal to an integer multiple of m2, node 24 will Report to node 14, where m2 can be a pre-set natural number;
[0056] 5) At the same time, the other nodes 21, 22 and 23 in the second layer will update the local model after completing k2=m2k3 times of local model training. and Report to the corresponding upper node 12 or node 13 respectively;
[0057] Layer 1:
[0058] 6) After node 14 at the first layer completes k2 times of local model training and updating, the resulting local model The model reported by the corresponding node 24 Perform local aggregation to obtain a local aggregation model When the number of local aggregations is not an integer multiple of m1, node 14 will Send it to node 24, and when the number of local aggregations is equal to an integer multiple of m1, node 13 will Report to the root node, where m1 can be a pre-set natural number;
[0059] 7) After node 12 in the first layer completes k2 times of local model training and updating, the local model generated as a result The model reported by the corresponding nodes 21 and 22 Perform local aggregation to obtain a local aggregation model When the number of local aggregations is not an integer multiple of m1, node 12 will Send it to nodes 21 and 22, and when the number of local aggregations is equal to an integer multiple of m1, node 12 will Report to the root node;
[0060] 8) After node 13 in the first layer completes k2 times of local model training and updating, the local model generated as a result The model reported by the corresponding node 23 Perform local aggregation to obtain a local aggregation model When the number of local aggregations is not an integer multiple of m1, node 13 will Send it to node 23, and when the number of local aggregations is equal to an integer multiple of m1, node 13 will Report to the root node;
[0061] 9) At the same time, after node 11 in the first layer completes k1=m1k2=m1m2k3 times of local model training and updating, it updates the local model Report to root node 10;
[0062] 10) Root node 10 receives models from nodes 11, 12, 13, and 14 After global aggregation, the updated model is issued.
[0063] 11) Repeat steps 1)-10) until the model converges.
[0064] Those skilled in the art will appreciate that the hierarchical tree-like federated learning network 200 is merely an example, and hierarchical tree-like federated learning networks with different structures and arrangements can be designed as needed, such as hierarchical tree-like federated learning networks with more or fewer lower-level nodes and more or fewer layers. The number of times the local model of each node in each layer is trained and the number of times the local aggregation is performed can also be set differently for each node as needed. For example, the values of each m and each k in the example of FIG2 can be set differently for each node as needed.
[0065] FIG3 shows a flowchart of an exemplary method 300 for node selection and / or communication resource allocation in a hierarchical tree-like federated learning network according to an embodiment of the present disclosure.
[0066] The method 300 is executed, for example, by a root node (eg, the root node 10 in FIG. 2 ) of a hierarchical tree-like federated learning network.
[0067] As shown in FIG3 , method 300 may include operation 301 , in which topology information of each of a plurality of communication nodes in a hierarchical tree-like federated learning network with an electronic device as a root node is obtained.
[0068] The method 300 may further include operation 303 , in which communication transmission resources are allocated and / or communication nodes participating in hierarchical tree-based federated learning are selected based on topology information of each of the plurality of communication nodes.
[0069] In some embodiments, the topology information of each communication node indicates the total data volume of nodes in the subtree with the communication node as the root node. The data volume of each node can be based on the number of samples used by the node to participate in hierarchical tree-based federated learning.
[0070] For each communication node, the data volume of each node in the subtree with the communication node as the root node can be obtained and the total data volume of the nodes in the subtree with the communication node as the root node can be determined.
[0071] The analysis is conducted in conjunction with the hierarchical tree-like federated learning network 200 shown in FIG2 . For example, for nodes 11, 12, 13, and 14, the total amount of data of all nodes in the subtree with each node as the root node can be calculated as follows: N 11 =n 11 N 12 =n12 +n 21 +n 22 N 13 =n 13 +n 23 N 14 =n 14 +n 24 +n 31
[0072] For node 11, N 11 is the total amount of data in the subtree with node 11 as the root node, n 11 is the data volume of node 11 itself. As shown in FIG2 , the subtree with node 11 as the root node only includes node 11 itself. Therefore, the total data volume of the subtree with node 11 as the root node is the data volume of node 11 itself.
[0073] For node 12, as shown in FIG2 , the subtree with node 12 as the root node includes three nodes, namely nodes 12, 21 and 22. Therefore, the total data volume N of the subtree with node 12 as the root node is 12 is the data volume n of the three nodes 12, 21 and 22 themselves 12 、n 21 and n 22 sum.
[0074] Similarly, the total data volume N of the subtree with node 13 as the root node 13 is the data volume n of nodes 13 and 23 themselves 13 and n 23 The total amount of data N in the subtree with node 14 as the root node 14 is the data volume n of nodes 14, 24, and 31 themselves 14 、n 24 and n 31 sum.
[0075] You can choose {N 11 ,N 12 ,N 13 ,N 14}Nodes with larger median values are selected as nodes to participate in federated learning, so as to ensure that the final global aggregated model is trained based on the largest number of data samples. 11 ,N 12 ,N 13 ,N 14}Communication resources are allocated to nodes with larger median values to achieve accurate and effective allocation of communication resources.
[0076] For example, a first threshold value related to the total data volume may be set. When the total data volume of nodes in a subtree with a communication node as the root node exceeds the first threshold value, communication transmission resources may be preferentially allocated to the communication node and / or the communication node may be selected to participate in hierarchical tree-based federated learning.
