Method and system for optimizing resolution of 4K ultra-high-definition set top box

Abstract

The invention discloses a resolution optimization method and system for a 4K ultra-high-definition intelligent set top box, and relates to the technical field of resolution optimization of set top boxes. The invention provides a resolution optimization method for a 4K ultra-high-definition intelligent set top box, and the method comprises the steps: obtaining at least one piece of login user information and target user information, and obtaining the historical resolution preference of at least one login user according to the at least one piece of login user information; determining that the set top box outputs a first preset resolution according to the historical resolution preference and the target user information; according to the network fluctuation change rate of the first preset time period, predicting a prediction probability that the video stream is stuck in a second preset time period, and judging whether the prediction probability exceeds a preset threshold; if yes, the set top box outputs a second preset resolution, and the second preset resolution is smaller than the first preset resolution; and if not, the set top box maintains outputting the first preset resolution.

Description

Technical Field

[0001] This application relates to the technical field of set-top box resolution optimization, and particularly to a method and system for optimizing the resolution of a 4K ultra-high definition set-top box. Background Art

[0002] 4K refers to a 4K resolution, that is, a pixel resolution of 4096×2160, which is 4 times the resolution of a 2K projector and a high-definition TV, belonging to an ultra-high definition resolution. At this resolution, users can see every detail and every close-up in the picture. In addition to the significant improvement in resolution, there are also changes in aspects such as frame rate, color depth, and color space.

[0003] Resolution is one of the very important parameter indicators for liquid crystal TVs. Usually, for screens of the same material and size, the higher the resolution of the TV, the clearer and more delicate the displayed image looks.

[0004] After the manufacturer produces, the optimal resolution of the TV will be set. Therefore, the resolution of the TV is usually set before leaving the factory, and it is usually the optimal resolution of the TV. However, for different types of video streams, the user's requirements for the resolution of the video stream may be different, and the resolution requirements of different family members may also be different. In the prior art, setting the resolution of the set-top box usually requires manual setting. Summary of the Invention

[0005] The main purpose of this application is to provide a method and system for optimizing the resolution of a 4K ultra-high definition smart set-top box, aiming to solve the technical problem in the prior art that for different users, their resolution requirements for different types of video streams may be different, and it is cumbersome to adjust the resolution of the set-top box.

[0006] To achieve the above object, in a first aspect, this application provides a method for optimizing the resolution of a 4K ultra-high definition smart set-top box, including:

[0007] Obtain at least one logged-in user information and target user information, where the logged-in user information is configured as the family library member information of the set-top box, and the target user information is configured as the user information of the user watching TV at the current moment;

[0008] Obtain the historical resolution preferences of at least one logged-in user according to the at least one logged-in user information;

[0009] Determine the first preset resolution output by the set-top box according to the historical resolution preference and the target user information;

[0010] Predict the prediction probability of video stream stuttering in a second preset time period according to the network fluctuation change rate in a first preset time period, and determine whether the prediction probability exceeds a preset threshold;

[0011] If so, the set-top box outputs a second preset resolution, where the second preset resolution is less than the first preset resolution;

[0012] If not, the set-top box maintains the output of the first preset resolution.

[0013] Optionally, the step of obtaining the historical resolution preferences of at least one logged-in user information according to the at least one logged-in user information includes:

[0014] According to the at least one logged-in user information, obtain the historical resolution usage records of each logged-in user;

[0015] According to the historical resolution usage records, obtain the type of video stream watched by each logged-in user, the corresponding time information, and the resolution of the video stream;

[0016] According to the type of video stream, the corresponding time information, and the resolution of the video stream, obtain the historical resolution preferences of at least one logged-in user.

[0017] Optionally, the step of obtaining the first preset resolution output by the set-top box according to the historical resolution preferences and the target user information includes:

[0018] Determine the association relationship between the target user and at least one logged-in user, where all logged-in users belong to a family library;

[0019] If the target user does not belong to the family library members, obtain the specification attributes of the TV, and the set-top box sets the highest resolution of the TV as the first resolution;

[0020] If there is exactly one person among the target users who belongs to the family library members, the set-top box sets the historical resolution preference of the corresponding logged-in user as the first preset resolution;

[0021] If at least two of the target users belong to the family library members, obtain all the logged-in user information in the family library;

[0022] According to the priorities of all the logged-in users in the family library, the set-top box sets the historical resolution preference of the logged-in user with the highest priority among the target users as the first preset resolution.

[0023] Optionally, the step of determining the association relationship between the target user and at least one logged-in user includes:

[0024] When the set-top box receives a resolution setting request or the target user attempts to use the set-top box, the set-top box obtains at least one of the input information, device pairing information, and biometric information of the target user;

[0025] Compare at least one of the input information, device pairing information, and biological information with the logged-in user information;

[0026] Determine whether the target user belongs to the logged-in user.

