Cross-domain self-adaptive cooperative system based on data chain

Abstract

The invention relates to the technical field of communication, and discloses a cross-domain adaptive cooperative system based on a data link, which comprises a physical adaptation layer, a protocol abstraction layer, an intelligent scheduling layer and a security enhancement layer, the physical adaptation layer is used for receiving and transmitting cross-domain equipment signals, realizing software radio dynamic waveform reconstruction of multiple frequency bands, generating digital baseband signals and transmitting the digital baseband signals to the protocol abstraction layer; the protocol abstraction layer is used for realizing grammar conversion of multiple protocols, generating a semantic data stream in a unified format from the digital baseband signal, and transmitting the semantic data stream to the intelligent scheduling layer; the intelligence layer calculates the service priority according to the semantic data flow, realizes self-adaptive broadband distribution and realizes data packet scheduling; and the security enhancement layer generates encrypted state sensing data from the data packet, interacts with the command and control system, and sends the encrypted state sensing data to the physical channel. According to the technical scheme, the problem of interoperation of heterogeneous systems can be solved, the spectrum efficiency is improved, and fine-grained access control of security domain transmission is realized.

Description

Technical Field

[0001] This invention relates to the field of communication technology, and more specifically, to a cross-domain adaptive collaborative system based on a data link. Background Technology

[0002] Under current technological conditions, several long-standing technical bottlenecks exist in cross-vendor, multi-protocol network environments, severely restricting the improvement of overall system performance and business flexibility. These bottlenecks are mainly manifested in the following aspects: First, different equipment manufacturers typically use proprietary communication protocols, leading to "protocol silos" between systems. This makes cross-platform data interaction difficult, hindering effective information sharing and business collaboration, and significantly limiting the overall scalability and interoperability of the system. Second, existing spectrum resource allocation mechanisms are rigid, often employing fixed time limits or static allocation strategies, which cannot adapt to dynamic fluctuations in service load, resulting in low spectrum resource utilization efficiency. Actual measurement data shows that the spectrum utilization rate of existing systems is generally below 40%, with prominent issues of resource waste and localized congestion. Furthermore, traditional security mechanisms rely on static boundary protection and fixed encryption strategies, which are insufficient to meet the needs of dynamic trust management in cross-security domain transmission scenarios. They cannot achieve fine-grained, adaptive security protection, resulting in significant security fragmentation.

[0003] Although existing technologies have proposed solutions such as multi-link aggregation, they still fail to effectively solve the problems of semantic consistency and context compatibility in the process of heterogeneous protocol conversion, making it difficult to guarantee the accuracy and business continuity of cross-protocol communication.

[0004] Therefore, there is an urgent need in this field for a new network architecture and method that can achieve efficient protocol conversion, dynamic resource allocation, and cross-domain secure collaboration in order to overcome the above-mentioned technical deficiencies. Summary of the Invention

[0005] To achieve the above objectives, this application provides a cross-domain adaptive collaborative system based on a data chain, implemented using a four-layer elastic architecture, including: a physical adaptation layer, a protocol abstraction layer, an intelligent scheduling layer, and a security enhancement layer; The physical adaptation layer is used to transmit and receive signals from cross-domain devices, realize software radio dynamic waveform reconstruction in multiple frequency bands, and generate digital baseband signals. After receiving the signal, the physical adaptation layer performs down-conversion, filtering, synchronization, demodulation and decoding on the signal to generate a digital bit stream, which is then transmitted to the protocol abstraction layer along with the accompanying metadata as a digital baseband signal. The protocol abstraction layer is used to implement the syntax conversion of multiple protocols, generate a semantic data stream in a unified format from the digital baseband signal, and pass it to the intelligent scheduling layer; The intelligent scheduling layer is used to calculate service priorities based on semantic data streams, realize adaptive bandwidth allocation, and schedule data packets. The security enhancement layer generates encrypted potential sensing data from data packets, interacts with the command and control system, and sends the encrypted potential sensing data to the physical channel.

[0006] The physical adaptation layer is implemented through a multi-band radio frequency front-end configuration, including: A software-defined electronic radio module was selected, and a multi-band antenna array was configured. A multi-mode baseband processing chipset is constructed using a hybrid architecture of FPGA and custom protocol conversion chip to achieve signal modulation and demodulation.

