WO2023246343A1 - 数据处理方法、装置、计算机设备、存储介质及产品 - Google Patents
数据处理方法、装置、计算机设备、存储介质及产品 Download PDFInfo
- Publication number
- WO2023246343A1 WO2023246343A1 PCT/CN2023/092635 CN2023092635W WO2023246343A1 WO 2023246343 A1 WO2023246343 A1 WO 2023246343A1 CN 2023092635 W CN2023092635 W CN 2023092635W WO 2023246343 A1 WO2023246343 A1 WO 2023246343A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- transmission
- time
- parameters
- moment
- network status
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0803—Configuration setting
- H04L41/0823—Configuration setting characterised by the purposes of a change of settings, e.g. optimising configuration for enhancing reliability
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/147—Network analysis or design for predicting network behaviour
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/14—Systems for two-way working
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Definitions
- the present application relates to the field of computer technology, specifically to a data processing method, a data processing device, a computer equipment, a storage medium and a data processing product.
- Embodiments of the present application provide a data processing method, device, equipment, storage medium and product, which can adjust the streaming media transmission strategy to adapt to the changing network status.
- embodiments of the present application provide a data processing method
- Executed by computer equipment including:
- embodiments of the present application provide a data processing method, executed by a computer device, including:
- the transmission parameters at the second time are adjusted, and the streaming media data is transmitted based on the adjusted transmission parameters at the second time.
- embodiments of the present application provide a data processing device, including:
- the first acquisition unit is used to acquire the transmission parameters configured when the target device performs streaming media data transmission at the first moment, and acquire the network status parameters of the transmission link used by the target device at the first moment;
- a first processing unit configured to predict the network status parameters of the transmission link at the second time based on the transmission parameters at the first time and the network status parameters at the first time; and for predicting the network status parameters of the transmission link at the second time based on the The network status parameters at the first time, the transmission parameters at the first time and the predicted network status parameters at the second time are used to determine the transmission reference information at the second time; and,
- a first sending unit configured to send transmission reference information at the second time to the target device, where the transmission reference information is used to instruct the target device to adjust the transmission parameters at the second time, and based on the adjusted
- the transmission parameters at the second time are used for streaming media data transmission, and the first time is earlier than the second time.
- embodiments of the present application provide a data processing device, including:
- the second acquisition unit is used to acquire the transmission parameters configured when streaming media data is transmitted at the first moment, and acquire the network status parameters of the used transmission link at the first moment;
- the second sending unit is configured to send the transmission parameters at the first time and the network status parameters at the first time to the server, so that the server can determine the transmission parameters at the first time and the network status parameters at the first time according to the transmission parameters at the first time and the network status parameters at the first time.
- Network status parameters predict the network status parameters of the transmission link at the second time, and based on the network status parameters at the first time, the transmission parameters at the first time and the predicted network at the second time Status parameter, determine and return the transmission reference information at the second moment;
- the second processing unit is configured to adjust the transmission parameters at the second time based on the transmission reference information at the second time returned by the server, and perform streaming media data transmission based on the adjusted transmission parameters at the second time.
- this application provides a computer device, including:
- a memory has a computer program stored in the memory.
- the computer program is executed by the processor, the above data processing method is implemented.
- the present application provides a computer-readable storage medium.
- the computer-readable storage medium stores a computer program.
- the computer program is adapted to be loaded by a processor and execute the above-mentioned data processing method.
- the present application provides a computer program product or computer program that includes computer instructions stored in a computer-readable storage medium.
- the processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the above-mentioned data processing method.
- Figure 1 is an architectural diagram of a data processing system provided by an embodiment of the present application
- Figure 2 is a flow chart of a data processing method provided by an embodiment of the present application.
- Figure 3 is a flow chart of another data processing method provided by an embodiment of the present application.
- Figure 4a is a training flow chart of a network state prediction model provided by an embodiment of the present application.
- Figure 4b is a model architecture diagram provided by an embodiment of the present application.
- Figure 5 is a flow chart of yet another data processing method provided by an embodiment of the present application.
- Figure 6a is an application scenario diagram of a data processing method provided by an embodiment of the present application.
- Figure 6b is a data processing flow chart in a smart traffic streaming media scenario provided by the embodiment of the present application.
- Figure 7 is a schematic structural diagram of a data processing device provided by an embodiment of the present application.
- FIG. 9 is a schematic structural diagram of a computer device provided by an embodiment of the present application.
- Machine learning is a multi-field interdisciplinary subject involving probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory and other disciplines. It specializes in studying how computers can simulate or implement human learning behavior to acquire new knowledge or skills, and reorganize existing knowledge structures to continuously improve their performance.
- Machine learning/deep learning is the core of AI and the fundamental way to make computers intelligent. It usually includes artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, teaching learning and other technologies.
- Deep learning The concept of deep learning originates from the research of artificial neural networks. For example, a multi-layer perceptron with multiple hidden layers is a deep learning structure. Deep learning discovers distributed feature representations of data by combining low-level features to form more abstract high-level representation attribute categories or features.
- the embodiments of this application mainly involve training an initial model through a training data set to obtain a network state prediction model, and using the network state prediction model to predict the network at the second moment based on the network state at the first moment and the transmission parameters at the first moment. state.
- the target device 101 obtains the network status parameters at the first moment.
- the network status parameters are used to indicate the network status of the transmission link used by the target device 101 in the streaming media transmission system.
- the network status parameters include at least one of the following: information Signal to Interference plus Noise Ratio (SINR), Received Signal Strength Indicator (RSSI), Reference Signal Receiving Quality (RSRP), Reference Signal Receiving Power (RSRQ) ), delay, throughput, physical layer shared channel transmission block size, modulation and coding strategy, data transmission rate.
- SINR Signal to Interference plus Noise Ratio
- RSSI Received Signal Strength Indicator
- RSRP Reference Signal Receiving Quality
- RSRQ Reference Signal Receiving Power
- the target device 101 After obtaining the network status parameters at the first time, the target device 101 sends the network status parameters at the first time and the transmission parameters at the first time to the server 102 .
- the target device 101 can also send alternative information to the server 102 to improve the prediction accuracy of the network status parameters at the second moment.
- the alternative information includes at least one of the following: location information and azimuth information of the target device 101. , latitude and longitude information.
- the first moment is earlier than the second moment. For example, the first moment is the current moment and the second moment is the next moment of the current moment.
- the server 102 receives the network status parameters at the first time and the transmission parameters at the first time sent by the target device 101, and based on the transmission parameters at the first time and the network status parameters at the first time, the server 102 determines the network status parameters used by the target device 101.
- the network status parameters of the transmission link at the second moment are predicted.
- the server 102 can call the network status prediction model to perform prediction processing on the transmission parameters and the network status parameters at the first moment to obtain the transmission link used by the target device 101 at the second moment. Network status parameters.
- the server 102 can obtain historical data, which includes network status parameters at different times and transmission parameters of the target device 101 at that time, and based on the historical data, the target device 101 at the first time The transmission parameters and the network status parameters at the first moment predict the network status parameters at the second moment.
- the server 102 determines the second time based on the network status parameters at the first time, the transmission parameters of the target device 101 at the first time, and the predicted network status parameters at the second time. Transmission reference information at two moments.
- the transmission reference information includes transmission quality when streaming media data is transmitted at different bit rates.
- the server 102 calls the transmission quality model to perform transmission quality detection on the network status at the first moment, the transmission parameters at the first moment, and the predicted network status parameters at the second moment, and obtains data corresponding to each code rate. Transmission quality information (such as transmission quality score, etc.), and determine the data transmission quality information corresponding to each code rate as the transmission reference information at the second moment. After obtaining the transmission reference information at the second time, the server 102 sends the transmission reference information at the second time to the target device 101 .
- the target device 101 After receiving the transmission reference information at the second time returned by the server 102, the target device 101 adjusts the transmission parameters at the second time based on the transmission reference information returned by the server 102, and adjusts the transmission parameters at the second time based on the adjusted second time.
- the transmission parameters are used for streaming media data transmission.
- the transmission reference information at the second moment indicates: the data transmission quality score of code rate 1 at the second moment is 45, the data transmission quality score of code rate 2 at the second moment is 94, and the data transmission quality score of code rate 3 at the second moment is 94.
- data transmission If the quality score is 75, the target device 101 adjusts the transmission code rate to code rate 2, and performs streaming media data transmission based on code rate 2 at the second moment.
- the transmission parameters at the first moment are obtained, the network status parameters of the transmission link at the first moment are obtained, and the transmission parameters at the first moment and the network status parameters at the first moment are sent to the server; the server is based on the first moment.
- the transmission parameters at the first moment and the network status parameters at the first moment are used to predict the network status parameters of the target device's transmission link at the second moment, and based on the network status parameters at the first moment, the transmission parameters at the first moment and the predicted Network status parameters at the second moment, determine and return the transmission reference information at the second moment; the transmission reference information is used to instruct the target device to adjust the transmission parameters at the second moment, and stream media data based on the adjusted transmission parameters at the second moment transmission. It can be seen that by predicting the network status parameters at the second time, the transmission reference information at the second time is generated, so that the target device can adjust the transmission parameters at the second time based on the transmission reference information at the second time to adapt to the changing network status.
- the embodiment of the present application proposes a more detailed data processing method.
- the data processing method proposed by the embodiment of the present application will be introduced in detail below with reference to the accompanying drawings.
- Figure 2 is a flow chart of a data processing method provided by an embodiment of the present application.
- the data processing method may be executed by a computer device, and the computer device may specifically be the server 102 shown in FIG. 1 .
- the data processing method may include but is not limited to S201-S204:
- the transmission parameters configured at the first moment include at least one of the following: the duration of each streaming media segment, the code rate of each streaming media segment, the size of the buffer, The number of streaming segments to be transferred.
- the code rate refers to the amount of data transmitted per unit time.
- the network status parameters of the target device's transmission link at the first moment are used to indicate the network status of the transmission link.
- the network status parameters include at least one of the following: Signal to Interference plus Noise Ratio (SINR), received signal strength Indication (Received Signal Strength Indicator, RSSI), Reference Signal Receiving Quality (RSRP), Reference Signal Receiving Power (RSRQ), delay, throughput, physical layer shared channel transmission block size, Modulation and coding strategies, data transmission rates.
- SINR Signal to Interference plus Noise Ratio
- RSSI received Signal Strength Indication
- RSRP Reference Signal Receiving Quality
- RSRQ Reference Signal Receiving Power
- the transmission parameters of the target device at the first moment and the network status parameters of the target device's transmission link at the first moment can be collected by the target device and provided to the server; they can also be collected by the communication operator. , and provided to the server.
- the network that communication operators are on is a 5G network.
- the first moment is earlier than the second moment.
- the first moment may be the current moment
- the second moment may be the next moment of the first moment.
- the network status parameters at the second moment may include throughput and delay at the second moment.
- the computer device can call the network status prediction model to perform prediction processing on the transmission parameters and the network status parameters at the first time to obtain the network status parameters of the transmission link at the second time.
- the network state prediction model is obtained by training the initial model using the training data set.
- the training data set is generated based on historical data.
- the training model specifically includes:
- the initial model may include but is not limited to a long short-term memory network model (Long Short-Term Memory Network Model).
- Memory Long Short-Term Memory Network Model
- LSTM Long Short-Term Memory Network Model
- RNN recurrent neural network model
- GRU Gated recurrent unit model
- the parameters in the initial model are adjusted based on the loss value to obtain the network state prediction model.
- the transmission reference information at the second moment is used to provide a reference for the target device to determine the transmission parameters at the second moment.
- the transmission reference information at the second moment may include data transmission quality information at the second moment at different code rates.
- the computer device can call the transmission quality model to perform transmission quality detection on the network status of the target device at the first moment, the transmission parameters at the first moment, and the predicted network status parameters at the second moment, and obtain each The data transmission quality information corresponding to the code rate is determined as the transmission reference information at the second moment.
- the transmission reference information is used to instruct the target device to adjust the transmission parameters at the second time and perform streaming media data transmission based on the adjusted transmission parameters at the second time.
- the transmission reference information at the second time indicates: the data transmission quality score of code rate 1 at the second time is 45, the data transmission quality score of code rate 2 at the second time is 94, and the data transmission quality score of code rate 3 at the second time is 94.
- the transmission quality score is 75, then the target device adjusts the transmission parameters at the second time based on the transmission reference information at the second time.
- the transmission parameter refers to the code rate, then the transmission code rate is adjusted to the code rate 2, and the At the second moment, streaming media data is transmitted based on code rate 2.
- the transmission parameters configured by the target device when transmitting streaming media data at the first moment are obtained, and the network status parameters of the transmission link used by the target device at the first moment are obtained.
- the transmission parameters at the first moment and the network status parameters at the first moment and predict the network status parameters of the transmission link at the second moment, based on the network status parameters at the first moment, the transmission parameters at the first moment and the predicted network status parameters at the second moment , determine the transmission reference information at the second moment, and send the transmission reference information at the second moment to the target device.
- the transmission reference information is used to instruct the target device to adjust the transmission parameters at the second moment, and perform the operation based on the adjusted transmission parameters at the second moment.
- the entire streaming media transmission network changes greatly.
- different types of prediction templates can be matched, so that the target device can predict the prediction based on the second moment.
- the transmission reference information adjusts the transmission parameters at the second moment to adapt to the changing network status and support video response under various network changes.
- FIG. 3 is a flow chart of another data processing method provided by an embodiment of the present application.
- the data The processing method may be executed by a computer device, and the computer device may specifically be the server 102 shown in FIG. 1 .
- the data processing method may include but is not limited to S301-S306:
- the data transmitted in the transmission link of the target device includes streaming media data.
- the streaming media data is divided into at least one streaming media segment during the transmission process.
- the transmission parameters of the target device at the first moment include at least one of the following: Items: the duration of each streaming media segment, the bit rate of each streaming media segment (including the bit rate of the streaming media segment that has been transmitted, and the bit rate of the streaming media segment to be transmitted), the size of the buffer, the streaming media segment to be transmitted quantity.
- the specific acquisition method is completed through a communication device outside the streaming media transmission system.
