Intelligent terminal multi-protocol adaptation communication system

Abstract

The application relates to the technical field of communication systems, in particular to a smart terminal multi-protocol adaptive communication system which comprises a protocol interface layer, a protocol resource pool, a smart adaptive middleware layer and a configuration management module. The smart adaptive middleware layer automatically identifies the opposite end communication protocol through a protocol sensing module; a strategy decision engine constructs protocol adaptation decisions into a constraint satisfaction problem with the target of minimizing energy consumption, time delay and maximizing reliability based on multi-source context information, and outputs an optimal decision vector; a unified scheduler performs resource isolation and unified scheduling on the pluggable protocol stack instances in the protocol resource pool through a virtualized container management unit according to the decision; and a security context management module realizes seamless migration of the security state during cross-protocol switching by generating and synchronizing security context tokens. The application realizes the automation and intelligence of the whole protocol adaptation process, dynamically optimizes resource utilization, guarantees security continuity, and supports remote policy updating and system evolution.

Description

Technical Field

[0001] This invention relates to the field of communication system technology, and more specifically, to a multi-protocol adaptation communication system for smart terminals. Background Technology

[0002] With the development of the Internet of Things and the Industrial Internet, smart terminals need to use various devices or cloud platforms with different communication protocols (such as MQTT, CoAP, HTTP / HTTPS, Modbus, Bluetooth, Zigbee, proprietary protocols, etc.) to interact with data.

[0003] Currently, the mainstream technical solutions for handling multi-protocol communication in smart terminals mainly include two types: one is firmware pre-installed single protocol stack, where the terminal supports one or a few protocols for a given application scenario. This approach has poor compatibility and cannot adapt to unknown or newly added devices and protocols. The other is static protocol stack stacking, where multiple independent protocol stacks are pre-integrated into the terminal's firmware. While the latter improves compatibility to some extent, it is essentially still a static and passive adaptation method. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a multi-protocol adaptation communication system for smart terminals, addressing the aforementioned deficiencies of the prior art.

[0005] The technical solution adopted by this invention to solve its technical problem is: A multi-protocol adaptation communication system for smart terminals, deployed on smart terminals, includes a protocol interface layer, a protocol resource pool, a smart adaptation middleware layer, and a configuration management module; The protocol interface layer is used to provide a standard data interface for communication with lower-layer hardware and upper-layer applications. The protocol resource pool is used to store multiple communication protocol stacks in the form of pluggable components; The intelligent adaptation middleware layer connects the protocol interface layer and the protocol resource pool, and is used for dynamic perception, decision-making, scheduling, and security management of communication protocols. The intelligent adaptation middleware layer includes a protocol perception module, a policy decision engine, a unified scheduler, and a context management module. The protocol perception module is used to automatically adapt to the communication protocol type and version of the end device or service. The policy decision engine is used to dynamically generate protocol adaptation decisions for the current communication session based on multi-dimensional context information. The unified scheduler is used to uniformly schedule and manage protocol stack instances and system communication resources in the protocol resource pool based on the protocol adaptation decisions. The security context management module is used to provide unified security services across protocols and manage the continuity of security status during protocol switching. The configuration management module is connected to the intelligent adaptation middleware layer and is used to provide a human-computer interaction interface for policy configuration and system monitoring.

[0006] Preferably, the protocol awareness module includes a deep detection unit, an active detection unit, and a protocol feature learning unit; The depth detection unit is used to listen to and analyze the initial communication messages and perform passive protocol identification based on the protocol feature library of the threshold. The active detection unit is used to construct and send low-power detection messages when passive identification fails and perform auxiliary identification based on the response insulation feature. The protocol feature learning unit is used to accumulate and analyze historical communication data and establish and update the dynamic mapping relationship between peer identifier and commonly used protocols.