[0007] Metadata consists of channel and signal-related data, including channel quality information, timestamps, frequencies, and bandwidths. The channel quality information includes signal-to-noise ratio (SNR), received signal strength (RSSI), and channel quality index (CQI).

[0008] Furthermore, the protocol abstraction layer consists of a syntax parser, a semantic mapping engine, and a dynamic wrapper, enabling protocol adaptation when performing syntax conversion of multiple protocols; The implementation of the protocol abstraction layer includes: The protocol recognition engine supports a syntax parser and includes a protocol feature code database with various protocol header features. Before building the semantic mapping engine, we define a military / industrial ontology model library, define entity relations as an ontology knowledge base, and set up a context-aware module for injecting environmental parameters. Build a dynamic encapsulation library supported by a protocol template library.

[0009] Furthermore, the process of achieving protocol adaptation includes: Multi-protocol syntax parsing includes: a protocol recognition engine identifies the protocol, and based on the recognition result, a syntax parser performs syntax parsing to generate a data object; Semantic mapping and consistency checks are performed on data objects to obtain data with a unified semantic model, which serves as a standardized semantic data package; The dynamic encapsulator, in conjunction with the protocol template library, encapsulates standardized semantic data packets into a format suitable for transmission, forming target protocol compliant data frames, which are then passed to the intelligent scheduling layer.

[0010] Furthermore, the intelligent scheduling layer implements adaptive bandwidth allocation through the following steps: Reinforcement training learning models; Dynamic TDMA scheduling is achieved by updating the time slot allocation table in real time based on the learning model, and avoiding signal conflicts caused by simultaneous transmission of multiple devices through the non-overlapping nature of time slot allocation.

[0011] Among them, reinforcement training learning models include: Define a deep reinforcement learning state space, which includes channel quality (CQI), traffic queue length, historical collision count, and current time slot allocation table; Define the action space, including bandwidth allocation, time slot location selection, and priority settings; Design a reward function, expressed as follows: ,in, For throughput coefficient, For delay coefficient, The conflict coefficient, For throughput, The end-to-end transmission delay of data packets. Number of conflicts.

[0012] The calculation methods for the parameters in the reward function include: , , .

[0013] Furthermore, the security enhancement layer's processing of data packets includes: Dynamic trust assessment is based on device fingerprints and behavior to achieve dynamic trust assessment; if the dynamic trust assessment shows that the device is trustworthy, hybrid encryption is performed, using the national cryptographic standard SM4 and quantum key distribution for hybrid encryption; Continuous anomaly detection is performed. If an anomaly is detected, a blocking mechanism is triggered, and feedback is sent to the intelligent scheduling layer to adjust the scheduling strategy. After processing, the data packet is sent to the physical adaptation layer for transmission.

[0014] According to the method described in this invention, the interoperability problem of heterogeneous systems is solved by parsing the protocol syntax and utilizing the semantic separation and transformation mechanism; spectrum efficiency is improved by using a multi-objective optimization model driven by the channel quality index; and fine-grained access control for secure domain transmission is achieved by creating dynamic trust assessment evidence. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of a cross-domain adaptive collaborative system structure provided according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the protocol adaptation process provided in an embodiment of the present invention. Detailed Implementation

[0016] The specific implementation of the present invention will now be described in detail with reference to the accompanying drawings.

[0017] This invention provides a cross-domain adaptive collaborative system based on a data chain, implemented using a four-layer elastic architecture, such as... Figure 1As shown, the four-layer elastic architecture includes: P110 physical adaptation layer, P120 protocol abstraction layer, P130 intelligent scheduling layer and P140 security enhancement layer; The P110 physical adapter layer is used to receive signals from cross-domain devices, realize software radio dynamic waveform reconstruction in multiple frequency bands, and generate digital baseband signals. The physical adaptation layer includes a multi-band antenna array, a software-defined radio, and a baseband processing chipset unit; it dynamically switches between multiple frequency bands via software-defined radio (SDR), supporting heterogeneous device access in HF / UHF / L / S / C / X / Ku bands.

