# US 2023/0276286 A1 · Method and apparatus for detecting event in communication network > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2023/0276286 A1 - Document ID: US20230276286A1 - Status: Published application - Kind: A1 - Application number: 18/043,573 - Filed: 2020-09-27 - Published: 2023-08-31 - Earliest claimed: 2020-09-27 - Assignees: Telefonaktiebolaget LM Ericsson (publ) - Applicants: Telefonaktiebolaget LM Ericsson (publ) - Inventors: Zhenhua HE; Min LIU; Yanning LIU - Web page: https://patentgrep.com/patent/US20230276286A1 ## CPC - H04W 24/10: Scheduling measurement reports {; Arrangements for measurement reports} - H04W 76/19: Connection re-establishment - H04L 41/142: using statistical or mathematical methods - H04L 41/16: using machine learning or artificial intelligence - G06F 11/0754: {by exceeding limits} - H04L 41/5009: Determining service level performance parameters or violations of service level contracts, e.g. violations of agreed response time or mean time between failures [MTBF] - H04W: WIRELESS COMMUNICATION NETWORKS (broadcast communication H04H; communication systems using wireless links for non-selective communication, e.g. wireless extensions H04M1/72) - H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M) - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract A method and apparatus for detecting event in a communication network. The method performed at a network node may comprise: obtaining at least one parameter associated to an event; determining, by a first classifier, whether a first measurement result of the at least one parameter satisfies at least one rule; classifying, by a second classifier, a second measurement result of the at least one parameter, when the first measurement result of the at least one parameter satisfies the at least one rule; and determining whether the event happens based on a classifying result of the second measurement result of the at least one parameter. More than one classifier may be cascaded, so as to reduce a misdetection rate. ## Claims ### Claim 1 A method performed at a network node, comprising: obtaining at least one parameter associated to an event; determining, by a first classifier, whether a first measurement result of the at least one parameter satisfies at least one rule; classifying, by a second classifier, a second measurement result of the at least one parameter, when the first measurement result of the at least one parameter satisfies the at least one rule; and determining whether the event happens based on a classifying result of the second measurement result of the at least one parameter. ### Claim 2 (depends on 1) The method according to claim 1, further comprising: determining a threshold used in the at least one rule for determining whether a first measurement result of the at least one parameter satisfies at least one rule, based on a training result of the second classifier. ### Claim 3 (depends on 2) The method according to claim 2, wherein the threshold comprises a vector. ### Claim 4 (depends on 2) The method according to claim 2, wherein the determining the threshold used in the at least one rule comprises: obtaining a plurality of samples, wherein each sample includes a historic measurement result of the at least one parameter, and each sample is previously labeled as the event or not the event; selecting a candidate threshold from a plurality of candidate thresholds; determining, by the first classifier, whether each sample satisfies the at least one rule including the selected candidate threshold; training the second classifier with at least one sample which satisfies the at least one rule including the selected candidate threshold; determining a classifying score of the second classifier associated to the selected candidate threshold, based on a training result; and determining the selected candidate threshold as the threshold used in the at least one rule, when the selected candidate threshold corresponds to a highest classifying score among the plurality of candidate thresholds. ### Claim 5 (depends on 4) The method according to claim 4, wherein the at least one sample comprises at least one sample labelled as the event, and at least one sample labelled as not the event. ### Claim 6 (depends on 1) The method according to claim 1, wherein the second classifier comprises a machine learning classifier. ### Claim 7 (depends on 1) The method according to claim 1, wherein the second classifier comprises a support vector machine (SVM) classifier. ### Claim 8 (depends on 7) The method according to claim 7, wherein a kernel of the SVM classifier comprises a Radial Basis Function (RBF). ### Claim 9 (depends on 1) The method according to claim 1, wherein the first measurement result, the second measurement result, or both the first measurement result and the second measurement result comprises a feature value of the at least one parameter. ### Claim 10 (depends on 9) The method according to claim 9, wherein the feature value is generated based on a plurality of measurement values of the at least one parameter; and wherein the at least one parameter is measured periodically. ### Claim 11 (depends on 10) The method according to claim 10, wherein the feature value comprises: a maximum value of the plurality of measurement values; a minimum value of the plurality of measurement values; a fluctuation degree of the plurality of measurement values; or any combination thereof. ### Claim 12 (depends on 1) The method according to claim 1, wherein the event comprises a near silent issue (NSI) of a base station. ### Claim 13 (depends on 1) The method according to claim 1, wherein the at least one parameter comprises at least one key performance indictor parameter. ### Claim 14 (depends on 13) The method according to claim 13, wherein the key performance indictor parameter comprises: a quantity number of connected users; a downlink user throughput; a downlink physical resource block usage; or any combination thereof. ### Claim 15 (depends on 1) The method according to claim 1, wherein the network node comprises an operation, administration and maintenance (OAM). ### Claim 16 A network node, comprising: a processor; and a memory, the memory containing instructions which, when executed by the processor, cause the network node to: obtain at least one parameter associated to an event; determine, by a first classifier, whether a first measurement result of the at least one parameter satisfies at least one rule; classify, by a second classifier, a second measurement result of the at least one parameter, when the first measurement result of the at least one parameter satisfies the at least one rule; and determine whether the event happens based on a classifying result of the second measurement result of the at least one parameter. ### Claim 17 (depends on 16) The network node according to claim 16, wherein the network node is further to: determine a threshold used in the at least one rule to determine whether a first measurement result of the at least one parameter satisfies at least one rule, based on a training result of the second classifier. ### Claim 18 A non-transitory computer-readable storage medium storing instructions which, when executed by at least one processor, cause the at least one processor to perform operations comprising: obtaining at least one parameter associated to an event; determining, by a first classifier, whether a first measurement result of the at least one parameter satisfies at least one rule; classifying, by a second classifier, a second measurement result of the at least one parameter, when the first measurement result of the at least one parameter satisfies the at least one rule; and determining whether the event happens based on a classifying result of the second measurement result of the at least one parameter. ### Claim 19 (depends on 17) The network node according to claim 17, wherein the threshold comprises a vector. ### Claim 20 (depends on 17) The network node according to claim 17, wherein to determine the threshold used in the at least one rule comprises: obtain a plurality of samples, wherein each sample includes a historic measurement result of the at least one parameter, and each sample is previously labeled as the event or not the event; select a candidate threshold from a plurality of candidate thresholds; determine, by the first classifier, whether each sample satisfies the at least one rule including the selected candidate threshold; train the second classifier with at least one sample which satisfies the at least one rule including the selected candidate threshold; determine a classifying score of the second classifier associated to the selected candidate threshold, based on a training result; and determine the selected candidate threshold as the threshold used in the at least one rule, when the selected candidate threshold corresponds to a highest classifying score among the plurality of candidate thresholds.