Dynamic power management system for low power wapi connections

Abstract

The present invention specifically relates to a dynamic power management system for low-power WAPI connection, comprising: a connection management unit, which is used to first start the wireless network interface when a mobile terminal attempts to access a WAPI network, and set the transmission power of a radio frequency module to a high-power working state to quickly complete the discovery and scanning of the network; an authentication management unit, which is used to configure a processor or an encryption engine to a high-frequency working state to quickly complete the authentication when performing X.509 certificate interaction and key negotiation when performing WAPI authentication, and if there are multiple authentication gaps, set the authentication gap period to configure the processor or encryption engine to a non-high-frequency working state, or enter a periodic dormant state to reduce power consumption; a data transmission management unit, which is used to dynamically adjust the power of the radio frequency module according to the real-time data transmission rate and flow demand after completing the WAPI network connection and authentication.

Description

Technical Field

[0001] The invention belongs to the field of power supply control, and in particular relates to a dynamic power supply management system for low-power WAPI connection. Background Art

[0002] When a mobile terminal uses WAPI (Wireless LAN Authentication and Privacy Infrastructure) to connect, compared with other wireless network protocols, it may cause additional energy consumption due to the complexity of its security mechanism. In order to reduce the power consumption problem caused by WAPI connection, the existing technology usually adopts the following strategies:

[0003] Optimize protocol processing: By independently optimizing the WAPI protocol stack, unnecessary data packet processing and protocol interaction can be reduced. For example, the implementation of encryption algorithms can be optimized to make them more efficient and reduce the load on the CPU and wireless chips. However, this control effect at the software level is relatively poor.

[0004] Hardware acceleration: Use a dedicated hardware encryption engine to process encryption algorithms such as AES to reduce the CPU burden. However, this only reduces the CPU burden and adds encryption design, which still consumes a lot of energy.

[0005] Energy-saving design: Low power consumption factors are taken into consideration during the design phase of mobile terminals, such as optimizing the RF circuit design and designing a dynamic cache module to reduce unnecessary read and write operations. The effect of optimizing the RF circuit alone is not obvious. Summary of the invention

[0006] The object of the present invention is to provide a dynamic power management system for low-power WAPI connections to solve the problems raised in the above-mentioned background technology.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0008] Dynamic power management system for low-power WAPI connections, including

[0009] The connection management unit is used to start the wireless network interface first and set the transmission power of the radio frequency module to a high-power working state to quickly complete the discovery and scanning of the network when the mobile terminal attempts to access the WAPI network;

[0010] And used to set the transmission power of the radio frequency module to a non-high-power working state, or enter a periodic dormant state to reduce power consumption if the response time exceeds a threshold while waiting for an access request or an authentication response;

[0011] The authentication management unit is used to configure the processor or encryption engine to a high-frequency working state to quickly complete the authentication when performing the interaction of X.509 certificates and key negotiation during WAPI authentication. If there are multiple authentication intervals, the authentication interval is set to configure the processor or encryption engine to a non-high-frequency working state, or enter a periodic dormant state to reduce power consumption;

[0012] The data transmission management unit is used to dynamically adjust the power of the radio frequency module according to the real-time data transmission rate and traffic requirements after completing the WAPI network connection and authentication.

[0013] Furthermore, the RF module includes a RF front end and an antenna.

[0014] Furthermore, the connection management unit is integrated into the wireless network interface controller chip.

[0015] Furthermore, the authentication management unit is integrated into the encryption coprocessor inside the CPU or uses an independent security chip.

[0016] Furthermore, the power of the RF module is dynamically adjusted according to the real-time data transmission rate and flow requirements. Specifically:

[0017] The data transmission management unit obtains the current data transmission rate, received signal strength indicator, and signal-to-noise ratio through the wireless network interface controller of the mobile terminal; monitors the current data transmission rate, received signal strength indicator, and signal-to-noise ratio in real time; monitors the network activities of the applications running on the terminal, and if the traffic of a certain application is increasing, it can be determined that the user's current traffic demand is increasing;

[0018] When the received signal strength indicator and the signal-to-noise ratio are both greater than the threshold and it is determined that the user's current traffic demand is not increasing, the radio frequency power is reduced; otherwise, the radio frequency power is maintained or increased.