[0018] The physical adaptation layer is implemented through multi-band RF front-end configuration: 1) A Software-Defined Radio Reception (SDR) module is selected, configured with antenna arrays in HF (2-30MHz), UHF (300 MHz - 3GHz), L-band (1 GHz - 2 GHz), S-band (2 GHz - 4 GHz), C-band (4 GHz - 8 GHz), X-band (8 GHz - 12GHz), and Ku-band (12 GHz - 18 GHz). It is evident that the SDR module supports multi-band radio frequency signal reception and transmission (SDR), compatible with multiple frequency bands such as HF / UHF / Ku; it receives digitally sampled analog signals, i.e., a complex sequence (I / Q data), representing the in-phase and quadrature components of the signal; simultaneously, the SDR module can select the optimal waveform based on channel conditions, realizing dynamic waveform reconstruction for software-defined radio.

[0019] 2) A multi-mode baseband processing chipset is constructed using a hybrid architecture of FPGA and custom protocol conversion chip (ASIC) to achieve signal modulation and demodulation.

[0020] The demodulation process parses the data according to a predefined frame structure. A frame typically includes a preamble (used for synchronization), a frame header (containing information such as frame length and modulation scheme), and a payload (the actual transmitted data).

[0021] In summary, after receiving the signal, the physical adaptation layer performs down-conversion, filtering, synchronization, demodulation, and decoding to generate a digital bitstream. The digital bitstream contains binary data of complete frames and accompanying metadata as a digital baseband signal, i.e., the raw data is transmitted to the protocol abstraction layer. The metadata is channel and signal-related metadata, including channel quality information (such as signal-to-noise ratio SNR, received signal strength RSSI, and channel quality index CQI), timestamps, frequency, and bandwidth.

[0022] The physical adaptation layer is a physical infrastructure implemented through a hardware platform. In addition to building a multi-band radio frequency front-end configuration through software-defined radio (SDR) modules and multi-mode baseband processing chipsets, it also deploys edge computing nodes: that is, deploying a resource scheduling server on each network node; installing a real-time operating system and starting a deterministic network protocol stack.

[0023] The P120 protocol abstraction layer is used to implement the syntax conversion of multiple protocols, generate a unified format semantic data stream from the digital baseband signal, and pass it to the intelligent scheduling layer. The protocol abstraction layer, based on an ontology knowledge base, implements a dual-engine syntax-semantic conversion for various military / industrial protocols. It consists of a syntax parser, a semantic mapping engine, and a dynamic wrapper.

[0024] Specifically, the syntax parser implements syntax conversion for 12 protocols, including Link16, TTNT, and HDL, through protocol-independent middleware; while the semantic mapping engine maintains the consistency of data meaning through an ontology knowledge base.

[0025] Before building the protocol abstraction layer, edge computing nodes need to be configured: this includes deploying resource height servers on each network node, installing real-time operation protocols, and starting the deterministic network protocol stack.

[0026] The implementation of the protocol abstraction layer requires first building a protocol recognition engine to support a syntax parser. The protocol recognition engine includes building a protocol feature code database containing various protocol header features such as Link16 / TTNT / HDL. Secondly, before building the semantic mapping engine, a military / industrial ontology model library defined using the OWL language is built, entity relations are defined as an ontology knowledge base, and a context-aware module that injects environment parameters is set to support the semantic mapping engine in generating standardized semantic data packets. Finally, a dynamic encapsulation library supported by a protocol template library is needed to realize the conversion of the target protocol.

[0027] The process of achieving protocol adaptation is as follows Figure 2 As shown, it includes the following steps: Step S210: After the protocol abstraction layer obtains the raw data provided by the physical adaptation layer, it performs multi-protocol syntax parsing: first, the protocol recognition engine performs protocol recognition, and then, based on the recognition results, the syntax parser performs syntax parsing to generate data objects; In this step, the protocol recognition engine detects the protocol type based on the header feature code, outputs the corresponding protocol label, and then the syntax parser extracts structured data from the raw data through a parallel parsing queue to generate data objects. The syntax parser uses protocol-independent middleware to implement syntax conversion for 12 protocols, including Link16, TTNT, and HDL, parsing the raw data into structured data.

[0028] Step S220: Perform semantic mapping and consistency checks on the data objects to obtain data with a unified semantic model, which serves as a standardized semantic data package; Semantic mapping is implemented through an ontology knowledge base, a semantic mapping engine, and a context-aware module. The ontology knowledge base defines a semantic model library and entity relationship information for the relevant domain, which provides a standard basis for the semantic mapping engine to convert data pairs into a row-unified model and detect semantic conflicts. The context-aware module provides environmental parameters, such as network latency, security level, and business priority configuration parameters.