- the target device sends the transmission parameters at the first moment and the network status parameters at the first moment to the communication device, and then the communication device forwards them to the computer device via the core network.
- the communication equipment may include but is not limited to one or more of the following: 4G (fourth-generation) base station, 5G (fifth-generation) base station, Road Side Unit (Roud Side Unit, RSU), Wireless Fidelity (Wireless Fidelity, WiFi) ); the core network can be either the core network or the 5G cloud core network.
- S303 can be executed directly; for example, the target device has structured the network status parameters and provides structured processing to the server. The subsequent network status parameters. If the obtained network status parameters have been structured, continue to execute S302.
- Structured processing is used to extract key information from network status parameters.
- Structured processing includes, for example, the following implementation methods:
- the computer device can perform feature extraction on the network status parameters; for example, perform feature extraction on the network status parameters based on semantics, and use the extracted semantic features as processed network status parameters.
- the computer device can filter the network status parameters based on the data filtering rules to obtain the processed network status parameters; for example, according to the configured filtering rules, filter out the network status parameters that do not meet the requirements, and obtain Processed network status parameters.
- the first moment is earlier than the second moment.
- the first moment may be the current moment
- the second moment may be the next moment of the first moment.
- Figure 4a is a training flow chart of a network state prediction model provided by an embodiment of the present application. As shown in Figure 4a, the training process of the network state prediction model is as follows:
- the computer device generates a training data set and verification data corresponding to the training data set based on historical data.
- the historical data in chronological order, includes the network status parameters of the target device from time 1 to time 10, and the transmission parameters of the target device from time 1 to time 10, then the computer device can record the target device from time 1 to time 9.
- the network status parameters of the target device and the transmission parameters of the target device from time 1 to time 9 are used as the training data set.
- the network status parameters of the target device from time 2 to time 10 and the transmission parameters of the target device from time 2 to time 10 are used as the training data set. Verification data corresponding to the data set.
- the computer device inputs the training data set into the initial model, obtains the prediction data output by the initial model, calculates the loss value between the prediction data and the verification data corresponding to the training data set, and performs calculation on the parameters in the initial model based on the loss value. Adjust (such as adjusting the number of layers of the neural network, the number of neurons in each layer of the neural network, etc.) to obtain a network state prediction model.
- the network state prediction model is output.
- the computer device can also use the network status prediction model to further mine multiple network status parameters and multiple transmission parameters at the same time, and then predict the network status parameters at the second time.
- a multi-terminal/multi-user streaming media transmission parameter (such as code rate) adaptive algorithm is provided, taking into account the coupling between multiple target devices, that is, between multiple users The coupling effect between them can improve the accuracy of the predicted network state parameters at the second moment and be more closely integrated with actual application scenarios.
- the computer device sends the predicted network status parameters at the second moment to the target device, so that the target device adjusts its own operating strategy based on the network status parameters at the second moment.
- the target device may be a vehicle-mounted terminal mounted in the target vehicle, and the computer device sends the predicted network status parameters at the second moment to the target device, so that the target device Based on the predicted network status parameters at the second moment, the target vehicle's driving strategy is adjusted (such as slowing down, ending remote control and taking over manually, or starting remote control, etc.).
- the target device may be a mobile communication device (such as a mobile phone), and the computer device sends the predicted network status parameters at the second moment to the target device, so that the target device Based on the predicted network status parameters at the second moment, the communication method is adjusted (such as switching from video call to voice call, switching video call resolution, etc.).
- the communication method is adjusted (such as switching from video call to voice call, switching video call resolution, etc.).
- the transmission quality model can be a reinforcement learning model (Deep Q Networks, DQN).
- the transmission quality model includes an environment module, a decision-making module and a transmission quality evaluation module.
- the process of computer equipment calling the transmission quality model to perform transmission quality detection includes:
- the configuration information includes N metric indicators and the weight of each metric indicator.
- N is a positive integer; the configuration information can be set by the object based on actual needs, or it can be set by default by the developer.
- the N metrics may include but are not limited to: code rate smoothing and buffering.
- the environment module can be used to indicate the predicted network status at the second moment.
- the data transmission quality information corresponding to the target code rate is determined based on the N metric indicators, the weight of each metric indicator, and the network status at the second moment indicated by the environment module.
- the target code rate may be any code rate that is not selected among the candidate code rates.
- the transmission reference information at the second moment is based on different code rates at the second moment.
- the data transmission quality score under the network status parameters is generated based on the network status parameters of the target device at the first moment and the transmission parameters at the first moment as a reference basis.
- Figure 4b is a model architecture diagram provided by an embodiment of the present application. As shown in Figure 4b, first input the data into the network status prediction model. The data includes the transmission parameters of the target device at the first moment and the network status parameters at the first moment. The network status at the second moment output by the network status prediction model is obtained. parameter.
- the network status parameters predicted by the network status prediction model at the second moment, the transmission parameters of the target device at the first moment, and the network status parameters at the first moment are used as environmental information and input into the transmission quality model.
- the transmission quality model updates the network status based on the network status parameters at the second moment predicted by the network status prediction model, the transmission parameters of the target device at the first moment, and the network status parameters at the first moment, and determines different parameters through the neural network
- the data transmission quality of the code rate (determined each time by the code rate adaptive decision) under the current network status, and then the transmission reference information at the second moment is obtained.
- the transmission quality model can use pensieve as the basic framework.
- the state variables are also adjusted according to the update of the network status parameters (from the update of the network status parameters at the first moment). The new is the network status parameter at the second moment).
- the number of layers of the neural network and the number of neurons shown in Figure 4b are only for examples and do not constitute actual limitations of the present application.
- the layers of the neural network can be modified based on the training results.
- the number of neurons and the number of neurons are adjusted to achieve better results in the transmission quality model.
- the embodiment of the present application can better extract network status features by performing structured processing on the network status parameters, thereby providing the accuracy of predicting the network status parameters at the second moment.
- the target device can adjust its own operating strategy based on the network status parameters at the second moment to adapt to the changing network status.
- it is more closely integrated with the actual scene, thereby improving the data transmission effect.
- FIG. 5 is a flow chart of yet another data processing method provided by an embodiment of the present application.
- the data processing method may be executed by a computer device, and the computer device may specifically be the target device 101 shown in FIG. 1 .
- the data processing method may include but is not limited to S501-S503:
- the transmission parameters configured for streaming media transmission include at least one of the following: the duration of each streaming media segment, the code rate of each streaming media segment, the size of the buffer, and the number of streaming media segments to be transmitted.
- the code rate refers to the amount of data transmitted per unit time.
- the network status parameter of the transmission link at the first moment is used to indicate the network status of the transmission link in the streaming media transmission system.
- the network status parameter includes at least one of the following: Signal to Interference plus Noise Ratio (SINR), Received Signal Strength Indicator (RSSI), Reference Signal Receiving Quality (RSRP), Reference Signal Receiving Power (RSRQ), delay, throughput, physical layer shared channel transmission Block size, modulation and coding strategy, data transfer rate.
- SINR Signal to Interference plus Noise Ratio
- RSSI Received Signal Strength Indicator
- RSRP Reference Signal Receiving Quality
- RSRQ Reference Signal Receiving Power
- the method by which the target device obtains the transmission parameters at the first moment may include: collecting the transmission parameters at the first moment, or obtaining the transmission parameters at the first moment configured by the object.
- the method by which the target device obtains the network status parameters of the used transmission link at the first moment may include: the target device collects the network status parameters of the used transmission link at the first moment, or the target device obtains the network status parameters from the communication operator or network The operator obtains the network status parameters of the used transmission link at the first moment.
- the target device collects the transmission parameters at the first moment in the streaming media transmission and the network status parameters at the first moment, and transmits the transmission parameters at the first moment and the network status parameters at the first moment through the uplink of the communication network.
- the network status parameters at the first moment are uploaded to the communication equipment.
- the communication equipment may include but is not limited to one or more of the following: 4G (fourth-generation) base station, 5G (fifth-generation) base station, Road Side Unit (Roud Side Unit, RSU) ), Wireless Fidelity (WiFi) equipment.
- the communication device then sends the transmission parameters at the first moment and the network status parameters at the first moment to the server through the core network.
- the core network can be either a core network or a 5G cloud core network.
- the server determines the transmission link based on the transmission parameters at the first moment and the network status parameters at the first moment. Predict the network status parameters at the second time, and determine and return the transmission reference information at the second time based on the network status parameters at the first time, the transmission parameters at the first time, and the predicted network status parameters at the second time.
- the server determines the transmission link based on the transmission parameters at the first moment and the network status parameters at the first moment. Predict the network status parameters at the second time, and determine and return the transmission reference information at the second time based on the network status parameters at the first time, the transmission parameters at the first time, and the predicted network status parameters at the second time.
- the specific methods for processing network status parameters include the following two methods:
- the target device sends the network status parameters at the first moment to the server, so that the server performs structured processing on the network status parameters at the first moment.
- the server performs structured processing on the network status parameters at the first moment.
- the target device performs structured processing on the network status parameters at the first moment, and sends the structured network status parameters to the server.
- the target device can generate a decision tree based on the network status parameters, and use the decision tree as the processed network status parameters.
- the target device can also perform feature extraction on the network status parameters; for example, perform feature extraction on the network status parameters based on semantics to obtain processed network status parameters.
- the target device can also filter network status parameters based on data filtering rules to obtain processed network status parameters; for example, filter out network status parameters that do not meet the requirements according to configured filtering rules to obtain processed network status parameters.
- the transmission parameters at the second moment include at least one of the following: the duration of the streaming media segment transmitted at the second moment, the code rate of the streaming media segment transmitted at the second moment, and the size of the buffer.
- the transmission reference information at the second moment is used to provide a reference for the target device to determine the transmission parameters at the second moment.
- the transmission reference information at the second moment may include data transmission quality information at the second moment at different code rates.
- the data transmitted in the transmission link includes streaming media data, and the streaming media data is divided into at least one streaming media segment during the transmission process.
- the transmission reference information at the second moment includes: the transmission quality of the streaming media segments with different bit rates at the second moment.
- the transmission reference information at the second moment may include the data transmission quality score of the streaming media segment at the second moment at the code rate.
- the transmission parameters at the second moment include: the code rate of the streaming media segment transmitted at the second moment.
- the target device adjusts the code rate of the streaming media segment transmitted at the second time to the target code rate based on the transmission reference information at the second time; wherein, among the various code rates indicated by the transmission reference information at the second time, the target code rate is The code rate whose transmission quality at the second moment is higher than the transmission quality threshold; for example, the target code rate may be the code rate with the highest data transmission quality score at the second moment among the candidate code rates.
- Figure 6a is an application scenario diagram of a data processing method provided by an embodiment of the present application.
- the application scenario is a smart transportation streaming media scenario.
- the scenario also includes a base station 603, a core network 604, and a remote control server.
- the remote control server 605 can be used to provide remote control services for the vehicle 601, and the transportation business 606 can include but is not limited to: unmanned truck collection, vehicle-road collaboration, real-time twins, etc.
- Figure 6b is a data processing flow chart in a smart traffic streaming media scenario provided by an embodiment of the present application.
- the target device or communication operator, such as a base station
- the target device After obtaining the network status parameters at the first moment, in S612, the target device performs structured processing on the network status parameters, and sends the structured network status parameters to the application server.
- the target device After specific implementation methods of performing structured processing on network status parameters, reference may be made to the implementation method in S302, which will not be described again here.
- the target device uploads the structured network status parameters and the transmission parameters at the first moment to the communication device.
- communication equipment may include but is not limited to: 4G/5G base stations, One or more of RSU and WiFi.
- the communication device uses relevant interfaces to forward the structured network status parameters and the transmission parameters of the target device at the first moment to the application server through the core network.
- the core network can be one of the 5G core network or the 5G cloud core network.
- the application server inputs the structured network status parameters and the transmission parameters of the target device at the first moment into the network status prediction model and the transmission quality model.
- the application server can preprocess the received data (such as generating a matrix) to make the data conform to the network status. Input requirements for prediction models, or transmission quality models.
- the application server can configure relevant parameters of the network status prediction model and transmission quality model according to actual needs, such as configuring input features, output features, prediction features, model parameters, prediction time, model training and testing ratio, number of iterations, model Loss function, model optimization function, number of machine learning neurons, etc.
- the application server can return the network status parameters at the second moment to the target device (vehicle terminal), so that the target device can adjust the vehicle driving strategy (such as slowing down, ending remote control and manually taking over, based on the network status parameters at the second moment). Or start remote control, etc.).
- vehicle driving strategy such as slowing down, ending remote control and manually taking over, based on the network status parameters at the second moment. Or start remote control, etc.
- the application server may send the transmission reference information at the second moment to the target device, so that the target device adjusts the transmission parameters at the second moment (such as increasing/lowering the camera bit rate to adjust the transmission based on the transmission reference information at the second moment). bit rate), and perform streaming media data transmission based on the adjusted transmission parameters at the second moment.
- Figure 7 is a schematic structural diagram of a data processing device provided by an embodiment of the present application.
- the device can be mounted on a computer device.
- the computer device can be the server 102 shown in Figure 1.
- the data processing device shown in Figure 7 can be used to perform some or all of the functions in the method embodiments described in Figures 2 and 3 above. Please refer to Figure 7.
- the detailed description of each unit is as follows:
- the first obtaining unit 701 is used to obtain the transmission parameters configured by the target device when transmitting streaming media data at the first moment, and obtain the network status parameters of the transmission link used by the target device at the first moment;
- the first processing unit 702 is configured to predict the network status parameters of the transmission link at the second time based on the transmission parameters at the first time and the network status parameters at the first time; and to predict the network status parameters of the transmission link at the first time based on the network status parameters at the first time. , the transmission parameters at the first moment and the predicted network status parameters at the second moment, determine the transmission reference information at the second moment;
- the first sending unit 703 is also configured to send transmission reference information at the second time to the target device.
- the transmission reference information is used to instruct the target device to adjust the transmission parameters at the second time, and perform streaming based on the adjusted transmission parameters at the second time.
- the first moment is earlier than the second moment.
