A safety authentication and trusted execution method of a PlantDGP cooperative control instruction set

Abstract

The application discloses a kind of PlantDGP cooperative control instruction set security authentication and trusted execution method, it is related to industrial internet security control technical field, including obtaining the interactive control instruction between PlantDGP node, and according to hardware execution state parameter extraction interactive control instruction energy fluctuation characteristics;Based on energy fluctuation characteristics, the phase shift region of interactive control instruction is identified, the instruction flow to be verified is obtained and context reconstruction is carried out, to obtain the first trusted instruction set;Extract the rollback feature of each instruction in the first trusted instruction set, and check the preset instruction execution path based on the rollback feature, to obtain the second trusted instruction set;Second trusted instruction set is imaged and run, and execution permission signal is generated according to mirror image running result;And be issued to PlantDGP node and carry out instruction execution instruction in multistage compilation chain processing destroys original field order, to affect the problem of the credibility of instruction.

Description

Technical Field

[0001] This invention relates to the field of industrial internet security control technology, and more specifically, to a secure authentication and trusted execution method for PlantDGP collaborative control instruction set. Background Technology

[0002] In existing agricultural automation control systems, PlantDGP nodes typically coordinate and execute tasks through interactive control commands. These commands are used to schedule and control the actions of each node, including single-node commands such as opening or closing irrigation valves and adjusting fertilizer dosages, as well as multi-node collaborative commands such as dynamically adjusting irrigation sequence based on soil moisture or crop growth status and coordinating the simultaneous operation of multiple agricultural machines. In this way, PlantDGP nodes can achieve information sharing, task synchronization, and collaborative decision-making in a distributed environment, thereby improving the automation level and operational accuracy of agricultural production.

[0003] However, during PlantDGP node collaborative control, instructions typically undergo processing through multiple levels of compilation chains and intermediate build tools before generation, transmission, and execution. During this process, some configuration generation tools or instruction structure processing tools may automatically rearrange instruction fields based on field names or internal indexes. For example, during configuration file conversion, structure serialization, or instruction encapsulation, the field order may be adjusted to improve data access efficiency or optimize storage structure. While this rearrangement usually does not affect the instruction's business logic or node operation functionality, and is therefore difficult to detect in routine functional testing, when the instruction field order is used for execution path verification, signature verification, or inter-node synchronization coordination, this automatic rearrangement may disrupt the original field order, thereby affecting the instruction's reliability and collaborative execution effectiveness.

[0004] To address the above problems, this invention proposes a solution. Summary of the Invention

[0005] To overcome the aforementioned deficiencies in the prior art, embodiments of the present invention provide a secure authentication and trusted execution method for the PlantDGP collaborative control instruction set. By performing energy fluctuation analysis and instruction reconstruction and verification on the interactive control instructions between PlantDGP nodes, the method addresses the problem that the original field order is destroyed during multi-level compilation chain processing, thereby affecting the trustworthiness of the instructions.

[0006] To achieve the above objectives, the present invention provides the following technical solution: A secure authentication and trusted execution method for a PlantDGP collaborative control instruction set includes the following steps: acquiring interactive control instructions between PlantDGP nodes and extracting energy fluctuation characteristics of the interactive control instructions based on hardware execution state parameters; identifying the phase offset region of the interactive control instructions based on the energy fluctuation characteristics to obtain the instruction stream to be verified and reconstructing the context of the instruction stream to be verified to obtain a first trusted instruction set; extracting the backtracking features of each instruction in the first trusted instruction set and verifying a preset instruction execution path based on the backtracking features to obtain a second trusted instruction set; performing a mirror run on the second trusted instruction set and generating an execution permission signal based on the mirror run result; binding the execution permission signal with the second trusted instruction set and sending it to the PlantDGP node for instruction execution.

[0007] In a preferred embodiment, the step of extracting energy fluctuation characteristics of interactive control commands based on hardware execution state parameters specifically involves: collecting hardware execution state parameters of the PlantDGP node when executing historical commands, and aligning the hardware execution state parameters by time to obtain a hardware state timing sequence; matching the interactive control commands with the hardware state timing sequence on the time axis to identify the hardware state change interval corresponding to each interactive control command; calculating the cumulative change in voltage fluctuation amplitude and the instantaneous change rate of current fluctuation frequency within the hardware state change interval; constructing a time distribution curve of energy fluctuation based on the cumulative change and the instantaneous change rate, and extracting energy fluctuation characteristics from the time distribution curve.