[0029] Step S230: Through target protocol encapsulation processing, the standardized semantic data packets are encapsulated into a format suitable for transmission by the dynamic encapsulator combined with the protocol template library to form a target protocol compliant data frame, which is then passed to the intelligent scheduling layer.

[0030] The protocol template library contains multiple target protocol encapsulation rules, while the dynamic encapsulation library can select the appropriate protocol version based on the target end, insert protocol-specific fields into the standardized semantic data packet, and generate a target protocol compliant data frame.

[0031] The P130 intelligent scheduling layer is used to calculate service priorities based on the semantic data stream, perform adaptive bandwidth allocation, and realize data packet scheduling. After the intelligent scheduling layer acquires the formatted data, it first classifies the data for services and extracts QoS requirements. At the same time, it combines the results of channel state monitoring (CQI, etc.) as input for reinforcement learning decision-making. The decision outputs a time slot allocation scheme, which is then executed by dynamic TDMA scheduling.

[0032] Specifically, the intelligent scheduling layer uses a deep reinforcement learning (DQN) algorithm to dynamically optimize time slot allocation; it includes: a dynamic TDMA scheduler, a reinforcement learning decision center, a QoS guarantee module, and a resource monitoring dashboard.

[0033] The process of achieving adaptive bandwidth allocation includes the following: 1) Reinforcement training of the learning model: Define the state space of Deep Reinforcement Learning (DQN) (such as Channel Quality Index (CQI), traffic queue length, historical collision count, current time slot allocation table, etc.). Define the action space, including bandwidth allocation, time slot location selection, and priority settings; Design a reward function: (1) in, For throughput coefficient, For delay coefficient, The conflict coefficient, based on a combination of business demand-driven approaches, theoretical modeling, and latency data verification, is assigned the following values: ; in, Throughput is the amount of data successfully transmitted per unit of time, which is obtained by counting the total number of bits of data packets successfully received per unit of time. End-to-end transmission delay of data packets, calculated using timestamps; : Number of collisions, i.e. the number of collisions caused by channel resource contention. In TDMA systems, collisions are identified by the time slot usage information fed back by the receiver. The calculation method is expressed as follows: (2) (3) (4) 2) Implement dynamic TDMA scheduling: First, the time slot allocation table is updated based on the output of the DQN algorithm to ensure that high-priority tasks get resources first. DQN algorithm optimizations include: The experience playback mechanism is determined using a priority sampling method, where the priority is calculated as follows: (5) in, TD error represents the difference between the current state value estimate and the target value; age is the sample storage duration; i is the training sequence number; Determine dynamics Greedy strategy: The exponential decay from 0.9 to 0.05 with each training round; Adjust the network parameter update cycle to synchronize with the TDMA scheduling cycle.

[0034] Secondly, a conflict resolution mechanism is employed: by ensuring non-overlapping time slot allocation, signal conflicts caused by simultaneous transmission from multiple devices are avoided. If time slot overlap is detected (e.g., a sudden high-priority task), the Dutch auction algorithm is triggered to reallocate the overlapping time slots.

[0035] The P140 security enhancement layer generates encrypted potential sensing data from the data packet, interacts with the command and control system, and sends the encrypted potential sensing data to the physical channel.

[0036] Before being sent to the physical channel, the data packets obtained after scheduling by the intelligent scheduling layer need to be processed by the security enhancement layer: 1) The security enhancement layer first performs dynamic trust assessment, that is, dynamic trust assessment is achieved based on device fingerprints and behavior; 2) If the dynamic trust assessment shows that the device is trustworthy, hybrid encryption is performed, using the national cryptographic SM4 and quantum key distribution for hybrid encryption; 3) Anomaly detection is continuously performed. If an anomaly is found, the blocking mechanism is triggered and feedback is given to the intelligent scheduling layer to adjust the scheduling strategy. After processing, the data packets are sent to the physical adaptation layer for transmission.

[0037] The security enhancement layer adopts a cross-domain zero-trust framework, supports hybrid encryption of the national cryptographic standard SM4 and quantum key distribution, and realizes dynamic trust assessment and millisecond-level anomaly blocking. It includes a hybrid encryption module, a zero-trust control center, a blockchain evidence storage node, and an anomaly detection engine.