[0077] In some embodiments, the topology information of each communication node indicates the total number of nodes in a subtree with the communication node as the root node. For each communication node, the total number of nodes in the subtree with the communication node as the root node can be obtained. A second threshold associated with the total number of nodes can be set. When the total number of nodes in the subtree with a communication node as the root node exceeds the second threshold, communication transmission resources can be preferentially allocated to the communication node and / or the communication node can be selected to participate in hierarchical tree-based federated learning.
[0078] Still referring to FIG2 , for example, for nodes 11, 12, 13 and 14, the total number of nodes M in the subtree with each node as the root node is obtained respectively. 11 、M 12 、M 13 and M 14 As follows: M 11 =1 M 12 =3 M 13 =2 M 14 =3
[0079] You can choose {M 11 ,M 12 ,M 13 ,M 14 Nodes with larger median values are selected as nodes participating in federated learning, because a larger number of nodes often corresponds to better data diversity. 11 ,N 12 ,N 13 ,N 14}Communication resources are allocated to nodes with larger median values to achieve accurate and effective allocation of communication resources.
[0080] Taking nodes 11 and 12 as an example, if node 11 is prioritized or communication with node 11 is guaranteed, only the model trained with node 11's data can be obtained. However, the model uploaded by node 12 after aggregation contains information from nodes 21, 22, and 12. Therefore, it is beneficial to prioritize node 12 for federated learning and / or prioritize communication resource allocation for node 12.
[0081] In some embodiments, the topology information of each communication node indicates the number of layers of a subtree with the communication node as the root node. For each communication node, the number of layers of each node in the subtree with the communication node as the root node can be obtained and the number of layers of the subtree with the communication node as the root node can be determined. A third threshold associated with the number of layers of the subtree with the communication node as the root node can be set. When the number of layers of the subtree with a communication node as the root node exceeds the third threshold, communication transmission resources can be preferentially allocated to the communication node and / or the communication node can be selected to participate in hierarchical tree-based federated learning.
[0082] Still in conjunction with Figure 2, the number of layers L of the subtree with each communication node as the root node can be determined 11 =1 L 12 =2 L 13 =2 L 14 =3
[0083] For example, for nodes 12 and 14, the total number of nodes in the subtree rooted at each node is the same: 3. However, the subtree involved in node 14 has more layers, and the uploaded model is often associated with richer data. For example, nodes 14 and 12 can be edge servers, nodes 21, 22, and 24 can be VMRs, and node 31 can be a small device such as a mobile phone. In this case, the subtree involved in node 14 may be associated with data from three different devices. Therefore, the model uploaded by node 14 may be associated with richer data. It is beneficial to prioritize node 14 for participating in federated learning and / or prioritize the allocation of communication resources to node 14.
[0084] In some embodiments, the topology information of each communication node indicates the number of important nodes in a subtree with the communication node as the root node. A fourth threshold associated with the number of important nodes can be set. When the number of important nodes in a subtree with a communication node as the root node exceeds the fourth threshold, communication transmission resources can be preferentially allocated to the communication node and / or the communication node can be selected to participate in hierarchical tree-based federated learning.
[0085] Each node may have a corresponding importance. A fifth threshold associated with the importance of the node may be set, and when the importance of a node exceeds the fifth threshold, the node may be determined as an important node.
[0086] The importance of each node in a subtree with a communication node as the root node can be obtained and the number of important nodes in the subtree with the communication node as the root node can be determined, for example, the number of nodes in the subtree with the communication node as the root node whose importance exceeds the fifth threshold.
[0087] The importance of a node can be based on the positional relationship between the node and the area of interest. For example, assume that the importance of a node is represented by 1-5 and is associated with the positional relationship between the node and the area of interest. If it is determined that the node is located in the area of interest, the importance of the node can be determined to be 5. If it is determined that the node is not in the area of interest but is near it (such as within a predetermined distance), the importance of the node can be determined to be 4. If it is determined that the node is not in the area of interest and is far away from the area of interest (such as exceeding a predetermined distance), the importance of the node can be determined to be 1. The fifth threshold can be set to 3, and when the importance of the node exceeds 3, the node can be determined to be an important node.
[0088] Those skilled in the art will appreciate that various factors may determine the importance of a node, and are not limited to the location relationship between the node and the region of interest. For example, the importance of a node may be associated with the type of node. For example, a node serving as an edge server may have a higher importance, while a node serving as a user device may have a lower importance.
[0089] In some other embodiments, the topology information of each communication node indicates the total importance of nodes in a subtree with the communication node as the root node.
[0090] The importance of each node in a subtree with a communication node as the root node can be obtained and the total importance of the nodes in the subtree with the communication node as the root node can be determined. For example, the importance of all nodes in the subtree with the communication node as the root node can be added to obtain the total importance of the communication node.
[0091] A sixth threshold associated with the total importance may be set. When the total importance of nodes in a subtree with a communication node as the root node exceeds the sixth threshold, communication transmission resources may be preferentially allocated to the communication node and / or the communication node may be selected to participate in hierarchical tree-based federated learning.