[0007] Preferably, the strategy decision engine includes: The multi-source context acquisition unit is used to collect network status data, terminal resource status data, service characteristic data, and security policy data in real time. The dynamic decision model unit is used to run a multi-objective optimization decision model. The multi-objective optimization decision model takes the data from the multi-source context acquisition unit as input, aims to minimize communication energy consumption and latency or maximize reliability, and outputs the optimal protocol type, protocol parameter configuration or cross-protocol combination strategy for the current session.

[0008] Preferably, the dynamic decision model unit is specifically used to construct the protocol decision as a constraint satisfaction problem, wherein the constraints include network bandwidth threshold, terminal remaining power threshold, maximum allowable service latency, and mandatory security level.

[0009] Preferably, the unified scheduler includes: The virtualization protocol container management unit is used to encapsulate each protocol stack instance loaded in the protocol resource pool into an independent lightweight container for operation, thereby achieving resource isolation and independent lifecycle management. The central resource scheduling unit is used to dynamically allocate and arbitrate the CPU computing resources, memory buffers, and network channel resources required by multiple protocol container instances based on the decision priority output by the policy decision engine and using a weighted fair queue algorithm.

[0010] Preferably, the virtualization protocol container management unit supports hot loading and hot unloading of protocol stack instances, and the protocol stack components in the protocol resource pool include at least two of the following: MQTT protocol stack, CoAP protocol stack, HTTP / HTTPS protocol stack, Modbus protocol stack, and Bluetooth protocol stack.

[0011] Preferably, the security context management module includes: The abstract security service interface is used to provide a unified encryption and authentication interface for upper-layer applications and automatically map to the security mechanisms of the underlying specific protocols. The security context token generation and synchronization unit is used to extract the security status information of the current protocol session and generate a security context token when the policy decision engine indicates a protocol switch, and then pass the security context token to the target protocol stack instance to achieve seamless migration of the security session. The collaborative security analysis unit is used to aggregate security event logs from various protocol stacks, perform cross-protocol correlation analysis, and trigger unified threat response actions.

[0012] Preferably, the system further includes a rule and model update channel, through which the configuration management module receives policy rule updates or parameter updates of the dynamic decision model from the remote management platform and dynamically configures the intelligent adaptation middleware layer.

[0013] Preferably, the smart terminal is an Internet of Things gateway, an industrial smart terminal, or an in-vehicle smart device.

[0014] The beneficial effects of this invention are as follows: 1. Through deep detection, active probing, and feature learning by the protocol awareness module, the system achieves automatic and accurate identification of unknown device protocols; the policy decision engine then automatically generates the optimal decision based on the real-time context. This combination automates the entire process of protocol discovery, selection, and configuration, enabling smart terminals to be plug-and-play and autonomously optimize, greatly reducing the manual intervention and professional barriers required for deployment, debugging, and maintenance, and improving overall operational efficiency.

[0015] 2. The strategy decision engine formalizes protocol decisions as a multi-objective constraint satisfaction problem, comprehensively considering key indicators such as energy consumption, latency, reliability, and security, and using real-time collected network status, terminal resources, and service characteristics for solution. This enables the system to dynamically make the most favorable adaptation to the current global state under different scenarios (such as pursuing low latency when battery is sufficient, and prioritizing reliability when network is poor), rather than adopting a fixed strategy, thereby achieving the optimal balance between service quality and resource utilization overall.

[0016] 3. Virtualized protocol containers are used to manage each protocol stack, enabling hot loading, hot unloading, and resource isolation of protocol components. This design allows the system to dynamically add, upgrade, or replace protocol stacks without restarting the service, providing strong functional scalability and maintainability. Meanwhile, the pluggable component specifications of the protocol resource pool and the centralized resource management of the unified scheduler further enhance modularity. Fault isolation between containers also ensures that anomalies in a single protocol stack will not propagate and affect the entire system, significantly improving the overall reliability and stability of the system.

[0017] 4. The Security Context Management module creatively introduces a security context token mechanism, which can extract and migrate critical security states during protocol switching, achieving seamless integration of authentication and encrypted sessions and avoiding duplicate authentication or security degradation caused by protocol switching. Furthermore, the abstract security service interface provides applications with a consistent secure invocation method, while the collaborative security analysis unit can perform cross-protocol log correlation analysis, realizing comprehensive and continuous security protection from a unified interface and state migration to overall threat response.

