Communication method and system for realizing three-in-one network based on intelligent CPE gateway

Abstract

The invention discloses a communication method and system for realizing three-in-one network based on an intelligent CPE gateway, and relates to the technical field of network communication, and the method comprises the steps: connecting a telecommunication network, a broadcast television network and the Internet through the intelligent CPE gateway, collecting bandwidth, delay and packet loss rate data in real time, and storing the data in a time sequence database; analyzing historical and real-time data based on a flow prediction model of machine learning, generating a network demand prediction result of a future time period, and dynamically allocating multiple network resources by using a reinforcement learning algorithm; according to the priority of the service type, splitting the data flow to different network links by adopting a multi-path transmission protocol, and meanwhile, carrying out localization processing on the high-time-sensitivity service through an edge computing node; and providing a user-defined interface, setting the bandwidth priority of the service, and feeding back a user manual strategy to the resource allocation model. According to the invention, efficient interaction of three-network data is realized through multi-network access and protocol conversion of the intelligent CPE gateway.

Description

Technical Field

[0001] The present invention relates to the field of network communication technologies, and particularly to a communication method and system for achieving triple play based on an intelligent CPE gateway. Background Art

[0002] Triple play, namely the integration of the telecommunications network, the broadcast television network, and the Internet, aims to achieve resource sharing and service optimization and has become an important development direction of modern communication technologies. However, despite certain progress in related research and technology applications, there are still several key issues that have not been resolved. The telecommunications network (such as the SIP protocol), the broadcast television network (such as the MPEG-TS protocol), and the Internet (such as the IP protocol) adopt different transmission standards and data formats. When existing gateway devices implement seamless data interaction among the three networks, they often face problems such as low protocol conversion efficiency, data loss, or increased latency, making it difficult to meet real-time requirements.

[0003] Triple play needs to dynamically allocate bandwidth, computing resources, etc. according to service requirements. However, when current technologies face sudden traffic changes or highly time-sensitive services (such as 4K videos and online games), the response speed and accuracy of resource scheduling algorithms are insufficient, resulting in a decline in service quality (QoS) or network congestion. These two problems severely restrict the actual application effect of triple play and have become technical bottlenecks that need to be solved urgently. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a communication method for achieving triple play based on an intelligent CPE gateway to solve the compatibility problems caused by multi-network protocol heterogeneity and the contradiction between dynamic resource allocation and service quality guarantee.

[0006] To solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a communication method for achieving triple play based on an intelligent CPE gateway, which includes simultaneously connecting to the telecommunications network, the broadcast television network, and the Internet through the intelligent CPE gateway, collecting bandwidth, latency, and packet loss rate data in real time, and storing them in a time series database;

[0008] Analyzing historical and real-time data using a traffic prediction model based on machine learning to generate a network demand prediction result for a future period, and dynamically allocating multi-network resources using a reinforcement learning algorithm;

[0009] According to the service type priority, splitting the data stream into different network links using a multi-path transmission protocol, and simultaneously performing local processing on highly time-sensitive services through edge computing nodes;

[0010] Provide a user-defined interface to set the bandwidth priority of the service and feedback the user's manual policy to the resource allocation model;

[0011] Build a hybrid encryption channel that combines quantum keys and classical encryption algorithms, intercept attack behaviors in real time in combination with the abnormal traffic detection model, and protect the key security based on the trusted execution environment;

[0012] When network interruption and performance degradation are detected, enable multi-level redundant links of satellite communication, ad hoc network, and low-power narrow-band Internet of Things in sequence, and trigger the service degradation strategy to ensure core services.

[0013] As a preferred solution of the communication method for realizing triple play based on the intelligent CPE gateway according to the present invention, wherein: the state space of the reinforcement learning algorithm includes network load, user priority label, and QoS constraint variables, and optimizes the resource allocation strategy by dynamically adjusting the Lagrange multiplier;

[0014] The QoS constraint variables reflect the delay threshold and bandwidth requirements of high-priority services in real time, and the Lagrange multiplier is dynamically updated according to the degree of violation of the QoS constraint variables to adjust the resource allocation weight.