[0008] In a preferred embodiment, the step of identifying the phase offset region of the interactive control command based on energy fluctuation characteristics to obtain the command stream to be verified specifically involves: mapping the energy fluctuation characteristics to the command execution cycle of the PlantDGP node, and marking the energy anomaly start point and energy anomaly end point on the time axis of the command execution cycle; using the energy anomaly start point and energy anomaly end point as boundaries, extracting all command fragments flowing through the PlantDGP node bus within the corresponding time period to obtain the first offset command stream; performing opcode parsing and operand tracing on each command in the first offset command stream to obtain the data dependency relationship and control flow jump relationship between commands; and performing context command completion on the first offset command stream based on the data dependency relationship and control flow jump relationship to obtain the command stream to be verified.

[0009] In a preferred embodiment, the step of performing context instruction completion on the first offset instruction stream based on data dependencies and control flow jump relationships to obtain the instruction stream to be verified specifically involves: in the first offset instruction stream, tracing back to the previous write instruction corresponding to its source operand using each instruction as the starting point to form a forward data dependency chain; tracing back to the next read instruction corresponding to its destination operand using each instruction as the starting point to form a backward data dependency chain; merging the forward and backward data dependency chains to obtain a first data dependency chain; identifying instructions to be removed from the first data dependency chain based on control flow jump relationships; removing instructions to be removed from the first offset instruction stream; and extracting missing context instruction fragments from the historical instruction cache of the PlantDGP node according to the first data dependency chain to fill in the missing context instruction fragments, thereby obtaining the instruction stream to be verified.

[0010] In a preferred embodiment, the context reconstruction of the instruction stream to be verified to obtain a first trusted instruction set specifically involves: extracting the timestamp of each instruction in the instruction stream to be verified within a preset execution cycle, and sorting the instructions according to the order of the timestamps to obtain a first instruction queue; identifying the time interval between adjacent instructions in the first instruction queue, and if the time interval is greater than a preset instruction gap threshold, inserting a no-operation instruction between adjacent instructions to obtain a second instruction queue; performing an instruction address space scan on the second instruction queue, and identifying address anomaly regions based on the scan results; reallocating continuous virtual address space to the address anomaly regions, and migrating the instructions in the address anomaly regions to the newly allocated virtual address space to obtain migrated instructions; merging the migrated instructions with the instructions in the first instruction queue that have not experienced anomalies to obtain the first trusted instruction set.

[0011] In a preferred embodiment, the step of extracting the backtracking features of each instruction in the first trusted instruction set and verifying the preset instruction execution path based on the backtracking features to obtain the second trusted instruction set specifically involves: traversing each instruction in the first trusted instruction set, extracting the source register set and destination register set corresponding to the instruction, and recording the changes in register states before and after instruction execution; constructing the reverse instruction sequence for each instruction based on the changes in register states, and comparing the reverse instruction sequence bit by bit with the preset instruction template library of the PlantDGP node; identifying the first path instruction and the second path instruction based on the comparison results, and moving the second path instruction into the pending instruction buffer pool; performing a second backtracking on the second path instruction in the pending instruction buffer pool to obtain the second path instruction verified by the second backtracking; merging the second path instruction verified by the second backtracking with the first path instruction, and reorganizing it according to the preset arrangement order of the first trusted instruction set to obtain the second trusted instruction set.

[0012] In a preferred embodiment, the step of mirroring the second trusted instruction set and generating an execution permission signal based on the mirroring result specifically involves: loading the second trusted instruction set in a preset isolated mirroring execution environment to simulate the actual node hardware state and input data; collecting the instruction execution sequence, resource access records, and output responses during the mirroring process in real time to obtain a mirroring execution log; comparing the mirroring execution log with the expected standard output instruction by instruction to identify whether there are abnormal jumps or illegal accesses; if the mirroring result is consistent with the expected standard output, generating a positive execution permission signal; otherwise, generating a rejection signal and recording the abnormal instruction.

