Abstract: An apparatus, method and computer program is described as obtaining forwarded traffic volume data and discarded traffic volume data of a point-to-multipoint communication network in each of a plurality of discrete time periods over an observation period; generating calibrated forwarded and discarded traffic volume data, wherein calibration data is dependent on a calibration level; filtering, by applying signal processing techniques, multiple instances of said calibrated forwarded and discarded traffic volume data; identifying first sample periods within said observation period; and determining impact scores for the forwarded and discarded traffic volume data for each of said one or more first sample periods.

Abstract: A system for diagnosing noise impacting impairments affecting a Digital Subscriber Line includes at least one processor and at least one memory storing instructions. The at least one memory and the instructions configured to, with the at least one processor, cause the system at least to, receive noise sequences for said Digital Subscriber Line; detect noise symptoms in said noise sequences and identify noise impacting impairments associated with said noise symptoms using a trained machine learning model; and determine an impact on said Digital Subscriber Line of a plurality of noise impacting impairments.

Abstract: An apparatus, method and computer program provide for obtaining channel response data including a channel frequency response of a channel over a frequency spectrum, wherein the channel frequency response is generated in response to a transmission over the channel or a simulation thereof; and generating an indication of channel impairments in response to applying the channel response data to a transformer-based machine-learning (ML) model trained to predict a channel impairment estimate.

Abstract: An apparatus, method and computer program is described comprising: obtaining channel response data comprising a first channel frequency response of a channel over a first frequency spectrum, where the first channel frequency response is generated in response to a transmission over the channel or a simulation thereof; and generating an estimate of a second channel frequency response of the channel over a second frequency spectrum in response to applying the channel response data to a machine-learning model, where the second frequency spectrum is different to the first frequency spectrum.

Abstract: Embodiments of the present disclosure relate to an apparatus including means configured to perform: obtaining a measured channel frequency response, and a measured noise power spectral density for a line of a wired network susceptible to an impairment; deriving, in case of an indication of an impairment present on the line, from the measured channel frequency response and the measured noise power spectral density, a first theoretical noise representation for the line with the impairment and a second theoretical noise representation for the line without the impairment; and determining information indicative of a location of the impairment in the line, by processing the measured noise power spectral density, the first theoretical noise representation, and the second theoretical noise representation with a neural network.

Abstract: An apparatus, method and computer program is described comprising: combining first features extracted from an echo signal using a convolutional encoder of a convolutional encoder-decoder having first weights, wherein the echo signal is obtained in response to a transmission over a channel or a simulation thereof; and using a convolutional decoder of the convolutional encoder-decoder to generate an estimate of a frequency response of the channel based on the echo signal, wherein the convolutional decoder has second weights.

Abstract: A system for diagnosing noise impacting impairments affecting a Digital Subscriber Line includes at least one processor and at least one memory storing instructions. The at least one memory and the instructions configured to, with the at least one processor, cause the system at least to, receive noise sequences for said Digital Subscriber Line; detect noise symptoms in said noise sequences and identify noise impacting impairments associated with said noise symptoms using a trained machine learning model; and determine an impact on said Digital Subscriber Line of a plurality of noise impacting impairments.

Abstract: Embodiments relate to an apparatus for monitoring a telecommunication network including one or more telecommunication channels. The apparatus includes a processor configured after executing computer code to design a convolutional neural network configured for determining an impairment type of a telecommunication channel as a function of a channel frequency response of the telecommunication channel, by selecting at least one of a number of convolutional layers, a number of filters for respective convolutional layers, and/or a size of filters for respective convolutional layers; and train the convolutional neural network based on training data specifying, for respective telecommunication channels, a channel frequency response and an associated impairment type.

Abstract: The apparatus includes a memory configured to store executable code; and a processor configured to execute the executable code and cause the apparatus to perform the operations of generating and training. The generating generates a dataset specifying, for a plurality of communication lines, i) a channel frequency response of a communication line having one or two bridged taps, and ii) a set of M lengths of bridged taps associated with the communication line, with M greater than one. The training trains, based on the dataset, a machine learning model, the machine learning model configured for determining, based on the channel frequency response of a communication line, a set of M lengths of bridged taps associated with the communication line.

Abstract: Embodiments relate to an apparatus (5, 7) comprising means (31, 32, P) configured for: generating (S1) a dataset specifying, for a plurality of communication lines: a channel frequency response of a communication line having one or two bridged taps, and a set of M lengths of bridged taps associated with the communication line, with M greater than one, training (S2), based on the dataset, a machine learning model configured for determining, based on the channel frequency response of a communication line, a set of M lengths of bridged taps associated with the communication line.

Abstract: Embodiments relate to an apparatus for monitoring a telecommunication network including one or more telecommunication channels. The apparatus includes a processor configured after executing computer code to design a convolutional neural network configured for determining an impairment type of a telecommunication channel as a function of a channel frequency response of the telecommunication channel, by selecting at least one of a number of convolutional layers, a number of filters for respective convolutional layers, and/or a size of filters for respective convolutional layers; and train the convolutional neural network based on training data specifying, for respective telecommunication channels, a channel frequency response and an associated impairment type.