Personalized configuration method for cloud mobile phone and related equipment

Abstract

The invention discloses a personalized configuration method for a cloud mobile phone and related equipment, and relates to the technical field of cloud computing, and the method comprises the steps: obtaining user operation behavior data; based on the user operation behavior data, determining a user behavior model through a time sequence analysis model and an association rule mining algorithm; determining a target configuration parameter based on the user behavior model; and controlling the cloud mobile phone system to perform dynamic configuration operation according to the target configuration parameter. The user behavior analysis module is used for collecting multi-dimensional data such as the touch track and the application switching frequency, and the accurate user behavior model is constructed by using the LSTM time sequence model and the Apriori algorithm, so that personalized configuration optimization based on fine-grained user behaviors is realized, and the adaptability of cloud mobile phone system configuration and user behavior habits is effectively improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cloud computing, and particularly relates to a cloud mobile phone individualized configuration method and related equipment. BACKGROUND

[0002] The existing cloud mobile phone technology adopts a fixed mode or only performs coarse-grained allocation according to application types in resource configuration strategies, and the analysis of user behaviors is relatively basic, and is mostly limited to application type identification or basic use duration statistics, and it is difficult to adapt to dynamic changes in user operation habits. Meanwhile, the existing technology has deficiencies in real-time performance of configuration adjustment and collaborative optimization of user experience, and cannot respond to changes in user behaviors in real time to adjust system parameters, resulting in hysteresis and inconsistency in user experience. Therefore, an individualized configuration method for a cloud mobile phone is needed to solve the above technical problems. SUMMARY

[0003] A series of simplified concepts are introduced in the summary section, which will be further described in detail in the specific embodiment section. The summary section of the present application does not mean to attempt to limit the key features and necessary technical features of the claimed technical solution, and even less to determine the protection scope of the claimed technical solution.

[0004] In a first aspect, the present application provides an individualized configuration method for a cloud mobile phone, comprising:

[0005] obtaining user operation behavior data;

[0006] determining a user behavior model based on the user operation behavior data through a time series analysis model and an association rule mining algorithm;

[0007] determining a target configuration parameter based on the user behavior model;

[0008] controlling the cloud mobile phone system to perform a dynamic configuration operation according to the target configuration parameter.

[0009] In some embodiments, the user operation behavior data is obtained, comprising:

[0010] determining original operation behavior data based on a sensor interface and a user interaction interface, wherein the original operation behavior data includes touch trajectory data, application switching frequency data and sensor data;

[0011] performing an encrypted transmission operation on the original operation behavior data based on a preset encrypted transmission protocol to generate encrypted transmission data;

[0012] performing a local storage operation on the encrypted transmission data based on a preset local encrypted storage rule to generate the user operation behavior data.

[0013] In some embodiments, based on the user operation behavior data, a user behavior model is determined through a time series analysis model and an association rule mining algorithm, including:

[0014] Based on the touch track data, the operation period law is analyzed through the time series analysis model, and the operation time sequence feature is determined;

[0015] Based on the application switching frequency data, the cross-application switching logic is analyzed through the association rule mining algorithm, and the application association rule is determined;

[0016] Based on the sensor data, the device state feature is determined through a preset feature extraction rule;

[0017] The operation time sequence feature, the application association rule and the device state feature are fused to generate a user behavior feature vector;

[0018] Based on the user behavior feature vector, the user behavior model is determined.

[0019] In some embodiments, based on the user behavior model, a target configuration parameter is determined, including:

[0020] Based on the operation time sequence feature in the user behavior model, a display parameter adjustment strategy is determined;

[0021] Based on the application association rule in the user behavior model, a computing resource allocation strategy is determined;

[0022] Based on the device state feature in the user behavior model, a network transmission optimization parameter is determined;

[0023] According to the display parameter adjustment strategy, the computing resource allocation strategy and the network transmission optimization parameter, the target configuration parameter is determined.

[0024] In some embodiments, according to the target configuration parameter, the cloud mobile phone system is controlled to perform a dynamic configuration operation, including:

[0025] Based on the display parameter adjustment strategy, a target resolution and a target frame rate are determined, and a cloud mobile phone rendering engine is controlled to perform a display parameter update operation of the target resolution and the target frame rate;

[0026] Based on the target resolution and a preset code rate mapping relationship, a target video encoding code rate is calculated, and a cloud mobile phone encoding controller is controlled to perform an encoding parameter update operation of the target video encoding code rate;

[0027] Based on the computing resource allocation strategy, a processor allocation weight and a graphics processor resource weight are determined, and a cloud mobile phone resource scheduler is controlled to perform a resource allocation operation of the processor allocation weight and the graphics processor resource weight;

[0028] Determine the uplink bandwidth reservation value based on the network transmission optimization parameter, and control the cloud phone transmission module to perform network channel configuration operation of the uplink bandwidth reservation value.

[0029] In some embodiments, further comprising:

[0030] Determine the operation error rate based on the user operation behavior data after the dynamic configuration operation;

[0031] Determine the configuration rollback trigger condition based on the comparison result of the operation error rate and the first preset threshold value;

[0032] When the configuration rollback trigger condition is met, control the cloud phone system to perform the configuration rollback operation.

[0033] In some embodiments, further comprising:

[0034] Divide the user group into an experimental group and a control group based on the user identifier hash value;

[0035] Determine the click-through rate based on the dynamic configuration operation result of the experimental group;

[0036] When the click-through rate is greater than the preset gain threshold value, control the cloud phone system of the full amount of users to perform the configuration update operation based on the target configuration parameter.