[0018] 5. A closed loop is formed between the rule and model update channels and the remote management platform. Administrators can remotely distribute new policy rules, optimized decision model parameters, and even entirely new protocol components, enabling the system deployed on a massive number of terminals to continuously upgrade its capabilities and optimize its strategies. Simultaneously, data learned locally on the terminals can be anonymously fed back to the cloud for training a more accurate global model. This mechanism gives the entire system the ability to sustainably learn, optimize, and deploy, continuously adapting to new protocols, new services, and new threats, thus achieving long-term system evolvability.

[0019] 6. It adopts a clear layered architecture (hardware abstraction layer, protocol interface layer, middleware layer, etc.) and standardized interface design, enabling it to flexibly adapt to various smart terminal forms such as IoT gateways, industrial smart terminals, and automotive devices. The hardware abstraction layer shields the differences in underlying hardware, and a unified data model interface standardizes upper-layer application interactions. The intelligent adaptation middleware, as the core, can be reused across different product lines. This highly modular and decoupled design significantly improves the portability of the technical solution, facilitating rapid integration and deployment in different industries and products, and accelerating industrial application. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the present invention will be further described below in conjunction with the accompanying drawings and embodiments. The drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Figure 1 This is a schematic diagram of the overall architecture of the smart terminal multi-protocol adaptation communication system according to an embodiment of this application.

[0021] Figure 2 This is a schematic diagram of the internal collaboration process of the intelligent adaptation middleware layer in an embodiment of this application. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, a clear and complete description will be provided below in conjunction with the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the protection scope of the present invention.

[0023] Preferred embodiments of the present invention, for example Figure 1 and Figure 2 As shown, the intelligent terminal multi-protocol adaptation communication system runs in the intelligent terminal and adopts a layered, modular architecture. From bottom to top, the intelligent terminal multi-protocol adaptation system includes a hardware abstraction layer, a protocol interface layer, a protocol resource pool, an intelligent adaptation middleware layer, a configuration management module, and a rule and model update channel. The intelligent terminal can be an IoT gateway, an industrial intelligent terminal, or an in-vehicle intelligent device. The hardware abstraction layer is the lowest layer of the system, used to adapt to heterogeneous hardware such as Ethernet PHY chips, serial port controllers (e.g., RS-465, CAN), and wireless communication modules (4G / 5G, LoRa), providing a unified hardware operation interface, such as send_data() and receive_data().

[0024] The protocol interface layer provides standard data interfaces for communication with lower-layer hardware and upper-layer applications, residing above the hardware abstraction layer. As an optional implementation, the protocol interface layer includes a hardware driver adaptation submodule and a unified data model interface. The hardware driver adaptation submodule encapsulates operations on specific hardware interfaces. For example, for different 4G modules, the hardware driver adaptation submodule implements a unified PPP dialing or AT command set adaptation interface, hiding hardware differences at the upper layer. The unified data model interface provides standardized data and service access points for upper-layer applications. The unified data model interface defines a unified device data model based on JSON Schema or Protocol Buffers. Regardless of the underlying protocol, the application layer reports data through fixed interfaces such as `report_telemetry(device_id, timestamp, key_value_data)` and receives downlink commands through `command_callback(cmd_id, parameters)`. The protocol interface layer is responsible for bidirectional conversion between the application data model and the underlying protocol message format.

[0025] The protocol resource pool is used to store various communication protocol stacks in the form of pluggable components. Each protocol stack component must be encapsulated as an independent dynamic link library (such as a .so or .dll file) or container image, and each protocol stack group must be encapsulated according to a unified protocol component specification. The protocol component must provide an initialization function, a data sending function, a data receiving callback function, a resource destruction function, and a protocol metadata description file. As an optional embodiment, the protocol metadata description file uses YAML or JSON format, declaring attributes such as the protocol stack's identifier, version number, default port, typical power consumption level, supported security mechanisms, and applicable network types.