[0015] As a preferred solution of the communication method for realizing triple play based on the intelligent CPE gateway according to the present invention, wherein: the multi-path transmission protocol dynamically calculates the sub-flow weight based on the link delay and bandwidth utilization rate, and triggers path switching according to the packet loss rate;

[0016] The path switching is linked with the network state and service priority, and the end-to-end delay of high-time-sensitive services is reduced by dynamically adjusting the priority allocation of the transmission path.

[0017] As a preferred solution of the communication method for realizing triple play based on the intelligent CPE gateway according to the present invention, wherein: the user-defined interface inputs the user policy into the reinforcement learning model through a feedback mechanism and generates a decision basis through an interpretability algorithm;

[0018] The interpretability algorithm generates a visualization report based on the correlation analysis of the network state and the user policy.

[0019] As a preferred solution of the communication method for realizing triple play based on the intelligent CPE gateway according to the present invention, wherein: the hybrid encryption channel dynamically switches the encryption level according to the network security state, and the quantum key is stored and updated through the trusted execution environment.

[0020] As a preferred solution of the communication method for realizing triple play based on the intelligent CPE gateway according to the present invention, wherein: the switching priority of the multi-level redundant link is based on the link security score and service level, and the fault scenario is pre-acted through the digital twin network.

[0021] As a preferred solution of the communication method for realizing the integration of three networks based on the intelligent CPE gateway of the present invention, wherein: the service degradation policy implements hierarchical flow limiting for services with different priorities, and uses a conflict-free replicated data type to maintain the local service state.

[0022] In a second aspect, the present invention provides a communication system for realizing the integration of three networks based on an intelligent CPE gateway, including a three-network access module that simultaneously connects to the telecommunications network, the radio and television network, and the Internet through the intelligent CPE gateway, real-time collects bandwidth, latency, and packet loss rate data, and stores it in a time series database;

[0023] A traffic prediction module analyzes historical and real-time data based on a machine learning-based traffic prediction model, generates a network demand prediction result for a future period, and dynamically allocates multi-network resources using a reinforcement learning algorithm;

[0024] A data splitting module splits data streams into different network links using a multi-path transmission protocol according to the priority of service types, and simultaneously performs local processing on high-time-sensitivity services through edge computing nodes;

[0025] A user optimization module provides a user-defined interface, sets the bandwidth priority of services, and feeds back the user's manual policy to the resource allocation model;

[0026] A security encryption module constructs a hybrid encryption channel that combines quantum keys and classical encryption algorithms, intercepts attack behaviors in real time in combination with an abnormal traffic detection model, and protects the key security based on a trusted execution environment;

[0027] A redundancy guarantee module, when detecting network interruption and performance degradation, sequentially enables multi-level redundant links of satellite communication, self-organizing network, and low-power narrow-band Internet of Things, and triggers a service degradation policy to guarantee core services.

[0028] In a third aspect, the present invention provides a computer device, including a memory and a processor, wherein: when the computer program is executed by the processor, it implements any step of the communication method for realizing the integration of three networks based on the intelligent CPE gateway as described in the first aspect of the present invention.

[0029] In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, wherein: when the computer program is executed by the processor, it implements any step of the communication method for realizing the integration of three networks based on the intelligent CPE gateway as described in the first aspect of the present invention.

[0030] The beneficial effects of the present invention are as follows: The present invention provides a communication method for realizing the integration of three networks based on an intelligent CPE gateway, which has significant technical advantages and application values. Through the multi-network access and protocol conversion functions of the intelligent CPE gateway, the compatibility problems caused by the protocol heterogeneity of the telecommunications network, the broadcasting television network, and the Internet are effectively solved, the efficient interaction of data among the three networks is realized, and the real-time performance and stability of communication are improved. Based on machine learning traffic prediction and reinforcement learning resource allocation algorithms, the use of bandwidth and computing resources is dynamically optimized, the contradiction between resource allocation and service quality guarantee is overcome, and high-time-sensitivity services (such as 4K videos and online games) are ensured to operate with low latency and high quality. By adopting multi-path transmission and edge computing technologies, the risk of network congestion is further reduced, and the data transmission efficiency is improved; the hybrid encryption channel and anomaly detection mechanism enhance network security and protect user data privacy. The user-defined interface realizes personalized service optimization and improves the user experience. In the face of network interruption or performance degradation, the multi-level redundant link and service degradation strategy ensure the continuity of core services. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other accompanying drawings without creative efforts based on these drawings.