- the first processing unit 702 is configured to predict the network status parameters of the transmission link at the second time based on the transmission parameters at the first time and the network status parameters at the first time, specifically for:
- the first processing unit 702 is also used to train the network state prediction model, specifically including:
- the parameters in the initial model are adjusted based on the loss value to obtain the network state prediction model.
- the first processing unit 702 is configured to call the transmission quality model and perform transmission quality detection on the network status parameters at the first moment, the transmission parameters at the first moment, and the predicted network status parameters at the second moment. , obtain the data transmission quality information corresponding to each code rate;
- the data transmission quality information corresponding to each code rate is determined as the transmission reference information at the second moment.
- the transmission quality model includes an environment module, a decision-making module and a transmission quality evaluation module; the first processing unit 702 is configured to obtain configuration information, where the configuration information includes N metric indicators and the value of each metric indicator. Weight, N is a positive integer; configure the environment module based on the network status parameters at the first moment, the transmission parameters at the first moment, and the predicted network status parameters at the second moment; through the decision Module, selects the target code rate as the transmission code rate at the second moment; through the transmission quality evaluation module, according to the N metric indicators, the weight of each metric indicator, and the second second time indicated by the environment module The network status at the moment is determined to determine the data transmission quality information corresponding to the target code rate.
- the first processing unit 702 is configured to perform structured processing on the network status parameters to obtain processed network status parameters; according to the transmission parameters at the first moment and the processed network Status parameters: predict the network status parameters of the transmission link at the second moment.
- the network status parameters include at least one of the following: signal-to-interference ratio, received signal strength indication, reference signal reception quality, reference signal reception power, delay, throughput, physical layer shared channel transmission block size, modulation With encoding strategy, data transfer rate.
- the data transmitted by the target device using the transmission link includes streaming media data.
- the streaming media data is divided into at least one streaming media segment during the transmission process.
- the transmission parameters of the target device at the first moment include At least one of the following: the duration of each streaming media segment, the bit rate of each streaming media segment, the size of the buffer, and the number of streaming media segments to be transmitted.
- some of the steps involved in the data processing methods shown in Figures 2 and 3 can be performed by various units in the data processing device shown in Figure 7.
- S201 shown in Figure 2 can be performed by the first obtaining unit 701 shown in Figure 7
- S202 and S203 can be performed by the first processing unit 702 shown in Figure 7
- S204 can be performed by the first sending unit 703 shown in Figure 7 implement.
- S301 shown in Figure 3 can be executed by the transceiver unit 701 shown in Figure 7
- S302, S303 and S305 can be executed by the first processing unit 702 shown in Figure 7
- S304 and S306 can be executed by the first sending unit 703 shown in Figure 7 implement.
- the units 7 can be separately or entirely combined into one or several additional units, or some of the units can be further divided into multiple functionally smaller units. It is composed of units, which can achieve the same operation without affecting the realization of the technical effects of the embodiments of the present application.
- the above units are divided based on logical functions. In practical applications, the function of one unit can also be realized by multiple units, or the functions of multiple units can be realized by one unit. In other embodiments of the present application, the data processing device may also include other units. In practical applications, these functions are also It can be implemented with the assistance of other units, and can be implemented collaboratively by multiple units.
- a general computing device such as a computer including a central processing unit (CPU), a random access storage medium (RAM), a read-only storage medium (ROM), and other processing elements and storage elements can be used.
- Run a computer program (including program code) capable of executing the steps involved in the corresponding methods shown in Figures 2 and 3 to construct the data processing device shown in Figure 7, and to implement the embodiments of the present application.
- the computer program can be recorded on, for example, a computer-readable recording medium, loaded into the above-mentioned computing device through the computer-readable recording medium, and run therein.
- Figure 8 is a schematic structural diagram of another data processing device provided by an embodiment of the present application.
- the device can be mounted on a computer device.
- the computer device can be the target device 101 shown in Figure 1.
- the data processing device shown in Figure 8 can be used to perform some or all of the functions in the method embodiment described in Figure 5 above. Please refer to Figure 8.
- the detailed description of each unit is as follows:
- the second obtaining unit 801 is used to obtain the transmission parameters configured when transmitting streaming media data at the first moment, and obtain the network status parameters of the used transmission link at the first moment;
- the second sending unit 802 is configured to send the transmission parameters at the first time and the network status parameters at the first time to the server, so that the server can perform the transmission link on the transmission link based on the transmission parameters at the first time and the network status parameters at the first time. Predict the network status parameters at the second moment, and determine and return the transmission reference information at the second moment based on the network status parameters at the first moment, the transmission parameters at the first moment, and the predicted network status parameters at the second moment;
- the second processing unit 803 is configured to adjust the transmission parameters at the second time based on the transmission reference information at the second time returned by the server, and perform streaming media data transmission based on the adjusted transmission parameters at the second time.
- the second sending unit 802 is configured to send the network status parameters at the first moment to the server, so that the server performs structured processing on the network status parameters at the first moment; or,
- the data transmitted in the transmission link includes streaming media data, and the streaming media data is divided into at least one streaming media segment during the transmission process;
- the transmission reference information at the second moment includes: different code rates
- the transmission parameters at the second moment include: the bit rate of the streaming media segment transmitted at the second moment;
- the second processing unit 803 is configured to adjust the code rate of the streaming media segment transmitted at the second time to the target code rate based on the transmission reference information at the second time;
- the target code rate is a code rate whose transmission quality at the second time is higher than the transmission quality threshold among each code rate indicated by the transmission reference information at the second time.
- some of the steps involved in the data processing method shown in Figure 5 can be performed by various units in the data processing device shown in Figure 8 .
- S501 shown in FIG. 5 may be executed by the second obtaining unit 801 shown in FIG. 8
- S502 may be executed by the second sending unit 802 shown in FIG. 8
- S503 may be executed by the second processing unit 803 shown in FIG. 8 .
- Each unit in the data processing device shown in Figure 8 can be separately or entirely combined into one or several additional units, or some of the units can be further split. It is composed of multiple functionally smaller units, which can achieve the same operation without affecting the realization of the technical effects of the embodiments of the present application.
- the above units are divided based on logical functions.
- the function of one unit can also be realized by multiple units, or the functions of multiple units can be realized by one unit.
- the data processing device may also include other units.
- these functions may also be implemented with the assistance of other units, and may be implemented by multiple units in cooperation.
- a general computing device such as a computer including a central processing unit (CPU), a random access storage medium (RAM), a read-only storage medium (ROM), and other processing elements and storage elements can be used.
- Run a computer program (including program code) capable of executing the steps involved in the corresponding method as shown in Figure 5 to construct the data processing device as shown in Figure 8 and implement the data processing method of the embodiment of the present application.
- the computer program can be recorded on, for example, a computer-readable recording medium, loaded into the above-mentioned computing device through the computer-readable recording medium, and run therein.
- Figure 9 is a schematic structural diagram of a computer device provided by an embodiment of the present application.
- the computer device at least includes a processor 901, a communication interface 902, and a memory 903.
- the processor 901, the communication interface 902 and the memory 903 can be connected through a bus or other means.
- the processor 901 or central processing unit (Central Processing Unit, CPU) is the computing core and control core of the computer device.
- CPU Central Processing Unit
- the CPU can parse various instructions in the computer device and process various data of the computer device, such as:
- the CPU can be used to parse the power on and off instructions sent by the user to the computer device, and control the computer device to perform power on and off operations; another example: the CPU can transmit various types of interactive data between the internal structures of the computer device, and so on.
- the communication interface 902 can optionally include standard wired interfaces and wireless interfaces (such as WI-FI, mobile communication interfaces, etc.), and can be used to send and receive data under the control of the processor 901; the communication interface 902 can also be used for internal data of computer equipment. transmission and interaction.
- Memory 903 (Memory) is a memory device in a computer device, used to store programs and data.
- the memory 903 here may include a built-in memory of the computer device, and of course may also include an extended memory supported by the computer device.
- the memory 903 provides storage space, and the storage space stores the operating system of the computer device, which may include but is not limited to: Android system, iOS system, Windows Phone system, etc. This application is not limited to this.
- Embodiments of the present application also provide a computer-readable storage medium (Memory).
- the computer-readable storage medium is a memory device in a computer device and is used to store programs and data. It can be understood that the computer-readable storage medium here may include a built-in storage medium in the computer device, and of course may also include an extended storage medium supported by the computer device.
- Computer-readable storage media provide storage space that stores the processing system of the computer device. Furthermore, one or more instructions suitable for being loaded and executed by the processor 901 are also stored in the storage space. These instructions may be one or more computer programs (including program codes).
- the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory (non-volatile memory), such as at least one disk memory; optionally, it can also be at least one located remotely The computer-readable storage medium of the aforementioned processor.
- the computer device may be the server 102 shown in FIG. 1 .
- the processor 901 performs the following operations by running the executable program code in the memory 903:
- the communication interface 902 obtain the transmission configuration configured by the target device when transmitting streaming media data at the first moment. parameters, and obtain the network status parameters of the transmission link used by the target device at the first moment;
- the transmission reference information is used to instruct the target device to adjust the transmission parameters at the second moment, and perform streaming media data transmission based on the adjusted transmission parameters at the second moment,
- the first moment is earlier than the second moment.
- the processor 901 predicts the network status parameters of the transmission link at the second time based on the transmission parameters at the first time and the network status parameters at the first time, and specifically executes:
- the network status prediction model is called to perform prediction processing on the transmission parameters and the network status parameters at the first moment to obtain the network status parameters of the transmission link at the second moment.
- the processor 901 also performs the following operations:
- Training the network state prediction model specifically includes:
- the parameters in the initial model are adjusted based on the loss value to obtain the network state prediction model.
- the processor 901 determines the transmission reference information at the second moment based on the network status parameters at the first moment, the transmission parameters at the first moment, and the predicted network status parameters at the second moment, and specifically executes :
- the data transmission quality information corresponding to each code rate is determined as the transmission reference information at the second moment.
- the transmission quality model includes an environment module, a decision-making module and a transmission quality evaluation module; the processor 901 calls the transmission quality model to calculate the network status parameters at the first moment and the transmission at the first moment. Parameters and the predicted network status parameters at the second moment are used for transmission quality detection to obtain data transmission quality information corresponding to each code rate. Specifically, perform:
- Obtain configuration information which includes N metric indicators and the weight of each metric indicator, where N is a positive integer;
- Configure the environment module based on the network status parameters at the first time, the transmission parameters at the first time, and the predicted network status parameters at the second time;
- the data transmission quality information corresponding to the target code rate is determined based on the N metric indicators, the weight of each metric indicator, and the network status at the second moment indicated by the environment module. .
- the processor 901 predicts the network status parameters of the transmission link at the second time based on the transmission parameters at the first time and the network status parameters at the first time. Specifically, implement:
- the network status parameters of the transmission link at the second time are predicted.
- the network status parameters include at least one of the following: signal-to-interference ratio, received signal strength indication, reference signal reception quality, reference signal reception power, delay, throughput, physical layer shared channel transmission block size , modulation and coding strategies, data transmission rates.
- the data transmitted by the target device using the transmission link includes streaming media data.
- the streaming media data is divided into at least one streaming media segment during the transmission process.
- the transmission parameters of the target device at the first moment include At least one of the following: the duration of each streaming media segment, the bit rate of each streaming media segment, the size of the buffer, and the number of streaming media segments to be transmitted.
- the computer device may be the target device 101 shown in FIG. 1 .
- the processor 1001 performs the following operations by running the executable program code in the memory 1003:
- the transmission parameters at the first time and the network status parameters at the first time are sent to the server through the communication interface 1002, so that the server determines the transmission link at the second time based on the transmission parameters at the first time and the network status parameters at the first time. Predict the network status parameters, and determine and return the transmission reference information at the second time based on the network status parameters at the first time, the transmission parameters at the first time, and the predicted network status parameters at the second time;
- the transmission parameters at the second time are adjusted, and the streaming media data is transmitted based on the adjusted transmission parameters at the second time.
- the processor 1001 sends the network status parameters at the first moment to the server through the communication interface 1002, specifically executing:
- the data transmitted in the transmission link includes streaming media data, and the streaming media data is divided into at least one streaming media segment during the transmission process;
- the transmission reference information at the second moment includes: different code rates
- the transmission parameters at the second moment include: the bit rate of the streaming media segment transmitted at the second moment;
- the processor 1001 adjusts the transmission parameters at the second time based on the transmission reference information at the second time returned by the server, and specifically executes:
- the target code rate is a code rate whose transmission quality at the second time is higher than the transmission quality threshold among each code rate indicated by the transmission reference information at the second time.
- Embodiments of the present application also provide a computer-readable storage medium.
- the computer-readable storage medium stores a computer-readable storage medium.
- the computer program is adapted to be loaded by the processor and execute the data processing method of the above method embodiment.
- An embodiment of the present application also provides a computer program product.
- the computer program product includes a computer program.
- the computer program is adapted to be loaded by a processor and execute the data processing method of the above method embodiment.
- Embodiments of the present application also provide a computer program product or computer program.
- the computer program product or computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium.
- the processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the above-mentioned data processing method.
- Modules in the device of the embodiment of the present application can be merged, divided, and deleted according to actual needs.
- the program can be stored in a computer-readable storage medium, and the readable storage medium can Including: flash disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk, etc.