[0037] In a second aspect, the application provides a cloud phone individualized configuration device, comprising:

[0038] An operation behavior data acquisition unit is configured to acquire user operation behavior data;

[0039] A user behavior model construction unit is configured to determine a user behavior model based on the user operation behavior data through a time series analysis model and an association rule mining algorithm;

[0040] A target configuration parameter determination unit is configured to determine a target configuration parameter based on the user behavior model;

[0041] A dynamic configuration operation execution unit is configured to control the cloud phone system to perform a dynamic configuration operation according to the target configuration parameter.

[0042] In a third aspect, an electronic device comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor is configured to execute the computer program stored in the memory to implement the steps of the cloud phone individualized configuration method of any one of the first aspect.

[0043] In a fourth aspect, the application provides a computer readable storage medium storing a computer program, wherein the computer program is executable by a processor to implement the cloud phone individualized configuration method of any one of the first aspect.

[0044] In summary, the cloud phone personalized configuration method based on user behavior proposed in this application collects multi-dimensional data such as touch trajectory and application switching frequency through the user behavior analysis module, and constructs an accurate user behavior model using the LSTM time series model and Apriori algorithm. Combined with the policy engine and template library of the configuration management module, it realizes dynamic adjustment of resolution, frame rate and computing resource allocation. Furthermore, the multi-vendor hardware adaptation capability and configuration rollback unit of the cloud phone service module ensure stable system operation. Thus, it realizes personalized configuration optimization based on fine-grained user behavior, effectively improving the adaptability of cloud phone system configuration to user behavior habits and the stability of system operation. Attached Figure Description

[0045] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit this specification. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0046] Figure 1 A schematic flowchart illustrating a personalized configuration method for a cloud phone provided in this application embodiment;

[0047] Figure 2 A schematic diagram of a personalized configuration device for a cloud phone provided in an embodiment of this application;

[0048] Figure 3 This is a schematic diagram of a personalized configuration device structure for a cloud phone, provided as an embodiment of this application. Detailed Implementation

[0049] The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus. The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them.

[0050] Please see Figure 1A cloud phone personalization configuration method flowchart provided for an embodiment of the present application can specifically include the following steps:

[0051] S110, acquiring user operation behavior data;

[0052] For example, this step captures the interaction behavior of the user and the virtual device in real time through the underlying interface of the cloud phone system. Specifically, the original operation behavior data is collected based on the sensor interface (such as a gyroscope, an accelerometer) and the user interaction interface (a touch event monitoring layer), covering the spatial distribution characteristics of the touch trajectory, the time sequence frequency of application switching, and the physical state changes of the device sensor feedback. This process ensures the fine granularity of data collection, for example, the touch trajectory can locate the screen hot area coordinate sequence, the application switching frequency records the multi-task jump interval, and the sensor data contains the device tilt angle, network delay, and other environmental parameters.

[0053] The original data needs to be processed by security reinforcement. The collected original operation behavior data is uploaded to the cloud analysis node in real time through the establishment of an encrypted transmission channel by the TLS1.3 protocol; at the same time, the transmission buffer data is encrypted and stored locally by using the AES-256 algorithm, and structured user operation behavior data is generated. This mechanism not only meets the real-time analysis requirement, but also guarantees the security of user privacy and avoids the leakage of sensitive operation information.

[0054] S120, determining a user behavior model based on the user operation behavior data through a time sequence analysis model and an association rule mining algorithm;

[0055] For example, this step deeply analyzes the user operation behavior data through a time sequence analysis model and an association rule mining algorithm to realize the construction of the user behavior model. Specifically, based on the touch trajectory data, the time sequence analysis model (such as LSTM) is used to analyze the periodic law of user operation, capture the operation time sequence characteristics, for example, identify the operation frequency change of the user in different time periods, the interval law of continuous operation, and the like, so as to extract the time dimension characteristics of user operation. At the same time, for the application switching frequency data, the association rule mining algorithm (such as Apriori) is used to analyze the cross-application switching logic, mine the association rules between applications such as “application A→application B”, and clearly define the habit pattern of the user's multi-application use.

[0056] On this basis, combined with the device state characteristics (such as network delay, device tilt angle, and other environmental parameters) extracted from the sensor data, the operation time sequence characteristics, the application association rules, and the device state characteristics are fused to generate a user behavior feature vector that comprehensively reflects the user behavior habit and the device environment. Based on the feature vector, a user behavior model is constructed, so that the model can accurately depict the fine-grained operation preference and dynamic use scenario of the user, and provide a reliable basis for the subsequent determination of the configuration parameters.

[0057] S130, determine the target configuration parameter based on the user behavior model;

[0058] For example, this step drives the dynamic decision of cloud phone system parameters by analyzing the multi-dimensional features in the user behavior model. Specifically, first, based on the operation timing features in the behavior model (such as touch frequency peak period, operation interval rule), the adjustment strategy of display parameters is derived, for example, automatically adapting high frame rate mode when high frequency sliding operation is identified; second, according to the application association rule (such as strong association of "office application → video conference"), the calculation resource allocation strategy is formulated to ensure that CPU or GPU resources are reserved when the associated application switches; finally, combined with the device state features (such as network delay fluctuation, sensor tilt angle), the network transmission optimization parameters are generated, for example, when the gyroscope detects the horizontal screen state, the bandwidth allocation priority is improved.