[0027] Optionally, the step of predicting the probability of video stream stuttering in the second preset time period according to the network fluctuation change rate in the first preset time period and determining whether the predicted probability exceeds the preset threshold includes:

[0028] Divide the first preset time period into a first target time period and a second target time period;

[0029] The network volatility of the first target time period is:

[0030]

[0031] where N is the number of data points in the first target time period, V i is the network performance index of the i-th data point, V i+1 is the network performance index of the (i + 1)-th data point;

[0032] The network volatility of the second target time period is:

[0033]

[0034] where M is the number of data points in the second target time period, V j is the network performance index of the j-th data point, V j+1 is the network performance index of the (j + 1)-th data point;

[0035] The calculation formula of the network fluctuation change rate is as follows:

[0036] Optionally, the formula of the network performance index is as follows:

[0037] V = ω 1 ×B′ + ω 2 ×(1 - D′) + ω 3 ×(1 - Q′)

[0038] where B′ represents the standardized value of bandwidth utilization, ω 1 represents the weight coefficient of bandwidth, D′ represents the standardized value of delay, ω 2 represents the weight coefficient of delay, Q′ represents the standardized value of packet loss rate, ω 3 represents the weight coefficient of packet loss rate.

[0039] Optionally, the step of predicting the probability of video stream stuttering in the second preset time period according to the network fluctuation change rate in the first preset time period and determining whether the predicted probability exceeds a preset threshold includes:

[0040] Obtain the required bandwidth amount and cache interval size of the video stream in the second preset time period;

[0041] Predict the predicted probability of video stream stuttering in the second preset time period according to the network fluctuation change rate, the required bandwidth amount, and the cache interval size.

[0042] Optionally, the step of predicting the probability of video stream stuttering in the second preset time period according to the network fluctuation change rate in the first preset time period and determining whether the predicted probability exceeds a preset threshold includes:

[0043] The calculation formula of the predicted probability is as follows:

[0044]

[0045] Among them, ΔL represents the network fluctuation change rate, F represents the required bandwidth amount, G represents the cache interval size, and β 0 is the intercept term, β 1 is the coefficient of the network fluctuation change rate, β 2 is the coefficient of the required bandwidth amount, β 3 is the coefficient of the cache area size.

[0046] Optionally, the step of if so, the set-top box outputs the second preset resolution, where the second preset resolution is less than the first preset resolution includes:

[0047] The set-top box outputs the second preset resolution and sends a notification message to the target user, and determines whether a cancellation instruction from the target user is received within a preset confirmation time period;

[0048] If not, the set-top box maintains the output of the second preset resolution;

[0049] If so, the set-top box returns to output the first preset resolution.

[0050] In a second aspect, the present application provides a 4K ultra-high-definition intelligent set-top box resolution optimization system, which is characterized by including:

[0051] An information acquisition module, which is configured to acquire at least one logged-in user information and target user information, where the logged-in user information is configured as the family library member information of the set-top box, and the target user information is configured as the user information of the user watching TV at the current moment.

[0052] A preference acquisition module, configured to acquire historical resolution preferences of at least one logged-in user according to the at least one logged-in user information;

[0053] A first preset resolution acquisition module, configured to determine that the set-top box outputs a first preset resolution according to the historical resolution preference and the target user information;

[0054] A prediction probability judgment module, configured to predict the prediction probability of video stream jamming in a second preset time period according to the network fluctuation change rate in a first preset time period, and judge whether the prediction probability exceeds a preset threshold;

[0055] A first execution module, configured to, if so, the set-top box outputs a second preset resolution, where the second preset resolution is less than the first preset resolution;

[0056] A second execution module, configured to, if not, the set-top box maintains the output of the first preset resolution.

[0057] Beneficial effects that the present application can achieve:

[0058] A 4K ultra-high-definition intelligent set-top box resolution optimization method and system proposed by an embodiment of the present application include: acquiring at least one logged-in user information and target user information, where the logged-in user information is configured as the family library member information of the set-top box, and the target user information is configured as the user information of the user watching TV at the current moment; acquiring historical resolution preferences of at least one logged-in user according to the at least one logged-in user information; determining that the set-top box outputs a first preset resolution according to the historical resolution preference and the target user information; predicting the prediction probability of video stream jamming in a second preset time period according to the network fluctuation change rate in a first preset time period, and judging whether the prediction probability exceeds a preset threshold; if so, the set-top box outputs a second preset resolution, where the second preset resolution is less than the first preset resolution; if not, the set-top box maintains the output of the first preset resolution. At least the following beneficial effects are included. By determining the initial output resolution of the set-top box according to the historical resolution preference of the logged-in user and the target user information, this method can more accurately meet the viewing needs of users, thereby improving the viewing experience of users. When it is predicted that network fluctuations may cause video stream jamming, the set-top box automatically adjusts the output resolution to reduce the jamming phenomenon, further ensuring that users can smoothly watch video content. Brief Description of the Drawings

[0059] Figure 1 It is a flowchart of the set-top box resolution optimization method according to an embodiment of the present application.

[0060] The realization, functional features and advantages of the purpose of the present application will be further described with reference to the embodiments and the accompanying drawings. Detailed implementation manners

[0061] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without making creative efforts fall within the protection scope of the present invention.

[0062] It should be noted that all directional indications (such as up, down, left, right, front, back,...) in the embodiments of the present invention are only used to explain the relative positional relationship and movement conditions between components in a specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indications will also change accordingly.

[0063] In the present invention, unless otherwise clearly defined and limited, the terms "connection", "fixation", etc. should be understood in a broad sense. For example, "fixation" can be a fixed connection, a detachable connection, or integrated; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two components or the interaction relationship between two components, unless otherwise clearly limited. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0064] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present invention, the descriptions of "first", "second", etc. are only for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one such feature. In addition, the meaning of "and / or" appearing throughout the text includes three parallel solutions. Taking "A and / or B" as an example, it includes solution A, solution B, or a solution where A and B are satisfied simultaneously. In addition, the technical solutions between the embodiments can be combined with each other, but it must be based on the fact that those of ordinary skill in the art can implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the protection scope required by the present invention.