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Environmental & Geological Engineering (AREA)
- Multimedia (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
本申请实施例公开了一种数据处理方法、装置、设备、存储介质及产品。其中方法包括:获取目标设备在第一时刻进行流媒体数据传输时配置的传输参数,并获取目标设备所使用的传输链路在第一时刻的网络状态参数,根据第一时刻的传输参数及第一时刻的网络状态参数,对传输链路在第二时刻的网络状态参数进行预测,基于第一时刻的网络状态参数、第一时刻的传输参数以及预测得到的第二时刻的网络状态参数,确定第二时刻的传输参考信息,向目标设备发送第二时刻的传输参考信息,传输参考信息用于指示目标设备调整第二时刻的传输参数,并基于调整后的所述第二时刻的传输参数进行流媒体数据传输,所述第一时刻早于所述第二时刻。
Description
本申请要求于2022年6月21日提交中国专利局、申请号为202210716471.7、申请名称为“一种数据处理方法、装置、设备、存储介质及产品”的中国专利申请的优先权。
本申请涉及计算机技术领域,具体涉及一种数据处理方法、一种数据处理装置、一种计算机设备、一种存储介质及一种数据处理产品。
发明背景
随着科技研究的进步,许多传统业务从线下被转移到了线上。许多线上业务是依赖流媒体传输进行的,例如线上会议、视频通话、智能驾驶等。实践发现,由于网络状态受地理位置、用户数量、网络波动等因素的影响不断变化,线上业务在进行过程中可能出现卡顿、数据传输质量较差的情况,如何调整流媒体传输策略来适应不断变化的网络状态成为当前研究的热门问题。
发明内容
本申请实施例提供了一种数据处理方法、装置、设备、存储介质及产品,能够调整流媒体传输策略来适应不断变化的网络状态。
一方面,本申请实施例提供了一种数据处理方法,
由计算机设备执行,包括:
获取目标设备在第一时刻进行流媒体数据传输时配置的传输参数,并获取所述目标设备所使用的传输链路在所述第一时刻的网络状态参数;
根据所述第一时刻的传输参数及所述第一时刻的网络状态参数,对所述传输链路在第二时刻的网络状态参数进行预测;
基于所述第一时刻的网络状态参数、所述第一时刻的传输参数以及预测得到的所述第二时刻的网络状态参数,确定所述第二时刻的传输参考信息;及,
向所述目标设备发送所述第二时刻的传输参考信息,所述传输参考信息用于指示所述目标设备调整所述第二时刻的传输参数,并基于调整后的所述第二时刻的传输参数进行流媒体数据传输,所述第一时刻早于所述第二时刻。
另一方面,本申请实施例提供了一种数据处理方法,由计算机设备执行,包括:
获取在第一时刻进行流媒体数据传输时配置的传输参数,并获取所使用的传输链路在所述第一时刻的网络状态参数;
向服务器发送所述第一时刻的传输参数及所述第一时刻的网络状态参数,以使所述服务器根据所述第一时刻的传输参数及所述第一时刻的网络状态参数,对所述传输链路在第二时刻的网络状态参数进行预测,并基于所述第一时刻的网络状态参数、所述第一时刻的传输参数以及预测得到的第二时刻的网络状态参数,确定并返回第二时刻的传输参考信息;及,
基于所述服务器返回的所述第二时刻的传输参考信息,调整第二时刻的传输参数,并基于调整后的第二时刻的传输参数进行流媒体数据传输。
另一方面,本申请实施例提供了一种数据处理装置,包括:
第一获取单元,用于获取目标设备在第一时刻进行流媒体数据传输时配置的传输参数,并获取所述目标设备所使用的传输链路在所述第一时刻的网络状态参数;
第一处理单元,用于根据所述第一时刻的传输参数及所述第一时刻的网络状态参数,对所述传输链路在第二时刻的网络状态参数进行预测;以及用于基于所述第一时刻的网络状态参数、所述第一时刻的传输参数以及预测得到的所述第二时刻的网络状态参数,确定所述第二时刻的传输参考信息;及,
第一发送单元,用于向所述目标设备发送所述第二时刻的传输参考信息,所述传输参考信息用于指示所述目标设备调整所述第二时刻的传输参数,并基于调整后的所述第二时刻的传输参数进行流媒体数据传输,所述第一时刻早于所述第二时刻。
另一方面,本申请实施例提供了一种数据处理装置,包括:
第二获取单元,用于获取在第一时刻进行流媒体数据传输时配置的传输参数,并获取所使用的传输链路在所述第一时刻的网络状态参数;
第二发送单元,用于向服务器发送所述第一时刻的传输参数及所述第一时刻的网络状态参数,以使所述服务器根据所述第一时刻的传输参数及所述第一时刻的网络状态参数,对所述传输链路在第二时刻的网络状态参数进行预测,并基于所述第一时刻的网络状态参数、所述第一时刻的传输参数以及预测得到的第二时刻的网络状态参数,确定并返回第二时刻的传输参考信息;
第二处理单元,用于基于所述服务器返回的所述第二时刻的传输参考信息,调整第二时刻的传输参数,并基于调整后的第二时刻的传输参数进行流媒体数据传输。
另一方面,本申请提供了一种计算机设备,包括:
处理器,用于加载并执行计算机程序;
存储器,该存储器中存储有计算机程序,该计算机程序被处理器执行时,实现上述数据处理方法。
另一方面,本申请提供了一种计算机可读存储介质,计算机可读存储介质存储有计算机程序,该计算机程序适于由处理器加载并执行上述数据处理方法。
另一方面,本申请提供了一种计算机程序产品或计算机程序,该计算机程序产品或计算机程序包括计算机指令,该计算机指令存储在计算机可读存储介质中。计算机设备的处理器从计算机可读存储介质读取该计算机指令,处理器执行该计算机指令,使得该计算机设备执行上述数据处理方法。
附图简要说明
图1为本申请实施例提供的一种数据处理系统的架构图;
图2为本申请实施例提供的一种数据处理方法的流程图;
图3为本申请实施例提供的另一种数据处理方法的流程图;
图4a为本申请实施例提供的一种网络状态预测模型的训练流程图;
图4b为本申请实施例提供的一种模型架构图;
图5为本申请实施例提供的再一种数据处理方法的流程图;
图6a为本申请实施例提供的一种数据处理方法的应用场景图;
图6b为本申请实施例提供的一种智慧交通流媒体场景下的数据处理流程图;
图7为本申请实施例提供的一种数据处理装置的结构示意图;
图8为本申请实施例提供的另一种数据处理装置的结构示意图;
图9为本申请实施例提供的一种计算机设备的结构示意图。
实施方式
下面将结合本申请实施例中的附图,对本申请实施例中的技术方案进行清楚、完整地描述。
本申请实施例涉及人工智能,下面对云技术以及人工智能的相关术语及概念进行简要介绍:
云技术(Cloud technology)是指在广域网或局域网内将硬件、软件、网络等系列资源统一起来,实现数据的计算、储存、处理和共享的一种托管技术。
人工智能(Artificial Intelligence,AI)是利用数字计算机或者数字计算机控制的机器模拟、延伸和扩展人的智能,感知环境、获取知识并使用知识获得最佳结果的理论、方法、技术及应用系统。人工智能软件技术主要包括计算机视觉技术、语音处理技术、自然语言处理技术以及机器学习/深度学习等几大方向。
机器学习是一门多领域交叉学科,涉及概率论、统计学、逼近论、凸分析、算法复杂度理论等多门学科。专门研究计算机怎样模拟或实现人类的学习行为,以获取新的知识或技能,重新组织已有的知识结构使之不断改善自身的性能。机器学习/深度学习是AI的核心,是使计算机具有智能的根本途径,通常包括人工神经网络、置信网络、强化学习、迁移学习、归纳学习、式教学习等技术。
深度学习:深度学习的概念源于人工神经网络的研究。例如,含多隐层的多层感知器就是一种深度学习结构。深度学习通过组合低层特征形成更加抽象的高层表示属性类别或特征,以发现数据的分布式特征表示。本申请实施例主要涉及通过训练数据集对初始模型进行训练,得到网络状态预测模型,并采用网络状态预测模型,基于第一时刻的网络状态和第一时刻的传输参数,预测第二时刻的网络状态。
基于上述对本申请实施例涉及的云技术和人工智能的相关介绍,下面简单介绍本申请实施例基于上述云技术和人工智能提出的数据处理方案,能够使得目标设备动态调整流媒体传输策略来适应不断变化的网络状态。
请参阅图1,图1为本申请实施例提供的一种数据处理系统的架构图。如图1所示,该数据处理系统可以包括:目标设备101,服务器102。本申请实施例提供的数据处理方法可由服务器102执行。
目标设备101可以包括但不限于:智能手机(如Android手机、IOS手机等)、平板电脑、便携式个人计算机、移动互联网设备(Mobile Internet Devices,简称MID)、车载终端等智能设备,本申请实施例对此不做限定。
服务器102可以是独立的物理服务器,也可以是多个物理服务器构成的服务器集群或者分布式系统,还可以是提供云服务、云数据库、云计算、云函数、云存储、网络服务、云通信、中间件服务、域名服务、安全服务、CDN(Content Delivery Network,内容分发网络)、以及大数据和人工智能平台等基础云计算服务的云服务器,本申请实施例对此不做限定。
需要说明的是,图1中目标设备101和服务器102之间可以通过有线通信或者无线通信方式进行直接或间接地连接,本申请在此不做限制。目标设备和服务器的数量仅用于举例,并不构成本申请的实际限定;例如,数据处理系统中还可以包括服务器104。
关于数据处理方案的描述如下:
(1)目标设备101获取第一时刻的网络状态参数,网络状态参数用于指示目标设备101在流媒体传输系统中所使用的传输链路的网络状态,网络状态参数包括以下至少一项:信干比(Signal to Interference plus Noise Ratio,SINR)、接收信号强度指示(Received Signal Strength Indicator,RSSI)、参考信号接收质量(Reference Signal Receiving Quality,RSRP)、参考信号接收功率(Reference Signal Receiving Power,RSRQ)、时延、吞吐量、物理层共享信道传输块大小、调制与编码策略、数据传输速率。
在获取到第一时刻的网络状态参数后,目标设备101将第一时刻的网络状态参数以及第一时刻的传输参数发送给服务器102。
其中,第一时刻的传输参数是目标设备101在第一时刻进行流媒体数据传输时配置的传输参数,传输参数包括以下至少一项:各个流媒体片段的时长、各个流媒体片段的码率、缓冲区的大小、待传输流媒体片段的数量。码率是指数据在单位时间内的传输量。
可选的,目标设备101还可以向服务器102发送备选信息,以提高第二时刻的网络状态参数的预测准确率,备选信息包括以下至少一项:目标设备101的位置信息、方位角信息、经纬度信息。第一时刻早于第二时刻,例如,第一时刻是当前时刻,第二时刻是当前时刻的下一时刻。
(2)服务器102接收目标设备101发送的第一时刻的网络状态参数以及第一时刻的传输参数,并根据第一时刻的传输参数及第一时刻的网络状态参数,对目标设备101所使用的传输链路在第二时刻的网络状态参数进行预测。
在一种实施方式中,服务器102可以调用网络状态预测模型,对第一时刻的传输参数及第一时刻的网络状态参数进行预测处理,得到目标设备101所使用的传输链路在第二时刻的网络状态参数。
在另一种实施方式中,服务器102可以获取历史数据,历史数据包括不同时刻下的网络状态参数,以及该时刻下目标设备101的传输参数,并基于历史数据,目标设备101在第一时刻的传输参数以及第一时刻的网络状态参数预测第二时刻的网络状态参数。
(3)服务器102在预测得到第二时刻的网络状态参数后,基于第一时刻的网络状态参数、目标设备101在第一时刻的传输参数以及预测得到的第二时刻的网络状态参数,确定第二时刻的传输参考信息。
传输参考信息包括以不同码率进行流媒体数据传输时的传输质量。在一种实施方式中,服务器102调用传输质量模型对第一时刻的网络状态、第一时刻的传输参数以及预测得到的第二时刻的网络状态参数进行传输质量检测,得到各个码率对应的数据传输质量信息(如传输质量评分等),并将各个码率对应的数据传输质量信息确定为第二时刻的传输参考信息。服务器102在得到第二时刻的传输参考信息后,向目标设备101发送第二时刻的传输参考信息。
(4)目标设备101在接收到服务器102返回的第二时刻的传输参考信息后,基于服务器返回的第二时刻的传输参考信息,调整第二时刻的传输参数,并基于调整后的第二时刻的传输参数进行流媒体数据传输。
例如,假设第二时刻的传输参考信息指示:码率1在第二时刻的数据传输质量评分为45,码率2在第二时刻的数据传输质量评分为94,码率3在第二时刻的数据传输
质量评分为75,则目标设备101将传输码率调整为码率2,并在第二时刻基于码率2进行流媒体数据传输。
本申请实施例中,获取第一时刻的传输参数,并获取传输链路在第一时刻的网络状态参数,向服务器发送第一时刻的传输参数及第一时刻的网络状态参数;服务器根据第一时刻的传输参数及第一时刻的网络状态参数,对目标设备传输链路在第二时刻的网络状态参数进行预测,并基于第一时刻的网络状态参数、第一时刻的传输参数以及预测得到的第二时刻的网络状态参数,确定并返回第二时刻的传输参考信息;传输参考信息用于指示目标设备调整第二时刻的传输参数,并基于调整后的第二时刻的传输参数进行流媒体数据传输。可见,通过预测第二时刻的网络状态参数,来生成第二时刻的传输参考信息,使得目标设备可以基于第二时刻的传输参考信息调整第二时刻的传输参数,来适应不断变化的网络状态。
基于上述数据处理方案,本申请实施例提出更为详细的数据处理方法,下面将结合附图对本申请实施例提出的数据处理方法进行详细介绍。
请参阅图2,图2为本申请实施例提供的一种数据处理方法的流程图。该数据处理方法可以由计算机设备执行,该计算机设备具体可以是图1所示的服务器102。如图2所示,该数据处理方法可包括但不限于S201-S204:
S201、获取目标设备在第一时刻进行流媒体数据传输时配置的传输参数,并获取目标设备所使用的传输链路在第一时刻的网络状态参数。
目标设备在流媒体传输系统中进行流媒体数据传输时,在第一时刻配置的传输参数,包括以下至少一项:各个流媒体片段的时长、各个流媒体片段的码率、缓冲区的大小、待传输流媒体片段的数量。码率是指数据在单位时间内的传输量。