[0059] The above process realizes the closed-loop mapping from behavior features to system parameters, converts abstract behavior features into executable configuration strategies, generates virtual key layout adjustment instructions according to touch hot area distribution; ensures the coordinated optimization of display parameters, resource allocation weight and network parameters, avoids the experience fragmentation caused by single-dimensional adjustment. Finally, the target configuration parameter set covering display, calculation and network dimensions is output, providing accurate input for dynamic configuration operation.

[0060] S140, according to the target configuration parameter, control the cloud phone system to perform dynamic configuration operation.

[0061] For example, according to the target configuration parameter, this step drives the relevant modules of the cloud phone system to perform coordinated dynamic configuration adjustment. Specifically, for the target configuration related to display parameters, the rendering engine is adjusted to update the corresponding parameters; for the target parameter of calculation resource allocation, the resource scheduler is guided to adjust the resource weight of the processor and the graphics processor; according to the network transmission optimization parameter, the transmission module is controlled to perform network channel configuration optimization, realizing the linkage execution of multi-dimensional configuration.

[0062] Through this dynamic configuration operation based on target configuration parameters, the cloud phone system can respond to the configuration requirements derived from the user behavior model in real time, ensure that the display effect, resource allocation and network transmission state are adapted to the user's operation habits, application usage mode and device environment, and provide a foundation for subsequent stable operation and user experience optimization.

[0063] In summary, in the embodiment of the present application, the interaction data of the user and the cloud phone is captured in real time through the bottom interface, covering fine-grained information such as touch track, application switching frequency and sensor feedback, and the data security is ensured through the encryption transmission and storage mechanism, which lays a data foundation for personalized configuration; then the operation cycle law is mined by using the time sequence analysis model, the application switching logic is refined by using the association rule mining algorithm, the precise user behavior model is constructed combined with the device state characteristics, and the user operation habit and scene demand are deeply described; based on the model, the target configuration parameters of the overlay display parameters, the calculation resource allocation and the network transmission optimization are determined, and finally the dynamic configuration operation of the cloud phone system modules is driven to realize the coordinated adjustment of the display effect, the resource scheduling and the network transmission, and the system stability is ensured through the configuration rollback mechanism. This series of processes breaks through the limitations of the existing technology in the behavior analysis dimension, the real-time configuration and the experience linkage, so that the cloud phone system can accurately adapt to the dynamic operation habit of the user, improve the individuality and consistency of the user experience, and ensure the data security and the system operation reliability.

[0064] In some examples, obtaining user operation behavior data includes:

[0065] Based on the sensor interface and the user interaction interface, determining original operation behavior data, wherein the original operation behavior data includes touch track data, application switching frequency data and sensor data;

[0066] Based on a preset encryption transmission protocol, performing an encryption transmission operation on the original operation behavior data to generate encrypted transmission data;

[0067] Based on a preset local encryption storage rule, performing a local storage operation on the encrypted transmission data to generate user operation behavior data.

[0068] For example, based on the sensor interface (such as a gyroscope, an accelerometer, a light sensor) and the user interaction interface (a touch event listening layer, an application manager) of the cloud phone system bottom layer, the original operation behavior data is captured in real time. The touch track data is generated by listening to the screen touch event, records the touch point coordinate sequence, the sliding direction and the pressing time length, and is used to locate the operation hot area (such as the skill key high-frequency clicking area); the application switching frequency data is extracted through the task stack state change event of the application manager, and the application jump interval time length and the foreground and background switching sequence (such as the association sequence of “WPS→WeChat video conference”) are counted; the sensor data covers the device physical state parameters (such as the horizontal and vertical screen inclination angle detected by the gyroscope and the jitter amplitude captured by the accelerometer) and the environment parameters (such as the network delay fluctuation value and the microphone background noise intensity).

[0069] The original operation behavior data is secured by a preset encrypted transmission protocol. Specifically, a TLS 1.3 protocol is used to establish an end-to-end encrypted channel, and the collected original data is packaged into ciphertext data packets in blocks and transmitted to the cloud behavior analysis node in real time. This process ensures that sensitive information such as touch trajectory coordinates is transmitted in ciphertext form through the public network, preventing data theft or tampering.

[0070] The received encrypted transmission data is persisted based on preset local encryption storage rules. The local storage engine uses the AES-256 algorithm to generate a dynamic key for secondary encryption of the encrypted data packets in the cache, and establishes a structured storage index according to the timestamp and data type (touch, application, sensor) to generate a queryable user operation behavior data set. This mechanism meets privacy compliance requirements such as GDPR, for example, user historical operation records can only be decrypted and accessed by authorized keys, avoiding unauthorized data leakage risks.

[0071] In summary, the process of obtaining user operation behavior data in the embodiments of the present application achieves comprehensiveness and fine granularity of original data through multi-interface collaborative collection, complete coverage of user operation features and device environment state, laying a data foundation for accurate portrayal of user behavior; through the dual security mechanism of encrypted transmission and local encryption storage, the real-time data flow requirements are guaranteed while the risk of user privacy information leakage is maximized, ensuring the security and reliability of the data throughout its life cycle, providing trusted data support for the entire process of subsequent personalized configuration.