[0065] Embodiment 1

[0066] Referring to Figure 1 , the first embodiment of the present application provides a method for optimizing the resolution of a 4K ultra-high-definition intelligent set-top box, including the following operation steps:

[0067] S10. Obtain at least one logged-in user information and target user information, where the logged-in user information is configured as the home library member information of the set-top box, and the target user information is configured as the user information of the user watching TV at the current moment.

[0068] Optionally, when the set-top box system starts up or the user logs in, the set-top box obtains the account information of the logged-in user through the user authentication module. The set-top box retrieves the user information associated with the logged-in user account from the user database, including user ID, user name, user level, user subscription information, usage habits, etc. Determine the target user for the current operation or viewing. The target user can be obtained by collecting image information, or by using usage habits such as time periods.

[0069] S20. Obtain the historical resolution preferences of at least one logged-in user according to the at least one logged-in user information.

[0070] Optionally, the set-top box accesses the user historical viewing record database. According to the logged-in user ID, retrieve the resolution settings selected by the user when watching videos in the past or the resolution records automatically adjusted by the system. Analyze these data to determine the historical resolution preferences of the logged-in user, such as being more inclined to high definition, ultra high definition, or 4K resolution.

[0071] S30. Determine the first preset resolution output by the set-top box according to the historical resolution preference and the target user information.

[0072] Optionally, combine the historical resolution preference of the logged-in user and the information of the target user to determine an optimal first preset resolution. This resolution aims to provide the best viewing experience for the target user while considering device performance and bandwidth limitations.

[0073] S40. According to the network fluctuation change rate in the first preset time period, predict the prediction probability of video stream jamming in the second preset time period, and determine whether the prediction probability exceeds the preset threshold.

[0074] Optionally, the set-top box monitors the network status in real time and records the network fluctuation change rate in the first preset time period, such as in the past 5 minutes. Analyze the network fluctuation data to identify network anomalies or trends. Use machine learning models or statistical methods to predict the probability of video stream jamming in the second preset time period, such as in the next 1 minute, 3 minutes, or 4 minutes, according to the relationship between historical network fluctuation data and video stream jamming situations.

[0075] S50. If the prediction probability exceeds the preset threshold, the set-top box outputs the second preset resolution, where the second preset resolution is less than the first preset resolution.

[0076] Optionally, if the predicted probability exceeds a preset threshold, indicating that the video stream is likely to experience stuttering, the preset threshold can be dynamically adjusted according to the actual situation. The set-top box automatically adjusts the output resolution to a second preset resolution, which is lower than the first preset resolution, to reduce the demand for network bandwidth. The selection of the second preset resolution should ensure that the video remains clearly visible while minimizing the risk of stuttering.

[0077] The step that if the predicted probability exceeds the preset threshold, the set-top box outputs the second preset resolution, where the second preset resolution is less than the first preset resolution, includes:

[0078] S510. If the predicted probability exceeds the preset threshold, obtain the video type and video duration of the current video stream;

[0079] S520. According to the video type and video duration, obtain the requirements of the video stream for coherence and clarity;

[0080] S530. If the requirement of the video stream for coherence is higher than the requirement for clarity, the set-top box outputs the second preset resolution;

[0081] S540. If the requirement of the video type for coherence is lower than the requirement for clarity, the set-top box outputs the first preset resolution.

[0082] Optionally, if the predicted probability exceeds the preset threshold, obtain the type of the video stream currently played by the user and the duration of a single video stream, and obtain the requirements of the current video stream for video coherence or smoothness. By analyzing the type and duration of the video stream, obtain the coherence requirement of the video stream, aiming to analyze the user's risk acceptance ability when the video stream may experience stuttering when the predicted probability exceeds the preset threshold. When the duration of the video stream is long, the user's requirement for the coherence of the plot development and the coherence of the funny plots set in variety shows may be relatively high to improve the user's viewing experience; when the duration of the video stream is short, such as short videos, for teaching videos, etc., the requirement for the coherence of the plot is relatively low, or for this type of teaching video, the user's requirement for the picture quality is usually higher than the requirement for the coherence of the video stream. For example: if the currently played video stream is a movie, variety show, documentary, etc., which has a long duration and a high requirement for coherence, coherence is more important than clarity, then output the second resolution; if the current video stream is a short video, teaching video, etc., which has a short duration and a low requirement for coherence, or the picture quality is higher than the coherence requirement, then output the first preset resolution.

[0083] S60. If the predicted probability does not exceed the preset threshold, the set-top box maintains the output of the first preset resolution.

[0084] Optionally, if the predicted probability does not exceed a preset threshold, it indicates that the network condition is stable and the possibility of video stream stuttering is small, within an acceptable range. The set-top box continues to maintain the output of the first preset resolution to provide the best viewing experience. The system continuously monitors the network condition to adjust the resolution setting in a timely manner when necessary.

[0085] Embodiment 2

[0086] Based on Embodiment 1, this embodiment provides a method for optimizing the resolution of a 4K ultra-high-definition smart set-top box, including the following operation steps:

[0087] S10. Obtain at least one logged-in user information and target user information, where the logged-in user information is configured as the home library member information of the set-top box, and the target user information is configured as the user information of the user watching TV at the current moment.