[0026] The intelligent adaptation middleware layer connects the protocol interface layer and the protocol resource pool. It dynamically senses, makes decisions, and schedules communication protocols for secure management, serving as the system's core control layer and coordinating lower-layer protocols with upper-layer applications. The intelligent adaptation middleware layer includes a protocol awareness module, a policy decision engine, a unified scheduler, and a context management module. During system startup or runtime, the configuration management module downloads new protocol component packages from the remote platform via the rule and model update channel. The system stores the protocol component packages in local storage, parses their metadata description files, and registers them in the protocol registry. When the intelligent adaptation middleware layer requests to use a component, it can dynamically load the component instance from storage. As an optional embodiment, the protocol stack components stored in the protocol resource pool include, but are not limited to: MQTT protocol stack (supporting v3.1.1 and v5.0), CoAP protocol stack (supporting RFC 7252), HTTP / HTTPS protocol stack, Modbus TCP / RTU protocol stack, Bluetooth BLE protocol stack, OPCUA protocol stack, and a private binary protocol stack.

[0027] The protocol awareness module automatically adapts to the communication protocol type and version of the end device or service. It includes a deep detection unit, an active detection unit, and a protocol feature learning unit. The deep detection unit listens for and analyzes initial communication packets, performing passive protocol identification based on a threshold-based protocol feature library. When a new data stream arrives, the deep detection unit captures the initial packet (e.g., the first three packets), extracts its fixed fields, port number, specific command codes (e.g., Modbus function codes), and performs rule matching with the built-in static protocol feature library. Example feature library rule: {Target port: 8883, Packet header bytes: 0x10 0x??} -> MQTT over TLS. If a match is successful, the identification result is immediately output.

[0028] The active probing unit is used to construct and send low-power probing messages when passive identification fails, and to perform auxiliary identification based on response insulation characteristics; for example, this unit is activated when the depth detection unit fails to identify within a preset timeout period (e.g., 200 milliseconds). It constructs and sends a series of low-power, stateless probing messages (e.g., MQTT Connect messages, CoAP Empty Confirmable Messages, HTTP OPTIONS requests) in priority order. Based on the peer's response (including rejection due to incorrect format, no response, or valid reply) and its characteristics, it comprehensively infers the most likely protocol type.

[0029] The protocol feature learning unit is used to accumulate and analyze historical communication data, establish and update a dynamic mapping relationship between peer identifiers and commonly used protocols, and continuously record the context information of successful historical communication sessions, forming a log: <Peer Identifier (IP / MAC / Device ID), final protocol used, time, service type>. These logs are analyzed using lightweight machine learning algorithms (such as online decision trees) to establish and maintain a dynamic protocol mapping table. For example, it learns that "the device with IP address 192.168.1.100 has a 95% probability of using Modbus TCP during the daily 8:00-18:00 period." Subsequent perception processes will prioritize querying this dynamic table to achieve rapid prediction.

[0030] The strategy decision engine dynamically generates protocol adaptation decisions for the current communication session based on multi-dimensional context information. The strategy decision engine includes a multi-source context acquisition unit and a dynamic decision model unit. The multi-source context acquisition unit collects network status data, terminal resource status data, service characteristic data, and security policy data in real time. As an optional embodiment, network status includes currently available network interfaces, signal strength (RSSI), bandwidth estimation, end-to-end latency, and packet loss rate. Terminal resource status data includes current CPU utilization, available memory space, remaining battery power and power supply status, and device temperature. Service characteristic data is declared by upper-layer applications or derived from data stream analysis, including data flow direction (uplink / downlink), packet size and frequency, and service priority tags (such as emergency alarms, general telemetry, and file downloads). Security policy data comes from preset policies in the configuration management module, including data confidentiality requirements (yes / no), data integrity requirements (yes / no), compliance requirements (such as mandatory use of national cryptographic algorithms), and protocol whitelists / blacklists.