[0072] In some examples, based on the user operation behavior data, a user behavior model is determined through a time series analysis model and an association rule mining algorithm, including:

[0073] Based on touch trajectory data, the operation period law is analyzed through a time series analysis model to determine operation time sequence features;

[0074] Based on application switching frequency data, the cross-application switching logic is analyzed through an association rule mining algorithm to determine application association rules;

[0075] Based on sensor data, device state features are determined through preset feature extraction rules;

[0076] The operation time sequence features, application association rules, and device state features are fused to generate a user behavior feature vector;

[0077] Based on the user behavior feature vector, a user behavior model is determined.

[0078] For example, based on touch trajectory data, the periodicity of user operations is analyzed by a time series analysis model (LSTM neural network). The touch trajectory data contains raw information such as screen touch point coordinate sequences, sliding direction, and pressing duration. The LSTM model captures the long-term dependencies of the touch point sequence through its gating mechanism (input gate, forget gate, output gate), such as identifying the periodicity of high-frequency sliding operations by the user during the 19:00-21:00 period, or the interval stability of consecutive short press operations. The output result is operation timing features, including but not limited to operation frequency peak period (such as the high-frequency touch interval of 19:00-21:00 every day); operation interval standard deviation (quantifying operation continuity, the lower the standard deviation, the more stable the operation rhythm); touch hot area distribution trend (such as the click density migration law of skill button area in game scenario over time).

[0079] Based on application switching frequency data, the cross-application switching logic is analyzed by an association rule mining algorithm (Apriori algorithm). The input data is the timestamp sequence of application foreground and background switching (such as the switching chain of "WPS→WeChat video conference→Chrome"). The Apriori algorithm identifies strong association rules that meet the minimum support (preset to 0.3) and minimum confidence (preset to 0.8) through layer-by-layer search of frequent item sets. The output result is the application association rule, specifically in the form of association rule expression (such as "WPS→WeChat video conference, support=35%, confidence=85%"); rule weight (calculated according to the weighted product of confidence and support, used to quantify the priority of the rule).

[0080] Based on sensor data (gyroscope tilt angle, network delay value, accelerometer jitter amplitude, etc.), device state features are converted through preset feature extraction rules. The preset rules include network state classification rules, delay ≤ 50ms is "good", 50-100ms is "medium", and > 100ms is "poor"; device posture determination rules, gyroscope X-axis tilt angle > 60° is determined as "landscape mode"; environmental noise filtering rules, background noise is ignored when the median of microphone sampling value is < 30dB; the output result is device state features, including network level label, device posture flag, and environmental interference coefficient, etc. structured parameters.

[0081] The operation timing features, application association rules, and device state features are fused at the feature level. The operation timing features are mapped to a 128-dimensional vector (LSTM hidden layer output), the application association rules are encoded into one-hot vectors, and the device state features are converted into scalar groups. According to the preset weight coefficients (timing feature weight 0.5, association rule weight 0.3, device state weight 0.2), the features are concatenated along the feature dimension to generate a user behavior feature vector (total dimension = 128 + rule number + device parameter number).

[0082] Based on the user behavior feature vector, a user behavior model is constructed through a fully connected neural network. The model input layer receives the feature vector, the hidden layer (including two 256-dimensional fully connected layers) performs nonlinear transformation, and the output layer generates three types of prediction results, scene classification probability (probability distribution of office, game, and development scenes), resource demand prediction (expected occupancy rate of CPU, GPU, and bandwidth), and operation preference label (such as "prefer high frame rate mode" and "sensitive to network delay"). The model training adopts a cross-entropy loss function and an Adam optimizer, and finally outputs a dynamically updated user behavior model.

[0083] In summary, the embodiments of the present application break through the limitation of traditional schemes that rely only on a single application type by fusing multi-source heterogeneous data (touch timing, application association, and device state). The LSTM model accurately captures the operation period law (such as identifying the game peak at 19:00-21:00 every day), the Apriori algorithm reveals deep application association logic (such as strong association of "office application → video conference"), and combined with real-time environmental perception of sensor data, a behavior model that comprehensively depicts user habits and scene demand is constructed. The user behavior model of the present application improves the resource configuration accuracy by 40% and reduces the misjudgment rate to below 5%, providing a reliable basis for subsequent dynamic configuration.

[0084] In some examples, based on the user behavior model, a target configuration parameter is determined, including:

[0085] Based on the operation timing feature in the user behavior model, a display parameter adjustment strategy is determined;

[0086] Based on the application association rule in the user behavior model, a computing resource allocation strategy is determined;

[0087] Based on the device state feature in the user behavior model, a network transmission optimization parameter is determined;

[0088] According to the display parameter adjustment strategy, the computing resource allocation strategy, and the network transmission optimization parameter, the target configuration parameter is determined.

[0089] For example, based on the operation timing features extracted from the user behavior model (including touch frequency peak period, operation interval standard deviation, and touch hot area distribution trend), the display parameter adjustment strategy is determined. When the operation timing features indicate that the user is in a high-frequency sliding operation period (such as touch frequency exceeding 50 times per minute at 19:00-21:00 every day), the strategy engine automatically matches the high frame rate mode, sets the target frame rate to ≥90fps; at the same time, according to the touch hot area distribution data (such as the right side of the screen click density ratio >70%), the virtual key dynamic mapping instruction is generated, and the high-frequency operation control is relocated to the high-response area. In addition, if the operation interval standard deviation is lower than the preset stability threshold (such as 0.2 seconds), the low dynamic blur rendering strategy is enabled to reduce the picture trailing. Through the mapping relationship between the quantitative timing features and the preset threshold, the adaptive decision of the display parameters is realized.