目标设备传输链路在第一时刻的网络状态参数用于指示该传输链路的网络状态,网络状态参数包括以下至少一项:信干比(Signal to Interference plus Noise Ratio,SINR)、接收信号强度指示(Received Signal Strength Indicator,RSSI)、参考信号接收质量(Reference Signal Receiving Quality,RSRP)、参考信号接收功率(Reference Signal Receiving Power,RSRQ)、时延、吞吐量、物理层共享信道传输块大小、调制与编码策略、数据传输速率。
在具体实现过程中,目标设备在第一时刻的传输参数,以及目标设备传输链路在第一时刻的网络状态参数可以是由目标设备采集,并提供给服务器;也可以是由通信运营商采集,并提供给服务器。例如,通信运营商所处的网络是5G网络。
S202、根据第一时刻的传输参数及第一时刻的网络状态参数,对传输链路在第二时刻的网络状态参数进行预测。
第一时刻早于第二时刻,例如,第一时刻可以是当前时刻,第二时刻可以是第一时刻的下一时刻。第二时刻的网络状态参数可以包括第二时刻的吞吐量和时延。
在一种实施方式中,计算机设备可以调用网络状态预测模型,对第一时刻的传输参数及第一时刻的网络状态参数进行预测处理,得到传输链路在第二时刻的网络状态参数。
其中,网络状态预测模型是采用训练数据集对初始模型进行训练得到的,训练数据集是基于历史数据生成的,训练模型具体包括:
获取历史数据,所述历史数据包括目标设备在不同时刻下发送的网络状态参数,
以及各个网络状态参数对应的传输参数;
基于所述历史数据生成训练数据集,并将所述训练数据集输入初始模型,得到所述初始模型输出的预测数据;其中,初始模型可以包括但不限于长短期记忆网络模型(Long Short-Term Memory,LSTM)、循环神经网络模型(Recurrent Neural Network,RNN)、门控循环单元模型(Gated Recurrent Unit,GRU)等时序模型;
计算所述预测数据与所述训练数据集对应的校验数据之间的损失值;
基于所述损失值对所述初始模型中的参数进行调整,得到所述网络状态预测模型。
S203、基于第一时刻的网络状态参数、第一时刻的传输参数以及预测得到的第二时刻的网络状态参数,确定第二时刻的传输参考信息。
第二时刻的传输参考信息用于为目标设备确定第二时刻的传输参数提供参考。第二时刻的传输参考信息可以包括不同码率在第二时刻的数据传输质量信息。
在一种实施方式中,计算机设备可以调用传输质量模型,对目标设备在第一时刻的网络状态、第一时刻的传输参数以及预测得到的第二时刻的网络状态参数进行传输质量检测,得到各个码率对应的数据传输质量信息,并将各个码率对应的数据传输质量信息,确定为第二时刻的传输参考信息。
S204、向目标设备发送第二时刻的传输参考信息。
传输参考信息用于指示目标设备调整第二时刻的传输参数,并基于调整后的第二时刻的传输参数进行流媒体数据传输。
例如,第二时刻的传输参考信息指示:码率1在第二时刻的数据传输质量评分为45,码率2在第二时刻的数据传输质量评分为94,码率3在第二时刻的数据传输质量评分为75,则目标设备基于该第二时刻的传输参考信息,调整第二时刻的传输参数,例如,该传输参数指码率,那么,将传输码率调整为码率2,并在第二时刻基于码率2进行流媒体数据传输。
本申请实施例中,获取目标设备在第一时刻进行流媒体数据传输时配置的传输参数,并获取目标设备所使用的传输链路在第一时刻的网络状态参数,根据第一时刻的传输参数及第一时刻的网络状态参数,对传输链路在第二时刻的网络状态参数进行预测,基于第一时刻的网络状态参数、第一时刻的传输参数以及预测得到的第二时刻的网络状态参数,确定第二时刻的传输参考信息,向目标设备发送第二时刻的传输参考信息,传输参考信息用于指示目标设备调整第二时刻的传输参数,并基于调整后的第二时刻的传输参数进行流媒体数据传输。可见,
1)提供了一种传输参数的自适应调整方法,由于该传输参数是为流媒体数据传输所配置的,是一种用于评价流媒体的指标,因此,该方法不仅包含了网络参数,还包含流媒体评价指标,通过将“网络+流媒体”结合,能够对流媒体(例如视频)的调整更加优化,例如对码率进行自适应调整,可以在一定的网络资源下实现更好的视频传输效果;
2)当应用于智慧交通这种应用场景时,由于目标设备的移动性强,整个流媒体传输网络的变化大,采用该方法,可匹配不同类型的预测模板,使得目标设备可以基于第二时刻的传输参考信息调整第二时刻的传输参数,来适应不断变化的网络状态,支持多种网络变化情况下的视频响应。
请参阅图3,图3为本申请实施例提供的另一种数据处理方法的流程图。该数据
处理方法可以由计算机设备执行,该计算机设备具体可以是图1所示的服务器102。如图3所示,该数据处理方法可包括但不限于S301-S306:
S301、获取目标设备在第一时刻进行流媒体数据传输时配置的传输参数,并获取目标设备所使用的传输链路在第一时刻的网络状态参数。
在一种实施方式中,目标设备传输链路中传输的数据包括流媒体数据,流媒体数据在传输过程中被划分为至少一个流媒体片段,目标设备在第一时刻的传输参数包括以下至少一项:各个流媒体片段的时长、各个流媒体片段的码率(包括已传输的流媒体片段的码率,以及待传输的流媒体片段的码率)、缓冲区的大小、待传输流媒体片段的数量。
在一个实施例中,具体的获取方式是通过流媒体传输系统之外的通信设备完成的。目标设备将在第一时刻的传输参数以及在第一时刻的网络状态参数发送给通信设备,再由通信设备经由核心网转发给计算机设备的。
其中,通信设备可以包括但不限于以下一个或多个:4G(fourth-generation)基站、5G(fifth-generation)基站、路侧单元(Roud Side Unit,RSU)、无线保真(Wireless Fidelity,WiFi);核心网可以是核心网或者5G云化核心网中的任一种。
需要说明的是,若获取的网络状态参数为已经进行过结构化处理的网络状态参数,则可以直接执行S303;例如,目标设备对网络状态参数进行过结构化处理,并向服务器提供结构化处理后的网络状态参数。若获取的网络状态参数为进行过结构化处理,则继续执行S302。
S302、对网络状态参数进行结构化处理,得到处理后的网络状态参数。
结构化处理用于对网络状态参数进行关键信息提取,结构化处理的形式有多种,例如,包括如下几种实施方式:
在一种实施方式中,计算机设备可以根据网络状态参数,生成决策树,并将决策树作为处理后的网络状态参数。
在另一种实施方式中,计算机设备可以对网络状态参数进行特征提取;例如,基于语义对网络状态参数进行特征提取,将提取后的语义特征作为处理后的网络状态参数。
在再一种实施方式中,计算机设备可以基于数据筛选规则,对网络状态参数进行筛选,得到处理后的网络状态参数;例如,根据配置的筛选规则,筛选出不符合要求的网络状态参数,得到处理后的网络状态参数。
S303、根据第一时刻的传输参数及处理后的网络状态参数,对传输链路在第二时刻的网络状态参数进行预测。
第一时刻早于第二时刻,例如,第一时刻可以是当前时刻,第二时刻可以是第一时刻的下一时刻。
在一种实施方式中,计算机设备可以调用网络状态预测模型,对第一时刻的传输参数及第一时刻的网络状态参数进行预测处理,得到传输链路在第二时刻的网络状态参数。
图4a为本申请实施例提供的一种网络状态预测模型的训练流程图,如图4a所示,网络状态预测模型的训练过程如下:
S401,计算机设备获取历史数据,将时间序列问题转化为监督学习问题,历史数据包括在不同时刻下的网络状态参数,以及各个网络状态参数对应的传输参数。
S402,计算机设备基于历史数据,生成训练数据集以及训练数据集对应的校验数据。
例如,历史数据,按照时间先后顺序,包括目标设备在时刻1-时刻10的网络状态参数,以及目标设备在时刻1-时刻10的传输参数,则计算机设备可以将目标设备在时刻1-时刻9的网络状态参数,以及目标设备在时刻1-时刻9的传输参数作为训练数据集,将目标设备在时刻2-时刻10的网络状态参数,以及目标设备在时刻2-时刻10的传输参数作为训练数据集对应的校验数据。
S403,在生成训练数据集后,通过训练数据集和校验数据,对初始模型进行训练。
具体地,计算机设备将训练数据集输入初始模型,得到初始模型输出的预测数据,并计算预测数据与训练数据集对应的校验数据之间的损失值,基于损失值对初始模型中的参数进行调整(如调整神经网络的层数、每层神经网络中神经元的数量等),得到网络状态预测模型。
其中,初始模型可以包括但不限于长短期记忆网络模型(Long Short-Term Memory,LSTM)、循环神经网络模型(Recurrent Neural Network,RNN)、门控循环单元模型(Gated Recurrent Unit,GRU)等时序模型。
在S404,输出网络状态预测模型。
在另一种实施方式中,计算机设备可以同时获取多个目标设备发送的第一时刻的传输参数和第一时刻的网络状态参数,并基于各个目标设备在第一时刻的传输参数和网络状态参数,对传输链路在第二时刻的网络状态参数进行预测。例如,若指示第一时刻网络拥塞的第一时刻的网络状态参数的数量超过数量阈值,则预测第一时刻的下一时刻(第二时刻)网络拥塞。
需要说明的是,计算机设备还可以通过网络状态预测模型,来对同一时刻下的多个网络状态参数及多个传输参数,进行进一步挖掘,进而预测第二时刻的网络状态参数。可以理解的是,通过上述方式,提供了一种多终端/多用户的流媒体传输参数(例如码率)自适应算法,将多个目标设备之间的耦合性考虑在内,即多用户之间的耦合效应,可以提高预测的第二时刻的网络状态参数准确性,与实际应用场景结合更加紧密。
S304、向目标设备发送预测得到的第二时刻的网络状态参数。
计算机设备向目标设备发送预测得到的第二时刻的网络状态参数,以使目标设备基于第二时刻的网络状态参数调整自身运行策略。
在一种实施方式中,在智慧交通流媒体传输系统中,目标设备可以是搭载在目标车辆中的车载终端,计算机设备向目标设备发送预测得到的第二时刻的网络状态参数,以使目标设备基于预测得到的第二时刻的网络状态参数,调整目标车辆的驾驶策略(如减速、结束远程控制并人为接管,或者开始远程控制等)。
在另一种实施方式中,在移动流媒体传输系统中,目标设备可以是可移动通信设备(如手机),计算机设备向目标设备发送预测得到的第二时刻的网络状态参数,以使目标设备基于预测得到的第二时刻的网络状态参数,调整通信方式(如由视频通话切换为语音通话,切换视频通话分辨率等)。
S305、基于第一时刻的网络状态参数、第一时刻的传输参数以及预测得到的第二时刻的网络状态参数,确定第二时刻的传输参考信息。
第二时刻的传输参考信息用于为目标设备确定第二时刻的传输参数提供参考。第
二时刻的传输参考信息可以直接用于指示候选码率中在第二时刻数据传输质量最优的码率,第二时刻的传输参考信息也可以包括不同码率在第二时刻的数据传输质量信息。
在一种实施方式中,计算机设备可以调用传输质量模型,对目标设备在第一时刻的网络状态、第一时刻的传输参数以及预测得到的第二时刻的网络状态参数进行传输质量检测,得到各个码率对应的数据传输质量信息,并将各个码率对应的数据传输质量信息,确定为第二时刻的传输参考信息。
其中,传输质量模型可以是强化学习模型(Deep Q Networks,DQN),传输质量模型包括环境模块、决策模块和传输质量评测模块,计算机设备调用传输质量模型进行传输质量检测的过程,包括:
①获取配置信息,配置信息包括N个度量指标,以及每个度量指标的权重,N为正整数;配置信息可以是对象基于实际需求设置的,也可以是由开发人员默认设置的。N个度量指标可以包括但不限于:码率平滑、缓冲。
将N个度量指标和每个度量指标的权重作为质量评测模块的评测规则。可以理解的是,若采用默认配置信息,则无需再对传输质量模型进行额外配置(即采用默认评测规则)。
②基于第一时刻的网络状态参数、第一时刻的传输参数以及预测得到的第二时刻的网络状态参数,配置环境模块,环境模块可用于指示预测的第二时刻的网络状态。
③通过决策模块,选择目标码率作为第二时刻的传输码率;
④通过传输质量评测模块,根据N个度量指标、每个度量指标的权重、以及环境模块所指示的第二时刻的网络状态,确定目标码率对应的数据传输质量信息。
例如,在预测的第二时刻的网络状态下,目标码率在指标A的评测得分为5,在指标B的评测得分为3,在指标C的评测得分为10,且指标A的权重为1,指标B的权重为3,指标C的权重为2,则目标码率对应的数据传输质量评分为:5+3*3+10*2=34。其中,目标码率可以是候选码率中未被选取码率中的任一个码率。
可以理解的是,通过重复执行上述③和④,可以得到各个候选码率对应的数据传输质量信息,进而得到第二时刻的传输参考信息。
由于目标设备在第一时刻的网络状态参数,以及第一时刻的传输参数,用于为确定第二时刻的传输参考信息提供参考,第二时刻的传输参考信息是基于不同码率在第二时刻的网络状态参数下的数据传输质量评分,以及将目标设备在第一时刻的网络状态参数,以及第一时刻的传输参数作为参考依据生成的。
图4b为本申请实施例提供的一种模型架构图。如图4b所示,首先将数据输入网络状态预测模型,数据包括目标设备在第一时刻的传输参数,以及在第一时刻的网络状态参数,得到网络状态预测模型输出的第二时刻的网络状态参数。
将网络状态预测模型预测的第二时刻的网络状态参数,目标设备在第一时刻的传输参数,以及在第一时刻的网络状态参数作为环境信息,输入传输质量模型。
传输质量模型基于网络状态预测模型预测的第二时刻的网络状态参数,目标设备在第一时刻的传输参数,以及在第一时刻的网络状态参数,对网络状态进行更新,并通过神经网络确定不同码率(每次由码率自适应决策确定)在当前网络状态下的数据传输质量,进而得到第二时刻的传输参考信息。
传输质量模型可以以pensieve为基本框架,状态变量除了目标设备在第一时刻的传输参数以外,还根据网络状态参数的更新进行调整(从第一时刻的网络状态参数更
新为第二时刻的网络状态参数)。
可以理解的是,图4b中所示的神经网络的层数,以及神经元的数量仅用于举例,并不构成本申请的实际限定,在实际应用中,可以基于训练结果对神经网络的层数,以及神经元的数量进行调整,以使传输质量模型达到更好的效果。
S306、向目标设备发送第二时刻的传输参考信息。
S306的具体实施方式可参考图2中S204的实施方式,在此不再赘述。
本申请实施例在图2实施例的基础上,通过对网络状态参数进行结构化处理,可以更好的提取网络状态特征,进而提供预测的第二时刻的网络状态参数的准确率。通过向目标设备发送预测得到的第二时刻的网络状态参数,使得目标设备可以基于第二时刻的网络状态参数调整自身运行策略,来适应不断变化的网络状态。此外,通过将多个目标设备之间的耦合性考虑在内,与实际场景结合更紧密,进而提升了数据传输效果。