[0092] Those skilled in the art will understand that the topology information of each communication node may indicate one or more of the total amount of data of the nodes in the subtree with the communication node as the root node, the total number of nodes in the subtree with the communication node as the root node, the number of layers of the subtree with the communication node as the root node, the number of important nodes in the subtree with the communication node as the root node, or the total importance of the nodes in the subtree with the communication node as the root node.
[0093] The topology information of each communication node may also indicate other information related to the topology of the subtree with the communication node as the root node. For example, in addition to indicating the number of layers of the subtree with the communication node as the root node, the topology information of each communication node may also indicate the layer number of each node in the subtree with the communication node as the root node. The topology information of each communication node may also indicate the priority or weight of each factor considered. For example, the total amount of data of the nodes in the subtree with the communication node as the root node will be given priority over the total number of nodes in the subtree with the communication node as the root node. Alternatively, the total amount of data of the nodes in the subtree with the communication node as the root node may be given a higher consideration weight.
[0094] Those skilled in the art will appreciate that communication resource allocation may include allocating communication resources such as time domain resources, frequency domain resources, etc. to uplinks, downlinks, and sidelinks.
[0095] The disclosed embodiments can effectively select nodes with large sample data volumes or high sample diversity to participate in hierarchical federated learning based on topological information related to subtrees with lower-level nodes as root nodes, thereby facilitating the efficient implementation of hierarchical federated learning in wireless communication networks, and enabling the resulting model to better adapt to the actual conditions of the communication network. The disclosed embodiments can more accurately allocate communication transmission resources based on topological information related to subtrees with lower-level nodes as root nodes, effectively improving resource utilization.
[0096] The techniques described in this disclosure can be applied to a variety of products.
[0097] For example, the electronic device according to the embodiments of the present disclosure may be implemented as various base stations or installed in a base station, or may be implemented as various user equipments or installed in various user equipments.
[0098] The communication method according to the embodiments of the present disclosure can be implemented by various base stations or user equipment; the methods and operations according to the embodiments of the present disclosure can be embodied as computer-executable instructions, stored in a non-temporary computer-readable storage medium, and can be executed by various base stations or user equipment to implement one or more functions described above.
[0099] The technology according to the embodiments of the present disclosure can be made into various computer program products, which can be used in various base stations or user equipments to implement one or more functions described above.
[0100] The base station referred to in this disclosure may be implemented as any type of base station, preferably, such as the macro gNB and ng-eNB defined in the 3GPP 5GNR standard. The gNB may be a gNB that covers cells smaller than a macro cell, such as a pico gNB, micro gNB, and home (femto) gNB. Alternatively, the base station may be implemented as any other type of base station, such as a NodeB, eNodeB, and base transceiver station (BTS). The base station may also include: a subject configured to control wireless communications, and one or more remote radio heads (RRHs) located separately from the subject, a wireless relay station, a drone tower, a control node in an automated factory, and the like.
[0101] The user equipment can be implemented as a mobile terminal (such as a smartphone, a tablet personal computer (PC), a notebook PC, a portable game terminal, a portable / dongle-type mobile router, and a digital camera) or an in-vehicle terminal (such as a car navigation device). The user equipment can also be implemented as a terminal that performs machine-to-machine (M2M) communication (also known as a machine-type communication (MTC) terminal), a drone, a sensor and actuator in an automated factory, etc. In addition, the user equipment can be a wireless communication module (such as an integrated circuit module including a single chip) installed on each of the above terminals.
[0102] The following briefly introduces examples of base stations and user equipment to which the technology of the present disclosure can be applied.
[0103] It should be understood that the term "base station" used in this disclosure has the full breadth of its usual meaning and includes at least a wireless communication station used as part of a wireless communication system or radio system to facilitate communication. Examples of base stations may be, for example, but not limited to, the following: one or both of a base transceiver station (BTS) and a base station controller (BSC) in a GSM communication system; one or both of a radio network controller (RNC) and a NodeB in a 3G communication system; an eNB in a 4G LTE and LTE-A system; and gNB and ng-eNB in a 5G communication system. In D2D, M2M, and V2V communication scenarios, a logical entity that has a control function for communication may also be referred to as a base station. In a cognitive radio communication scenario, a logical entity that plays a role in spectrum coordination may also be referred to as a base station. In an automated factory, a logical entity that provides network control functions may be referred to as a base station.
[0104] First application example of base station
[0105] FIG4 is a block diagram illustrating a first example of a schematic configuration of a base station to which the techniques of the present disclosure may be applied. In FIG4 , the base station may be implemented as gNB 1400. gNB 1400 includes multiple antennas 1410 and a base station device 1420. Base station device 1420 and each antenna 1410 may be connected to each other via an RF cable.
[0106] Antenna 1410 includes multiple antenna elements, such as multiple antenna arrays for massive MIMO. Antenna 1410 can be arranged in a matrix antenna array, for example, and used by base station device 1420 to transmit and receive wireless signals. For example, multiple antennas 1410 can be compatible with multiple frequency bands used by gNB 1400.
[0107] The base station device 1420 includes a controller 1421 , a memory 1422 , a network interface 1423 , and a wireless communication interface 1425 .
