# US 12,003,797 B2 · Method and system for adaptive data transmission > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,003,797 B2 - Document ID: US12003797B2 - Status: Granted - Kind: B2 - Application number: 17/125,212 - Filed: 2020-12-17 - Published: 2021-04-08 - Granted: 2024-06-04 - Earliest claimed: 2018-06-20 - Assignees: Naver Corporation - Applicants: Naver Corporation - Inventors: Sungtak Cho; Dongju Lee; Young Soo Lee; Hyemin Lee; SungHo Kim; Joon-kee Chang - Web page: https://patentgrep.com/patent/US12003797B2 ## CPC - H04N 21/2402: {Monitoring of the downstream path of the transmission network, e.g. bandwidth available (traffic monitoring in data switching networks H04L43/00; monitoring data switching networks utilization H04L43/0876)} - G06N 20/00: Machine learning - H04L 65/612: for unicast - H04L 65/752: adapting media to network capabilities - H04L 65/80: Responding to QoS - H04N 21/2187: Live feed - H04N 21/23406: {involving management of server-side video buffer} - H04N 21/2385: Channel allocation (H04N21/266 takes precedence); Bandwidth allocation (H04N21/24 takes precedence {; allocation of channels according to the instantaneous demands of the users in time-division multiplex systems H04J3/1682; admission control, resource allocation in open networks H04L12/5692; arrangements for maintenance or administration in data switching networks involving bandwidth and capacity management H04L41/0896; negotiating bandwidth in wireless networks H04W28/16}) - H04N 21/64738: {Monitoring network characteristics, e.g. bandwidth, congestion level (data switched network analysis H04L41/14; monitoring functioning in data switched networks H04L43/0817; flow control in packet networks H04L47/10)} - G06N 3/084: Backpropagation, e.g. using gradient descent - G06N 5/01: Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound - G06N 20/20: Ensemble learning - H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M) ## Abstract A video transmission control method for adaptive data transmission includes predicting a network state through machine learning using information related to video transmission as a feature in a real-time video streaming environment; and determining an option for controlling video transmission on the basis of the predicted network state. ## Claims ### Claim 1 A video transmission control method performed by a server implemented by a computer, comprising: designing a plurality of rule-based algorithms of video transmission options, where each of the rule-based algorithms is optimized differently; estimating a network state through machine learning using information related to a video transmission as a feature in a real-time video streaming environment; and determining an option for controlling the video transmission based on the estimated network state, wherein the determining of the option comprises selecting a rule-based algorithm corresponding to the estimated network state from among the plurality of rule-based algorithms in each of which a video transmission option including at least two of a bitrate, frames per second (FPS), and a buffer flush rate are differently set based on the network state, wherein the selected rule-based algorithm utilizes the bitrate, and at least one of the FPS and the buffer flush rate as controllable parameters when transmitting video data, wherein the bitrate is adjusted in a first manner in one of the rule-based algorithms and the bitrate is adjusted in a second manner in another one of the rule-based algorithms, where the first manner is different from the second manner, and wherein in each of the rule-based algorithms, a different ratio between two of the following elements is relied upon: (i) estimated bandwidth; (ii) network stability; (iii) bandwidth delta amplitude; (iv) bandwidth delta confidence; and (v) transmission buffer configuration. ### Claim 2 (depends on 1) The video transmission control method of claim 1, wherein the estimating of the network state comprises estimating an element that represents a current network state and an element that represents a future network state through the machine learning. ### Claim 3 (depends on 1) The video transmission control method of claim 1, wherein the estimating of the network state comprises: estimating a current network state through the machine learning using transmission buffer data comprising at least one of a buffer size that represents a size of not-transmitted data in a transmission buffer, a buffer duration that represents a playback time of the not-transmitted data in the transmission buffer, and a throughput that represents a data transmission amount per time unit. ### Claim 4 (depends on 3) The video transmission control method of claim 3, wherein the estimating of the network state further comprises estimating a future network state through the machine learning using the transmission buffer data and transmission buffer data corresponding to a desired previous time. ### Claim 5 (depends on 3) The video transmission control method of claim 3, wherein the machine learning further includes using at least one of connection type information and communication provider information as information related to a communication network. ### Claim 6 (depends on 1) The video transmission control method of claim 1, wherein the estimating of the network state comprises estimating a current network state that comprises at least one of an estimated bandwidth and transmission buffer data, and a future network state that comprises at least one of a network stability, a bandwidth delta amplitude, a bandwidth delta, and a bandwidth delta confidence, through the machine learning. ### Claim 7 (depends on 1) The video transmission control method of claim 1, wherein the determining of the option comprises selecting a rule-based algorithm of an option corresponding to the estimated network state from among a plurality of rule-based algorithms in each of which a video transmission option is differently set based on the network state using a method of classifying the network state through the machine learning using a plurality of decision trees. ### Claim 8 (depends on 1) The video transmission control method of claim 1, wherein: the determining of the option comprises determining at least two of a bitrate, frames per second (FPS), and a buffer flush rate as a controllable parameter when transmitting video data based on the estimated network state, and the determining of the option also comprises calculating at least one parameter among the bitrate, the FPS, and the buffer flush rate as the video transmission option according to the estimated network state, using a regression scheme using a machine learning scheme that has a function to learn based on experience. ### Claim 9 (depends on 1) A non-transitory computer-readable record medium storing a program that, when executed by a computer, causes the computer to execute the video transmission control method according to claim 1. ### Claim 10 (depends on 8) The video transmission control method of claim 8, wherein the determining of the