请参阅图5,图5为本申请实施例提供的再一种数据处理方法的流程图。该数据处理方法可以由计算机设备执行,该计算机设备具体可以是图1所示的目标设备101。如图5所示,该数据处理方法可包括但不限于S501-S503:
S501、获取在第一时刻进行流媒体数据传输时配置的传输参数,并获取所使用的传输链路在第一时刻的网络状态参数。
针对流媒体传输配置的传输参数包括以下至少一项:各个流媒体片段的时长、各个流媒体片段的码率、缓冲区的大小、待传输流媒体片段的数量。码率是指数据在单位时间内的传输量。
传输链路在第一时刻的网络状态参数用于指示在流媒体传输系统中传输链路的网络状态,网络状态参数包括以下至少一项:信干比(Signal to Interference plus Noise Ratio,SINR)、接收信号强度指示(Received Signal Strength Indicator,RSSI)、参考信号接收质量(Reference Signal Receiving Quality,RSRP)、参考信号接收功率(Reference Signal Receiving Power,RSRQ)、时延、吞吐量、物理层共享信道传输块大小、调制与编码策略、数据传输速率。
目标设备获取第一时刻的传输参数的方式可以包括:采集第一时刻的传输参数,或者,获取对象配置的第一时刻的传输参数。
目标设备获取所使用的传输链路在第一时刻的网络状态参数的方式可以包括:目标设备采集所使用的传输链路在第一时刻的网络状态参数,或者,目标设备从通信运营商或者网络运营商获取所使用的传输链路在第一时刻的网络状态参数。
S502、向服务器发送第一时刻的传输参数及第一时刻的网络状态参数。
在一种实施方式中,目标设备采集在流媒体传输中第一时刻的传输参数以及在第一时刻的网络状态参数,并通过通信网络的上行链路,将第一时刻的传输参数,以及在第一时刻的网络状态参数上传给通信设备,通信设备可以包括但不限于以下一个或多个:4G(fourth-generation)基站、5G(fifth-generation)基站、路侧单元(Roud Side Unit,RSU)、无线保真(Wireless Fidelity,WiFi)设备。通信设备再通过核心网将第一时刻的传输参数,以及在第一时刻的网络状态参数发送给服务器,核心网可以是核心网或者5G云化核心网中的任一种。
进而,服务器根据第一时刻的传输参数及第一时刻的网络状态参数,对传输链路
在第二时刻的网络状态参数进行预测,并基于第一时刻的网络状态参数、第一时刻的传输参数以及预测得到的第二时刻的网络状态参数,确定并返回第二时刻的传输参考信息。具体实施方式可参考图2或者图3中的实施方式,在此不再赘述。
对于处理网络状态参数的具体方式,包括如下2种:
在一种实施方式中,目标设备向服务器发送第一时刻的网络状态参数,使服务器对第一时刻的网络状态参数进行结构化处理。服务器对第一时刻的网络状态参数进行结构化处理的实施方式可参考S302,在此不再赘述。
在另一种实施方式中,目标设备对第一时刻的网络状态参数进行结构化处理,并将结构化处理后的网络状态参数发送至服务器。具体地,目标设备可以根据网络状态参数,生成决策树,并将决策树作为处理后的网络状态参数。目标设备也可以对网络状态参数进行特征提取;例如,基于语义对网络状态参数进行特征提取,得到处理后的网络状态参数。目标设备还可以基于数据筛选规则对网络状态参数进行筛选,得到处理后的网络状态参数;例如,根据配置的筛选规则筛选出不符合要求的网络状态参数,得到处理后的网络状态参数。
S503、基于服务器返回的第二时刻的传输参考信息,调整第二时刻的传输参数,并基于调整后的第二时刻的传输参数进行流媒体数据传输。
第二时刻的传输参数包括以下至少一项:第二时刻传输的流媒体片段的时长、第二时刻传输的流媒体片段的码率、缓冲区的大小。第二时刻的传输参考信息用于为目标设备确定第二时刻的传输参数提供参考。第二时刻的传输参考信息可以包括不同码率在第二时刻的数据传输质量信息。
在一种实施方式中,传输链路中传输的数据包括流媒体数据,流媒体数据在传输过程中被划分为至少一个流媒体片段。第二时刻的传输参考信息包括:不同码率的流媒体片段在第二时刻的传输质量。例如,第二时刻的传输参考信息可以包括码率的流媒体片段在第二时刻的数据传输质量评分。第二时刻的传输参数包括:第二时刻传输的流媒体片段的码率。目标设备基于第二时刻的传输参考信息,将第二时刻传输的流媒体片段的码率调整为目标码率;其中,目标码率为第二时刻的传输参考信息指示的各个码率中,在第二时刻的传输质量高于传输质量阈值的码率;例如,目标码率可以是候选码率中在第二时刻的数据传输质量评分最高的码率。
图6a为本申请实施例提供的一种数据处理方法的应用场景图。如图6a所示,该应用场景为智慧交通流媒体场景,该场景除了包括搭载了车载终端(即目标设备)的车辆601和应用服务器602外,还包括基站603、核心网604、远控服务器605、交通业务606。其中,远控服务器605可以用于为车辆601提供远程控制服务,交通业务606可以包括但不限于:无人集卡、车路协同、实时孪生等。
图6b为本申请实施例提供的一种智慧交通流媒体场景下的数据处理流程图。如图6b所示,S611中,目标设备(或者通信运营商,如基站)采集目标设备在第一时刻的传输参数,以及传输链路在第一时刻的网络状态参数。
在得到第一时刻的网络状态参数后,S612中,目标设备对网络状态参数进行结构化处理,并将结构化处理后的网络状态参数发送给应用服务器。对网络状态参数进行结构化处理的具体实施方式可参考S302中的实施方式,在此不再赘述。
S613中,通过5G上行链路,目标设备将结构化处理后的网络状态参数和在第一时刻的传输参数上传给通信设备。其中,通信设备可以包括但不限于:4G/5G基站、
RSU、WiFi中的一种或多种。
S614中,通信设备,利用相关接口,通过核心网,将结构化处理后的网络状态参数和目标设备在第一时刻的传输参数转发给应用服务器。核心网可以是5G核心网或者5G云化核心网中的一种。
S615中,应用服务器将结构化处理后的网络状态参数和目标设备在第一时刻的传输参数,输入到网络状态预测模型和传输质量模型。
在一种实施方式中,应用服务器接收到结构化处理后的网络状态参数和目标设备在第一时刻的传输参数后,可以对接收的数据进行预处理(如生成矩阵),使得数据符合网络状态预测模型,或者传输质量模型的输入要求。
应用服务器可以根据实际需求,对网络状态预测模型和传输质量模型进行相关参数配置,如配置输入特征、输出特征、预测特征、模型的参数、预测时间、模型训练和测试的比例、迭代次数、模型损失函数、模型优化函数、机器学习神经元的数量等等。
S616中,将结构化处理后的网络状态参数和目标设备在第一时刻的传输参数输入网络状态预测模型,输出第二时刻的网络状态参数;将结构化处理后的网络状态参数和目标设备在第一时刻的传输参数输入传输质量模型,输出第二时刻的传输参考信息。
S617中,应用服务器可以向目标设备(车载终端)返回第二时刻的网络状态参数,以使目标设备基于第二时刻的网络状态参数,调整车辆驾驶策略(如减速、结束远程控制并人为接管,或者开始远程控制等)。
S618中,应用服务器可以向目标设备发送第二时刻的传输参考信息,以使目标设备基于第二时刻的传输参考信息,调整第二时刻的传输参数(如提高/降低摄像头码率,以调整传输码率),并基于调整后的第二时刻的传输参数进行流媒体数据传输。
上述详细阐述了本申请实施例的方法,为了便于更好地实施本申请实施例的上述方案,相应地,下面提供了本申请实施例的装置。
请参见图7,图7为本申请实施例提供的一种数据处理装置的结构示意图,该装置可以搭载在计算机设备上,该计算机设备具体可以是图1所示的服务器102。图7所示的数据处理装置可以用于执行上述图2和图3所描述的方法实施例中的部分或全部功能。请参见图7,各个单元的详细描述如下:
第一获取单元701,用于获取目标设备在第一时刻进行流媒体数据传输时配置的传输参数,并获取目标设备所使用的传输链路在第一时刻的网络状态参数;
第一处理单元702,用于根据第一时刻的传输参数及第一时刻的网络状态参数,对传输链路在第二时刻的网络状态参数进行预测;以及用于基于第一时刻的网络状态参数、第一时刻的传输参数以及预测得到的第二时刻的网络状态参数,确定第二时刻的传输参考信息;
第一发送单元703,还用于向目标设备发送第二时刻的传输参考信息,传输参考信息用于指示目标设备调整第二时刻的传输参数,并基于调整后的第二时刻的传输参数进行流媒体数据传输,第一时刻早于第二时刻。
在一种实施方式中,第一处理单元702用于,根据第一时刻的传输参数及第一时刻的网络状态参数,对传输链路在第二时刻的网络状态参数进行预测,具体用于:
调用网络状态预测模型,对第一时刻的传输参数及第一时刻的网络状态参数进行
预测处理,得到传输链路在第二时刻的网络状态参数。
在一种实施方式中,第一处理单元702还用于对网络状态预测模型进行训练,具体包括:
获取历史数据,历史数据包括在不同时刻下的网络状态参数,以及各个网络状态参数对应的传输参数;
基于历史数据生成训练数据集,并将训练数据集输入初始模型,得到初始模型输出的预测数据;
计算预测数据与训练数据集对应的校验数据之间的损失值;
基于损失值对初始模型中的参数进行调整,得到所述网络状态预测模型。
在一种实施方式中,第一处理单元702用于,调用传输质量模型,对第一时刻的网络状态参数、第一时刻的传输参数以及预测得到的第二时刻的网络状态参数进行传输质量检测,得到各个码率对应的数据传输质量信息;
将各个码率对应的数据传输质量信息,确定为所述第二时刻的传输参考信息。
在一种实施方式中,传输质量模型包括环境模块、决策模块和传输质量评测模块;第一处理单元702用于,获取配置信息,所述配置信息包括N个度量指标,以及每个度量指标的权重,N为正整数;基于所述第一时刻的网络状态参数、所述第一时刻的传输参数以及预测得到的所述第二时刻的网络状态参数,配置所述环境模块;通过所述决策模块,选择目标码率作为所述第二时刻的传输码率;通过所述传输质量评测模块,根据所述N个度量指标、每个度量指标的权重、以及所述环境模块所指示的第二时刻的网络状态,确定所述目标码率对应的数据传输质量信息。
在一种实施方式中,第一处理单元702用于,对所述网络状态参数进行结构化处理,得到处理后的网络状态参数;根据所述第一时刻的传输参数及所述处理后的网络状态参数,对所述传输链路在第二时刻的网络状态参数进行预测。
在一种实施方式中,网络状态参数包括以下至少一项:信干比、接收信号强度指示、参考信号接收质量、参考信号接收功率、时延、吞吐量、物理层共享信道传输块大小、调制与编码策略、数据传输速率。
在一种实施方式中,目标设备使用所述传输链路中传输的数据包括流媒体数据,流媒体数据在传输过程中被划分为至少一个流媒体片段,目标设备在第一时刻的传输参数包括以下至少一项:各个流媒体片段的时长、各个流媒体片段的码率、缓冲区的大小、待传输流媒体片段的数量。
根据本申请的一个实施例,图2和图3所示的数据处理方法所涉及的部分步骤可由图7所示的数据处理装置中的各个单元来执行。例如,图2中所示的S201可由图7所示的第一获取单元701执行,S202和S203可由图7所示的第一处理单元702执行,S204可由图7所示的第一发送单元703执行。图3中所示的S301可由图7所示的收发单元701执行,S302、S303和S305可由图7所示的第一处理单元702执行,S304和S306可由图7所示的第一发送单元703执行。图7所示的数据处理装置中的各个单元可以分别或全部合并为一个或若干个另外的单元来构成,或者其中的某个(些)单元还可以再拆分为功能上更小的多个单元来构成,这可以实现同样的操作,而不影响本申请的实施例的技术效果的实现。上述单元是基于逻辑功能划分的,在实际应用中,一个单元的功能也可以由多个单元来实现,或者多个单元的功能由一个单元实现。在本申请的其它实施例中,数据处理装置也可以包括其它单元,在实际应用中,这些功能也
可以由其它单元协助实现,并且可以由多个单元协作实现。
根据本申请的另一个实施例,可以通过在包括中央处理单元(CPU)、随机存取存储介质(RAM)、只读存储介质(ROM)等处理元件和存储元件的例如计算机的通用计算装置上运行能够执行如图2和图3中所示的相应方法所涉及的各步骤的计算机程序(包括程序代码),来构造如图7中所示的数据处理装置,以及来实现本申请实施例的数据处理方法。计算机程序可以记载于例如计算机可读记录介质上,并通过计算机可读记录介质装载于上述计算装置中,并在其中运行。
基于同一发明构思,本申请实施例中提供的数据处理装置解决问题的原理与有益效果与本申请方法实施例中数据处理方法解决问题的原理和有益效果相似,可以参见方法的实施的原理和有益效果,为简洁描述,在这里不再赘述。
请参见图8,图8为本申请实施例提供的另一种数据处理装置的结构示意图,该装置可以搭载在计算机设备上,该计算机设备具体可以是图1所示的目标设备101。图8所示的数据处理装置可以用于执行上述图5所描述的方法实施例中的部分或全部功能。请参见图8,各个单元的详细描述如下:
第二获取单元801,用于获取在第一时刻进行流媒体数据传输时配置的传输参数,并获取所使用的传输链路在所述第一时刻的网络状态参数;
第二发送单元802,用于向服务器发送第一时刻的传输参数及第一时刻的网络状态参数,以使服务器根据第一时刻的传输参数及第一时刻的网络状态参数,对传输链路在第二时刻的网络状态参数进行预测,并基于第一时刻的网络状态参数、第一时刻的传输参数以及预测得到的第二时刻的网络状态参数,确定并返回第二时刻的传输参考信息;
第二处理单元803,用于基于服务器返回的第二时刻的传输参考信息,调整第二时刻的传输参数,并基于调整后的第二时刻的传输参数进行流媒体数据传输。
在一种实施方式中,第二发送单元802用于,向所述服务器发送第一时刻的网络状态参数,使服务器对第一时刻的网络状态参数进行结构化处理;或者,
对第一时刻的网络状态参数进行结构化处理,并将结构化处理后的网络状态参数发送至服务器。
在一种实施方式中,在所述传输链路中传输的数据包括流媒体数据,流媒体数据在传输过程中被划分为至少一个流媒体片段;第二时刻的传输参考信息包括:不同码率的流媒体片段在第二时刻的传输质量;第二时刻的传输参数包括:第二时刻传输的流媒体片段的码率;
第二处理单元803用于,基于第二时刻的传输参考信息,将第二时刻传输的流媒体片段的码率调整为目标码率;
其中,目标码率为第二时刻的传输参考信息指示的各个码率中,在第二时刻的传输质量高于传输质量阈值的码率。