[0031] The dynamic decision model unit is used to run a multi-objective optimization decision model. The multi-objective optimization decision model takes data from the multi-source context acquisition unit as input, aims to minimize communication energy consumption and latency or maximize reliability, and outputs the optimal protocol type, protocol parameter configuration or cross-protocol combination strategy for the current session. Specifically, the dynamic decision model unit is used to construct the protocol decision as a constraint satisfaction problem, where the constraints include network bandwidth threshold, terminal remaining power threshold, maximum allowable latency of the service and mandatory security level.

[0032] The unified scheduler is used to uniformly schedule and manage protocol stack instances and system communication resources in the protocol resource pool based on the decisions of the policy decision engine. The unified scheduler includes a virtual protocol container management unit and a central resource scheduling unit. The virtual protocol container management unit encapsulates each protocol stack instance loaded in the protocol resource pool into an independent lightweight container, achieving resource isolation and independent lifecycle management. The central resource scheduling unit dynamically allocates and arbitrates the CPU computing resources, memory buffers, and network channel resources required by multiple protocol container instances based on the decision priority output by the policy decision engine, using a weighted fair queue algorithm. The virtualized protocol container management unit supports hot loading and hot unloading of protocol stack instances. The protocol stack components in the protocol resource pool include at least two of the following: MQTT protocol stack, CoAP protocol stack, HTTP / HTTPS protocol stack, Modbus protocol stack, and Bluetooth protocol stack. The security context management module provides unified security services across protocols and manages the continuity of security state during protocol switching. It includes an abstract security service interface, a security context token generation and synchronization unit, and a collaborative security analysis unit. The abstract security service interface provides unified encryption and authentication interfaces for upper-layer applications and dynamically maps them to the security mechanisms of specific underlying protocols. Examples include `generate_secure_session(security_profile)` and `encrypt_data(data, level)`. When an application specifies a security level (e.g., "HIGH"), this interface automatically maps it to the specific security mechanism of the underlying protocol, such as mapping it to MQTT's TLS 1.3 (with mutual authentication) or CoAP's OSCORE (object security). The Security Context Token Generation and Synchronization unit extracts the security status information of the current protocol session and generates a security context token when the policy decision engine instructs a protocol switch. This token is then passed to the target protocol stack instance to achieve seamless migration of the security session. The Collaborative Security Analysis unit aggregates security event logs from various protocol stacks (such as authentication failures, abnormal message frequencies, and key update requests), performs cross-protocol correlation analysis, and triggers unified threat response actions. For example, if a Modbus "abnormal function code request" from the same IP is found to be highly correlated with a "scanning attempt" targeting the HTTP management port in time, it can be identified as a collaborative attack. Based on the analysis results, this unit can trigger a unified threat response, such as automatically increasing the mandatory security level of the session corresponding to that IP, or issuing an aggregated alert to the administrator through the configuration management module.

[0033] The configuration management module connects to the intelligent adaptation middleware layer and provides a human-computer interaction interface for policy configuration and system monitoring. The configuration management module receives policy rule updates or parameter updates from the remote management platform through the rule and module update channel, and dynamically configures the intelligent adaptation middleware layer.