[0038] The security enhancement layer is implemented based on the establishment of a zero-trust security architecture, including the following: 1) A hybrid encryption engine is adopted, which includes a national cryptographic algorithm and a quantum key distribution scheme. In terms of data transmission, SM4 encryption (CBC mode, key length 256 bits) is used; in terms of key distribution, a QKD system based on the BB84 protocol is used, with a key generation rate of ≥1Mbps.

[0039] 2) Implement dynamic trust assessment: The security enhancement layer defines a trust scoring model, expressed as: (6) in, Fingerprint matching degree (based on PUF physical no-cloning function); Scoring of behavioral patterns (hidden Markov model for anomaly detection); To establish a historical record of trust.

[0040] 3) Blockchain-based transmission and storage, using an improved PBFT consensus algorithm to reduce latency.

[0041] This invention provides specific implementation security for military collaborative communication scenarios, including joint battlefield communication among multiple services (NATO satellites (STANAG 7085), ground radar, naval carrier-based aircraft (TTNT protocol), army command vehicles (Link16 protocol), and airborne early warning aircraft), enabling real-time sharing of high-precision battlefield maps with end-to-end latency <100ms and anti-jamming success rate ≥95%. Specifically, the implementation of the cross-domain adaptive collaborative system includes the following: 1) System Configuration: RF front end: USRP X310 SDR×3 (HF / UHF / Ku band), Xilinx Zynq UltraScale+ RFSoC baseband processor; Encryption module: Quantum key distributor (QDK-300, key rate 1Mbps), SM4 / 9 national standard hardware acceleration card; Software parameters: Protocol library version: MIL-STD-1.2 (including 12 protocols such as Link16 / TTNT / STANAG); Scheduling algorithm: DQN; Security policy: Dynamic trust threshold 0.85, blocking response time ≤10ms.

[0042] 2) The implementation steps are as follows: S1: Physical layer-by-layer adaptation, including automatic selection of the optimal frequency band; waveform anti-interference switching: that is, when narrowband interference is detected, it instantly switches to the Chirp-Spread Spectrum (CSS) waveform; S2: Protocol conversion process, including Link16 to STANAG 7085; semantic verification: unit consistency is detected through ontology model; S3: Dynamic resource scheduling, including the following steps: (1) Problem modeling, including: DQN decision inputs include state space (CQI, queue length, number of historical collisions); action space (allocate 20MHz, allocate 10MHz+retransmission, preemption priority); output time slot allocation scheme, including time window, bandwidth, service type; design a reward function based on throughput, latency, number of collisions, etc., the reward function is expressed as: ,in .

[0043] (2) Design the network structure and use a deep neural network (DNN) to define an approximate Q function, where the input is the normalized state information and the output is the Q value of each action; specifically, the network structure includes: input layer (state dimension), first fully connected layer (128 neurons), second fully connected layer (64 neurons), and output layer (action dimension).

[0044] S4: Security protection, including the following: each device generates a 256-bit fingerprint (60% hardware features + 40% behavioral patterns) to achieve dynamic fingerprint authentication; hybrid encryption is used; data transmission uses SM4-CBC encryption (key update cycle of 30 seconds); key distribution uses quantum channel to distribute 256-bit keys (BB84 protocol).

[0045] The beneficial technical effects achieved by the specific embodiments provided by this invention include: solving core problems such as protocol silos, resource sharing, and security fragmentation in heterogeneous network environments through a four-layer elastic architecture and cross-domain collaborative technology; achieving real-time contribution of high-precision battlefield maps (4K resolution, 30fps frame rate); end-to-end testing <100ms; and an anti-interference success rate ≥95%. Therefore, the cross-domain adaptive collaborative design system for general data links provided by this invention is applicable to fields such as military collaboration, industrial internet, and smart cities, providing a systematic solution for highly reliable, secure, and real-time signal communication; and laying a theoretical foundation for data link networking in complex dynamic environments in subsequent engineering projects, possessing significant theoretical research value and engineering application value.

[0046] The above-disclosed embodiments are merely a few specific examples of the present invention. However, the present invention is not limited thereto, and any variations that can be conceived by those skilled in the art should fall within the protection scope of the present invention.

