Telefonaktiebolaget Lm Ericsson (publ)

Péter Kersch

Sándor Rácz

Geza Szabo

László Toka

Tamas Borsos

Peter Vaderna

Zsófia Kallus

Gabor Nemeth

Attila Korösi

Andras Valkó

Attila Körösi

Eszter Bencsne Bokanyi

Péter Bognár

Tamas Zambo

There is provided a method for assessing network performance. The method includes acquiring network measurements indicative of any changes in a network following a change to a configuration of the network and data indicative of one or more factors capable of causing the changes in the network. The one or more factors are independent of the change to the configuration of the network. The method includes analysing the acquired network measurements and data to identify a contribution of the one or more factors to a key performance indicator (KPI) and a contribution of the change to the configuration of the network to the KPI. The KPI is predicted by a machine learning model and is a measure of the network performance following the change to the configuration of the network.

Network performance assessment

The correlation of per-session network logs with trouble reports provided by the users/user devices via a trouble reporting app generates labeled training information used to train Machine Learning (ML) models that are subsequently used to infer, explain, and address service issues associated with an application service session. As such, the level of automation for Quality of Experience (QoE) management increases, while the accuracy of such QoE management increases and a subscriber's experience improves.

Correlating network data with instant app feedback for ML-based customer experience management

The present disclosure is directed to aspects of assessing performance of a session between an OTT service provider endpoint and an OTT service recipient endpoint in regard of a particular OTT service. The performance assessment is based on a series of data records generated for the session. Data record generation includes monitoring, at a network site between the two endpoints and during successive periods of time, at least one transport layer connection of the session. The monitoring is performed to determine, for each period of time, an activity metric indicative of a pending transport layer activity of the session in the respective period of time. This activity metric is then evaluated in connection with service performance assessment.

Transport layer monitoring and performance assessment for OTT services

Methods and apparatuses for generating real-time or near-real-time transport modality statistics by correlating mobile network and public transport location data are provided. In an example, a processing node may perform a method using one or more processing circuits, the method including identifying paths traveled by multiple vehicles and determining paths traveled by multiple subscriber telecommunications devices of one or more network or service providers. In addition, the method may include correlating the paths traveled by the multiple vehicles and the paths traveled by the multiple subscriber telecommunications devices to determine which one or more modes of transportation each subscriber telecommunications device used over the path traveled by the subscriber telecommunications device. Corresponding apparatus, computer programs, and signals are also described.

Technique for generating near real-time transport modality statistics

Handover bursts are identified for source and target cell pairs along a given public transport line, and each of the handover bursts are tagged with anonymized identifiers of participating users. By searching for handover bursts with a significant share of common user identifiers, it is possible to recursively connect bursts corresponding to a particular public transport vehicle trip. Further, by projecting the estimated location of all connected handover bursts to the path of the given public transport line, it is possible to interpolate stop times and locations for the given vehicle along the entire trip.

Timetables for public transport lines from mobile network handovers

A traffic analysis system monitors data traffic in a communication network. In the data traffic, flows are detected which are based on the Hypertext Transfer Protocol (HTTP). For each of the flows, a data record is created. The data record comprises at least a User Agent identifier from a message header of a HTTP message of the flow and a device identifier of a user equipment transmitting the flow. The data records are analyzed to determine a mapping of at least one User Agent identifier in the data records to a corresponding device category.

Traffic analysis for HTTP user agent based device category mapping

A technique for predicting network load in a communication network is described. For multiple subscriptions, a network load snapshot is provided. The network load snapshot comprises multiple network load records. Each network load record reflects network load caused by an associated subscription in the network at a particular point of time. Each network load record comprises at least one network load value reflecting a network load caused by the corresponding subscription and at least one network load influencing factor reflecting a parameter associated with the subscription which has the potential of influencing the network load value. The network load snapshot is transformed into a predicted network load snapshot comprising at least one predicted network load record, each predicted network load record reflecting estimated network load caused by the corresponding subscription at a particular point of time in the future. The transformation of a network load snapshot into a corresponding predicted network load snapshot is done based on both the network load values and the associated network load influencing factors.

Growing community of inventors

Budapest, Hungary

Péter Kersch