[0034] As an optional embodiment, taking a scenario of intelligent IoT accessing a new device and optimizing communication as an example, the internal logic of the system is as follows: First, device access and sensing are performed. The gateway receives the data stream from the new industrial instrument through the RS-485 interface. The deep detection unit of the protocol sensing module listens to and analyzes the initial message and finds no match in the static feature library. The active detection unit then starts and sends Modbus RTU query frames in sequence. The instrument responds correctly to the Modbus query, and the protocol is identified as Modbus RTU. The second step is context acquisition and decision-making. The multi-source context acquisition unit of the policy decision engine simultaneously acquires: the current main network (fiber optic) is stable, the gateway CPU load is low, and the service is "periodic data reading". The dynamic decision model unit runs, solving for "maximizing reliability" as the main objective, and outputs the decision: use the ModbusRTU protocol to communicate with the instrument locally, and use the MQTT protocol (QoS 1) for uplink to the cloud platform. The third step is resource scheduling and execution. Based on the decision, the unified scheduler commands the virtualization protocol container management unit to start (or wake up) the Modbus RTU and MQTT protocol stack containers. The central resource scheduling unit allocates computing and network resources to these two containers. The protocol interface layer reads the meter data through the Modbus RTU container, converts it into a unified data model, and then sends it to the cloud platform via the MQTT container. In the fourth step, when network conditions change (e.g., fiber optic failure, switching to a 4G backup link with poor signal), the multi-source context acquisition unit detects a decrease in bandwidth and an increase in latency. The policy decision engine re-decides, deciding to switch the uplink protocol from MQTT to the lighter CoAP (Cooperative Access Protocol) to save bandwidth. The security context management module immediately extracts the TLS security context of the current MQTT session and generates a token, which is then passed to the newly started CoAP over DTLS container, achieving a seamless migration of the encrypted session. Finally, throughout the entire process, the administrator can monitor in real time on the configuration management module interface. The platform can issue optimized policy models for "meter data reporting under 4G network" through the update channel, which are applied immediately by the system to achieve continuous performance improvement.

[0035] The implementation principle of the intelligent terminal multi-protocol adaptation communication system in this application embodiment is as follows: When an intelligent terminal connects to a new device or the environment changes, the system first automatically and accurately identifies the peer protocol through the protocol awareness module; then, the policy decision engine collects multi-dimensional context such as network, terminal, service, and security in real time, runs a multi-objective optimization model, and dynamically generates the optimal protocol adaptation decision that meets the constraints; based on the decision, the unified scheduler schedules the corresponding protocol stack instances in the protocol resource pool in a containerized manner and intelligently allocates system resources; during the dynamic switching of protocols, the security context management module ensures seamless migration of secure sessions through a token mechanism. The entire process, supported by the monitoring of the configuration management module and the remote update channel, forms a self-optimizing and continuously evolving closed-loop system.

[0036] It should be understood that those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims.

Claims

1. A multi-protocol adaptation system for intelligent terminals, characterized in that, It includes a hardware abstraction layer, a protocol interface layer, a protocol resource pool, an intelligent adaptation middleware layer, and a configuration management module; The hardware abstraction layer is compatible with at least one heterogeneous hardware, such as an Ethernet PHY chip, a serial port controller, and a wireless communication module, and provides a unified hardware operation interface to the upper layer. The protocol interface layer, located above the hardware abstraction layer, is used to provide communication interfaces with lower-layer hardware and standard data interfaces with upper-layer applications. The protocol interface layer includes a hardware driver adaptation submodule and a unified data model interface. The unified data model interface defines a unified device data model based on JSON Schema or Protocol Buffers. The protocol resource pool is used to store various communication protocol stacks in the form of pluggable components. Each protocol stack component is encapsulated as an independent dynamic link library or container image and follows a unified protocol component specification. The protocol component specification includes: providing initialization functions, data sending functions, data receiving callback functions, resource destruction functions, and a protocol metadata description file in YAML or JSON format. The protocol metadata description file declares the identifier, version number, default port, typical power consumption level, supported security mechanisms, and applicable network types of the protocol stack component. The protocol resource pool is used to store multiple communication protocol stacks in the form of pluggable components; The intelligent adaptation middleware layer connects the protocol interface layer and the protocol resource pool, and is used for dynamic perception, decision-making, scheduling and security management of communication protocols; the intelligent adaptation middleware layer includes a protocol perception module, a policy decision engine, a unified scheduler and a context management module. The protocol awareness module is used to automatically adapt to the communication protocol type and version of the end device or service; the protocol awareness module includes a deep detection unit, an active detection unit, and a protocol feature learning unit. The depth detection unit is used to listen to and analyze the initial communication message and perform passive protocol identification based on the protocol feature library of the threshold; the active detection unit is used to construct and send a low-power detection message when passive identification fails, and perform auxiliary identification based on the response insulation feature; the protocol feature learning unit is used to accumulate and analyze historical communication data, and establish and update the dynamic mapping relationship between the peer identifier and commonly used protocols. The deep detection unit captures the first few packets of the initial communication message, extracts fixed fields, port numbers, and specific command codes, and performs rule matching with the built-in static protocol feature library; if the active detection unit fails to identify within a preset timeout period, it constructs and sends low-power, stateless detection messages in order of priority; the protocol feature learning unit continuously records the context information of historical successful communication sessions, establishes and maintains a dynamic protocol mapping table through an online decision tree, and uses it for rapid prediction of subsequent perception. The strategy decision engine is used to dynamically generate protocol adaptation decisions for the current communication session based on multi-dimensional context information; The unified scheduler is used to uniformly schedule and manage protocol stack instances and system communication resources in the protocol resource pool according to the protocol adaptation decision; the unified scheduler includes: The virtualization protocol container management unit is used to encapsulate each protocol stack instance loaded in the protocol resource pool in an independent lightweight container to run, thereby achieving resource isolation and independent lifecycle management, and supporting hot loading and hot unloading of protocol stack instances. The central resource scheduling unit is used to dynamically allocate and arbitrate the CPU computing resources, memory buffers, and network channel resources required by multiple protocol container instances based on the decision priority output by the strategy decision engine and using a weighted fair queue algorithm. The security context management module is used to provide unified security services across protocols and manage the continuity of security state during protocol switching; The configuration management module is connected to the intelligent adaptation middleware layer and is used to provide a human-computer interaction interface for policy configuration and system monitoring.