[0019] Furthermore, the power of the RF module is dynamically adjusted according to the real-time data transmission rate and flow requirements. Specifically:

[0020] The data transmission management unit obtains the current data transmission rate, received signal strength indicator, and signal-to-noise ratio through the wireless network interface controller of the mobile terminal; monitors the current data transmission rate, received signal strength indicator, and signal-to-noise ratio in real time; monitors the network activities of multiple applications running on the terminal, and predicts the user's future traffic demand based on their correlation. If the traffic of a certain application is increasing, it can be determined that the user's current traffic demand is increasing;

[0021] When the received signal strength indicator and the signal-to-noise ratio are both greater than the threshold and it is determined that the user's current traffic demand is not increasing, the radio frequency power is reduced; otherwise, the radio frequency power is maintained or increased.

[0022] Furthermore, the network activities of multiple applications running on the terminal are monitored, and the user's future traffic demand is predicted based on their correlation. Steps:

[0023] Obtain real-time network activity data of each application on the terminal through the operating system API, including each application's ID, data transmission rate, cumulative traffic, start and end timestamps, packet size, and connection status;

[0024] Organize the collected multi-dimensional data to construct a multi-dimensional data cube. Each dimension represents the time, application ID, traffic rate, and packet size attributes, and each cell of the cube records the specific traffic data under the corresponding dimension.

[0025] Preprocess the data; create network behavior characteristics that represent the application, including average traffic, maximum traffic, minimum traffic, traffic fluctuation coefficient, and traffic growth rate in adjacent time periods;

[0026] Construct cross-features to mine the interactive effects between applications; extract potential common factors from multidimensional data cubes through principal component analysis (PCA), which can capture the common change trends and patterns between application traffic;

[0027] Apply Apriori association rule mining algorithm to mine association rules between application traffic;

[0028] According to the extracted common factors and association rules, a prediction model is constructed.

[0029] The historical data is divided into a training set and a validation set. The model is trained using the training set to learn how traffic demand evolves over time and the coupling effects between different applications, and then predict users' future traffic demand.

[0030] Furthermore, the specific process of extracting potential common factors from multidimensional data cubes through principal component analysis is as follows:

[0031] Convert the multidimensional data cube into a numerical matrix, in which rows represent different observation samples and columns represent different features. Calculate the covariance matrix of the data matrix, perform eigenvalue decomposition on the covariance matrix, and obtain a set of eigenvalues ​​and corresponding eigenvectors. Sort the eigenvalues ​​by size, and select the eigenvectors corresponding to the first k largest eigenvalues. These eigenvectors constitute a new basis. Project the original data into the new space formed by these k eigenvectors to obtain a new data matrix in which each column represents a principal component, the so-called potential common factor.

[0032] Furthermore, the specific process of applying the Apriori association rule mining algorithm to mine association rules between application traffic is as follows:

[0033] Treat application traffic activities occurring simultaneously within a time window as one transaction;

[0034] Initialize the minimum support threshold;

[0035] Traverse the transaction data set once, count the support of each application, and only keep those application item sets that meet the minimum support requirement;

[0036] Next, all possible frequent k-item sets, i.e., combinations of two or more applications, are generated, and the transaction dataset is traversed again to calculate the support of each candidate k-item set;

[0037] Based on the support threshold, remove the candidate k-item sets that do not meet the conditions and retain the frequent k-item sets;

[0038] Generate candidate sets of itemsets based on frequent k-itemsets, but only retain those candidate sets whose subsets have appeared in the frequent k-itemsets to avoid invalid traversal;

[0039] Recursive execution: The above process is iterated continuously until no more frequent item sets can be found;

[0040] Generate association rules from the frequent itemsets found. The form of the association rule is "if applications A and B are active at the same time, then application C is also active";

[0041] A minimum confidence threshold is set for each association rule; only those rules that meet the minimum confidence requirement are retained.

[0042] Beneficial effect: Compared with the prior art, the technical solution of the present application not only optimizes the power management during the WAPI connection process from the hardware level, but also uses advanced data analysis methods and prediction models at the software level to realize a more comprehensive and efficient low-power WAPI connection dynamic power management system, thereby significantly improving the overall energy consumption performance of mobile terminals when using WAPI connections. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 This is a connection block diagram of the low-power WAPI connected dynamic power management system used in this application. DETAILED DESCRIPTION

[0044] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0045] The present application discloses a dynamic power management system for low-power WAPI connections, such as Figure 1 As shown, including:

[0046] A connection management unit, configured to, when the mobile terminal attempts to access the WAPI network, first start the wireless network interface and set the transmit power of the radio frequency (RF) module to a high-power working state to quickly complete the discovery and scanning of the network; and to, during the waiting period for the access request or authentication response, if the response time exceeds a threshold, set the transmit power of the radio frequency (RF) module to a non-high-power working state, or enter a periodic dormant state to reduce power consumption;

[0047] In mobile terminals (such as smartphones, tablets, etc.), the radio frequency (RF) module that implements WAPI (Wireless LAN Authentication and Privacy Infrastructure) connection is one of the hardware components responsible for wireless communication and is also the main power-consuming hardware. It includes the following parts:

[0048] RF Front-End:

[0049] Includes a power amplifier (PA) to increase the power of the transmitted signal so that it can travel a long enough distance in the air;

[0050] Low Noise Amplifier (LNA), which amplifies weak wireless signals while keeping the noise level as low as possible when receiving signals;

[0051] Filters and duplexers are used to distinguish and filter transmit and receive signals to ensure that they do not interfere with each other in the same frequency band;

[0052] Tuning switch and antenna switch to control the path and direction of the signal;

[0053] RF mixer and local oscillator (LO) are used to convert the frequency of the signal to achieve up-conversion and down-conversion of the signal.

[0054] antenna:

[0055] The device for transmitting and receiving wireless signals is an important component of the RF module and is usually designed as a built-in antenna or an external antenna.

[0056] Radio Frequency Integrated Circuit (RFIC):

[0057] Contains the above front-end devices and other RF functional blocks, such as synthesizers and voltage-controlled oscillators (VCOs), which are integrated in one package to provide complete RF signal processing capabilities.

[0058] The authentication management unit is used to configure the processor or encryption engine to a high-frequency working state to quickly complete the authentication when performing the interaction of X.509 certificates and key negotiation during WAPI authentication. If there are multiple authentication intervals, the authentication interval is set to configure the processor or encryption engine to a non-high-frequency working state, or enter a periodic dormant state to reduce power consumption;

[0059] When actually deploying and implementing the WAPI security protocol, encryption operations can be completed through a hardware encryption engine. In order to improve encryption speed and performance, especially for real-time encrypted transmission of large amounts of data, modern network devices use hardware encryption engines, such as encryption coprocessors integrated in the CPU or dedicated encryption chips. This enables more efficient execution of encryption algorithms required in the WAPI protocol, such as the SM4 block cipher algorithm. Therefore, there is an OR relationship between the processor and the encryption engine in the configuration, that is, if the encryption engine is integrated in the CPU, then only the processor needs to be configured, otherwise a dedicated encryption chip needs to be configured and the processor does not need to be configured;

[0060] A data transmission management unit, which is used to dynamically adjust the power of the radio frequency (RF) module according to the real-time data transmission rate and flow demand after completing the WAPI network connection and authentication;

[0061] Dynamically adjust the power of the RF module according to the real-time data transmission rate and traffic requirements. Specifically:

[0062] The data transmission management unit obtains the current data transmission rate, received signal strength indication (RSSI), and signal-to-noise ratio (SINR) through the wireless network interface controller (WNIC) of the mobile terminal; monitors the current data transmission rate, received signal strength indication (RSSI), and signal-to-noise ratio (SINR) in real time; monitors the network activities of applications running on the terminal, and if the traffic of a certain application is increasing, it can be determined that the user's current traffic demand is increasing;

[0063] When the received signal strength indication (RSSI) and the signal-to-noise ratio (SINR) are both greater than the threshold and it is determined that the user's current traffic demand is not increasing, the radio frequency power is reduced; otherwise, the radio frequency power is maintained or increased.

[0064] Hardware components and functions of dynamic power management system for low power WAPI connection:

[0065] 1. The connection management unit is integrated into the wireless network interface controller (WNIC) chip, such as Wi-Fi SoC chips from Broadcom, Qualcomm or Intel, and has a programmable power management module that can adjust the transmission power of the radio frequency (RF) module according to software instructions;

[0066] 2. The authentication management unit is integrated into the encryption coprocessor inside the CPU or an independent security chip (such as TrustZone technology, dedicated encryption engine) to execute. For example, the ARM TrustZone CryptoCell series.

[0067] 3. The function of the data transmission management unit is dynamic power adjustment, which is also integrated in the WNIC chip and can perform intelligent adjustment based on real-time feedback information (such as RSSI, SINR, data transmission rate, etc.).