根据本申请的一个实施例,图5所示的数据处理方法所涉及的部分步骤可由图8所示的数据处理装置中的各个单元来执行。例如,图5中所示的S501可由图8所示的第二获取单元801执行,S502可由图8所示的第二发送单元802执行,S503可由图8所示的第二处理单元803执行。图8所示的数据处理装置中的各个单元可以分别或全部合并为一个或若干个另外的单元来构成,或者其中的某个(些)单元还可以再拆分
为功能上更小的多个单元来构成,这可以实现同样的操作,而不影响本申请的实施例的技术效果的实现。上述单元是基于逻辑功能划分的,在实际应用中,一个单元的功能也可以由多个单元来实现,或者多个单元的功能由一个单元实现。在本申请的其它实施例中,数据处理装置也可以包括其它单元,在实际应用中,这些功能也可以由其它单元协助实现,并且可以由多个单元协作实现。
根据本申请的另一个实施例,可以通过在包括中央处理单元(CPU)、随机存取存储介质(RAM)、只读存储介质(ROM)等处理元件和存储元件的例如计算机的通用计算装置上运行能够执行如图5中所示的相应方法所涉及的各步骤的计算机程序(包括程序代码),来构造如图8中所示的数据处理装置,以及来实现本申请实施例的数据处理方法。计算机程序可以记载于例如计算机可读记录介质上,并通过计算机可读记录介质装载于上述计算装置中,并在其中运行。
基于同一发明构思,本申请实施例中提供的数据处理装置解决问题的原理与有益效果与本申请方法实施例中数据处理方法解决问题的原理和有益效果相似,可以参见方法的实施的原理和有益效果,为简洁描述,在这里不再赘述。
请参阅图9,图9为本申请实施例提供的一种计算机设备的结构示意图,如图9所示,计算机设备至少包括处理器901、通信接口902和存储器903。其中,处理器901、通信接口902和存储器903可通过总线或其他方式连接。其中,处理器901(或称中央处理器(Central Processing Unit,CPU))是计算机设备的计算核心以及控制核心,其可以解析计算机设备内的各类指令以及处理计算机设备的各类数据,例如:CPU可以用于解析用户向计算机设备所发送的开关机指令,并控制计算机设备进行开关机操作;再如:CPU可以在计算机设备内部结构之间传输各类交互数据,等等。通信接口902可选的可以包括标准的有线接口、无线接口(如WI-FI、移动通信接口等),受处理器901的控制可以用于收发数据;通信接口902还可以用于计算机设备内部数据的传输以及交互。存储器903(Memory)是计算机设备中的记忆设备,用于存放程序和数据。可以理解的是,此处的存储器903既可以包括计算机设备的内置存储器,当然也可以包括计算机设备所支持的扩展存储器。存储器903提供存储空间,该存储空间存储了计算机设备的操作系统,可包括但不限于:Android系统、iOS系统、Windows Phone系统等等,本申请对此并不作限定。
本申请实施例还提供了一种计算机可读存储介质(Memory),计算机可读存储介质是计算机设备中的记忆设备,用于存放程序和数据。可以理解的是,此处的计算机可读存储介质既可以包括计算机设备中的内置存储介质,当然也可以包括计算机设备所支持的扩展存储介质。计算机可读存储介质提供存储空间,该存储空间存储了计算机设备的处理系统。并且,在该存储空间中还存放了适于被处理器901加载并执行的一条或多条的指令,这些指令可以是一个或多个的计算机程序(包括程序代码)。需要说明的是,此处的计算机可读存储介质可以是高速RAM存储器,也可以是非不稳定的存储器(non-volatile memory),例如至少一个磁盘存储器;可选的,还可以是至少一个位于远离前述处理器的计算机可读存储介质。
在一个实施例中,计算机设备具体可以是图1所示的服务器102。处理器901通过运行存储器903中的可执行程序代码,执行如下操作:
通过通信接口902,获取目标设备在第一时刻进行流媒体数据传输时配置的传输
参数,并获取目标设备所使用的传输链路在第一时刻的网络状态参数;
根据第一时刻的传输参数及第一时刻的网络状态参数,对传输链路在第二时刻的网络状态参数进行预测;
基于第一时刻的网络状态参数、第一时刻的传输参数以及预测得到的第二时刻的网络状态参数,确定第二时刻的传输参考信息;
通过通信接口902,向目标设备发送第二时刻的传输参考信息,传输参考信息用于指示目标设备调整第二时刻的传输参数,并基于调整后的第二时刻的传输参数进行流媒体数据传输,第一时刻早于第二时刻。
作为一种可选的实施例,处理器901根据第一时刻的传输参数及第一时刻的网络状态参数,对传输链路在第二时刻的网络状态参数进行预测,具体执行:
调用网络状态预测模型,对第一时刻的传输参数及第一时刻的网络状态参数进行预测处理,得到传输链路在第二时刻的网络状态参数。
作为一种可选的实施例,处理器901还执行如下操作:
对所述网络状态预测模型进行训练,具体包括:
获取历史数据,所述历史数据包括在不同时刻下的网络状态参数,以及各个网络状态参数对应的传输参数;
基于所述历史数据生成训练数据集,并将所述训练数据集输入初始模型,得到所述初始模型输出的预测数据;
计算所述预测数据与所述训练数据集对应的校验数据之间的损失值;
基于所述损失值对所述初始模型中的参数进行调整,得到所述网络状态预测模型。
作为一种可选的实施例,处理器901基于第一时刻的网络状态参数、第一时刻的传输参数以及预测得到的第二时刻的网络状态参数,确定第二时刻的传输参考信息,具体执行:
调用传输质量模型,对第一时刻的网络状态参数、第一时刻的传输参数以及预测得到的第二时刻的网络状态参数进行传输质量检测,得到各个码率对应的数据传输质量信息;
将各个码率对应的数据传输质量信息,确定为所述第二时刻的传输参考信息。
作为一种可选的实施例,传输质量模型包括环境模块、决策模块和传输质量评测模块;处理器901调用传输质量模型,对所述第一时刻的网络状态参数、所述第一时刻的传输参数以及预测得到的所述第二时刻的网络状态参数进行传输质量检测,得到各个码率对应的数据传输质量信息,具体执行:
获取配置信息,所述配置信息包括N个度量指标,以及每个度量指标的权重,N为正整数;
基于所述第一时刻的网络状态参数、所述第一时刻的传输参数以及预测得到的所述第二时刻的网络状态参数,配置所述环境模块;
通过所述决策模块,选择目标码率作为所述第二时刻的传输码率;
通过所述传输质量评测模块,根据所述N个度量指标、每个度量指标的权重、以及所述环境模块所指示的第二时刻的网络状态,确定所述目标码率对应的数据传输质量信息。
作为一种可选的实施例,处理器901根据所述第一时刻的传输参数及所述第一时刻的网络状态参数,对所述传输链路在第二时刻的网络状态参数进行预测,具体执行:
对所述网络状态参数进行结构化处理,得到处理后的网络状态参数;
根据所述第一时刻的传输参数及所述处理后的网络状态参数,对所述传输链路在第二时刻的网络状态参数进行预测。
作为一种可选的实施例,网络状态参数包括以下至少一项:信干比、接收信号强度指示、参考信号接收质量、参考信号接收功率、时延、吞吐量、物理层共享信道传输块大小、调制与编码策略、数据传输速率。
作为一种可选的实施例,目标设备使用传输链路中传输的数据包括流媒体数据,流媒体数据在传输过程中被划分为至少一个流媒体片段,目标设备在第一时刻的传输参数包括以下至少一项:各个流媒体片段的时长、各个流媒体片段的码率、缓冲区的大小、待传输流媒体片段的数量。
在另一个实施例中,计算机设备具体可以是图1所示的目标设备101。处理器1001通过运行存储器1003中的可执行程序代码,执行如下操作:
获取在第一时刻进行流媒体数据传输时配置的传输参数,并获取所使用的传输链路在所述第一时刻的网络状态参数;
通过通信接口1002向服务器发送第一时刻的传输参数及第一时刻的网络状态参数,以使服务器根据第一时刻的传输参数及第一时刻的网络状态参数,对传输链路在第二时刻的网络状态参数进行预测,并基于第一时刻的网络状态参数、第一时刻的传输参数以及预测得到的第二时刻的网络状态参数,确定并返回第二时刻的传输参考信息;
基于服务器返回的第二时刻的传输参考信息,调整第二时刻的传输参数,并基于调整后的第二时刻的传输参数进行流媒体数据传输。
作为一种可选的实施例,处理器1001通过通信接口1002向服务器发送第一时刻的网络状态参数,具体执行:
通过通信接口1002向服务器发送第一时刻的网络状态参数,使服务器对第一时刻的网络状态参数进行结构化处理;或者,
对第一时刻的网络状态参数进行结构化处理,并将结构化处理后的网络状态参数发送至服务器。
作为一种可选的实施例,传输链路中传输的数据包括流媒体数据,流媒体数据在传输过程中被划分为至少一个流媒体片段;第二时刻的传输参考信息包括:不同码率的流媒体片段在第二时刻的传输质量;第二时刻的传输参数包括:第二时刻传输的流媒体片段的码率;
处理器1001基于服务器返回的第二时刻的传输参考信息,调整第二时刻的传输参数,具体执行:
基于第二时刻的传输参考信息,将第二时刻传输的流媒体片段的码率调整为目标码率;
其中,目标码率为第二时刻的传输参考信息指示的各个码率中,在第二时刻的传输质量高于传输质量阈值的码率。
基于同一发明构思,本申请实施例中提供的计算机设备解决问题的原理与有益效果与本申请方法实施例中数据处理方法解决问题的原理和有益效果相似,可以参见方法的实施的原理和有益效果,为简洁描述,在这里不再赘述。
本申请实施例还提供一种计算机可读存储介质,计算机可读存储介质中存储有计
算机程序,计算机程序适于被处理器加载并执行上述方法实施例的数据处理方法。
本申请实施例还提供一种计算机程序产品,该计算机程序产品包括计算机程序,计算机程序适于被处理器加载并执行上述方法实施例的数据处理方法。
本申请实施例还提供一种计算机程序产品或计算机程序,该计算机程序产品或计算机程序包括计算机指令,该计算机指令存储在计算机可读存储介质中。计算机设备的处理器从计算机可读存储介质读取该计算机指令,处理器执行该计算机指令,使得该计算机设备执行上述的数据处理方法。
本申请实施例方法中的步骤可以根据实际需要进行顺序调整、合并和删减。
本申请实施例装置中的模块可以根据实际需要进行合并、划分和删减。
本领域普通技术人员可以理解上述实施例的各种方法中的全部或部分步骤是可以通过程序来指令相关的硬件来完成,该程序可以存储于一计算机可读存储介质中,可读存储介质可以包括:闪存盘、只读存储器(Read-Only Memory,ROM)、随机存取器(Random Access Memory,RAM)、磁盘或光盘等。
以上所揭露的仅为本申请一种较佳实施例而已,当然不能以此来限定本申请之权利范围,本领域普通技术人员可以理解实现上述实施例的全部或部分流程,并依本申请权利要求所作的等同变化,仍属于申请所涵盖的范围。
Claims (20)
- 一种数据处理方法,由计算机设备执行,包括:获取目标设备在第一时刻进行流媒体数据传输时配置的传输参数,并获取所述目标设备所使用的传输链路在所述第一时刻的网络状态参数;根据所述第一时刻的传输参数及所述第一时刻的网络状态参数,对所述传输链路在第二时刻的网络状态参数进行预测;基于所述第一时刻的网络状态参数、所述第一时刻的传输参数以及预测得到的所述第二时刻的网络状态参数,确定所述第二时刻的传输参考信息;及,向所述目标设备发送所述第二时刻的传输参考信息,所述传输参考信息用于指示所述目标设备调整所述第二时刻的传输参数,并基于调整后的所述第二时刻的传输参数进行流媒体数据传输,所述第一时刻早于所述第二时刻。
- 如权利要求1所述的方法,其中,所述根据所述第一时刻的传输参数及所述第一时刻的网络状态参数,对所述传输链路在第二时刻的网络状态参数进行预测,包括:调用网络状态预测模型,对所述第一时刻的传输参数及所述第一时刻的网络状态参数进行预测处理,得到所述传输链路在第二时刻的网络状态参数。
- 如权利要求2所述的方法,还包括:对所述网络状态预测模型进行训练,具体包括:获取历史数据,所述历史数据包括在不同时刻下的网络状态参数,以及各个网络状态参数对应的传输参数;基于所述历史数据生成训练数据集,并将所述训练数据集输入初始模型,得到所述初始模型输出的预测数据;计算所述预测数据与所述训练数据集对应的校验数据之间的损失值;基于所述损失值对所述初始模型中的参数进行调整,得到所述网络状态预测模型。
- 如权利要求1所述的方法,其中,所述基于所述第一时刻的网络状态参数、所述第一时刻的传输参数以及预测得到的所述第二时刻的网络状态参数,确定所述第二时刻的传输参考信息,包括:调用传输质量模型,对所述第一时刻的网络状态参数、所述第一时刻的传输参数以及预测得到的所述第二时刻的网络状态参数进行传输质量检测,得到各个码率对应的数据传输质量信息;将各个码率对应的数据传输质量信息,确定为所述第二时刻的传输参考信息。
- 如权利要求4所述的方法,其中,所述传输质量模型包括环境模块、决策模块和传输质量评测模块,所述调用传输质量模型,对所述第一时刻的网络状态参数、所述第一时刻的传输参数以及预测得到的所述第二时刻的网络状态参数进行传输质量检测,得到各个码率对应的数据传输质量信息,包括:获取配置信息,所述配置信息包括N个度量指标,以及每个度量指标的权重,N为正整数;基于所述第一时刻的网络状态参数、所述第一时刻的传输参数以及预测得到的所述第二时刻的网络状态参数,配置所述环境模块;通过所述决策模块,选择目标码率作为所述第二时刻的传输码率;通过所述传输质量评测模块,根据所述N个度量指标、每个度量指标的权重、以及所述环境模块所指示的第二时刻的网络状态,确定所述目标码率对应的数据传输质量信息。
- 如权利要求1所述的方法,其中,所述根据所述第一时刻的传输参数及所述第一时刻的网络状态参数,对所述传输链路在第二时刻的网络状态参数进行预测,包括:对所述网络状态参数进行结构化处理,得到处理后的网络状态参数;根据所述第一时刻的传输参数及所述处理后的网络状态参数,对所述传输链路在第二时刻的网络状态参数进行预测。
- 如权利要求1所述的方法,其中,所述网络状态参数包括以下至少一项:信干比、接收信号强度指示、参考信号接收质量、参考信号接收功率、时延、吞吐量、物理层共享信道传输块大小、调制与编码策略、数据传输速率。
- 如权利要求1-7中任一项所述的方法,其中,所述目标设备使用所述传输链路传输的数据包括流媒体数据,所述流媒体数据在传输过程中被划分为至少一个流媒体片段;所述目标设备在所述第一时刻的传输参数包括以下至少一项:各个流媒体片段的时长、各个流媒体片段的码率、缓冲区的大小、待传输流媒体片段的数量。
- 一种数据处理方法,由计算机设备执行,包括:获取在第一时刻进行流媒体数据传输时配置的传输参数,并获取所使用的传输链路在所述第一时刻的网络状态参数;向服务器发送所述第一时刻的传输参数及所述第一时刻的网络状态参数,以使所述服务器根据所述第一时刻的传输参数及所述第一时刻的网络状态参数,对所述传输链路在第二时刻的网络状态参数进行预测,并基于所述第一时刻的网络状态参数、所述第一时刻的传输参数以及预测得到的第二时刻的网络状态参数,确定并返回第二时刻的传输参考信息;及,基于所述服务器返回的所述第二时刻的传输参考信息,调整第二时刻的传输参数,并基于调整后的第二时刻的传输参数进行流媒体数据传输。