[0088] S20. According to the at least one logged-in user information, obtain the historical resolution preferences of at least one logged-in user.

[0089] Optionally, the step of obtaining the historical resolution preferences of at least one logged-in user according to the at least one logged-in user information includes:

[0090] S201. According to the at least one logged-in user information, obtain the historical resolution usage records of each logged-in user.

[0091] Specifically, the set-top box system identifies and obtains all the user information of the currently logged-in users through the user authentication module, including user IDs, user names, etc. The system accesses the user behavior database or the viewing history database, which store various behavior data of users when using the set-top box in the past. According to the user ID of each logged-in user, the system retrieves and extracts the historical resolution usage records of that user. These records may include information such as the resolution selected by the user each time when watching a video, the resolution automatically adjusted by the system, and the timestamp of the resolution adjustment.

[0092] S202. According to the historical resolution usage records, obtain the type of video stream watched by each logged-in user, the corresponding time information, and the resolution of the video stream.

[0093] Specifically, the system analyzes the extracted historical resolution usage records and further refines the information in each record. For each record, the system identifies and extracts the type of video stream the user watched, such as movies, TV series, sports events, documentaries, etc., which helps analyze the user's resolution preference for different types of video content in the follow-up. The system also extracts the time information corresponding to each record, including the start time of viewing, the end time, and the time points of resolution adjustment. These time information help analyze the user's resolution usage habits at different time periods, such as weekdays, weekends, holidays, etc. The system extracts the actual resolution of the video stream in each record, which is the direct basis for analyzing the user's resolution preference.

[0094] S203. Obtain the historical resolution preference of at least one logged-in user according to the type to which the video stream belongs, the corresponding time information, and the resolution of the video stream.

[0095] Specifically, the system comprehensively analyzes the extracted video stream type, time information, and resolution data. By counting the frequency of resolutions selected by the user when watching different types of video content, the system can identify the user's resolution preference for specific types of video content. For example, the user may be more inclined to select high-definition or 4K resolution when watching movies, while selecting standard-definition resolution when watching news programs. The system can also analyze the user's resolution usage habits at different time periods to identify any possible time correlations. For example, the user may be more inclined to select high resolution to watch videos at night or on weekends.

[0096] Based on the above analysis, the system obtains the historical resolution preference of at least one logged-in user, and these preferences can be used to determine the output resolution of the set-top box in the follow-up to provide a better user experience.

[0097] S30. Determine the set-top box to output the first preset resolution according to the historical resolution preference and the target user information.

[0098] Optionally, the step of obtaining the first preset resolution output by the set-top box according to the historical resolution preference and the target user information includes:

[0099] S301. Judge the association relationship between the target user and at least one logged-in user, where all logged-in users belong to a home library.

[0100] Specifically, the set-top box system receives target user information, which is usually obtained through user login, device pairing, or other authentication methods, or the target user watching the video at the current moment can also be obtained through image acquisition. The system accesses the home library database, which stores all the login user information associated with the set-top box, including user ID, username, family member relationship, etc. The system compares the target user information with the user information in the home library to determine whether the target user is a member of the home library and their association with other users in the home library.

[0101] Optionally, the step of determining the association between the target user and at least one logged-in user includes:

[0102] S3011. When the set-top box receives a resolution setting request or the target user attempts to use the set-top box, the set-top box obtains at least one of the target user's input information, device pairing information, and biometric information.

[0103] Specifically, when the set-top box receives a resolution setting request, or the target user attempts to access or use the set-top box through a remote control, mobile phone APP, voice control, etc., the system starts the user authentication process. The set-top box displays an input interface, prompting the user to enter a username, password, PIN code, or other authentication information. The user provides this information through a remote control, mobile phone keyboard, or voice input, etc. If the set-top box supports the device pairing function, the system checks whether the target device (such as a mobile phone, tablet) has been paired with the set-top box. The pairing information may include the device ID, pairing code, or encrypted pairing credentials. If the set-top box is equipped with biometric technology, such as a fingerprint scanner or a facial recognition camera, the system obtains the biometric information of the target user, such as a fingerprint image or a facial image.

[0104] S3012. Compare at least one of the input information, device pairing information, and biometric information with the logged-in user information.

[0105] Specifically, the set-top box system compares the information such as the username and password entered by the user with the logged-in user information stored in the home library database to verify its correctness. If device pairing information is used, the system checks whether the ID or pairing code of the target device matches the records in the database to confirm the validity of the device pairing relationship. If biometric information is used, the system processes the obtained biometric information using a biometric algorithm and compares it with the biometric template in the database to verify the user's identity.

[0106] S3013. Determine whether the target user belongs to the logged-in user.

[0107] Specifically, if the input information, device pairing information, or biometric information of the target user exactly matches that of a certain logged-in user in the database, the system will determine that the target user belongs to that logged-in user and allow them to access or use the relevant functions of the set-top box. If the information of the target user does not match any logged-in user in the database, the system will determine that the target user does not belong to any logged-in user, that is, it is determined that the target user does not belong to the family library members.

[0108] S302. If the target user does not belong to the family library members, obtain the specification attributes of the TV, and the set-top box sets the highest resolution of the TV as the first resolution.

[0109] Specifically, if the system determines that the target user is not a family library member, then it cannot directly obtain the user's historical resolution preference. In this case, the system queries the specification attributes of the TV connected to the set-top box, which usually includes information such as the TV model and the maximum supported resolution. The system takes the highest resolution of the TV as the first preset resolution and configures the set-top box to output the video stream at this resolution.