[0032] Figure 1 It is a flowchart of the communication method for realizing the integration of three networks based on an intelligent CPE gateway in Embodiment 1.

[0033] Figure 2 It is a schematic diagram of the communication system for realizing the integration of three networks based on an intelligent CPE gateway in Embodiment 1.

[0034] Figure 3 It is a flowchart of the QaSAL reinforcement learning algorithm in Embodiment 1.

[0035] Figure 4 It is a flowchart of redundant link switching and service degradation in Embodiment 1. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0036] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings of the specification.

[0037] Many specific details are set forth in the following description in order to provide a thorough understanding of the present invention, but the present invention may be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0038] Secondly, as used herein, "an embodiment" or "embodiment" refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The appearances of "in one embodiment" in different places in this specification do not all refer to the same embodiment, nor are they separate or alternative embodiments that are mutually exclusive of other embodiments.

[0039] Example 1, referring to Figures 1 to 4 , is the first embodiment of the present invention. This embodiment provides a communication method for achieving triple play based on an intelligent CPE gateway, including the following steps:

[0040] S1. Simultaneously connect to the telecommunications network, the radio and television network, and the Internet through the intelligent CPE gateway, and collect network status data including bandwidth, latency, and packet loss rate data in real time, and store it in a time series database.

[0041] The CPE gateway accesses the telecommunications network (1Gbps RJ45), the radio and television network (500Mbps coaxial cable), and the Internet (2.5Gbps GPON fiber). Collect bandwidth, latency, and packet loss rate data in real time, and expand the state space to the following dimensions: QoS constraint variables: latency threshold (≤30ms) for high-priority services (such as video conferencing), bandwidth requirements for radio and television video streams (≥50Mbps). Environmental variables: weather attenuation coefficient of the satellite link (attenuation of 0.8 in rain and snow), Wi-Fi channel occupancy rate (obtained through 802.11k reports). User policy: custom priority label (such as "TV → platinum level").

[0042] Use the time series database InfluxDB, store it in the network_metrics format, and update it once per second.

[0043] S2. Analyze historical and real-time data using a machine learning-based traffic prediction model to generate a network demand prediction result for a future period, and dynamically allocate multi-network resources using a reinforcement learning algorithm.

[0044] Specifically, the reinforcement learning algorithm based on the QaSAL framework dynamically allocates resources and explicitly optimizes QoS constraints. The traffic prediction model uses a two-layer LSTM neural network. It inputs the network state data of the past 24 hours (288 samples), normalizes the network state data to the interval [0, 1], and inputs it into the two-layer LSTM model (each layer of the LSTM has 128 neurons, uses the ReLU activation function, and is optimized using the historical data of the past 1 year (about 8,760 hours) during training, and the learning rate is set to 0.001). It outputs the traffic prediction value for the next 1 hour for subsequent resource allocation. The reinforcement learning uses a deep Q-network (DQN), the experience replay buffer stores 100,000 historical records, and the learning rate is set to one ten-thousandth. Input layer: 300 nodes, receiving the state space vector.

[0045] Set the state space vector, and the expression is,

[0046] S = [L, ΔD, C, λ];

[0047] Among them, S is the state space vector of time, L is the network load, ΔD is the delay deviation, C is the channel occupancy rate, and λ is the Lagrange multiplier.

[0048] Update the Lagrange multiplier, and the expression is,

[0049] λ t+1 = max(λ t + η·ΔD t , 0);

[0050] Among them, λ t+1 is the updated Lagrange multiplier, max is the judgment function, λ t is the Lagrange multiplier at the current time, η is the learning rate, and ΔD t is the delay deviation.

[0051] The composition of the reward function combines the bandwidth utilization rate (weight 60%), the reciprocal of the delay (weight 30%), and the delay deviation penalty (weight 10%), and the expression is,

[0052]

[0053] Among them, R t is the reward value at time step t, α is the weight of the bandwidth utilization rate, U t is the bandwidth utilization rate, β is the weight of the reciprocal of the delay, D t is the current delay, γ is the weight of the delay deviation penalty, ΔD t is the delay deviation, and λ t is the Lagrange multiplier.