[0068] In addition to the above chips that specifically handle wireless communication and encryption functions, there is also an important component responsible for coordinating and controlling the power distribution and state transition of the overall system, which is the power management integrated circuit (PMIC). It can be a separate chip or integrated into SoC, such as Qualcomm Snapdragon series, Apple A series, Huawei Kirin series, etc.

[0069] Independent PMIC chips, such as various power tube solutions produced by Dialog Semiconductor, TI (Texas Instruments) or Maxim Integrated.

[0070] Dynamically adjust the power of the RF module according to the real-time data transmission rate and traffic requirements. Specifically:

[0071] The data transmission management unit obtains the current data transmission rate, received signal strength indication (RSSI), and signal-to-noise ratio (SINR) through the wireless network interface controller (WNIC) of the mobile terminal; monitors the current data transmission rate, received signal strength indication (RSSI), and signal-to-noise ratio (SINR) in real time; monitors the network activities of multiple applications running on the terminal, and predicts the user's future traffic demand based on their correlation. If the traffic of a certain application is increasing, it can be determined that the user's current traffic demand is increasing;

[0072] When the received signal strength indicator (RSSI) and the signal-to-noise ratio (SINR) are both greater than the threshold and it is determined that the user's current traffic demand is not increasing, the radio frequency power is reduced; otherwise, the radio frequency power is maintained or increased;

[0073] Monitor the network activities of multiple applications running on the terminal and predict the user's future traffic demand based on their correlation. Follow the detailed steps below:

[0074] 1. Data collection:

[0075] Obtain real-time network activity data of each application on the terminal through the operating system API, including each application's ID, data transmission rate, cumulative traffic, start and end timestamps, packet size, and connection status;

[0076] 2. Construct a multidimensional data cube:

[0077] Organize the collected multi-dimensional data to construct a multi-dimensional data cube (OLAP Cube), where each dimension represents the time, application ID, traffic rate, and packet size attributes, and each cell of the cube records the specific traffic data under the corresponding dimension;

[0078] 3. Data preprocessing:

[0079] Clean the data, eliminate outliers and noise, fill in missing values, and normalize or standardize the data;

[0080] 4. Feature Engineering:

[0081] Create network behavior characteristics that represent the application, including average traffic, maximum traffic, minimum traffic, traffic fluctuation coefficient, and traffic growth rate in adjacent time periods;

[0082] Construct cross-features to mine the interaction effects between applications, such as the combination of applications that are active at the same time;

[0083] 5. Common factor analysis:

[0084] Principal component analysis (PCA) is used to extract potential common factors from multidimensional data cubes. These factors can capture the common trends and patterns of application traffic. Principal component analysis (PCA) is a commonly used data analysis and preprocessing technology, mainly used for dimensionality reduction of high-dimensional data. PCA transforms the original data into a new coordinate system through linear transformation. The axes of the new coordinate system are called "principal components". These principal components are orthogonal bases sorted from large to small according to the variance of the data contained. The specific process of extracting potential common factors from multidimensional data cubes through principal component analysis (PCA) is as follows:

[0085] The multidimensional data cube is transformed into a numerical matrix, where the rows represent different observation samples (such as traffic data in each time window or period) and the columns represent different features (such as traffic rates and packet sizes of different applications).

[0086] Calculate the covariance matrix of the data matrix. The covariance matrix can reflect the linear relationship between each feature. For high-dimensional data, the covariance matrix can help understand whether there is covariance or correlation between each feature.

[0087] Perform eigenvalue decomposition on the covariance matrix to obtain a set of eigenvalues ​​and corresponding eigenvectors. The eigenvalues ​​reflect the size of the original data variance contained in the newly generated principal components, and the eigenvectors indicate the direction of the new coordinate system.

[0088] According to the order of eigenvalues, the eigenvectors corresponding to the first k largest eigenvalues ​​are selected, and these eigenvectors form a new basis. The choice of k is generally determined by the proportion of explained data variance (such as retaining more than 90% of the variance) or actual needs.

[0089] The original data is projected into the new space formed by these k eigenvectors to obtain a new data matrix. Each column represents a principal component, the so-called latent common factor. These principal components are linear combinations of the original features, which retain most of the variation information of the data and reduce the dimension of the data.