- 如权利要求9所述的方法,其中,所述向服务器发送所述第一时刻的网络状态参数,包括:向所述服务器发送所述第一时刻的网络状态参数,使所述服务器对所述第一时刻的网络状态参数进行结构化处理;或者,对所述第一时刻的网络状态参数进行结构化处理,并将结构化处理后的网络状态参数发送至服务器。
- 如权利要求9所述的方法,其中,在所述传输链路中传输的数据包括流媒体数据,所述流媒体数据在传输过程中被划分为至少一个流媒体片段;所述第二时刻的传输参考信 息包括:不同码率的流媒体片段在所述第二时刻的传输质量;所述第二时刻的传输参数包括:所述第二时刻传输的流媒体片段的码率;所述基于所述服务器返回的所述第二时刻的传输参考信息,调整第二时刻的传输参数,包括:基于所述第二时刻的传输参考信息,将所述第二时刻传输的流媒体片段的码率调整为目标码率;其中,所述目标码率为所述第二时刻的传输参考信息指示的各个码率中,在所述第二时刻的传输质量高于传输质量阈值的码率。
- 一种数据处理装置,包括:第一获取单元,用于获取目标设备在第一时刻进行流媒体数据传输时配置的传输参数,并获取所述目标设备所使用的传输链路在所述第一时刻的网络状态参数;第一处理单元,用于根据所述第一时刻的传输参数及所述第一时刻的网络状态参数,对所述传输链路在第二时刻的网络状态参数进行预测;以及用于基于所述第一时刻的网络状态参数、所述第一时刻的传输参数以及预测得到的所述第二时刻的网络状态参数,确定所述第二时刻的传输参考信息;及,第一发送单元,用于向所述目标设备发送所述第二时刻的传输参考信息,所述传输参考信息用于指示所述目标设备调整所述第二时刻的传输参数,并基于调整后的所述第二时刻的传输参数进行流媒体数据传输,所述第一时刻早于所述第二时刻。
- 如权利要求12所述的装置,其中,所述第一处理单元用于,调用网络状态预测模型,对所述第一时刻的传输参数及所述第一时刻的网络状态参数进行预测处理,得到所述传输链路在第二时刻的网络状态参数。
- 如权利要求12所述的装置,其中,所述第一处理单元用于,调用传输质量模型,对所述第一时刻的网络状态参数、所述第一时刻的传输参数以及预测得到的所述第二时刻的网络状态参数进行传输质量检测,得到各个码率对应的数据传输质量信息;将各个码率对应的数据传输质量信息,确定为所述第二时刻的传输参考信息。
- 如权利要求12-14中任一项所述的装置,其中,所述目标设备使用所述传输链路传输的数据包括流媒体数据,所述流媒体数据在传输过程中被划分为至少一个流媒体片段;所述目标设备在所述第一时刻的传输参数包括以下至少一项:各个流媒体片段的时长、各个流媒体片段的码率、缓冲区的大小、待传输流媒体片段的数量。
- 一种数据处理装置,包括:第二获取单元,用于获取在第一时刻进行流媒体数据传输时配置的传输参数,并获取所使用的传输链路在所述第一时刻的网络状态参数;第二发送单元,用于向服务器发送所述第一时刻的传输参数及所述第一时刻的网络状态参数,以使所述服务器根据所述第一时刻的传输参数及所述第一时刻的网络状态参数,对所述传输链路在第二时刻的网络状态参数进行预测,并基于所述第一时刻的网络状态参数、所述第一时刻的传输参数以及预测得到的第二时刻的网络状态参数,确定并返回第二 时刻的传输参考信息;第二处理单元,用于基于所述服务器返回的所述第二时刻的传输参考信息,调整第二时刻的传输参数,并基于调整后的第二时刻的传输参数进行流媒体数据传输。
- 如权利要求16所述的装置,其中,在所述传输链路中传输的数据包括流媒体数据,所述流媒体数据在传输过程中被划分为至少一个流媒体片段;所述第二时刻的传输参考信息包括:不同码率的流媒体片段在所述第二时刻的传输质量;所述第二时刻的传输参数包括:所述第二时刻传输的流媒体片段的码率;所述第二处理单元用于,基于所述第二时刻的传输参考信息,将所述第二时刻传输的流媒体片段的码率调整为目标码率;其中,所述目标码率为所述第二时刻的传输参考信息指示的各个码率中,在所述第二时刻的传输质量高于传输质量阈值的码率。
- 一种计算机设备,包括:存储装置和处理器;存储器,所述存储器中存储有计算机程序;处理器,用于加载所述计算机程序实现权利要求1-8中任一项所述的数据处理方法;或用于加载所述计算机程序实现权利要求9-11中任一项所述的数据处理方法。
- 一种计算机可读存储介质,存储有计算机程序,所述计算机程序适于被处理器加载并执行权利要求1-8中任一项所述的数据处理方法;或加载并执行权利要求9-11中任一项所述的数据处理方法。
- 一种计算机程序产品,包括计算机程序,所述计算机程序适于被处理器加载并执行权利要求1-8中任一项所述的数据处理方法;或加载并执行权利要求9-11中任一项所述的数据处理方法。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210716471.7 | 2022-06-21 | ||
| CN202210716471.7A CN117319228A (zh) | 2022-06-21 | 2022-06-21 | 一种数据处理方法、装置、设备、存储介质及产品 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2023246343A1 true WO2023246343A1 (zh) | 2023-12-28 |
Family
ID=89245009
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2023/092635 Ceased WO2023246343A1 (zh) | 2022-06-21 | 2023-05-08 | 数据处理方法、装置、计算机设备、存储介质及产品 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN117319228A (zh) |
| WO (1) | WO2023246343A1 (zh) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119676758A (zh) * | 2024-10-31 | 2025-03-21 | 南通先进通信技术研究院有限公司 | 一种基于5g信道质量动态调整流媒体传输参数的方法 |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118695027A (zh) * | 2024-06-11 | 2024-09-24 | 国网陕西省电力有限公司电力科学研究院 | 适用于公网信号下视频传输的卡顿优化方法 |
| CN119676311B (zh) * | 2025-02-21 | 2025-05-23 | 上海假面信息科技有限公司 | 一种数据处理方法、系统及存储介质 |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104954279A (zh) * | 2014-03-28 | 2015-09-30 | 华为技术有限公司 | 一种传输控制方法、装置及系统 |
| CN112367490A (zh) * | 2020-09-11 | 2021-02-12 | 北京邮电大学 | 面向交互视频传输质量提升的混合学习方法、装置及设备 |
| US20210105517A1 (en) * | 2018-06-20 | 2021-04-08 | Naver Corporation | Method and system for adaptive data transmission |
| CN114079652A (zh) * | 2022-01-19 | 2022-02-22 | 阿里巴巴达摩院(杭州)科技有限公司 | 数据处理方法 |
-
2022
- 2022-06-21 CN CN202210716471.7A patent/CN117319228A/zh active Pending
-
2023
- 2023-05-08 WO PCT/CN2023/092635 patent/WO2023246343A1/zh not_active Ceased
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104954279A (zh) * | 2014-03-28 | 2015-09-30 | 华为技术有限公司 | 一种传输控制方法、装置及系统 |
| US20210105517A1 (en) * | 2018-06-20 | 2021-04-08 | Naver Corporation | Method and system for adaptive data transmission |
| CN112367490A (zh) * | 2020-09-11 | 2021-02-12 | 北京邮电大学 | 面向交互视频传输质量提升的混合学习方法、装置及设备 |
| CN114079652A (zh) * | 2022-01-19 | 2022-02-22 | 阿里巴巴达摩院(杭州)科技有限公司 | 数据处理方法 |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119676758A (zh) * | 2024-10-31 | 2025-03-21 | 南通先进通信技术研究院有限公司 | 一种基于5g信道质量动态调整流媒体传输参数的方法 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117319228A (zh) | 2023-12-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2023246343A1 (zh) | 数据处理方法、装置、计算机设备、存储介质及产品 | |
| Zhang et al. | Deep learning empowered task offloading for mobile edge computing in urban informatics | |
| Wang et al. | A novel reputation-aware client selection scheme for federated learning within mobile environments | |
| EP4078899A1 (en) | Systems and methods for enhanced feedback for cascaded federated machine learning | |
| CN112104867B (zh) | 一种视频处理方法、视频处理装置、智能设备及存储介质 | |
| CN109068350B (zh) | 一种无线异构网络的终端自主选网系统及方法 | |
| Ren et al. | Fine-grained elastic partitioning for distributed DNN towards mobile web AR services in the 5G era | |
| CN113626104B (zh) | 边云架构下基于深度强化学习的多目标优化卸载策略 | |
| CN112135325A (zh) | 网络切换方法、装置、存储介质以及终端 | |
| CN110533106A (zh) | 图像分类处理方法、装置及存储介质 | |
| Dangi et al. | 5G network traffic control: a temporal analysis and forecasting of cumulative network activity using machine learning and deep learning technologies | |
| US20250039473A1 (en) | Video Streaming Control Based on Network Conditions | |
| CN119012298B (zh) | 一种多频运营商网络参数智能协同和调整系统及方法 | |
| CN115549750A (zh) | 卫星接入选择方法和装置、电子设备和存储介质 | |
| CN116933873A (zh) | 样本确定方法、装置及设备 | |
| CN118784550A (zh) | 路由决策模型训练方法、装置、计算机设备及存储介质 | |
| CN116528296A (zh) | 物联网场景下的资源分配方法、装置、设备及存储介质 | |
| CN116957067A (zh) | 公共安全事件预测模型的强化联邦学习方法及装置 | |
| CN117979353A (zh) | 数据处理方法及装置、设备、存储介质、智能交通系统 | |
| CN114257988A (zh) | 超可靠车联网中面向不完美csi的资源分配方法及系统 | |
| Xu et al. | BlockHybrid: Accelerating Object Detection Pipelines With Hybrid Block-Wise Execution | |
| CN114363827B (zh) | 网络资源分配方法及装置、存储介质及电子设备 | |
| CN116963193A (zh) | 数据传输方法、装置、设备、存储介质及产品 | |
| CN116418691A (zh) | 一种基于深度学习的移动网络实时带宽预测方法及系统 | |
| Zhi et al. | Autonomic Mobility Management of Heterogeneous Ubiquitous Networks: A Deep Reinforcement Learning Method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 23825993 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 23825993 Country of ref document: EP Kind code of ref document: A1 |