[0054] Policy optimization (DQN parameter update), and the expression is,

[0055]

[0056] where θ t+1 is the updated policy parameter, and θ t is the policy parameter at the current time t, is the gradient of the Q-value function with respect to the parameter θ, S t is the state space vector, a t is the executed action, θ is the policy parameter, and t is the time index.

[0057] The LSTM model predicts the future traffic demand (such as 800 Mbps in a telecommunications network). An enhanced state space is constructed (including delay deviation, Wi-Fi channel occupancy rate, etc.). After predicting the traffic demand, a reinforcement learning algorithm based on the QaSAL framework (specifically implemented as a deep Q-network, DQN) is used to dynamically allocate the resources of the three networks. The DQN learns how to reasonably allocate bandwidth among the telecommunications network, the broadcasting television network, and the Internet by simulating the network state and allocation policies, ensuring that the delay of high-priority services (such as video conferencing) does not exceed 30 ms. The allocation effect is monitored, and the policy is dynamically adjusted to meet the QoS requirements.

[0058] It should be noted that: The LSTM accurately predicts the traffic trend by analyzing historical and real-time data, providing a basis for resource allocation. The DQN combines the QaSAL framework and dynamically adjusts the allocation policy through a reward mechanism and Lagrange multipliers to ensure the performance of high-priority services while maximizing the bandwidth utilization rate. It is more flexible than traditional static allocation and can adapt to traffic mutations and changes in user requirements.

[0059] S3. According to the priority of the service type, use the multi-path transmission protocol to split the data stream to different network links, and at the same time perform local processing on high-time-sensitive services through edge computing nodes;

[0060] Specifically, the multi-path transmission protocol (MPTCP) is used to split the data, and the MAC layer parameters are dynamically adjusted to optimize the coexistence performance. The MPTCP path weight is 70% for delay and 30% for bandwidth utilization rate.

[0061] The MAC layer is optimized as follows. The contention window (CW) of 5G NR-U is dynamically adjusted according to the network state (such as shortening the window when the collision rate exceeds 20%). High-priority services (such as video conferencing) are preferentially allocated to low-latency links (telecommunications network). When detecting Wi-Fi channel conflicts, the contention window of 5G NR-U is dynamically adjusted (from the default 15 to 7). The edge node performs real-time transcoding on the video stream (H.265 compression, with the delay controlled within 5 ms).

[0062] It should be noted that the dynamic adjustment of MAC layer parameters improves the coexistence fairness between 5G and Wi-Fi and reduces channel competition conflicts.

[0063] S4. Provide a user-defined interface to set the bandwidth priority of services and feedback the user's manual policy to the resource allocation model;

[0064] Specifically, the user sets the bandwidth priority through the interface, and the policy is fed back to the Deep Q model. The user interface is a Web interface (Vue3 framework) that supports priority setting (high / medium / low). The CPE gateway starts the built-in Web server (port 8080) and loads the interface developed with the Vue3 framework. The user accesses http: / / cpe.local:8080 through a browser, and the interface displays the current service list (such as "Video", "Browsing", "Download") and priority options (high / medium / low). The user selects "Video → High Priority" and clicks the "Save" button, and the interface records the corresponding priority value 1 for "Video".

[0065] Based on XGBoost, the training data is 100,000 network status records in the past month to obtain a pre-trained SHAP model. SHAP calculates the impact of each feature on bandwidth allocation. For example, it outputs "Wi-Fi occupancy rate 60% contributes to a 30% delay overrun". A visualization report (HTML format) is generated, displaying the text "Excessive Wi-Fi occupancy rate leads to delay overrun" and pushed to the user through the Web interface.

[0066] The DQN recalculates the resource allocation action according to the updated state space (including the user priority 1). For example, adjust the bandwidth of the telecommunications network from 50% (500 Mbps) to 60% (600 Mbps), the Internet from 40% to 35%, and the radio and television remains 10%. Execute the allocation, update the CPE gateway routing table to ensure that the video stream obtains 600 Mbps of bandwidth. Monitor the network status after adjustment. For example, the video stream delay drops from 30 ms to 25 ms, and the bandwidth utilization rate rises to 80%. The results are displayed through the Web interface, such as "Video priority increased, bandwidth increased to 600 Mbps, delay 25 ms". If the user adjusts again (such as setting "Download" to medium priority), repeat steps 2-5 and update every 5 minutes.