[0090] Analyze the loads of the principal components (i.e., the eigenvector elements) to understand which original features play an important role in constructing each principal component. This helps to discover common trends and patterns between traffic data of different applications.

[0091] Through principal component analysis, complex multidimensional data cubes can be simplified into a few principal components. These principal components can not only better describe the main change trends of the original multidimensional data, but can also be used to build prediction models to predict users' future traffic needs.

[0092] 6. Association rule mining:

[0093] The Apriori association rule mining algorithm is used to mine association rules between application traffic, revealing which applications' network activities will occur at the same time and their joint impact on the overall traffic demand; the Apriori association rule mining algorithm is a classic algorithm widely used in the field of data mining, mainly used to discover frequent item sets in large-scale data sets and strong association rules generated based on these frequent item sets. Association rules are usually in the form of "A->B", which means that if A occurs, the probability of B occurring is higher. Among them, "support" and "confidence" are two important indicators to measure the quality of association rules. The specific process is as follows:

[0094] Treat application traffic activities that occur simultaneously within a time window as a transaction; for example, a collection of applications that are active in the same time period, each collection is a transaction;

[0095] Initialize the minimum support threshold (min_support), which is a measure of how frequently an application combination appears;

[0096] Traverse the transaction data set once, count the support of each application (item), and only keep those application item sets that meet the minimum support requirements;

[0097] Next, generate all possible frequent k-item sets (k>=2), that is, combinations of two or more applications, traverse the transaction dataset again, and calculate the support of each candidate k-item set;

[0098] Based on the support threshold, remove the candidate k-item sets that do not meet the conditions and retain the frequent k-item sets;

[0099] Generate candidate sets of (k+1) itemsets based on the frequent k-itemsets, but only retain those candidate sets whose subsets have appeared in the frequent k-itemsets to avoid invalid traversal;

[0100] Recursive execution: The above process is iterated continuously until no more frequent item sets can be found;

[0101] Generate association rules from the found frequent itemsets. The form of association rules is "if applications A and B are active at the same time, then application C is also active" (A->C or A, B->C);

[0102] Set a minimum confidence threshold (min_confidence) for each association rule. Confidence refers to the probability of the conclusion event occurring when the premise event in the rule occurs. Only those rules that meet the minimum confidence requirement are retained.

[0103] The Apriori algorithm can be used to discover association patterns between applications. For example, some applications may have significant linkage in network usage. When the traffic of one application increases, the traffic of one or more other applications may also increase accordingly. These association rules help to further explore users' network usage habits and can be incorporated into prediction models to more accurately predict future user traffic demand.

[0104] 7. Time series prediction model construction:

[0105] Based on the extracted common factors and association rules, a prediction model is constructed. The historical data is divided into a training set and a validation set. The model is trained using the training set to learn the evolution of traffic demand over time and the coupling effect between different applications. Then the user's future traffic demand is predicted.

[0106] The effects of the solution proposed in this application compared with the prior art are mainly reflected in the following aspects:

[0107] WAPI connects full cycle dynamic power management innovation:

[0108] Connection management unit: When a mobile terminal attempts to access a WAPI network, it flexibly adjusts the transmission power of the radio frequency (RF) module. It is set to a high-power working state during the initial rapid discovery and scanning of the network. During the waiting period for the access request or authentication response, if a timeout occurs, it is reduced to a non-high-power state or enters a dormant state, effectively reducing unnecessary power consumption.

[0109] Authentication management unit: When performing WAPI authentication, it intelligently schedules encryption operations and configures the working state of the processor or encryption engine according to the requirements of the authentication stage, ensuring high-speed authentication while making full use of the authentication interval to enter low-power mode;

[0110] Data transmission management unit: After the connection is completed, the RF power is dynamically adjusted according to the real-time data transmission rate, RSSI, SINR and predicted traffic demand to achieve more refined power consumption control; this unit not only pays attention to current traffic demand, but also monitors the network activity of multiple applications and predicts future traffic demand, further improving the accuracy and foresight of power management.

[0111] Deep integration and collaborative optimization:

[0112] The system solution proposed in this application fully integrates multiple components such as wireless network interface controller (WNIC), encryption engine, CPU and power management integrated circuit (PMIC), realizes low power consumption management at the hardware level, so that each component can work together seamlessly according to the actual application scenario and reduce invalid power consumption;

[0113] Data-driven and intelligent prediction:

[0114] Data mining and machine learning technologies are introduced to gain in-depth insights into the correlation between applications through the construction of multi-dimensional data cubes, feature engineering, common factor analysis, and association rule mining. Time series prediction models are then built to accurately predict future traffic demand and ensure that the RF module power adjustment strategy is more predictable and adaptable.

