# US 10,582,526 B2 · System and method for measuring end-to-end channel capacity entropy > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 10,582,526 B2 - Document ID: US10582526B2 - Status: Granted - Kind: B2 - Application number: 15/827,166 - Filed: 2017-11-30 - Published: 2019-05-30 - Granted: 2020-03-03 - Earliest claimed: 2017-11-30 - Assignees: Verizon Patent and Licensing Inc. - Applicants: Verizon Patent and Licensing Inc. - Inventors: Bjorn Olof Erland Kalderen; Haim S. Ner; Said Soulhi - Web page: https://patentgrep.com/patent/US10582526B2 ## CPC - H04L 41/083: for increasing network speed - H04B 7/0842: {Weighted combining} - H04L 1/0009: {by adapting the channel coding (H04L1/1812 takes precedence)} - H04L 1/0026: {Transmission of channel quality indication} - H04L 41/147: for predicting network behaviour - H04L 41/5012: {determining service availability, e.g. which services are available at a certain point in time} - H04L 43/0888: Throughput - H04L 47/127: by using congestion prediction - H04L 47/283: in response to processing delays, e.g. caused by jitter or round trip time [RTT] - H04Q 3/0062: {Provisions for network management} - H04W 16/22: Traffic simulation tools or models - H04W 24/02: Arrangements for optimising operational condition - H04W 24/08: Testing, {supervising or monitoring} using real traffic - H04W 28/02: Traffic management, e.g. flow control or congestion control - H04W 72/542: using measured or perceived quality - H04Q 2213/13526: resource management - H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M) - H04B: TRANSMISSION - H04Q: SELECTING (switches, relays, selectors H01H; wireless communication networks H04W) - H04W: WIRELESS COMMUNICATION NETWORKS (broadcast communication H04H; communication systems using wireless links for non-selective communication, e.g. wireless extensions H04M1/72) ## Abstract A network device predicts end-to-end channel capacity entropy to permit use of optimal throughput settings in a network. The network device stores class definitions for a network condition; identifies multiple input features to correlate with the class definitions; generates a multiclass classification model that produces an importance score for each of the multiple input features, wherein the importance score reflects the contribution of an input feature to the network condition; selects two or more of the multiple input features with highest importance scores as influential features; predicts the behavior of the influential features to identify a current class, from the class definitions, for the network condition over an end-to-end communication channel; and sends an estimated network condition, based on the current class, to a device for traffic optimization. ## Claims ### Claim 1 A method comprising: storing, by a computing device, class definitions for a network condition, wherein the class definitions for the network condition include multiple congestion levels ranging from least-congested to most-congested; identifying, by the computing device, multiple input features to correlate with the class definitions, wherein the multiple input features include measurable performance indicators for transmission of data over a network; generating, by the computing device, a multi class classification model that produces an importance score for each of the multiple input features, wherein each importance score reflects the contribution of one of the multiple input features to the network condition; selecting, by the computing device, two or more of the multiple input features with highest importance scores as influential features; predicting, by the computing device, the behavior of the influential features to identify a current class, from the class definitions, for the network condition over an end-to-end communication channel; and sending, by the computing device, an estimated network condition, based on the current class, to a device for traffic optimization. ### Claim 2 (depends on 1) The method of claim 1, wherein the class definitions for the network condition include congestion levels based on one or more of physical resource block utilization and perceived throughput. ### Claim 3 (depends on 1) The method of claim 1, wherein the congestion levels are based on one or more of application layer round trip times or burst rates. ### Claim 4 (depends on 1) The method of claim 1, wherein selecting two or more of the multiple input features includes assigning one of a raw score, a weighted score, or a percentile score to each of the multiple input features. ### Claim 5 (depends on 1) The method of claim 1, further comprising: configuring, by an end device, a transmission throughput for a session based on the estimated network condition. ### Claim 6 (depends on 1) The method of claim 1, wherein at least some of the multiple input features have non-linear relationships with the network condition. ### Claim 7 (depends on 1) The method of claim 1, wherein the input features include key performance indicators for a radio access network (RAN). ### Claim 8 (depends on 1) The method of claim 1, wherein the input features include an initial block error ratio, an average number of radio resource controls (RRCs) in connected mode; a physical resource block utilization value, a layer 1 throughput, and a radio link control (RL retransmission rate. ### Claim 9 (depends on 1) The method of claim 1, wherein the predicting includes generating a prediction model using backpropagation through time to updated weights in a recursive neural network of the influential features. ### Claim 10 (depends on 1) The method of claim 1, wherein generating the multiclass classification model includes using machine learning with boosted decision trees. ### Claim 11 A computing device comprising: a memory configured to store instructions; and a processor configured to execute the instructions to: store class definitions for a network condition, wherein the class definitions for the network condition include multiple congestion levels ranging from least-congested to most-congested, identify multiple input features to correlate with the class definitions, wherein each of the multiple input features include measurable performance indicators for transmission of data over a network, generate a multiclass classification model that produces an importance score for each of the multiple input features, wherein each importance score reflects the contribution of one of the multiple input features to the network condition, select two or more of the multiple input features with highest importance scores as influential features, predict the behavior of the influential features to identify a current class, from the class definitions, for the network condition over an end-to-end communication channel, and send an estimated network condition, based on the current class, to a device for traffic optimization. ### Claim 12 (depends on 11) The computing device of claim 11, wherein the class definitions for the network condition include congestion levels based on one or more of physical resource block utilization and perceived throughput. ### Claim 13 (depends on 11) The computing device of claim 11, wherein the congestion levels are based on one or more of application layer round trip times or burst rates. ### Claim 14 (depends on 11) The computing device of claim 11, wherein at least some of the multiple input features have non-linear relationships with the network condition. ### Claim 15 (depends on 11) The computing device of claim 11, wherein the processor is further configured to: configure transmission throughput for a session based on the estimated network condition. ### Claim 16 (depends on 11) The computing device of claim 11, wherein the input features include key performance indicators for a radio access network (RAN) and at least one other network. ### Claim 17 (depends on 11) The computing device of claim 11, wherein, when generating the multiclass classification model, the processor is further configured to apply machine leaning with boosted decision trees, and wherein, when predicting the behavior of the influential features, the processor is further configured to generate a prediction model using backpropagation through time to update weights in a recursive neural network of the influential features. ### Claim 18 A system, comprising: a computing device configured to: store Mass definitions for a network condition, wherein the class definitions for the network condition include multiple congestion levels ranging from least-congested to most-congested, identify multiple input features to correlate with the class definitions, wherein each of the multiple input features include measurable performance indicators for transmission of data over a network, generate a multiclass classification model that produces an importance score for each of the multiple input features, wherein each importance score reflects the contribution of one of the multiple input features to the network condition, select two or more of the multiple input features with highest importance scores as influential features, predict the behavior of the influential features to identify a current class, from the class definitions, for the network condition over an end-to-end communication channel, and send an estimated network condition, based on the current class, to a network device for traffic optimization. ### Claim 19 (depends on 18) The system of claim 18, further comprising: a network optimizing device configured to configure transmission throughput for an end-to-end session based on the estimated network condition. ### Claim 20 (depends on 18) The system of claim 18, wherein at least some of the multiple input features have non-linear relationships with the network condition.