[0067] It should be noted that the user feedback forms a closed-loop optimization, enhancing policy transparency and conforming to the dynamic rules of QaSAL.

[0068] S5. Build a hybrid encryption channel that combines quantum keys and classical encryption algorithms, intercept attack behaviors in real time in combination with an abnormal traffic detection model, and protect the key security based on a trusted execution environment;

[0069] Specifically, build a quantum-classical hybrid encryption channel to dynamically respond to network attacks.

[0070] Start the BB84 protocol in the quantum module of the CPE gateway, and establish a quantum channel with the receiving end (such as a base station or a cloud server) through an 850nm single-photon source. Generate an initial quantum key (256 bits) and store it in the secure memory area of Intel SGX. Configure the AES-256 encryption module, use the quantum key as input, and enable the CBC mode to encrypt the three-network data stream (such as video conference data of the telecommunications network). The CPE gateway reads traffic data (bandwidth, latency, packet loss rate) from InfluxDB every second and extracts IP header information (such as source IP, port number). Input the data into the XGBoost model to calculate the current traffic characteristics. For example, the number of SYN requests per second reaches 600 times (exceeding the normal value of 500 times). If the XGBoost detects that the abnormal traffic ratio exceeds 5% (such as a SYN Flood attack), it is marked as a DDoS attack and triggers an encryption switch. Pause the AES-256 regular encryption, and call the BB84 protocol to regenerate a new quantum key (at a rate of 1Kbps, generating a 256-bit key in about 2 seconds).

[0071] Load the new key into the AES-256 module, encrypt all data streams, and control the switching time within 3 seconds. According to the XGBoost output, identify the source IP of the attack, add a discard instruction to the firewall rules of the CPE gateway to block abnormal data packets. Update the firewall log. Store the quantum key in the 1MB secure memory of Intel SGX and enable hardware encryption isolation. Update the key through the BB84 protocol every 10 minutes, and the old key is automatically destroyed (overwritten with zeros) to prevent physical side-channel attacks (such as power consumption analysis).

[0072] Check the status of the encrypted data stream. For example, the video conference latency is restored from 50ms to 25ms, and the packet loss rate is reduced from 10% to 1%. If a continuous attack is detected, maintain the quantum encryption mode; otherwise, switch back to the regular AES-256 encryption after 10 minutes.

[0073] It should be noted that the dynamic encryption policy balances security and efficiency, and the attack response is linked with QoS constraints.

[0074] S6. When a network interruption and performance degradation are detected, successively enable multi-level redundant links of satellite communication, self-organizing network, and low-power wide-area Internet of Things, and trigger a service degradation strategy to ensure core services.

[0075] Specifically, the digital twin pre-acts the fault scenario to optimize the redundant link switching decision. The digital twin simulates the impact of network topology, traffic load, and weather on satellite links. The digital twin model is run on the NS-3 simulation tool, loading the current network topology (3 main link nodes) and real-time traffic load (800 Mbps for the telecommunications network, 2 Gbps for the Internet, and 400 Mbps for radio and television). Weather data is input (obtained through an external API, for example, the attenuation coefficient for rain and snow weather is 0.8) to simulate the performance of satellite links. 10 Monte Carlo simulations are run (1 second each time) to evaluate the bandwidth allocation and latency performance after switching to satellite communication or ad hoc networking.

[0076] According to the digital twin results, calculate the fairness index (JFI) for each redundant link. The calculation method is the ratio of the sum of the squares of the throughputs of each service to the total sum of squares. For example, the simulation results of satellite communication show that JFI = 0.85 (higher than 0.8), and the JFI of ad hoc networking is 0.9. Select the link with the highest JFI and sufficient bandwidth to meet the requirements, such as ad hoc networking (50 Mbps, latency 50 ms).