[0108] The controller 1421 may be, for example, a CPU or DSP, and operates various higher-layer functions of the base station device 1420. For example, the controller 1421 generates data packets based on the data in the signal processed by the wireless communication interface 1425 and transmits the generated packets via the network interface 1423. The controller 1421 may bundle data from multiple baseband processors to generate bundled packets and transmit the generated bundled packets. The controller 1421 may have logic functions for performing control such as radio resource control, radio bearer control, mobility management, admission control, and scheduling. This control may be performed in conjunction with a nearby gNB or core network node. The memory 1422 includes RAM and ROM and stores programs executed by the controller 1421 and various types of control data (such as terminal lists, transmission power data, and scheduling data).
[0109] Network interface 1423 is a communication interface for connecting base station device 1420 to core network 1424 (e.g., 5G core network). Controller 1421 can communicate with a core network node or another gNB via network interface 1423. In this case, gNB 1400 and the core network node or other gNB can be connected to each other via logical interfaces (such as NG interfaces and Xn interfaces). Network interface 1423 can also be a wired communication interface or a wireless communication interface for wireless backhaul. If network interface 1423 is a wireless communication interface, network interface 1423 can use a higher frequency band for wireless communication than the frequency band used by wireless communication interface 1425.
[0110] The wireless communication interface 1425 supports any cellular communication scheme (such as 5G NR) and provides wireless connectivity to terminals located in the cell of the gNB 1400 via the antenna 1410. The wireless communication interface 1425 may typically include, for example, a baseband (BB) processor 1426 and RF circuitry 1427. The BB processor 1426 can perform, for example, encoding / decoding, modulation / demodulation, and multiplexing / demultiplexing, and various types of signal processing at various layers (e.g., the physical layer, MAC layer, RLC layer, PDCP layer, and SDAP layer). In place of the controller 1421, the BB processor 1426 may perform some or all of the aforementioned logical functions. The BB processor 1426 may be a memory that stores communication control programs, or a module including a processor configured to execute programs and associated circuitry. Program updates can modify the functionality of the BB processor 1426. This module may be a card or blade inserted into a slot in the base station device 1420. Alternatively, it may be a chip mounted on the card or blade. Meanwhile, the RF circuit 1427 may include, for example, a mixer, a filter, and an amplifier, and transmits and receives wireless signals via the antenna 1410. Although FIG4 shows an example in which one RF circuit 1427 is connected to one antenna 1410, the present disclosure is not limited to this illustration, and one RF circuit 1427 may be connected to multiple antennas 1410 at the same time.
[0111] As shown in Figure 4 , the wireless communication interface 1425 may include multiple BB processors 1426. For example, multiple BB processors 1426 may be compatible with multiple frequency bands used by gNB 1400. As shown in Figure 4 , the wireless communication interface 1425 may include multiple RF circuits 1427. For example, multiple RF circuits 1427 may be compatible with multiple antenna elements. While Figure 4 illustrates an example in which the wireless communication interface 1425 includes multiple BB processors 1426 and multiple RF circuits 1427, the wireless communication interface 1425 may also include a single BB processor 1426 or a single RF circuit 1427.
[0112] In the gNB 1400 shown in FIG4 , one or more units included in the processing circuits 1001, 2001, 3001, or 4001 (e.g., the transmitting unit 1003, the receiving unit 2002, the receiving unit 3003, etc.) may be implemented in the wireless communication interface 1425. Alternatively, at least a portion of these components may be implemented in the controller 1421. For example, the gNB 1400 may include a portion (e.g., the BB processor 1426) or the entirety of the wireless communication interface 1425 and / or include a module of the controller 1421, and one or more components may be implemented in the module. In this case, the module may store a program that allows the processor to function as one or more components (in other words, a program that allows the processor to perform the operations of one or more components) and may execute the program. As another example, the program that allows the processor to function as one or more components may be installed in the gNB 1400, and the wireless communication interface 1425 (e.g., the BB processor 1426) and / or the controller 1421 may execute the program. As described above, the gNB 1400, base station device 1420, or module may be provided as a device including one or more components, and a program for allowing a processor to function as one or more components may be provided. In addition, a readable medium having the program recorded therein may be provided.
[0113] Second application example of base station
[0114] FIG5 is a block diagram illustrating a second example of a schematic configuration of a base station to which the techniques of this disclosure can be applied. FIG5 illustrates a base station as a gNB 1530. gNB 1530 includes multiple antennas 1540, a base station device 1550, and an RRH 1560. RRH 1560 and each antenna 1540 can be connected to each other via an RF cable. Base station device 1550 and RRH 1560 can be connected to each other via a high-speed line, such as an optical fiber cable.
[0115] Antenna 1540 includes multiple antenna elements, such as multiple antenna arrays for massive MIMO. Antenna 1540 can be arranged in a matrix antenna array, for example, and used by base station device 1550 to transmit and receive wireless signals. For example, multiple antennas 1540 can be compatible with multiple frequency bands used by gNB 1530.
[0116] Base station device 1550 includes a controller 1551, a memory 1552, a network interface 1553, a wireless communication interface 1555, and a connection interface 1557. Controller 1551, memory 1552, and network interface 1553 are the same as controller 1421, memory 1422, and network interface 1423 described with reference to FIG.