2. The intelligent terminal multi-protocol adaptation system according to claim 1, characterized in that, The strategy decision engine includes: The multi-source context acquisition unit is used to collect network status data, terminal resource status data, service characteristic data, and security policy data in real time. The dynamic decision model unit is used to run a multi-objective optimization decision model. The multi-objective optimization decision model takes the data from the multi-source context acquisition unit as input, aims to minimize communication energy consumption and latency or maximize reliability, and outputs the optimal protocol type, protocol parameter configuration or cross-protocol combination strategy for the current session.

3. The intelligent terminal multi-protocol adaptation system according to claim 2, characterized in that, The dynamic decision model unit is specifically used to construct the protocol decision as a constraint satisfaction problem, wherein the constraints include network bandwidth threshold, terminal remaining power threshold, maximum allowable service latency, and mandatory security level.

4. The intelligent terminal multi-protocol adaptation system according to claim 1, characterized in that, The virtualization protocol container management unit supports hot loading and hot unloading of protocol stack instances. The protocol stack components in the protocol resource pool include at least two of the following: MQTT protocol stack, CoAP protocol stack, HTTP / HTTPS protocol stack, Modbus protocol stack, and Bluetooth protocol stack.

5. A multi-protocol adaptation system for intelligent terminals according to claim 1, characterized in that, The security context management module includes: The abstract security service interface is used to provide a unified encryption and authentication interface for upper-layer applications and dynamically maps to the security mechanisms of the underlying specific protocols. The security context token generation and synchronization unit is used to extract the security status information of the current protocol session and generate a security context token when the policy decision engine indicates a protocol switch, and then pass the security context token to the target protocol stack instance to achieve seamless migration of the security session. The collaborative security analysis unit is used to aggregate security event logs from various protocol stacks, perform cross-protocol correlation analysis, and trigger unified threat response actions.

6. A multi-protocol adaptation system for intelligent terminals according to claim 2, characterized in that, It also includes a rules and model update channel. The configuration management module receives policy rule updates or parameter updates of the dynamic decision model from the remote management platform through the rules and model update channel, and dynamically configures the intelligent adaptation middleware layer.

7. A multi-protocol adaptation system for intelligent terminals according to any one of claims 1-6, characterized in that, The smart terminal is an IoT gateway, an industrial smart terminal, or an in-vehicle smart device.

Images