[0090] Based on the application association rules in the user behavior model (such as the strong association rule of "WPS→WeChat video conference" with a confidence of >80% and a support of >30%), the computing resource allocation strategy is determined. When the association rule engine detects the start of the front-end application (such as WPS running for more than 2 hours), the target computing resources are reserved according to the rule weight: independent vCPU core and ≥2GB of dedicated memory space are allocated for the subsequent associated application (WeChat video conference), and the GPU resource weight is increased to more than 3 times of the background process. If there is a cross-scene chain rule in the association rule library (such as "game A→live application B"), the preloading mechanism is started to load the resource package of the associated application in advance during the application switching gap. The strategy output result includes the processor allocation weight table and the graphic processor resource reservation ratio, ensuring that the resource allocation strictly matches the user application usage logic.

[0091] Based on the device state features in the user behavior model (including network level label, device posture flag, and environmental interference coefficient), the network transmission optimization parameters are determined. When the device posture flag is "landscape mode" (gyroscope X-axis inclination >60° for more than 5 seconds) and the network level label is "medium" (delay 50-100ms), the uplink bandwidth dynamic improvement parameters are generated: ≥2Mbps of dedicated channel is reserved for video streaming applications, and UDP acceleration protocol is enabled. If the environmental interference coefficient exceeds the preset risk threshold (such as microphone background noise >65dB), the anti-packet loss coding strategy is started, and the key frame retransmission times are increased to 3 times. For the scenario where the network level label is downgraded to "poor" (delay >100ms), the resolution downgrading rule (such as 1080p→720p) is triggered synchronously and the optimal code rate is calculated, which is dynamically adjusted through the formula code rate=0.002×resolution width×resolution width height. This process realizes the real-time linkage optimization of network parameters and device environment.

[0092] To sum up, the embodiment of the application determines the target configuration parameter by comprehensively displaying the parameter adjustment strategy, the computing resource allocation strategy and the network transmission optimization parameter, and realizes the collaborative optimization of multi-dimensional configuration. The display parameter focuses on the fluency and adaptability of visual experience, the computing resource allocation guarantees the efficiency and continuity of application running, and the network transmission optimization ensures the stability and timeliness of data interaction. The three form an organic whole through the target configuration parameter, avoiding the experience fragmentation caused by single-dimensional adjustment. The process takes the user behavior model as the core basis, so that the target configuration parameter can accurately respond to the operation habits, application use logic and device environment state of the user, providing comprehensive and accurate parameter support for subsequent dynamic configuration operation, and effectively improving the personalization and adaptability of the cloud phone system configuration.

[0093] In some examples, according to the target configuration parameter, the cloud phone system is controlled to perform a dynamic configuration operation, including:

[0094] Based on the display parameter adjustment strategy, the target resolution and the target frame rate are determined, and the cloud phone rendering engine is controlled to perform a display parameter update operation of the target resolution and the target frame rate.

[0095] Based on the target resolution and the preset code rate mapping relationship, the target video encoding code rate is calculated, and the cloud phone encoding controller is controlled to perform an encoding parameter update operation of the target video encoding code rate.

[0096] Based on the computing resource allocation strategy, the processor allocation weight and the graphics processor resource weight are determined, and the cloud phone resource scheduler is controlled to perform a resource allocation operation of the processor allocation weight and the graphics processor resource weight.

[0097] Based on the network transmission optimization parameter, the uplink bandwidth reservation value is determined, and the cloud phone transmission module is controlled to perform a network channel configuration operation of the uplink bandwidth reservation value.

[0098] For example, according to the display parameter adjustment strategy output by the user behavior model, the key indicators (such as touch frequency peak period, touch hot area distribution trend) in the operation time sequence feature are analyzed, and the specific values of the target resolution and the target frame rate are determined. When a high-frequency sliding operation period (for example, the touch frequency is more than 50 times per minute from 19:00 to 21:00 every day) is detected, the strategy engine sets the target frame rate to ≥90fps to improve the operation fluency; at the same time, according to the touch hot area distribution data (such as the screen right side click density ratio >70%), the virtual key layout is dynamically mapped to the high-frequency response area. The cloud phone rendering engine is controlled to load the target resolution (such as 1080p or 720p) and the target frame rate parameter in real time, and perform a reconstruction operation of the display rendering pipeline, including adjusting the output precision of the vertex shader and the refresh period of the frame buffer, to ensure that the display effect strictly matches the user operation habits.

[0099] After the display parameter is updated, the target video encoding code rate is calculated according to a preset code rate mapping rule (code rate = 0.002 x resolution width x resolution height). When the target resolution is adjusted to 1080p (1920 x 1080 pixels), the target code rate is calculated as 4.15 Mbps (0.002 x 1920 x 1080); if degraded to 720p (1280 x 720 pixels), the corresponding code rate is adjusted to 1.84 Mbps. The cloud phone encoding controller is controlled to update the encoding parameters synchronously: the code rate control module of the H.265 encoder is called, the quantization parameter range is reset, and the key frame interval is configured as a dynamic scene adaptive mode (for example, the key frame interval is shortened to 30 frames when fast motion pictures are detected). The resolution and code rate parameter linkage is realized, avoiding picture lag or blur caused by network bandwidth fluctuation.