[0117] The wireless communication interface 1555 supports any cellular communication scheme (such as 5G NR) and provides wireless communication to terminals located in the sector corresponding to the RRH 1560 via the RRH 1560 and the antenna 1540. The wireless communication interface 1555 may generally include, for example, a BB processor 1556. The BB processor 1556 is identical to the BB processor 1426 described with reference to FIG. 6 , except that the BB processor 1556 is connected to the RF circuit 1564 of the RRH 1560 via the connection interface 1557. As shown in FIG. 5 , the wireless communication interface 1555 may include multiple BB processors 1556. For example, multiple BB processors 1556 may be compatible with multiple frequency bands used by the gNB 1530. Although FIG. 5 illustrates an example in which the wireless communication interface 1555 includes multiple BB processors 1556, the wireless communication interface 1555 may also include a single BB processor 1556.
[0118] The connection interface 1557 is an interface for connecting the base station device 1550 (wireless communication interface 1555) to the RRH 1560. The connection interface 1557 may also be a communication module for connecting the base station device 1550 (wireless communication interface 1555) to the RRH 1560 for communication in the high-speed line.
[0119] The RRH 1560 includes a connection interface 1561 and a wireless communication interface 1563 .
[0120] The connection interface 1561 is an interface for connecting the RRH 1560 (wireless communication interface 1563) to the base station device 1550. The connection interface 1561 may also be a communication module for communication in the above-mentioned high-speed line.
[0121] The wireless communication interface 1563 transmits and receives wireless signals via the antenna 1540. The wireless communication interface 1563 may generally include, for example, an RF circuit 1564. The RF circuit 1564 may include, for example, a mixer, a filter, and an amplifier, and transmits and receives wireless signals via the antenna 1540. Although FIG. 5 shows an example in which one RF circuit 1564 is connected to one antenna 1540, the present disclosure is not limited to this illustration, and one RF circuit 1564 may be connected to multiple antennas 1540 simultaneously.
[0122] As shown in FIG5 , the wireless communication interface 1563 may include multiple RF circuits 1564. For example, the multiple RF circuits 1564 may support multiple antenna elements. Although FIG5 shows an example in which the wireless communication interface 1563 includes multiple RF circuits 1564, the wireless communication interface 1563 may also include a single RF circuit 1564.
[0123] In the gNB 1500 shown in FIG5 , one or more units included in the processing circuits 1001, 2001, 3001, or 4001 (e.g., the transmitting unit 1003, the receiving unit 2002, the receiving unit 3003, etc.) may be implemented in the wireless communication interface 1525. Alternatively, at least a portion of these components may be implemented in the controller 1521. For example, the gNB 1500 may include a portion (e.g., the BB processor 1526) or the entirety of the wireless communication interface 1525 and / or include a module of the controller 1521, and one or more components may be implemented in the module. In this case, the module may store a program that allows the processor to function as one or more components (in other words, a program that allows the processor to perform the operations of one or more components) and may execute the program. As another example, the program that allows the processor to function as one or more components may be installed in the gNB 1500, and the wireless communication interface 1525 (e.g., the BB processor 1526) and / or the controller 1521 may execute the program. As described above, gNB 1500, base station device 1520, or a module may be provided as a device including one or more components, and a program for allowing a processor to function as one or more components may be provided. In addition, a readable medium having the program recorded therein may be provided.
[0124] First application example of user equipment
[0125] FIG. 6 is a block diagram illustrating an example of a schematic configuration of a smartphone 1600 to which the technology of the present disclosure can be applied.
[0126] The smart phone 1600 includes a processor 1601, a memory 1602, a storage device 1603, an external connection interface 1604, a camera 1606, a sensor 1607, a microphone 1608, an input device 1609, a display device 1610, a speaker 1611, a wireless communication interface 1612, one or more antenna switches 1615, one or more antennas 1616, a bus 1617, a battery 1618 and an auxiliary controller 1619.
[0127] The processor 1601 may be, for example, a CPU or a system on a chip (SoC), and controls the functions of the application layer and other layers of the smartphone 1600. The processor 1601 may include or function as any of the processing circuits 1001, 2001, 3001, and 4001 described with reference to the drawings. The memory 1602 includes RAM and ROM, and stores data and programs executed by the processor 1601. The storage device 1603 may include storage media such as semiconductor memories and hard disks. The external connection interface 1604 is an interface for connecting external devices (such as memory cards and universal serial bus (USB) devices) to the smartphone 1600.
[0128] The camera 1606 includes an image sensor (such as a charge coupled device (CCD) and a complementary metal oxide semiconductor (CMOS)) and generates a captured image. The sensor 1607 may include a group of sensors such as a measurement sensor, a gyroscope sensor, a geomagnetic sensor, and an acceleration sensor. The microphone 1608 converts the sound input to the smartphone 1600 into an audio signal. The input device 1609 includes, for example, a touch sensor, a keypad, a keyboard, a button, or a switch configured to detect a touch on the screen of the display device 1610, and receives an operation or information input from the user. The display device 1610 includes a screen (such as a liquid crystal display (LCD) and an organic light emitting diode (OLED) display) and displays the output image of the smartphone 1600. The speaker 1611 converts the audio signal output from the smartphone 1600 into sound.