[0013] In a preferred embodiment, binding the execution permission signal with the second trusted instruction set and sending it to the PlantDGP node for instruction execution specifically involves: obtaining the timestamp and node identifier from the execution permission signal, and appending the timestamp and node identifier to the instruction sequence header of the second trusted instruction set to form an instruction packet to be sent; encrypting the instruction packet into segments, and adding a cyclic redundancy check code to each segment to obtain an encrypted instruction segment sequence; encapsulating the encrypted instruction segment sequence into packets according to a preset network transmission protocol between PlantDGP nodes, and embedding the sequence number of the current segment and the total number of segments into each data packet; sending data packets sequentially, and waiting for the node to return a reception confirmation signal during the sending process; and recording the completion log of this instruction sending after all data packets have received a reception confirmation signal.

[0014] The secure authentication and trusted execution method for the PlantDGP cooperative control instruction set provided by this invention has at least the following technical effects and advantages: This invention acquires the interaction control commands between nodes and extracts the energy fluctuation characteristics of the commands based on hardware execution state parameters, capturing minute changes in the transmission and execution of commands between nodes from a physical execution perspective. Subsequently, it identifies the phase offset regions of the commands based on the energy fluctuation characteristics and reconstructs the context of the corresponding command stream to be verified, obtaining a first trusted instruction set, thereby restoring the command order and execution semantics that may have been rearranged or lost. Next, it extracts the backtracking features of each command in the first trusted instruction set and verifies the preset command execution path based on these features, generating a second trusted instruction set to ensure the legality and consistency of the command sequence on the execution path. Then, it loads the second trusted instruction set into an isolated mirror environment for mirror operation, simulating the actual node hardware state and input conditions, collecting the execution process in real time and generating execution permission signals to verify command execution behavior and detect anomalies. Finally, it binds the execution permission signals with the second trusted instruction set and sends them to the PlantDGP node for execution, ensuring the reliability and security of command execution between nodes. By taking the above steps, we can effectively prevent multi-level compilation chains or intermediate tools from disrupting the order of fields when processing instructions, improve the reliability of instructions and the reliability of collaborative execution, and ensure the accurate, efficient and safe execution of agricultural automation control tasks in a distributed node environment. Attached Figure Description

[0015] Figure 1 This is a flowchart illustrating a secure authentication and trusted execution method for a PlantDGP collaborative control instruction set according to the present invention.

[0016] Figure 2 A system diagram of a secure authentication and trusted execution method for a PlantDGP cooperative control instruction set according to the present invention is provided. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0018] Example 1, Figure 1 This invention presents a secure authentication and trusted execution method for the PlantDGP cooperative control instruction set, comprising the following steps: S1, obtain the interaction control commands between PlantDGP nodes, and extract the energy fluctuation characteristics of the interaction control commands based on the hardware execution status parameters; In this embodiment, the energy fluctuation characteristics of the interactive control commands are extracted based on the hardware execution state parameters, specifically as follows: Collect hardware execution status parameters of PlantDGP nodes when executing historical instructions, and align the hardware execution status parameters by time to obtain a hardware status timing sequence. The interactive control commands are matched with the hardware state timing sequence on the time axis to identify the hardware state change range corresponding to each interactive control command. Calculate the cumulative change in voltage fluctuation amplitude and the instantaneous rate of change in current fluctuation frequency within the range of hardware state changes; The time distribution curve of energy fluctuation is constructed based on the cumulative change and the instantaneous rate of change, and the energy fluctuation characteristics are extracted from the time distribution curve.

[0019] It should be noted that PlantDGP inter-node interactive control commands refer to control instructions initiated by one node to another in the PlantDGP collaborative control network to trigger, coordinate, or adjust the operational behavior of agricultural production equipment. These commands typically include opcodes, target node identifiers, control parameters, and execution timing identifiers, instructing the target node to perform control actions such as equipment start / stop, actuator adjustment, sensor data acquisition triggering, or status synchronization. During the execution of commands, hardware components such as processing units, register units, and bus interfaces will experience corresponding changes in their operational status. These changes can be collected through hardware monitoring interfaces to form hardware execution status parameters.