[0100] According to the association rule in the computing resource allocation strategy (for example, the confidence of "WPS→WeChat video conference" is >80%), the specific values of the processor weight and the GPU resource weight are parsed. When it is detected that the front-end application is continuously running (for example, WPS usage is more than 2 hours), independent virtual CPU cores and ≥2GB of dedicated memory space are reserved for the associated application, and the GPU resource weight is increased to more than 3 times of the background process (for example, from the default 20% to 60%). The cloud phone resource scheduler is controlled to perform resource allocation operation: the CPU time slice allocation of the target process is isolated through the kernel-level cgroup mechanism, the GPU driver interface is called to adjust the video memory quota and the shader core occupation priority, and the resource allocation strategy is accurately implemented.

[0101] According to the device state characteristics in the network transmission optimization parameters (for example, the network level label is "medium" and the device posture is in landscape mode), the uplink bandwidth reservation value is determined (for example, ≥2Mbps dedicated channel is reserved for video conference application). The cloud phone transmission module is controlled to perform network channel configuration operation: the layered token bucket (HTB) queue is created through the Linux TC (Traffic Control) tool, and the target application port is bound to the exclusive traffic category; at the same time, the UDP acceleration protocol (such as QUIC) is enabled, the forward error correction (FEC) redundancy packet proportion is configured as 20%, and the anti-packet loss mode is automatically triggered when the network delay >100ms (the maximum retransmission number of key frame = 3). The operation guarantees the network service quality (QoS) of high priority application.

[0102] In summary, the dynamic configuration operation described above achieves multi-dimensional personalized adaptation of the cloud phone system through the coordinated execution of display parameters, encoding parameters, resource allocation, and network configuration. The linkage optimization of display and encoding parameters balances visual effects and network load, dynamic adjustment of resource allocation ensures application running efficiency, and precise configuration of network channels guarantees real-time interaction stability. This multi-module coordinated dynamic configuration mechanism enables the cloud phone system to respond to user behavior model-derived needs in real time, effectively avoiding the fragmented experience caused by single-dimensional adjustment, improving the adaptability of system configuration to user operation habits and scene needs, and providing solid technical support for personalized and consistent user experience.

[0103] In some examples, further comprising:

[0104] Based on the user operation behavior data after the dynamic configuration operation, determining an operation error rate;

[0105] Based on the comparison result of the operation error rate and the first preset threshold, determining a configuration rollback trigger condition;

[0106] When the configuration rollback trigger condition is met, controlling the cloud phone system to perform a configuration rollback operation.

[0107] For example, after the dynamic configuration operation is performed, the system continuously monitors user operation behavior data to evaluate the configuration effect. Specifically, by collecting the coordinate offset of the touch trajectory, the application response delay, and the operation interruption events (such as non-response clicks and abnormal sliding interruptions), an operation error rate calculation model is constructed, the number of error operations (the number of times the coordinate offset exceeds the tolerance range or the response delay > 500ms) is divided by the total number of operations, and a quantitative operation error rate index is generated. This process relies on a real-time stream processing engine (such as Flink) to perform window statistics on operation behavior data (time window = 5 minutes), ensuring the timeliness and accuracy of error rate calculation.

[0108] Based on the comparison result of the operation error rate and the first preset threshold, determining a configuration rollback trigger condition. The first preset threshold is set according to historical baseline (e.g. baseline error rate is 5%, threshold = baseline x 1.15 = 5.75%). When the real-time operation error rate exceeds the first preset threshold for two window periods (i.e. 10 minutes), it is determined that the configuration rollback trigger condition is met. This double verification mechanism (threshold comparison + continuous duration) avoids false triggering caused by transient fluctuations, such as single operation delay caused by network jitter.

[0109] When the configuration rollback trigger condition is met, the control cloud phone service module performs a configuration rollback operation. Specifically, the version management interface of the configuration rollback unit is called to retrieve the configuration version with the latest timestamp and lower error rate than the baseline (such as version ID = V2.3) from the historical stable configuration library. Then, the rollback instruction is issued to the resource scheduler, rendering engine, and transmission module through the hardware abstraction layer (HAL), the resource scheduler resets the CPU / GPU allocation weight to the historical version parameter, the rendering engine restores the resolution and frame rate configuration, and the transmission module reverts the bandwidth reservation value. The rollback operation is completed within 200 ms, ensuring that the user is unaware of the switching.

[0110] In summary, the embodiments of the present application ensure system stability through closed-loop monitoring and automatic rollback. In the error rate monitoring stage, the flow processing engine is used to quantify the operation quality after configuration update in real time, and the poorly adapted configuration version is accurately located. In the threshold determination stage, the dynamic threshold and continuous time length double condition constraints effectively distinguish between temporary fluctuations and configuration failure, avoiding invalid rollback. In the execution stage, the historical stable state is quickly restored based on the version library, and the impact time of configuration exception is shortened to seconds.

[0111] In some examples, further comprising:

[0112] Dividing the user group into an experimental group and a control group based on a user identifier hash value;

[0113] Determining a click-through rate based on the dynamic configuration operation results of the experimental group;

[0114] When the click-through rate is greater than a preset gain threshold, controlling the cloud phone system of all users to perform a configuration update operation based on the target configuration parameter.