[0129] The wireless communication interface 1612 supports any cellular communication scheme (such as 4G LTE or 5G NR, etc.) and performs wireless communication. The wireless communication interface 1612 may generally include, for example, a BB processor 1613 and an RF circuit 1614. The BB processor 1613 may perform, for example, encoding / decoding, modulation / demodulation, and multiplexing / demultiplexing, and perform various types of signal processing for wireless communication. Meanwhile, the RF circuit 1614 may include, for example, a mixer, a filter, and an amplifier, and transmit and receive wireless signals via an antenna 1616. The wireless communication interface 1612 may be a chip module on which the BB processor 1613 and the RF circuit 1614 are integrated. As shown in FIG6 , the wireless communication interface 1612 may include multiple BB processors 1613 and multiple RF circuits 1614. Although FIG6 shows an example in which the wireless communication interface 1612 includes multiple BB processors 1613 and multiple RF circuits 1614, the wireless communication interface 1612 may also include a single BB processor 1613 or a single RF circuit 1614.
[0130] In addition, in addition to the cellular communication scheme, the wireless communication interface 1612 can support other types of wireless communication schemes, such as a short-range wireless communication scheme, a near field communication scheme, and a wireless local area network (LAN) scheme. In this case, the wireless communication interface 1612 may include a BB processor 1613 and an RF circuit 1614 for each wireless communication scheme.
[0131] Each of the antenna switches 1615 switches the connection destination of the antenna 1616 between a plurality of circuits (eg, circuits for different wireless communication schemes) included in the wireless communication interface 1612 .
[0132] Antenna 1616 includes multiple antenna elements, such as multiple antenna arrays for massive MIMO. Antenna 1616 can be arranged in an antenna array matrix, for example, and is used for wireless communication interface 1612 to transmit and receive wireless signals. Smartphone 1600 may include one or more antenna panels (not shown).
[0133] In addition, the smartphone 1600 may include an antenna 1616 for each wireless communication scheme. In this case, the antenna switch 1615 may be omitted from the configuration of the smartphone 1600.
[0134] The bus 1617 connects the processor 1601, the memory 1602, the storage device 1603, the external connection interface 1604, the camera 1606, the sensor 1607, the microphone 1608, the input device 1609, the display device 1610, the speaker 1611, the wireless communication interface 1612, and the auxiliary controller 1619. The battery 1618 supplies power to the various blocks of the smartphone 1600 shown in FIG6 via feeders, which are partially shown as dashed lines in the figure. The auxiliary controller 1619 operates the minimum necessary functions of the smartphone 1600, for example, in sleep mode.
[0135] In the smartphone 1600 shown in FIG6 , one or more units included in the processing circuits 1001, 2001, 3001, or 4001 (e.g., the transmitting unit 1003, the receiving unit 2002, the receiving unit 3003, etc.) may be implemented in the wireless communication interface 1612. Alternatively, at least a portion of these components may be implemented in the processor 1601 or the auxiliary controller 1619. As an example, the smartphone 1600 includes a portion (e.g., the BB processor 1613) or the entirety of the wireless communication interface 1612, and / or a module including the processor 1601 and / or the auxiliary controller 1619, and one or more components may be implemented in the module. In this case, the module may store a program that allows the processor to function as one or more components (in other words, a program for allowing the processor to perform the operations of one or more components) and may execute the program. As another example, a program for allowing the processor to function as one or more components may be installed in the smartphone 1600, and the wireless communication interface 1612 (e.g., BB processor 1613), the processor 1601, and / or the auxiliary controller 1619 may execute the program. As described above, the smartphone 1600 or module may be provided as a device including one or more components, and a program for allowing the processor to function as one or more components may be provided. In addition, a readable medium having the program recorded therein may be provided.
[0136] Second application example of user equipment
[0137] 7 is a block diagram showing an example of a schematic configuration of a car navigation device 1720 to which the technology of the present disclosure can be applied. Car navigation device 1720 includes a processor 1721, a memory 1722, a global positioning system (GPS) module 1724, a sensor 1725, a data interface 1726, a content player 1727, a storage medium interface 1728, an input device 1729, a display device 1730, a speaker 1731, a wireless communication interface 1733, one or more antenna switches 1736, one or more antennas 1737, and a battery 1738.
[0138] The processor 1721 may be, for example, a CPU or an SoC, and controls a navigation function and other functions of the car navigation device 1720. The memory 1722 includes a RAM and a ROM, and stores data and programs executed by the processor 1721.
[0139] The GPS module 1724 uses GPS signals received from GPS satellites to measure the position (such as latitude, longitude, and altitude) of the car navigation device 1720. The sensor 1725 may include a group of sensors such as a gyroscope sensor, a geomagnetic sensor, and an air pressure sensor. The data interface 1726 is connected to, for example, the vehicle network 1741 via a terminal not shown, and obtains data generated by the vehicle (such as vehicle speed data).
[0140] The content player 1727 reproduces content stored in a storage medium (such as a CD or DVD) inserted into the storage medium interface 1728. The input device 1729 includes, for example, a touch sensor, button, or switch configured to detect a touch on the screen of the display device 1730, and receives operations or information input from the user. The display device 1730 includes a screen such as an LCD or OLED display and displays images of the navigation function or reproduced content. The speaker 1731 outputs sounds of the navigation function or reproduced content.