[0077] The CPE gateway activates the ad hoc network through pre-configured interfaces (satellite module: Ku-band transceiver; ad hoc networking: Wi-Fi Mesh module). Switch the data stream from the interrupted Internet link (2 Gbps) to the ad hoc network (50 Mbps), and control the switching time within 500 ms. Update the routing table to ensure that traffic is transmitted through the ad hoc network. For example, video conferencing data preferentially uses the new link. Check the current service priorities (obtained from S4, for example, video conferencing is of high priority and file download is of low priority). Set a bandwidth limit of 50 Mbps for high-priority services (such as video conferencing), and use the token bucket algorithm (bucket depth 5 MB) to limit the flow. Send a pause instruction to low-priority services (such as file downloads) to interrupt their data transmission. Notify the user of the downgraded status through the user interface, such as "The network has switched to ad hoc networking, and low-priority services are paused."

[0078] It should be noted that the digital twin pre-act improves the switching reliability, ensures the continuity of core services, and fits the dynamic environment adaptability of QaSAL.

[0079] This embodiment also provides a communication system for integrating the three networks based on an intelligent CPE gateway, including:

[0080] A three-network access module that simultaneously connects to the telecommunications network, radio and television network, and Internet through the intelligent CPE gateway, and real-time collects data on bandwidth, latency, and packet loss rate, and stores it in the time-series database;

[0081] A traffic prediction module that analyzes historical and real-time data based on a machine learning-based traffic prediction model to generate a prediction result of network demand for future periods, and dynamically allocates multi-network resources using a reinforcement learning algorithm;

[0082] A data shunting module, according to the priority of service types, splits data streams into different network links by using a multi-path transmission protocol, and at the same time performs local processing on high-time-sensitivity services through edge computing nodes;

[0083] A user optimization module, provides a user-defined interface, sets the bandwidth priority of services, and feeds back the user's manual policy to the resource allocation model;

[0084] A security encryption module, constructs a hybrid encryption channel that combines quantum keys and classical encryption algorithms, combines an abnormal traffic detection model to intercept attack behaviors in real time, and protects the key security based on a trusted execution environment;

[0085] A redundancy guarantee module, when detecting network interruption and performance degradation, successively enables multi-level redundant links of satellite communication, self-organizing network and low-power narrow-band Internet of Things, and triggers a service degradation strategy to guarantee core services.

[0086] This embodiment also provides a computer device, applicable to the case of implementing a communication method for triple-play based on an intelligent CPE gateway, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the communication method for triple-play based on an intelligent CPE gateway as proposed in the above embodiment.

[0087] This computer device can be a terminal. This computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of this computer device is used to provide computing and control capabilities. The memory of this computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of this computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a carrier network, NFC (Near Field Communication) or other technologies. The display screen of this computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of this computer device can be a touch layer covered on the display screen, or a button, a trackball or a touchpad set on the outer shell of the computer device, or an external keyboard, a touchpad or a mouse, etc.

[0088] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the communication method for realizing the integration of three networks based on an intelligent CPE gateway as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, abbreviated as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, abbreviated as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, abbreviated as EPROM), programmable read-only memory (Programmable Red-Only Memory, abbreviated as PROM), read-only memory (Read-Only Memory, abbreviated as ROM), magnetic memory, flash memory, magnetic disk or optical disc.

[0089] In summary, through the multi-network access and protocol conversion functions of the intelligent CPE gateway, the present invention effectively solves the compatibility problems caused by the protocol heterogeneity of the telecommunications network, the radio and television network, and the Internet, realizes the efficient interaction of the data of the three networks, and improves the real-time performance and stability of communication. Based on the machine learning traffic prediction and reinforcement learning resource allocation algorithms, the use of bandwidth and computing resources is dynamically optimized, the contradiction between resource allocation and service quality guarantee is overcome, and it is ensured that high-time-sensitive services (such as 4K video, online games) run with low latency and high quality. The multi-path transmission and edge computing technologies are adopted to further reduce the risk of network congestion and improve the data transmission efficiency; the hybrid encryption channel and anomaly detection mechanism enhance the network security and protect the user data privacy. The user-defined interface realizes personalized service optimization and improves the user experience. In the face of network interruption or performance degradation, the multi-level redundant link and service degradation strategy ensure the continuity of the core services.