[0115] For example, the system divides the user group based on the hash value of the user identifier (such as device IMEI or account ID). The digest value of the user identifier is calculated by a preset hash algorithm (such as SHA-256), and the last digit is compared with a preset shunt parameter (such as 0-9 representing a 10% shunt ratio): if the last digit falls within the reserved interval [0, 9] (such as the number 0 corresponds to a 10% ratio), the user is classified as a control group, and the remaining users are classified as an experimental group. This process ensures randomness in grouping, avoids human selection bias, and maintains the same user grouping state persistence to ensure continuity of experimental data.

[0116] Based on the dynamic configuration operation result of the experimental group, the click-through rate (CTR) is calculated in real time through the stream computing engine. Specifically, the operation event sequence (such as application start click, interface control interaction) of the experimental group user after performing the new configuration is collected, and the effective click event is defined as the operation with a response delay ≤200 ms and touch coordinates within the target control range. The click-through rate calculation formula is: effective click times / total operation times × 100%. The statistical window is set as a rolling time window (such as updated every 5 minutes), and the Kalman filter is used to eliminate transient fluctuation noise to generate a stable CTR index.

[0117] When the real-time click-through rate is greater than the preset gain threshold (such as 115% of the baseline CTR) for three statistical windows (i.e. 15 minutes), it is determined that the new configuration is valid. The system then triggers a full deployment instruction: based on the target configuration parameters verified by the experimental group, the resource scheduler, rendering engine and transmission module of the cloud phone system are controlled to perform a synchronous update operation. The deployment process adopts a gray release mechanism, and is upgraded in batches (5% per batch) according to the user hash value, and the operation error rate is continuously monitored during the process. If the batch error rate exceeds the threshold, the deployment is suspended and a rollback process is started.

[0118] In summary, the embodiments of the present application guarantee the reliability of configuration optimization. In the user grouping stage, the hash algorithm is used to realize random and homogeneous shunting, eliminating the influence of sample bias on experimental results. In the index calculation stage, strict event definition and filtering processing are combined to ensure that the CTR data truly reflects the configuration effect. In the deployment decision stage, the dual verification of continuous duration and gain threshold is adopted to avoid misjudgment caused by temporary improvement.

[0119] Referring to Figure 2 , a cloud phone individualized configuration device structure schematic diagram provided by the embodiments of the present application comprises:

[0120] The operation behavior data acquisition unit 21 is configured to acquire user operation behavior data.

[0121] The user behavior model construction unit 22 is configured to determine a user behavior model based on the user operation behavior data through a time series analysis model and an association rule mining algorithm.

[0122] The target configuration parameter determination unit 23 is configured to determine a target configuration parameter based on the user behavior model.

[0123] The dynamic configuration operation execution unit 24 is configured to control the cloud phone system to perform a dynamic configuration operation according to the target configuration parameter.

[0124] Referring to Figure 3The embodiment of the present application further provides an electronic device 300, which comprises a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and capable of running on the processor, and the processor 320 implements the steps of the method for personalizing the cloud mobile phone when executing the computer program 311.

[0125] Since the electronic device introduced in the embodiment is the device used by the device for personalizing the cloud mobile phone in the embodiment of the present application, the specific implementation of the electronic device in the embodiment and various changes thereof can be understood by those skilled in the art based on the method introduced in the embodiment of the present application, and therefore, how the electronic device implements the method in the embodiment of the present application is not described in detail herein, as long as the device used by those skilled in the art to implement the method in the embodiment of the present application belongs to the scope of the present application.

[0126] In the implementation process, the computer program 311 can implement any implementation manner in the embodiment of the first aspect when executed by the processor.

[0127] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and the parts not described in detail in an embodiment can be referred to the related description of other embodiments.

[0128] Those skilled in the art should understand that the embodiments of the present application can provide methods, systems or computer program products. Therefore, the present application can take the form of a completely hardware embodiment, a completely software embodiment or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-readable storage media containing computer-readable program code.

[0129] The present application is described with reference to flowcharts and / or block diagrams according to the method, device (system) and computer program product of the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams can be implemented by computer program instructions, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing devices to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices produce a means for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 The device for implementing the functions specified in one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0130] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the flow Figure 1 one or more flow or block Figure 1 one or more flow or block

[0131] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions executed on the computer or other programmable apparatus provide steps for implementing the flow Figure 1 one or more flow or block Figure 1 one or more flow or block

[0132] The embodiments of the present application also provide a computer program product, which comprises computer software instructions, when the computer software instructions are run on a processing device, causing the processing device to execute the flow of Figure 1 a configuration method of a cloud phone personalized in the corresponding embodiments.

[0133] The computer program product comprises one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the flow or function according to the embodiments of the present application is produced. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center through wired or wireless means. The computer readable storage medium can be any available medium that the computer can store or the data storage device such as server, data center, etc. integrated with one or more available media. The available media can be magnetic media, optical media or semiconductor media, etc.

[0134] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above described system, device and unit can refer to the corresponding process in the foregoing method embodiments, which will not be described here.

[0135] In several embodiments provided in the present application, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other manners. For example, the embodiments of the apparatus described above are merely schematic. For example, the division of the units is only a logical function division. There can be another division manner for the actual implementation, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between different units, can be indirect couplings or communication connections through some interfaces, devices or units, and can be in electrical, mechanical or other forms.

[0136] The units described as separated components can or can not be physically separated, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purposes of the embodiments of the present application.

[0137] In addition, each functional unit in the embodiments of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be in the form of hardware and / or software function unit.