[0141] The wireless communication interface 1733 supports any cellular communication scheme (such as 4G LTE or 5G NR) and performs wireless communication. The wireless communication interface 1733 may generally include, for example, a BB processor 1734 and an RF circuit 1735. The BB processor 1734 may perform, for example, encoding / decoding, modulation / demodulation, and multiplexing / demultiplexing, and perform various types of signal processing for wireless communication. Meanwhile, the RF circuit 1735 may include, for example, a mixer, a filter, and an amplifier, and transmit and receive wireless signals via an antenna 1737. The wireless communication interface 1733 may also be a chip module on which the BB processor 1734 and the RF circuit 1735 are integrated. As shown in FIG7 , the wireless communication interface 1733 may include multiple BB processors 1734 and multiple RF circuits 1735. Although FIG7 shows an example in which the wireless communication interface 1733 includes multiple BB processors 1734 and multiple RF circuits 1735, the wireless communication interface 1733 may also include a single BB processor 1734 or a single RF circuit 1735.
[0142] In addition, in addition to the cellular communication scheme, the wireless communication interface 1733 can support other types of wireless communication schemes, such as short-range wireless communication schemes, near field communication schemes, and wireless LAN schemes. In this case, for each wireless communication scheme, the wireless communication interface 1733 can include a BB processor 1734 and an RF circuit 1735.
[0143] Each of the antenna switches 1736 switches a connection destination of the antenna 1737 between a plurality of circuits included in the wireless communication interface 1733 , such as circuits for different wireless communication schemes.
[0144] The antenna 1737 includes multiple antenna elements, such as multiple antenna arrays for massive MIMO, and can be arranged into an antenna array matrix, for example, and used for the wireless communication interface 1733 to transmit and receive wireless signals.
[0145] In addition, the car navigation device 1720 may include an antenna 1737 for each wireless communication scheme. In this case, the antenna switch 1736 may be omitted from the configuration of the car navigation device 1720.
[0146] The battery 1738 supplies power to the respective blocks of the car navigation device 1720 shown in Fig. 7 via a feeder line, which is partially shown as a dotted line in the figure. The battery 1738 accumulates the power supplied from the vehicle.
[0147] In the car navigation device 1720 shown in FIG7 , one or more units included in the processing circuits 1001, 2001, 3001, or 4001 (e.g., the transmitting unit 1003, the receiving unit 2002, the receiving unit 3003, etc.) may be implemented in the wireless communication interface 1733. Alternatively, at least a portion of these components may be implemented in the processor 1721. As an example, the car navigation device 1720 includes a portion (e.g., the BB processor 1734) or the entirety of the wireless communication interface 1733, and / or includes a module of the processor 1721, and one or more components may be implemented in the module. In this case, the module may store a program that allows the processor to function as one or more components (in other words, a program for allowing the processor to perform the operations of one or more components) and may execute the program. As another example, a program for allowing the processor to function as one or more components may be installed in the car navigation device 1720, and the wireless communication interface 1733 (e.g., the BB processor 1734) and / or the processor 1721 may execute the program. As described above, the car navigation device 1720 or module may be provided as a device including one or more components, and a program for allowing a processor to function as one or more components may be provided. In addition, a readable medium having the program recorded therein may be provided.
[0148] The technology of the present disclosure can also be implemented as an in-vehicle system (or vehicle) 1740 including a car navigation device 1720, an in-vehicle network 1741, and one or more blocks of a vehicle module 1742. The vehicle module 1742 generates vehicle data (such as vehicle speed, engine speed, and fault information) and outputs the generated data to the in-vehicle network 1741.
[0149] The exemplary embodiments of the present disclosure are described above with reference to the accompanying drawings, but the present disclosure is certainly not limited to the above examples. Those skilled in the art may obtain various changes and modifications within the scope of the appended claims, and it should be understood that these changes and modifications will naturally fall within the technical scope of the present disclosure.
[0150] For example, a plurality of functions included in one unit in the above embodiments may be implemented by separate devices. Alternatively, a plurality of functions implemented by a plurality of units in the above embodiments may be implemented by separate devices, respectively. In addition, one of the above functions may be implemented by a plurality of units. Needless to say, such a configuration is included in the technical scope of the present disclosure.
[0151] In this specification, the steps described in the flowchart include not only processing executed in time series in the order described, but also processing executed in parallel or individually rather than necessarily in time series. In addition, even in the steps processed in time series, it goes without saying that the order can be changed as appropriate.
[0152] Although the present disclosure and its advantages have been described in detail, it should be understood that various changes, substitutions and transformations can be made without departing from the spirit and scope of the present disclosure as defined by the appended claims. Moreover, the terms "comprises," "comprising," or any other variations thereof in the embodiments of the present disclosure are intended to cover non-exclusive inclusions, such that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article, or device. In the absence of further restrictions, an element defined by the statement "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or device comprising the element.
Claims
1. An electronic device, comprising: A processing circuit, wherein the processing circuit is configured to: Acquire topology information of each of a plurality of communication nodes in a hierarchical tree-like federated learning network with the electronic device as a root node; and According to the topology information of each of the plurality of communication nodes, communication transmission resources are allocated and / or communication nodes participating in hierarchical tree-like federated learning are selected.
