Nokia Technologies Oy

Alcatel Lucent

Alcatel-lucent USA Inc.

Qi Liao

Zulfiquar Sayeed

Edward Grinshpun

David Faucher

Sameer Sharma

Ilaria Malanchini

Suresh Kalyanasundaram

Afef Feki

Veronique Capdevielle

Fahad Syed Muhammad

Mathieu Boussard

Alessandro Lieto

Systems, methods, apparatuses, and computer program products for dependency-aware quality of service (QoS) profile provisioning. The method may include obtaining information on at least one application data flow, and dependency information between one application and at least one other application. The method may also include receiving, from a network, information on an updated quality of service profile for the at least one application, and information on another updated quality of service profile for the at least one other application. The updated quality of service profile and the another updated quality of service profile may be selected based on the dependency graph.

Dependency-aware quality of service profile provisioning

A deep transfer reinforcement learning (DTRL) method based on transfer learning within a deep reinforcement learning (DRL) framework is provided to accelerate the GoB optimization decisions when experiencing environment changes in the same source radio network agent or when being applied from a source radio network agent to a target radio network agent. The transferability of the knowledge embedded in a pre-trained neural network model as a Q-approximator is exploited, and a mechanism to transfer parameters from a source agent to a target agent is provided, where the transferability criterion is based on the similarity measure between the source and target domain.

Apparatus, method and computer program for accelerating grid-of-beams optimization with transfer learning

An apparatus for use by a communication network control element or function configured to control a setting of parameters of a self-organizing communication network, the apparatus comprising at least one processing circuitry, and at least one memory for storing instructions to be executed by the processing circuitry, wherein the at least one memory and the instructions are configured to, with the at least one processing circuitry, cause the apparatus at least: to obtain a pre-trained network optimization model indicating a mapping between a communication network environment forming an input of the model, optimization actions or decisions forming an output of the model, and system performance indicators forming a reward, to cause sending, to at least one communication network element or function forming a part of the self-organizing communication network, a request for providing similarity data for a similarity analysis allowing to determine a similarity between a part of the self-organizing communication network for which the pre-trained network optimization model is derived and a part of the self-organizing communication network of the at least one communication network element or function to which the request is sent, to perform the similarity analysis for determining a similarity between the part of the self-organizing communication network for which the pre-trained network optimization model is derived and each part of the self-organizing communication network for which similarity data for the similarity analysis are received, to determine, on the basis of the similarity analysis, at least a part of the pre-trained network optimization model to be provided to the at least one communication network element or function forming a part of the self-organizing communication network from which the similarity data are received, and to cause sending of the determined part of the pre-trained network optimization model to the at least one communication network element or function forming a part of the self-organizing communication network from which the similarity data are received.

Procedure for optimization of self-organizing network

An application type of a bearer is classified by computing statistics vectors of bearer metrics and locating points on a label map corresponding to the statistics vectors to obtain application type information. The application type information is exported to a network node to control an operation of application. The bearer metrics include bearer identifier information and bearer condition information, where the bearer condition information includes channel condition information and cell congestion level information. The bearers are paired, such that uplink and downlink bearers for a same application are identified, so that paired bearers are classified together. The label map is produced using previously classified bearer information to calculate cluster centroids and cluster regions that define portions of the map for particular application types. The bearer is classified by determining which cluster region is closest to points on the label map that are associated with the statistics vectors for a particular bearer.

Method and system for controlling an operation of an application by classifying an application type using data bearer characteristics

Fast congestion detection is implemented in a wireless radio access network (RAN) to mitigate congestion by classifying bearers into bearer types using bearer metrics and monitoring bearers for type specific anomalies to obtain anomaly data. Congestion is detected based on the anomaly data to obtain congestion information that is exported to a network node to mitigate network congestion. Bearer metrics include radio link protocol (RLC) buffer size information, physical resource block (PRB) utilization information, and transmission burst interval information. The transmission burst interval information includes information on time intervals between successive time periods when a RLC buffer is empty. The bearers are classified into Type I and Type II bearers. Type I bearers are bearers displaying bursty data transfer behavior, and Type II bearers are bearers displaying a continuous stream of data transfer behavior.

System and method for mitigating network congestion using fast congestion detection in a wireless radio access network (RAN)

A smoothed transport block size is forecasted by predicting future value information based on historical time series data obtained at an e-Node B. The historical time series data includes historical transport block size information and historical modulation and coding scheme information. A mapping function is used to correlate the future value information with historical transport block size information. Once the mapping function is determined, the mapping function forecasts the average transport block sizes by inputting the future value information into the mapping function. The smoothed transport block sizes and the future value information is then transmitted to an application server and/or an application client at a user equipment to control an operation of an application.

Growing community of inventors

Stuttgart, Germany

Qi Liao