Abstract: Disclosed is a method comprising receiving a plurality of local growing neural gas models from a plurality of distributed trainers, wherein a local growing neural gas model of the plurality of local growing neural gas models represents a local state model of at least one radio access network node; and training a global growing neural gas model based on the plurality of local growing neural gas models, wherein the global growing neural gas model represents a global state model of a plurality of radio access network nodes.

Abstract: Disclosed are various example embodiments which may be configured to: obtain, for each of one or more network entities, a time series of values of a measurement parameter for respective timestamps, the time series including measured values and a special numerical value at one or more timestamps; parse the time series of values to determine which numerical value in the time series of values corresponds to the special numerical value; assign flags to the values in the time series of values, wherein a flag assigned to a value obtained at a given timestamp indicates whether the measurement was or not available at the given timestamp in the time series of values.

Abstract: Information regarding disjoint transport paths of a transport network is acquired (S501). Based on the acquired information, a user plane node of a plurality of user plane nodes of a core network is selected (S503), from which at least one of the disjoint transport paths is reachable, wherein the user plane node is selected as an endpoint of at least one tunnel out of at least two tunnels to be established for a packet data unit, PDU, session to carry traffic between the user plane node and an access node of an access network, wherein the at least two tunnels are to be set up by mapping the at least two tunnels to the disjoint transport paths.

Abstract: Various networks may benefit from suitable management techniques and systems. For example long term evolution networks and other wireless communication networks may benefit from a quality of experience aware transport self-organizing network framework. A method can include monitoring transport connectivity with respect to at least one base station (e.g. eNB) over a mobile backhaul. The method can also include detecting at least one degradation or anomaly in the transport connectivity of the at least one base station. The method can further include taking an appropriate network management action in response to the detected degradation or anomaly.

Abstract: Flow statistics for a buffer that receives data from a plurality of other entities are collected, and an available bandwidth for an interface coupled to an output side of the buffer is determined. From those is determined an arrival rate such that the buffer would not be overloaded. A flow control message is sent to each of the entities that limits aggregate flow from the entities so as not to exceed the arrival rate. One embodiment uses a scaling factor in the flow control messages, the same scaling factor for each entity on any given scheduling interval. An entity receives the flow control message and from it determines an allowed amount CRallowed of packet data units PDUs for that scheduling interval. The lower of CRallowed and an allocation of granted credits CR is used to compute a send rate, which the entity uses as its maximum rate for the interval.

Abstract: This invention relates to the forming of a communication network. The invention offers a common arrangement and a method to handle all tasks in the process of forming a communication network. The arrangement is divided into several modules, each carrying out certain tasks. The modules interwork with each other. A user selects the modules needed for forming a network. The selection depends on the network that is formed. Routine tasks have been automated in the arrangement. An iterative forming of a network is possible. The arrangement offers an interface to existing networks.