Abstract:
The present invention relates to a method and apparatus where a reinforcement learning agent ensures quality of an initial control operation of an environment on the basis of reinforcement learning, wherein a first action calculated by using an algorithm is selected at an initial learning stage, and a second action calculated by using a Q function is selected when the initial learning stage is ended.
Abstract:
A service stream control system includes at least one area controller configured to control a traffic path and content cashing within a domain (intra-area), a service content coordinator configured to determine and manage a caching location of content located in the domain, and a service stream controller configured to set an optimum path in accordance with a network traffic state by interworking with the at least one area controller and the service content coordinator, and forward target content corresponding to a content request with reference to a service stream forwarding graph generated based on the set path. Therefore, it is possible to enhance efficiency of a content-centric network (CCN).
Abstract:
The present invention related to a method for federated learning method interworking with a mobile core system, the method comprising: querying terminal information of each individual terminal among a plurality of terminals; querying network performance information; selecting participating terminals among the plurality of terminals on the basis of the terminal information and the network performance information; transmitting respective parameters to the participating terminals and requesting local learning; and integrating the parameters.
Abstract:
An apparatus for collecting adaptive flow statistics data includes a node type collector to access a network node and collect flows; a flow type calculator to calculate two or more flow types according to a flow duration length and request the node type collector to collect flow statistics data of a corresponding type of a flow at every collection period that corresponds to each of the two or more flow types; and a flow statistics storage to store the two or more flow types calculated by the flow type calculator, and store the flow statistics data, collected by the node type collector, for each of the two or more flow types in a table.
Abstract:
A system for controlling a carrier virtual network is disclosed. The system includes at least one virtual network controller for controlling resources of a carrier virtual network based on customer request information received from a customer network controller, and at least one physical network controller for controlling resources of a carrier physical network based on a service profile received from the at least one virtual network controller. The at least one virtual network controller interworks with the at least one physical network controller in a one to one, one to N (N is an integer equal to or larger than 2), or N to one correspondence.