[0115] In summary, compared with the existing technology, the technical solution of this application not only optimizes the power management during the WAPI connection process from the hardware level, but also uses advanced data analysis methods and prediction models at the software level to achieve a more comprehensive and efficient low-power WAPI connection dynamic power management system, thereby significantly improving the overall energy consumption performance of mobile terminals when using WAPI connections.

[0116] The embodiments that require protection in this application include:

[0117] The dynamic power management system for low-power WAPI connection includes: a connection management unit, which is used to first start the wireless network interface when the mobile terminal attempts to access the WAPI network, set the transmission power of the radio frequency module to a high-power working state to quickly complete the discovery and scanning of the network; and is used to set the transmission power of the radio frequency module to a non-high-power working state, or enter a periodic dormant state to reduce power consumption if the response time exceeds a threshold during the waiting period for the access request or the authentication response;

[0118] The authentication management unit is used to configure the processor or encryption engine to a high-frequency working state to quickly complete the authentication when performing the interaction of X.509 certificates and key negotiation during WAPI authentication. If there are multiple authentication intervals, the authentication interval is set to configure the processor or encryption engine to a non-high-frequency working state, or enter a periodic dormant state to reduce power consumption;

[0119] The data transmission management unit is used to dynamically adjust the power of the radio frequency module according to the real-time data transmission rate and traffic requirements after completing the WAPI network connection and authentication.

[0120] Preferably, the radio frequency module includes a radio frequency front end and an antenna.

[0121] Preferably, the connection management unit is integrated into a wireless network interface controller chip.

[0122] Preferably, the authentication management unit is integrated into the encryption coprocessor inside the CPU or uses an independent security chip.

[0123] Preferably, the power of the RF module is dynamically adjusted according to the real-time data transmission rate and flow requirements, specifically:

[0124] The data transmission management unit obtains the current data transmission rate, received signal strength indicator, and signal-to-noise ratio through the wireless network interface controller of the mobile terminal; monitors the current data transmission rate, received signal strength indicator, and signal-to-noise ratio in real time; monitors the network activities of the applications running on the terminal, and if the traffic of a certain application is increasing, it can be determined that the user's current traffic demand is increasing;

[0125] When the received signal strength indicator and the signal-to-noise ratio are both greater than the threshold and it is determined that the user's current traffic demand is not increasing, the radio frequency power is reduced; otherwise, the radio frequency power is maintained or increased.

[0126] Preferably, the power of the RF module is dynamically adjusted according to the real-time data transmission rate and flow requirements, specifically:

[0127] The data transmission management unit obtains the current data transmission rate, received signal strength indicator, and signal-to-noise ratio through the wireless network interface controller of the mobile terminal; monitors the current data transmission rate, received signal strength indicator, and signal-to-noise ratio in real time; monitors the network activities of multiple applications running on the terminal, and predicts the user's future traffic demand based on their correlation. If the traffic of a certain application is increasing, it can be determined that the user's current traffic demand is increasing;

[0128] When the received signal strength indicator and the signal-to-noise ratio are both greater than the threshold and it is determined that the user's current traffic demand is not increasing, the radio frequency power is reduced; otherwise, the radio frequency power is maintained or increased.

[0129] Preferably, the network activities of multiple applications running on the terminal are monitored, and the future traffic demand of the user is predicted based on their correlation, in the steps of:

[0130] Obtain real-time network activity data of each application on the terminal through the operating system API, including each application's ID, data transmission rate, cumulative traffic, start and end timestamps, packet size, and connection status;

[0131] Organize the collected multi-dimensional data to construct a multi-dimensional data cube. Each dimension represents the time, application ID, traffic rate, and packet size attributes, and each cell of the cube records the specific traffic data under the corresponding dimension.

[0132] Preprocess the data; create network behavior characteristics that represent the application, including average traffic, maximum traffic, minimum traffic, traffic fluctuation coefficient, and traffic growth rate in adjacent time periods;

[0133] Construct cross-features to mine the interactive effects between applications; extract potential common factors from multidimensional data cubes through principal component analysis (PCA), which can capture the common change trends and patterns between application traffic;

[0134] Apply Apriori association rule mining algorithm to mine association rules between application traffic;

[0135] According to the extracted common factors and association rules, a prediction model is constructed.