Claims

1. A cross-domain adaptive collaborative system based on data links, characterized in that, It is implemented based on a four-layer elastic architecture, including: physical adaptation layer, protocol abstraction layer, intelligent scheduling layer and security enhancement layer; The physical adaptation layer is used to transmit and receive cross-domain device signals, realize software radio dynamic waveform reconstruction of multiple frequency bands, and generate digital baseband signals. After receiving the signal, the physical adaptation layer performs down-conversion, filtering, synchronization, demodulation and decoding processing to generate a digital bit stream, which is transmitted to the protocol abstraction layer along with the accompanying metadata as a digital baseband signal. The protocol abstraction layer is used to implement the syntax conversion of multiple protocols, generate a unified format semantic data stream from the digital baseband signal, and pass it to the intelligent scheduling layer; The intelligent scheduling layer is used to calculate service priorities based on the semantic data stream, realize adaptive bandwidth allocation, and schedule data packets. The security enhancement layer generates encrypted potential sensing data from the data packet, interacts with the command and control system, and sends the encrypted potential sensing data to the physical channel.

2. The cross-domain adaptive collaborative system according to claim 1, characterized in that, The physical adaptation layer is implemented through a multi-band radio frequency front-end configuration, including: A software-defined electronic radio module was selected, and a multi-band antenna array was configured. A multi-mode baseband processing chipset is constructed using a hybrid architecture of FPGA and custom protocol conversion chip to achieve signal modulation and demodulation.

3. The cross-domain adaptive collaborative system according to claim 1, characterized in that, The metadata is channel and signal-related data, including channel quality information, timestamps, frequencies, and bandwidth; the channel quality information includes: signal-to-noise ratio (SNR), received signal strength (RSSI), and channel quality index (CQI).

4. The cross-domain adaptive collaborative system according to claim 1, characterized in that, The protocol abstraction layer consists of a syntax parser, a semantic mapping engine, and a dynamic wrapper, enabling protocol adaptation during multi-protocol syntax conversion. The implementation of the protocol abstraction layer includes: The protocol recognition engine supports a syntax parser and includes a protocol feature code database with various protocol header features. Before building the semantic mapping engine, we define a military / industrial ontology model library, define entity relations as an ontology knowledge base, and set up a context-aware module for injecting environmental parameters. Build a dynamic encapsulation library supported by a protocol template library.

5. The cross-domain adaptive collaborative system according to claim 4, characterized in that, The process of achieving protocol adaptation includes: Multi-protocol syntax parsing includes: a protocol recognition engine identifies the protocol, and based on the recognition result, a syntax parser performs syntax parsing to generate a data object; Semantic mapping and consistency checks are performed on data objects to obtain data with a unified semantic model, which serves as a standardized semantic data package; The dynamic encapsulator, in conjunction with the protocol template library, encapsulates standardized semantic data packets into a format suitable for transmission, forming target protocol compliant data frames, which are then passed to the intelligent scheduling layer.

6. The cross-domain adaptive collaborative system according to claim 1, characterized in that, The intelligent scheduling layer implements adaptive bandwidth allocation through the following steps: Reinforcement training learning models; Dynamic TDMA scheduling is achieved by updating the time slot allocation table in real time based on the learning model, and avoiding signal conflicts caused by simultaneous transmission of multiple devices through the non-overlapping nature of time slot allocation.

7. The cross-domain adaptive collaborative system according to claim 6, characterized in that, The reinforcement training learning model includes: Define a deep reinforcement learning state space, which includes channel quality (CQI), traffic queue length, historical collision count, and current time slot allocation table; Define the action space, including bandwidth allocation, time slot location selection, and priority settings; Design a reward function, expressed as follows: ,in, For throughput coefficient, For delay coefficient, The conflict coefficient, For throughput, The end-to-end transmission delay of data packets. Number of conflicts.

8. The cross-domain adaptive collaborative system according to claim 7, characterized in that, The calculation method for the parameters in the reward function includes: ， ， 。 9. The cross-domain adaptive collaborative system according to claim 1, characterized in that, The security enhancement layer processes data packets including: Dynamic trust assessment is based on device fingerprints and behavior to achieve dynamic trust assessment; if the dynamic trust assessment shows that the device is trustworthy, hybrid encryption is performed, using the national cryptographic standard SM4 and quantum key distribution for hybrid encryption; Continuous anomaly detection is performed. If an anomaly is detected, a blocking mechanism is triggered, which is fed back to the intelligent scheduling layer to adjust the scheduling strategy. After processing, the data packet is sent to the physical adaptation layer for transmission.

Images