# US 12,216,639 B2 · Methods for processing data samples in communication networks > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,216,639 B2 - Document ID: US12216639B2 - Status: Granted - Kind: B2 - Application number: 18/035,576 - Filed: 2020-12-04 - Published: 2023-12-07 - Granted: 2025-02-04 - Earliest claimed: 2020-12-04 - Assignees: Telefonaktiebolaget LM Ericsson (publ) - Applicants: Telefonaktiebolaget LM Ericsson (publ) - Inventors: Henrik Rydén; Joel Berglund; Alexandros Palaios; Reza Moosavi; Abdulrahman Alabbasi; Ricardo da Silva Souza - Web page: https://patentgrep.com/patent/US12216639B2 ## CPC - G06F 16/2365: {Ensuring data consistency and integrity} - H04L 41/16: using machine learning or artificial intelligence - G06N 3/0455: Auto-encoder networks; Encoder-decoder networks - G06N 3/082: modifying the architecture, e.g. adding, deleting or silencing nodes or connections - G06N 3/09: Supervised learning - G06N 20/00: Machine learning - H04L 41/0806: for initial configuration or provisioning, e.g. plug-and-play - H04L 43/02: Capturing of monitoring data - H04W 24/08: Testing, {supervising or monitoring} using real traffic - G06N 3/045: Combinations of networks - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - H04W: WIRELESS COMMUNICATION NETWORKS (broadcast communication H04H; communication systems using wireless links for non-selective communication, e.g. wireless extensions H04M1/72) ## Abstract A computer-implemented method for acquiring new data samples and for maintaining a set of data samples in a database, wherein the set of data samples are configured to form input to a function associated with a predictive performance, the method including obtaining at least one relevance metric (M), where the relevance metric is indicative of an increase in the predictive performance of the function when using a data sample as input together with the set of data samples compared to when not using the data sample, obtaining a relevance criterion (C), where the relevance criterion identifies relevant data samples in a set of data samples based on the at least one relevance metric, signaling the at least one relevance metric (M) and the relevance criterion (C) to a data collecting network node. ## Claims ### Claim 1 A computer-implemented method, performed in a data processing network node, for acquiring new data samples and for maintaining a set of data samples in a database, wherein the set of data samples are configured to form input to a function associated with a predictive performance, the method comprising obtaining at least one relevance metric, where the relevance metric is indicative of an increase in the predictive performance of the function when using a data sample as input together with the set of data samples compared to when not using the data sample, obtaining a relevance criterion, where the relevance criterion identifies relevant data samples in a set of data samples based on the at least one relevance metric, signaling the relevance criterion to a data collecting network node, and receiving one or more data samples from the data collecting network node, where the received data samples are associated with relevance metrics that satisfy the relevance criterion; and training a machine learning model arranged to indicate a novelty metric associated with a data sample in a set of collected data samples, wherein the machine learning model constitutes a relevance metric of the at least one relevance metric. ### Claim 2 (depends on 1) The method according to claim 1, comprising signaling the at least one relevance metric to the data collecting network node. ### Claim 3 (depends on 1) The method according to claim 1, wherein the relevance metric is pre-configured at the data collecting network node. ### Claim 4 (depends on 1) The method according to claim 1, comprising training a machine learning model arranged to indicate the predictive performance associated with a data sample in a set of collected data samples, wherein the machine learning model constitutes a relevance metric. ### Claim 5 (depends on 4) The method according to claim 4, comprising updating the machine learning model based on the received one or more data samples. ### Claim 6 (depends on 1) The method according to claim 1, comprising updating the function associated with a predictive performance based on the received one or more data samples. ### Claim 7 (depends on 1) The method according to claim 1, wherein the at least one relevance metric comprises a distance metric configured to quantify a distance between one or more data samples in a set of collected data samples and a set of samples in the database. ### Claim 8 (depends on 1) The method according to claim 1, wherein the at least one relevance metric comprises a reconstruction error metric value obtained from an autoencoder model. ### Claim 9 (depends on 1) The method according to claim 1, wherein the at least one relevance metric comprises a prediction error metric value obtained from a machine learning (ML) model. ### Claim 10 (depends on 1) The method according to claim 1, wherein the at least one relevance metric is based on a cyclo-stationarity operation configured to indicate a similarity between one or more data samples in a set of collected data samples and a set of basic patterns. ### Claim 11 (depends on 10) The method according to claim 10, wherein the cyclo-stationarity operation is an auto-correlation function or a cross-correlation function. ### Claim 12 (depends on 1) The method according to claim 1, comprising signaling the at least one relevance metric and the relevance criterion over a Physical Downlink Control Channel (PDCCH) of a third generation partnership program (3GPP) communications network. ### Claim 13 (depends on 1) The method according to claim 1, comprising signaling the at least one relevance metric and the relevance criterion as a Radio Resource Control (RRC) reconfiguration message of a third generation partnership program (3GPP) communications network. ### Claim 14 (depends on 1) The method according to claim 1, comprising signaling the at least one relevance metric and the relevance criterion as a dedicated message over a Physical Downlink Shared Channel (PDSCH) of a third generation partnership program (3GPP) communications network. ### Claim 15 (depends on 1) The method according to claim 1, comprising receiving one or more relevance metrics associated with the data samples received from the data collecting network node. ### Claim 16 (depends on 1) A computer program comprising program code means for performing a method according to claim 1 when said program is run on a computer or on processing circuitry of a network node. ### Claim 17 (depends on 16) A computer program product comprising a computer program according to claim 16, and a computer readable means on which the computer program is stored. ### Claim 18 A computer-implemented method, performed in a data collecting network node, for acquiring new data samples and for maintaining a set of data samples in a database at a data processing network node, wherein the set of data samples forms input to a function associated with a predictive performance, the method comprising obtaining at least one relevance metric, where the relevance metric is indicative of an increase in the predictive performance of the function when using the new data sample as input, obtaining a relevance criterion from the data processing network node, where the relevance criterion identifies relevant data samples in a set of data samples based on the at least one relevance metric, collecting a set of collected data samples by the data collecting network node, selecting one or more relevant data samples from the set of collected data samples based on the relevance metric and on the relevance criterion, where the relevant data samples are associated with relevance metrics meeting the relevance criterion, and transmitting the one or more selected data samples to the data processing network node and training a machine learning model arranged to indicate a novelty metric associated with a data sample in a set of collected data samples, wherein the machine learning model constitutes a relevance metric of the at least one relevance metric. ### Claim 19 (depends on 18) A network node, comprising: processing circuitry; a network interface coupled to the processing circuitry; and a memory coupled to the processing circuitry, wherein the memory comprises machine readable computer program instructions that, when executed by the processing circuitry, causes the network node to perform a method according to claim 18.