[0136] The historical data is divided into a training set and a validation set. The model is trained using the training set to learn how traffic demand evolves over time and the coupling effects between different applications, and then predict users' future traffic demand.

[0137] Preferably, the specific process of extracting potential common factors from a multidimensional data cube through principal component analysis is as follows:

[0138] Convert the multidimensional data cube into a numerical matrix, in which rows represent different observation samples and columns represent different features. Calculate the covariance matrix of the data matrix, perform eigenvalue decomposition on the covariance matrix, and obtain a set of eigenvalues ​​and corresponding eigenvectors. Sort the eigenvalues ​​by size, and select the eigenvectors corresponding to the first k largest eigenvalues. These eigenvectors constitute a new basis. Project the original data into the new space formed by these k eigenvectors to obtain a new data matrix in which each column represents a principal component, the so-called potential common factor.

[0139] Preferably, the specific process of using the Apriori association rule mining algorithm to mine association rules between application program traffic is as follows:

[0140] Treat application traffic activities occurring simultaneously within a time window as one transaction;

[0141] Initialize the minimum support threshold;

[0142] Traverse the transaction data set once, count the support of each application, and only keep those application item sets that meet the minimum support requirement;

[0143] Next, all possible frequent k-item sets, i.e., combinations of two or more applications, are generated, and the transaction dataset is traversed again to calculate the support of each candidate k-item set;

[0144] Based on the support threshold, remove the candidate k-item sets that do not meet the conditions and retain the frequent k-item sets;

[0145] Generate candidate sets of itemsets based on frequent k-itemsets, but only retain those candidate sets whose subsets have appeared in the frequent k-itemsets to avoid invalid traversal;

[0146] Recursive execution: The above process is iterated continuously until no more frequent item sets can be found;

[0147] Generate association rules from the frequent itemsets found. The form of the association rule is "if applications A and B are active at the same time, then application C is also active";

[0148] Set a minimum confidence threshold for each association rule; only those rules that meet the minimum confidence requirement are retained.

[0149] It is known from common technical knowledge that the present invention can be implemented by other embodiments that do not deviate from its spirit or essential features. The above disclosed embodiments are only illustrative in all respects, and not all changes within the scope of the present invention or within the scope equivalent to the present invention are included in the present invention.

Claims

1. A dynamic power management system for low-power WAPI connections, characterized in that: include The connection management unit is used to start the wireless network interface first and set the transmission power of the radio frequency module to a high-power working state to quickly complete the discovery and scanning of the network when the mobile terminal attempts to access the WAPI network; And used to set the transmission power of the radio frequency module to a non-high-power working state, or enter a periodic dormant state to reduce power consumption if the response time exceeds a threshold while waiting for an access request or an authentication response; The authentication management unit is used to configure the processor or encryption engine to a high-frequency working state to quickly complete the authentication when performing the interaction of X.509 certificates and key negotiation during WAPI authentication. If there are multiple authentication intervals, the authentication interval is set to configure the processor or encryption engine to a non-high-frequency working state, or enter a periodic dormant state to reduce power consumption; The data transmission management unit is used to dynamically adjust the power of the radio frequency module according to the real-time data transmission rate and traffic demand after completing the WAPI network connection and authentication. Specifically: the data transmission management unit obtains the current data transmission rate, received signal strength indication, and signal-to-noise ratio through the wireless network interface controller of the mobile terminal; monitors the current data transmission rate, received signal strength indication, and signal-to-noise ratio in real time; monitors the network activities of the applications running on the mobile terminal, and if the traffic of a certain application is increasing, it is determined that the user's current traffic demand is increasing; When the received signal strength indicator and the signal-to-noise ratio are both greater than the threshold and it is determined that the user's current traffic demand is not increasing, the radio frequency power is reduced; otherwise, the radio frequency power is maintained or increased.

2. The low-power WAPI connection dynamic power management system according to claim 1, characterized in that: The RF module includes a RF front end and an antenna.

3. The low-power WAPI connection dynamic power management system according to claim 1, characterized in that: The connection management unit is integrated into the wireless network interface controller chip.

4. The low-power WAPI connection dynamic power management system according to claim 1, characterized in that: The authentication management unit is integrated into the encryption coprocessor inside the CPU or uses an independent security chip.

Images