[0090] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A communication method for realizing the integration of three networks based on an intelligent CPE gateway, characterized in that: including Simultaneously connect to the telecommunications network, radio and television network, and Internet through an intelligent CPE gateway, collect bandwidth, latency, and packet loss rate data in real time, and store it in a time series database; Analyze historical and real-time data using a machine learning-based traffic prediction model to generate network demand prediction results for future periods, and dynamically allocate multi-network resources using a reinforcement learning algorithm; According to the priority of service types, use a multi-path transmission protocol to split data streams to different network links, and at the same time perform local processing on high-time-sensitivity services through edge computing nodes; Provide a user-defined interface to set the bandwidth priority of the service and feedback the user's manual policy to the resource allocation model; Construct a hybrid encryption channel that combines quantum keys and classical encryption algorithms, intercept attack behaviors in real time in combination with an abnormal traffic detection model, and protect the key security based on a trusted execution environment; When network interruption and performance degradation are detected, sequentially enable multi-level redundant links of satellite communication, ad hoc network, and low-power wide-area Internet of Things, and trigger a service degradation strategy to ensure core services; 2. The communication method for realizing triple play based on the intelligent CPE gateway according to claim 1, characterized in that: The state space of the reinforcement learning algorithm includes network load, user priority tags, and QoS constraint variables, and optimizes the resource allocation strategy by dynamically adjusting the Lagrange multiplier; The QoS constraint variables reflect the latency threshold and bandwidth requirements of high-priority services in real time, and the Lagrange multiplier is dynamically updated according to the degree of violation of the QoS constraint variables to adjust the resource allocation weight; 3. The communication method for realizing triple play based on an intelligent CPE gateway according to claim 2, wherein: The multi-path transmission protocol dynamically calculates the sub-flow weight based on link latency and bandwidth utilization, and triggers path switching according to the packet loss rate; Path switching is linked with network status and service priority. By dynamically adjusting the priority allocation of the transmission path, the end-to-end latency of high-time-sensitivity services is reduced; 4. The communication method for realizing triple play based on an intelligent CPE gateway as claimed in claim 3, wherein: The user-defined interface inputs the user policy into the reinforcement learning model through a feedback mechanism and generates decision-making basis through an interpretability algorithm; The interpretability algorithm generates a visual report based on the correlation analysis of network status and user policy; 5. The communication method for realizing triple play based on an intelligent CPE gateway as claimed in claim 4, wherein: The hybrid encryption channel dynamically switches the encryption level according to the network security status, and the quantum key is stored and updated through a trusted execution environment; 6. The communication method for realizing triple play based on an intelligent CPE gateway according to claim 5, wherein: The switching priority of the multi-level redundant link is based on link security scores and service levels, and pre-acts the fault scenario through a digital twin network; 7. The communication method for realizing triple play based on the intelligent CPE gateway according to claim 6, characterized in that: The service degradation strategy implements hierarchical flow control for services with different priorities, and maintains the local service status using conflict-free replicated data types; 8. A communication system that realizes the integration of three networks based on an intelligent CPE gateway, based on the communication method for realizing the integration of three networks based on an intelligent CPE gateway according to any one of claims 1 to 7, characterized in that: including A triple-network access module that simultaneously connects to the telecommunications network, radio and television network, and Internet through an intelligent CPE gateway, collects bandwidth, latency, and packet loss rate data in real time, and stores it in a time series database; A traffic prediction module that analyzes historical and real-time data using a machine learning-based traffic prediction model to generate network demand prediction results for future periods, and dynamically allocates multi-network resources using a reinforcement learning algorithm; A data shunting module that splits data streams to different network links using a multi-path transmission protocol according to the priority of service types, and at the same time performs local processing on high-time-sensitivity services through edge computing nodes; A user optimization module that provides a user-defined interface to set the bandwidth priority of the service and feedback the user's manual policy to the resource allocation model; A secure encryption module that constructs a hybrid encryption channel by coordinating quantum keys and classical encryption algorithms, intercepts attack behaviors in real time in combination with an abnormal traffic detection model, and protects the key security based on a trusted execution environment; A redundancy guarantee module that, when network interruption and performance degradation are detected, sequentially enables multi-level redundant links of satellite communication, ad hoc network, and low-power narrowband Internet of Things, and triggers a service degradation strategy to guarantee core services.

9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that: When the processor executes the computer program, the steps of the communication method for achieving triple play based on an intelligent CPE gateway according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, the steps of the communication method for achieving triple play based on an intelligent CPE gateway according to any one of claims 1 to 7 are implemented.

Images