[0110] S303. If exactly one of the target users belongs to the family library members, the set-top box sets the historical resolution preference of the corresponding logged-in user as the first preset resolution.

[0111] Specifically, if the system determines that only one of the target users is a family library member, then it directly obtains the user's historical resolution preference. The system determines the most frequently used or preferred resolution based on the user's historical resolution usage records. The set-top box sets this resolution as the first preset resolution and configures to output the video stream at this resolution.

[0112] S304. If at least two of the target users belong to the family library members, obtain all the logged-in user information in the family library.

[0113] Specifically, if the system determines that two or more of the target users are family library members, then it needs to further analyze to determine which user's resolution preference to use. The system accesses the family library database to obtain the user information of all family library members, including their historical resolution preferences, viewing habits, priorities, etc.

[0114] S305. According to the priorities of all the logged-in users in the family library, the set-top box sets the historical resolution preference of the logged-in user with the highest priority among the target users as the first preset resolution.

[0115] Specifically, the system sorts all members in the home library according to a preset priority rule or algorithm. These priorities may be based on the user's viewing frequency, the status of family members (such as parents, children), user-set preferences, etc. The system selects the user with the highest priority and obtains their historical resolution preference. The set-top box sets the resolution preference of this user as the first preset resolution and configures to output the video stream at this resolution. In this way, even if multiple home library members use the set-top box simultaneously, the system can select a suitable resolution according to the priority rule to meet the needs of most users.

[0116] S40. Predict the prediction probability of video stream jitter in the second preset time period according to the network fluctuation change rate in the first preset time period, and determine whether the prediction probability exceeds the preset threshold.

[0117] Optionally, the step of predicting the prediction probability of video stream jitter in the second preset time period according to the network fluctuation change rate in the first preset time period and determining whether the prediction probability exceeds the preset threshold includes:

[0118] S401. Divide the first preset time period into a first target time period and a second target time period.

[0119] Specifically, determine the total length of the first preset time period. According to the analysis needs, divide the first preset time period into two consecutive sub-time periods, namely the first target time period and the second target time period. The division method can be of equal length or determined according to changes in network traffic, user behavior patterns, or other relevant factors.

[0120] For example, if the first preset time period is 2 minutes, the first 1 minute can be selected as the first target time period and the second 1 minute as the second target time period.

[0121] S402. The network volatility of the first target time period is:

[0122]

[0123] Where N is the number of data points in the first target time period. The data points can be network performance metrics recorded every certain time, such as every 10s or 20s; V i is the network performance metric of the i-th data point, and V i+1 is the network performance metric of the (i + 1)-th data point; the network performance metrics include bandwidth, latency, packet loss rate, etc. The absolute value part in the formula represents the relative change rate of the network performance metric between two adjacent data points. Calculate the relative change rate for all adjacent data points in the entire first target time period and find its average value to obtain the network volatility of the first target time period.

[0124] S403. The network volatility of the second target time period is as follows:

[0125]

[0126] where M is the number of data points within the second target time period, V j is the network performance metric of the j-th data point, and V j+1 is the network performance metric of the (j + 1)-th data point. The calculation principles of the network volatility for the first target time period and the second target time period are the same.

[0127] Optionally, the formula for the network performance metric is as follows:

[0128] V = ω 1 × B′ + ω 2 × (1 - D′) + ω 3 × (1 - Q′)

[0129] where B′ represents the standardized value of the bandwidth utilization rate, and ω 1 represents the weight coefficient of the bandwidth, D′ represents the standardized value of the latency, and ω 2 represents the weight coefficient of the latency, Q′ represents the standardized value of the packet loss rate, and ω 3 represents the weight coefficient of the packet loss rate.

[0130] Specifically, the bandwidth utilization rate is an indicator for measuring the network bandwidth usage efficiency, usually expressed as the ratio of the actually used bandwidth to the total bandwidth. The standardized value B′ is obtained by converting the original bandwidth utilization rate data through a certain standardization method (such as min-max standardization, Z-score standardization, etc.) to a unified standard range (such as between 0 and 1) to facilitate comparison and weighting between different metrics. Standardization can eliminate the influence of different metric dimensions, making the network performance metric more objective and accurate. The weight coefficient ω 1 represents the importance degree of the bandwidth utilization rate in the network performance metric.

[0131] Latency is the time required for a data packet to travel from the sender to the receiver and is an important indicator for measuring the network response speed. The standardized value D′ is obtained by converting the original latency data through a standardization method to a unified standard range. In the formula, 1 - D′ is used to represent the negative impact of latency on network performance, that is, the smaller the latency, the better the network performance. The weight coefficient ω 2 represents the importance degree of latency in the network performance metric. According to the sensitivity of the network application to latency, the value of ω 2 can be adjusted to reflect the impact of latency on network performance.

[0132] The packet loss rate is the ratio of the number of lost data packets to the total number of data packets during data transmission, and it is an important indicator to measure network reliability. The standardized value Q′ is obtained by converting the original packet loss rate data into a unified standard range through a standardization method. In the formula, 1 - Q′ is used to represent the negative impact of the packet loss rate on network performance, that is, the smaller the packet loss rate, the better the network performance. The weight coefficient ω 3 represents the importance of the packet loss rate among network performance indicators. According to the tolerance of network applications to the packet loss rate, ω 3 can be adjusted to reflect the impact of the packet loss rate on network performance.