option comprises determining all three of the bitrate, the FPS and the buffer flush rate as controllable parameters when transmitting video data based on the estimated network state. ### Claim 11 (depends on 1) The video transmission control method of claim 1, wherein at least two of the rule-based algorithms include buffer flush rates that differ from each other. ### Claim 12 (depends on 1) The video transmission control method of claim 1, wherein the estimating of the network state comprises: estimating a current network state that comprises at least one of an estimated bandwidth and transmission buffer data, estimating a future network state that comprises an estimate of network stability; and wherein the selecting of the rule-based algorithm is based, at least in part, on the estimated network stability. ### Claim 13 A video transmission control system implemented by a computer, comprising: at least one processor configured to execute computer-readable instructions comprised in a memory, wherein the at least one processor comprises: a network state estimator configured to estimate a network state through machine learning using information related to a video transmission as a feature in a real-time video streaming environment; and a video transmission controller configured to determine an option for controlling the video transmission based on the estimated network state, wherein the determining of the option comprises selecting a rule-based algorithm of an option corresponding to the estimated network state from among a plurality of rule-based algorithms in each of which a video transmission option including at least two of a bitrate, frames per second (FPS), and a buffer flush rate are differently set based on the network state, wherein each of the plurality of rule-based algorithms of video transmission options has been designed to include a strategy optimized for a different network state and/or network environment, wherein the selected rule-based algorithm utilizes the bitrate, and at least one of the FPS and the buffer flush rate as controllable parameters when transmitting video data, wherein the bitrate is adjusted in a first manner in one of the rule-based algorithms and the bitrate is adjusted in a second manner in another one of the rule-based algorithms, where the first manner is different from the second manner, and wherein in each of the rule-based algorithms, a different ratio between two of the following elements is relied upon: (i) estimated bandwidth; (ii) network stability; (iii) bandwidth delta amplitude; (iv) bandwidth delta confidence; and (v) transmission buffer configuration. ### Claim 14 (depends on 13) The video transmission control system of claim 13, wherein the estimating of the network state includes estimating a current network state through the machine learning using transmission buffer data comprising at least one of a buffer size that represents a size of not-transmitted data in a transmission buffer, a buffer duration that represents a playback time of the not-transmitted data in the transmission buffer, and a throughput that represents a data transmission amount per time unit. ### Claim 15 (depends on 14) The video transmission control system of claim 14, wherein the estimating of the network state further includes estimating a future network state through the machine learning using the transmission buffer data and transmission buffer data corresponding to a desired previous time. ### Claim 16 (depends on 14) The video transmission control system of claim 14, wherein the machine learning further includes using at least one of connection type information and communication provider information as information related to a communication network. ### Claim 17 (depends on 13) The video transmission control system of claim 13, wherein the estimating of the network state includes estimating a current network state that comprises at least one of an estimated bandwidth and transmission buffer data, and a future network state that comprises at least one of a network stability, a bandwidth delta amplitude, a bandwidth delta, and a bandwidth delta confidence, through the machine learning. ### Claim 18 (depends on 13) The video transmission control system of claim 13, wherein the determining of the option includes selecting a rule-based algorithm of an option corresponding to the estimated network state from among a plurality of algorithms in each of which a video transmission option is differently set based on the network state using a method of classifying the network state through the machine learning using a plurality of decision trees. ### Claim 19 (depends on 13) The video transmission control system of claim 13, wherein: the determining of the option includes determining at least two of a bitrate, frames per second (FPS), and a buffer flush rate as a controllable parameter when transmitting video data based on the estimated network state, and the video transmission controller is also configured to calculate at least one parameter among the bitrate, the FPS, and the buffer flush rate as the video transmission option according to the estimated network state, using a regression scheme using a machine learning scheme that has a function to learn based on experience. ### Claim 20 (depends on 19) The video transmission control system of claim 19, wherein the determining of the option comprises determining all three of the bitrate, the FPS and the buffer flush rate as controllable parameters when transmitting video data based on the estimated network state. ### Claim 21 A video transmission control method performed by a server implemented by a computer, comprising: designing a plurality of rule-based algorithms of video transmission options, where each of the rule-based algorithms is optimized differently; estimating a network state through machine learning using information related to a video transmission as a feature in a real-time video streaming environment; and determining an option for controlling the video transmission based on the estimated network state, wherein the determining of the option comprises selecting a rule-based algorithm corresponding to the estimated network state from among the plurality of rule-based algorithms in each of which a video transmission option including at least two of a bitrate, frames per second (FPS), and a buffer flush rate are differently set based on the network state, wherein the selected rule-based algorithm utilizes the FPS and the bitrate as controllable parameters when transmitting video data, and further wherein the bitrate is adjusted in a first manner in one of the rule-based algorithms and the bitrate is adjusted in a second manner in another one of the rule-based algorithms, where the first manner is different from the second manner; and wherein in each of the rule-based algorithms, a different ratio between two of the following elements is relied upon: (i) estimated bandwidth; (ii) network stability; (iii) bandwidth delta amplitude; (iv) bandwidth delta confidence; and (v) transmission buffer configuration.