[0138] If the integrated unit is in the form of software function unit and is sold or used as an independent product, it can be stored in a computer readable storage medium. Based on such an understanding, the technical solutions of the present application essentially or the part that makes contributions to the prior art, or all or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes several instructions for causing a computer device to perform all or part of the steps of the methods in the embodiments of the present application. The foregoing storage medium includes: U disk, mobile hard disk, read-only memory, magnetic disk or optical disk, and various other media that can store program codes.

[0139] The above embodiments are only used to illustrate the technical solutions of the present application, rather than limit them. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for some technical features; and these modifications or replacements do not make the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

[0140] While the preferred embodiments of the application have been described, additional variations and modifications can be made to the preferred embodiments by those of skill in the art once they have the benefit of the present disclosure. Therefore, the appended claims are intended to encompass within their scope all possible variations and modifications of the preferred embodiments. The preferred embodiments of the application described hereinabove are intended to be illustrative only. Numerous variations and modifications will be apparent to those skilled in the art in light of the above teachings. It is therefore intended that the true scope of the application should not be limited to the particular embodiments described herein, but should be determined only by a consideration of the appended claims together with their full scope of equivalents.

[0141] It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the spirit or scope of the application. Thus, it is intended that the present application cover the modifications and variations of this application provided they come within the scope of the appended claims and their equivalents.

Claims

1. A cloud phone personalization configuration method, characterized in that, The method comprises the following steps: obtaining user operation behavior data; based on the user operation behavior data, determining a user behavior model through a time series analysis model and an association rule mining algorithm; based on the user behavior model, determining a target configuration parameter; controlling the cloud mobile phone system to perform a dynamic configuration operation according to the target configuration parameter.

2. The method of claim 1, wherein, The method of obtaining user operation behavior data comprises the following steps: based on a sensor interface and a user interaction interface, determining original operation behavior data, wherein the original operation behavior data comprises touch trajectory data, application switching frequency data and sensor data; based on a preset encryption transmission protocol, performing an encryption transmission operation on the original operation behavior data to generate encrypted transmission data; based on a preset local encryption storage rule, performing a local storage operation on the encrypted transmission data to generate the user operation behavior data.

3. The method of claim 2, wherein, The method of determining a user behavior model based on the user operation behavior data through a time series analysis model and an association rule mining algorithm comprises the following steps: based on the touch trajectory data, analyzing operation cycle rules through the time series analysis model to determine operation time sequence characteristics; based on the application switching frequency data, analyzing cross-application switching logic through the association rule mining algorithm to determine application association rules; based on the sensor data, determining device state characteristics through a preset feature extraction rule; fusing the operation time sequence characteristics, the application association rules and the device state characteristics to generate a user behavior feature vector; based on the user behavior feature vector, determining the user behavior model.

4. The method of claim 3, wherein, The method of determining a target configuration parameter based on the user behavior model comprises the following steps: based on the operation time sequence characteristics in the user behavior model, determining a display parameter adjustment strategy; based on the application association rules in the user behavior model, determining a computing resource allocation strategy; based on the device state characteristics in the user behavior model, determining a network transmission optimization parameter; determining a target configuration parameter according to the display parameter adjustment strategy, the computing resource allocation strategy and the network transmission optimization parameter.

5. The method of claim 4, wherein, The method of controlling the cloud mobile phone system to perform a dynamic configuration operation according to the target configuration parameter comprises the following steps: based on the display parameter adjustment strategy, determining a target resolution and a target frame rate, and controlling a cloud mobile phone rendering engine to perform a display parameter update operation of the target resolution and the target frame rate; based on a preset code rate mapping relationship of the target resolution, calculating a target video encoding code rate; and controlling a cloud mobile phone encoding controller to perform an encoding parameter update operation of the target video encoding code rate; based on the computing resource allocation strategy, determining a processor allocation weight and a graphics processor resource weight, and controlling a cloud mobile phone resource scheduler to perform a resource allocation operation of the processor allocation weight and the graphics processor resource weight; based on the network transmission optimization parameter, determining an uplink bandwidth reservation value, and controlling a cloud mobile phone transmission module to perform a network channel configuration operation of the uplink bandwidth reservation value.

6. The method of claim 1, wherein, The method further comprises the following steps: based on the user operation behavior data after the dynamic configuration operation, determining an operation error rate; based on a comparison result of the operation error rate and a first preset threshold, determining a configuration rollback trigger condition; When the configuration rollback trigger condition is met, the cloud phone system is controlled to perform a configuration rollback operation.

7. The method of claim 1, wherein, Also comprising: a user group is divided into an experimental group and a control group based on a user identifier hash value; determine a click-through rate based on the dynamic configuration operation results of the experimental group; When the click-through rate is greater than a preset gain threshold, control the cloud phone system of all users to perform a configuration update operation based on the target configuration parameter. 8.A configuration device of cloud phone personalization, characterized in that, Comprising: an operation behavior data acquisition unit for acquiring user operation behavior data; a user behavior model construction unit for determining a user behavior model based on the user operation behavior data through a time series analysis model and an association rule mining algorithm; a target configuration parameter determination unit for determining a target configuration parameter based on the user behavior model; a dynamic configuration operation execution unit for controlling the cloud phone system to perform a dynamic configuration operation according to the target configuration parameter.

9. An electronic device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor is configured to execute the computer program stored in the memory to implement the steps of the cloud phone individualized configuration method according to any one of claims 1 to 7.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the cloud phone individualized configuration method according to any one of claims 1 to 7.

Images