[0133] S404. The calculation formula for the network fluctuation change rate is as follows:

[0134] The above formula represents the change rate of the network volatility in the second target time period relative to the first target time period. If ΔL is positive, it means the network volatility is increasing; if ΔL is negative, it means the network volatility is decreasing.

[0135] To facilitate the understanding of the steps of S401 - S404, an example is given as follows:

[0136] S401. Time period division:

[0137] The first preset time period: Select the past 2 minutes as the analysis object. Divide these 2 minutes equally into two 1 - minute time periods, namely the first target time period (0 - 60 seconds) and the second target time period (61 - 120 seconds).

[0138] S402 & S403. Network volatility calculation:

[0139] Data point recording: Within each target time period, record the network performance indicators every 10 seconds, including bandwidth utilization, latency, and packet loss rate.

[0140] Suppose the data is as follows (for example only, the actual data will vary according to network conditions):

[0141] The first target time period: Bandwidth utilization [0.4, 0.45, 0.5, 0.55, 0.6, 0.65], latency (assume it has been standardized) [0.5, 0.48, 0.52, 0.47, 0.51, 0.49], packet loss rate (assume it has been standardized) [0.01, 0.02, 0.015, 0.01, 0.025, 0.02].

[0142] Second target time period: Bandwidth utilization rate [0.6, 0.62, 0.65, 0.7, 0.75, 0.8], latency (assuming standardized) [0.55, 0.53, 0.57, 0.52, 0.56, 0.54], packet loss rate (assuming standardized) [0.02, 0.015, 0.025, 0.03, 0.035, 0.03].

[0143] Network performance metric calculation:

[0144] Normalize the bandwidth utilization rate, latency, and packet loss rate for each data point (e.g., using min-max normalization), and calculate the network performance metric V according to the weight coefficients (assuming ω 1 = 0.5, ω 2 = 0.3, ω 3 = 0.2).

[0145] Network volatility calculation:

[0146] Within each target time period, calculate the relative change rate of the network performance metric between adjacent data points and find its average value.

[0147] Substitute the data of the first target time period into the formula of the network performance metric to obtain the corresponding network performance metrics within the first target time period as: 0.548, 0.577, 0.591, 0.632, 0.642, 0.674.

[0148] Similarly, obtain the corresponding network performance metrics within the first target time period as: 0.631, 0.648, 0.649, 0.688, 0.700, 0.732.

[0149] Substitute into the network volatility calculation formulas for the first target time period and the second target time period respectively. Calculate the network volatility of the first target time period as 0.0424, and the network volatility of the second target time period as 0.0304. Substitute into the network volatility change rate formula to calculate the network volatility change rate as 28.3%.

[0150] Optionally, the step of predicting the probability of video stream stuttering in the second preset time period according to the network volatility change rate of the first preset time period and determining whether the predicted probability exceeds the preset threshold further includes:

[0151] S411. Obtain the required bandwidth amount and cache interval size of the video stream within the second preset time period.

[0152] Specifically, the required bandwidth amount refers to the network bandwidth amount required for the video stream to maintain smooth playback within a second preset time period. It can be estimated by analyzing parameters such as the coding rate, resolution, and frame rate of the video stream. The cache interval size refers to the amount of video data that the video player pre-downloads and stores to cope with network fluctuations. The size of the cache interval affects the anti-freezing ability of the video stream during network fluctuations.

[0153] S412. Predict the prediction probability of video stream freezing within the second preset time period according to the network fluctuation change rate, the required bandwidth amount, and the cache interval size.

[0154] Optionally, the calculation formula for the prediction probability is as follows:

[0155]

[0156] Wherein, ΔL represents the network fluctuation change rate, which is actually the network fluctuation change rate in the first preset time period. The larger the network fluctuation change rate, the more unstable the network is, and the higher the probability of video stream freezing may be; F represents the required bandwidth amount. The larger the required bandwidth amount, the higher the requirement for the network, and the greater the probability of video stream freezing, especially in the case of limited network bandwidth; G represents the cache interval size. The larger the cache interval, the longer the video player can maintain playback during network fluctuations, thereby reducing the probability of freezing; β 0 is the intercept term, indicating the baseline value of the prediction probability when all input variables are 0; β 1 is the coefficient of the network fluctuation change rate, β 2 is the coefficient of the required bandwidth amount, β 3 is the coefficient of the cache area size. β 0 β 1 β 2 and β 3 are the parameters of the model and are obtained through training data.

[0157] For easy understanding, the determination process examples of β 0 β 1 β 2 β 3 are as follows:

[0158] First of all, a set of training data needs to be prepared. This set of data should contain multiple observations, and each observation has a set of input variables (in this example, the network fluctuation change rate ΔL, the required bandwidth amount F, and the cache interval size G) and a corresponding output variable (that is, whether the video stream freezes, which can be a binary classification variable, such as 0 indicating no freezing and 1 indicating freezing).

[0159] Select the logistic regression model as the prediction model. The logistic regression model is suitable for binary classification problems and can output a probability value between 0 and 1, representing the probability of belonging to a certain class.

[0160] Before starting the training, it is necessary to initialize the model coefficients β 0 、β 1 、β 2 and β 3 . These coefficients can be randomly initialized or set initial values using some prior knowledge.