2. The electronic device according to claim 1, wherein: The topology information of each communication node indicates at least one of the following: The total amount of data of nodes in the subtree with the communication node as the root node; The total number of nodes in the subtree with the communication node as the root node; The number of layers of the subtree with the communication node as the root node; The number of important nodes in the subtree with the communication node as the root node; and The total importance of the nodes in the subtree with the communication node as the root node.
3. The electronic device according to claim 2, wherein: The processing circuit is further configured to preferentially allocate communication transmission resources to a communication node and / or select the communication node to participate in hierarchical tree-like federated learning when a communication node satisfies at least one of the following conditions: The total amount of data of the nodes in the subtree with the communication node as the root node exceeds a first threshold; The total number of nodes in the subtree with the communication node as the root node exceeds a second threshold; The number of layers of the subtree with the communication node as the root node exceeds a third threshold; The number of important nodes in the subtree with the communication node as the root node exceeds a fourth threshold, wherein each important node has a corresponding importance, and when the importance of a node exceeds a fifth threshold, the node is determined to be an important node; and The total importance of the nodes in the subtree with the communication node as the root node exceeds a sixth threshold.
4. The electronic device according to claim 2, wherein: The operation of obtaining topology information of each of a plurality of communication nodes in the hierarchical tree-shaped federated learning network with the electronic device as a root node further includes at least one of the following: Acquire the data volume of each node in the subtree with the communication node as the root node and determine the total data volume of the nodes in the subtree with the communication node as the root node; Obtain the total number of nodes in the subtree with the communication node as the root node; Obtaining the number of layers of each node in the subtree with the communication node as the root node and determining the number of layers of the subtree with the communication node as the root node; Obtaining the importance of each node in the subtree with the communication node as the root node and determining the number of important nodes in the subtree with the communication node as the root node; and The importance of each node in the subtree with the communication node as the root node is obtained, and the total importance of the nodes in the subtree with the communication node as the root node is determined.
5. The electronic device according to claim 3 or 4, wherein: The amount of data for each node is based on the number of samples that the node uses to participate in hierarchical tree-based federated learning.
6. The electronic device according to claim 3 or 4, wherein: The importance of a node is based on the positional relationship between the node and the region of interest.
7. A method performed by an electronic device, comprising: Acquire topology information of each of a plurality of communication nodes in a hierarchical tree-like federated learning network with the electronic device as a root node; as well as According to the topology information of each of the plurality of communication nodes, communication transmission resources are allocated and / or communication nodes participating in hierarchical tree-like federated learning are selected.
8. The method of claim 7, wherein: The topology information of each communication node indicates at least one of the following: The total amount of data of nodes in the subtree with the communication node as the root node; The total number of nodes in the subtree with the communication node as the root node; The number of layers of the subtree with the communication node as the root node; The number of important nodes in the subtree with the communication node as the root node; and The total importance of the nodes in the subtree with the communication node as the root node.
9. The method of claim 8, further comprising, when a communication node satisfies at least one of the following conditions, preferentially allocating communication transmission resources to the communication node and / or selecting the communication node to participate in hierarchical tree-like federated learning: The total amount of data of the nodes in the subtree with the communication node as the root node exceeds a first threshold; The total number of nodes in the subtree with the communication node as the root node exceeds a second threshold; The number of layers of the subtree with the communication node as the root node exceeds a third threshold; and The number of important nodes in the subtree with the communication node as the root node exceeds a fourth threshold, wherein each important node has a corresponding importance, and when the importance of a node exceeds a fifth threshold, the node is determined to be an important node; and The total importance of the nodes in the subtree with the communication node as the root node exceeds a sixth threshold.
10. The electronic device according to claim 8, wherein: The operation of obtaining topology information of each of a plurality of communication nodes in the hierarchical tree-shaped federated learning network with the electronic device as a root node further includes at least one of the following: Acquire the data volume of each node in the subtree with the communication node as the root node and determine the total data volume of the nodes in the subtree with the communication node as the root node; Obtain the total number of nodes in the subtree with the communication node as the root node; Obtaining the number of layers of each node in the subtree with the communication node as the root node and determining the number of layers of the subtree with the communication node as the root node; Obtaining the importance of each node in the subtree with the communication node as the root node and determining the number of important nodes in the subtree with the communication node as the root node; and The importance of each node in the subtree with the communication node as the root node is obtained, and the total importance of the nodes in the subtree with the communication node as the root node is determined.
11. The electronic device according to claim 9 or 10, wherein: The amount of data for each node is based on the number of samples that the node uses to participate in hierarchical tree-based federated learning.
12. The electronic device according to claim 9 or 10, wherein: The importance of a node is based on the positional relationship between the node and the region of interest.
13. A computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processing circuit of an electronic device, the computer program performs the method according to any one of claims 7 to 12.
Citation Information
Patent Citations
Federated learning method and device based on tree topological structure
CN110728376A
Federal learning system deployed in edge computing network and learning method thereof
CN113010305A
Model determination system and method, and computer readable storage medium
CN114785731A
Calibration method and wireless communication device for achieving channel reciprocity
US20190132208A1