[0161] For each observation in the training data, calculate the predicted probability using the current coefficient values. To evaluate the prediction effect of the model, a loss function needs to be calculated. In logistic regression, the commonly used loss function is the logarithmic loss function (LogLoss). The logarithmic loss function measures the difference between the model's predicted probability and the actual label.

[0162] To find the coefficient values that minimize the loss function, use the gradient descent algorithm for optimization. The gradient descent algorithm iteratively updates the coefficient values to reduce the loss function. In each iteration, calculate the gradient of the loss function with respect to each coefficient and update the coefficient values along the opposite direction of the gradient.

[0163] Repeat the gradient descent step until the convergence condition is met. The convergence condition can be that the value of the loss function reaches a certain threshold or the update amount of the coefficient values is less than a certain threshold.

[0164] When the model converges, obtain the final coefficient values β 0 、β 1 、β 2 and β 3 . These coefficient values reflect the degree and direction of the influence of the input variables on the output variables.

[0165] S50. If the predicted probability exceeds the preset threshold, the set-top box outputs the second preset resolution, where the second preset resolution is less than the first preset resolution.

[0166] Optionally, the step of the set-top box outputting the second preset resolution when the predicted probability exceeds the preset threshold, where the second preset resolution is less than the first preset resolution, includes:

[0167] S501. The set-top box outputs the second preset resolution and sends a notification message to the target user, and determines whether a cancellation instruction from the target user is received within the preset confirmation period.

[0168] Specifically, when exceeding a preset threshold, the set-top box automatically adjusts its output resolution to a second preset resolution. The second preset resolution is lower than the first preset resolution to reduce the demand of the video stream for network bandwidth, thereby reducing the occurrence of lag.

[0169] While adjusting the resolution, the set-top box sends a notification message to the target user. The notification message can be sent through the display interface of the set-top box, remote control prompts, mobile APP push, etc. The notification message should include the reason for the resolution adjustment (such as lag prediction caused by network fluctuations), the adjusted resolution value, and the method or instruction to cancel the adjustment. The set-top box sets a preset confirmation period for waiting for the response of the target user. During the preset confirmation period, the set-top box continuously monitors whether it receives a cancellation instruction sent by the target user.

[0170] S502: If no cancellation instruction from the target user is received within the preset confirmation period, the set-top box maintains the output of the second preset resolution.

[0171] Specifically, if the set-top box does not receive a cancellation instruction sent by the target user within the preset confirmation period, it is considered that the target user accepts the current resolution adjustment or has not noticed the notification message. The set-top box continues to output the second preset resolution until the network condition improves or the target user actively makes an adjustment.

[0172] S503: If a cancellation instruction from the target user is received within the preset confirmation period, the set-top box returns to output the first preset resolution.

[0173] Specifically, if the set-top box receives a cancellation instruction sent by the target user within the preset confirmation period, it is considered that the target user does not wish to continue maintaining the current resolution adjustment and hopes to restore to the original resolution. The set-top box immediately adjusts its output resolution to the first preset resolution. Ensure that the video stream can be presented to the target user in the best quality, while continuing to monitor network fluctuations for future adjustments if necessary.

[0174] By first reducing the video stream resolution and then confirming with the user, the situation where the second preset time period approaches the first preset time period is reduced, so that during the confirmation process by the user, video stream lag is caused.

[0175] S60: If the prediction probability does not exceed the preset threshold, the set-top box maintains the output of the first preset resolution.

[0176] Embodiment 3

[0177] Based on Embodiment 1, this embodiment provides a 4K ultra-high-definition smart set-top box resolution optimization system, including:

[0178] An information acquisition module, which is configured to acquire at least one logged-in user information and target user information, where the logged-in user information is configured as the family library member information of the set-top box, and the target user information is configured as the user information of the user watching TV at the current moment;

[0179] A preference acquisition module, which is configured to acquire the historical resolution preferences of at least one logged-in user according to the at least one logged-in user information;

[0180] A first preset resolution acquisition module, which is configured to determine that the set-top box outputs a first preset resolution according to the historical resolution preference and the target user information;

[0181] A prediction probability judgment module, which is configured to predict the prediction probability of video stream stuttering in a second preset time period according to the network fluctuation change rate in a first preset time period, and judge whether the prediction probability exceeds a preset threshold;

[0182] A first execution module, which is configured to, if so, the set-top box outputs a second preset resolution, where the second preset resolution is less than the first preset resolution;

[0183] A second execution module, which is configured to, if not, the set-top box maintains the output of the first preset resolution.

[0184] The above are only the preferred embodiments of the present application, and do not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present application, or directly or indirectly applied in other related technical fields, shall be included in the patent protection scope of the present application by the same token.

Claims

1. A 4K ultra-high-definition smart set-top box resolution optimization method, characterized in that: include: Acquire at least one logged-in user information and target user information, wherein the logged-in user information is configured as family library member information of a set-top box, and the target user information is configured as user information of a user currently watching TV; Acquire a historical resolution preference of at least one logged-in user according to the at least one logged-in user information; Determining a first preset resolution output by a set-top box according to the historical resolution preference and the target user information; Predicting the probability of video streaming freezes within a second preset time period based on the network fluctuation change rate within the first preset time period, and determining whether the predicted probability exceeds a preset threshold; If yes, the set-top box outputs a second preset resolution, wherein the second preset resolution is smaller than the first preset resolution; If not, the set-top box maintains outputting the first preset resolution.

2. The 4K ultra-high-definition smart set-top box resolution optimization method according to claim 1, characterized in that: The step of acquiring the historical resolution preference of at least one logged-in user information according to the at least one logged-in user information comprises: According to the at least one logged-in user information, obtaining a historical resolution usage record of each logged-in user; According to the historical resolution usage record, the type of video stream watched by each logged-in user, the corresponding time information and the resolution of the video stream are obtained; According to the type of the video stream, the corresponding time information, and the resolution of the video stream, a historical resolution preference of at least one logged-in user is obtained.

3. The 4K ultra-high-definition smart set-top box resolution optimization method according to claim 1, characterized in that: The step of acquiring the first preset resolution output by the set-top box according to the historical resolution preference and the target user information comprises: Determining an association relationship between the target user and at least one logged-in user, wherein all logged-in users belong to one family library; If the target user is not a member of the family library, the specification attributes of the TV are obtained, and the set-top box sets the highest resolution of the TV to the first resolution; If there is only one target user who is a member of the family library, the set-top box sets the historical resolution preference of the corresponding logged-in user to the first preset resolution; If at least two of the target users are members of the family library, then obtain information of all logged-in users in the family library; According to the priorities of all logged-in users in the family library, the set-top box sets the historical resolution preference of the logged-in user with the highest priority among the target users as the first preset resolution.

4. The 4K ultra-high-definition smart set-top box resolution optimization method according to claim 3, characterized in that: The step of determining the association relationship between the target user and at least one logged-in user includes: When the set-top box receives a resolution setting request or the target user attempts to use the set-top box, the set-top box obtains at least one of input information, device pairing information, and biometric information of the target user; comparing at least one of the input information, the device pairing information, and the biometric information with the logged-in user information; Determine whether the target user is a logged-in user.

5. The 4K ultra-high-definition smart set-top box resolution optimization method according to claim 1, characterized in that: The step of predicting the probability of video stream freeze in the second preset time period according to the network fluctuation change rate in the first preset time period, and judging whether the predicted probability exceeds a preset threshold, comprises: Dividing the first preset time period into a first target time period and a second target time period; The network volatility of the first target time period is: Where N is the number of data points in the first target time period, V i is the network performance index of the ith data point, V i+1 is the network performance indicator of the i+1th data point; The network volatility of the second target time period is: Where M is the number of data points in the second target time period, V j is the network performance index of the jth data point, V j+1 is the network performance indicator of the j+1th data point; The calculation formula of the network fluctuation change rate is as follows:

6. The 4K ultra-high-definition smart set-top box resolution optimization method according to claim 5, characterized in that: The formula of the network performance indicator is as follows: V=ω1×B′+ω2×(1-D′)+ω3×(1-Q′) Among them, B′ represents the normalized value of bandwidth utilization, ω1 represents the weight coefficient of bandwidth, D′ represents the normalized value of delay, ω2 represents the weight coefficient of delay, Q′ represents the normalized value of packet loss rate, and ω3 represents the weight coefficient of packet loss rate.

7. The 4K ultra-high-definition smart set-top box resolution optimization method according to claim 1, characterized in that: The step of predicting the probability of video stream freeze in the second preset time period according to the network fluctuation change rate in the first preset time period, and judging whether the predicted probability exceeds a preset threshold, comprises: Obtaining the required bandwidth and cache interval size of the video stream within a second preset time period; The predicted probability of video streaming freeze within the second preset time period is predicted based on the network fluctuation change rate, the required bandwidth and the cache interval size.

8. The 4K ultra-high-definition smart set-top box resolution optimization method according to claim 7, characterized in that: The step of predicting the probability of video stream freeze in the second preset time period according to the network fluctuation change rate in the first preset time period, and judging whether the predicted probability exceeds a preset threshold, comprises: The formula for calculating the predicted probability is as follows: Wherein, ΔL represents the network fluctuation rate, F represents the required bandwidth, G represents the cache interval size, β0 is the intercept term, β1 is the coefficient of the network fluctuation rate, β2 is the coefficient of the required bandwidth, and β3 is the coefficient of the cache area size.

9. The 4K ultra-high-definition smart set-top box resolution optimization method according to claim 1, characterized in that: If so, the set-top box outputs a second preset resolution, wherein the second preset resolution is smaller than the first preset resolution, comprising: The set-top box outputs the second preset resolution and sends a notification message to the target user to determine whether a cancellation instruction from the target user is received within a preset confirmation time period; If not, the set-top box maintains outputting the second preset resolution; If yes, the set-top box returns to output the first preset resolution.

10. A 4K ultra-high-definition smart set-top box resolution optimization system, characterized in that: include: An information acquisition module, configured to acquire at least one logged-in user information and target user information, wherein the logged-in user information is configured as family library member information of a set-top box, and the target user information is configured as user information of a user currently watching TV; A preference acquisition module, configured to acquire a historical resolution preference of at least one logged-in user according to the at least one logged-in user information; A first preset resolution acquisition module, configured to determine a first preset resolution output by a set-top box according to the historical resolution preference and target user information; A prediction probability judgment module, which is configured to predict the prediction probability of video stream freeze in a second preset time period according to the network fluctuation change rate in the first preset time period, and determine whether the prediction probability exceeds a preset threshold; A first execution module, configured to, if yes, cause the set-top box to output a second preset resolution, wherein the second preset resolution is smaller than the first preset resolution; The second execution module is configured to, if not, the set-top box maintains outputting the first preset resolution.

Images