[0020] Hardware execution status parameters typically include changes in processor core power supply voltage, power supply current, instruction cycle usage, bus access activity intensity, and register read / write triggering. These parameters reflect the actual hardware-level operating status of the node during the execution of historical instructions, thus providing basic data for subsequent identification of energy changes caused during instruction execution.

[0021] Secondly, based on the unified clock counting mechanism within the node, the time stamp corresponding to the interactive control instruction when it enters the execution phase is recorded, and this time stamp is used as a reference for the start time of instruction execution. Subsequently, the hardware execution state parameters are arranged in the order of acquisition time to form a continuous hardware state timing sequence, and the time stamp corresponding to each state sampling point is retained in this sequence. On this basis, taking the start time of the interactive control instruction as the benchmark and combining the execution cycle range of the instruction in the processing unit, the state sampling points within the corresponding time interval are retrieved forward and backward in the hardware state timing sequence to determine the hardware state change range corresponding to the instruction execution activity. For multiple instructions executed consecutively, they are matched sequentially according to the order in which they enter the execution phase, so that each interactive control instruction can correspond to its change range in the hardware state timing sequence.

[0022] After determining the hardware state change range corresponding to a certain interactive control command, the voltage state data continuously collected within this range is analyzed segment by segment. By comparing the degree of voltage change between adjacent sampling points, the fluctuation amplitude formed by each voltage rise or fall is recorded, and all fluctuation amplitudes within this range are accumulated and statistically analyzed in chronological order to obtain the cumulative change in voltage fluctuation amplitude. At the same time, the periodic change data of current changes collected within the same change range is identified. By observing the changes in peaks and troughs of current between consecutive sampling points, the number of current fluctuations per unit time is counted, and the changing trend of the number of fluctuations in adjacent time periods is compared to obtain the rate of change of current fluctuation frequency within the current range, thus forming the instantaneous rate of change of current fluctuation frequency.

[0023] Furthermore, after obtaining the cumulative change in voltage fluctuation amplitude and the instantaneous rate of change in current fluctuation frequency, the cumulative degree of voltage change and the degree of change in current fluctuation frequency reflected at each sampling time point within the hardware state change interval are used as the sequential reference. This ensures that each time position can reflect the intensity of energy change at that moment. Subsequently, the energy change information is arranged continuously in chronological order, and the change trends in adjacent time periods are smoothly connected, so that voltage change and current fluctuation form a continuous change trajectory on the time axis, resulting in a time distribution curve.

[0024] Finally, after obtaining the time distribution curve of energy fluctuations, the change pattern of the curve on the time axis is segmented and analyzed. By identifying the positions in the curve where there are obvious increases, decreases or abrupt changes, the time segments where energy changes are more significant are determined. Subsequently, the amplitude, duration and continuity of the curve changes in each segment are statistically analyzed, and regions with more concentrated energy fluctuations or regions with relatively stable energy changes are identified based on the concentration of curve changes. On this basis, the characteristics of the amplitude, duration and density of changes reflected in each segment are summarized to obtain the characteristics of energy fluctuations.

[0025] S2, based on the energy fluctuation characteristics, identify the phase offset region of the interactive control command, obtain the command stream to be verified, and reconstruct the context of the command stream to be verified to obtain the first trusted command set; In this embodiment, the phase offset region of the interactive control command is identified based on energy fluctuation characteristics to obtain the command stream to be verified, specifically as follows: The energy fluctuation characteristics are mapped to the instruction execution cycle of the PlantDGP node, and the start point and end point of the energy anomaly are marked on the time axis of the instruction execution cycle. Using the energy anomaly start point and energy anomaly end point as boundaries, extract all instruction segments flowing through the PlantDGP node bus within the corresponding time period to obtain the first offset instruction stream. For each instruction in the first offset instruction stream, perform opcode parsing and operand tracing to obtain the data dependencies and control flow jump relationships between instructions; Based on data dependencies and control flow jump relationships, context instruction completion is performed on the first offset instruction stream to obtain the instruction stream to be verified.

[0026] It should be noted that, firstly, based on the instruction execution cycle records within the PlantDGP node, the entry and completion times of each instruction in the processing unit are identified, and a corresponding instruction execution cycle time sequence is formed accordingly. Then, the time positions recorded in the energy fluctuation characteristics are aligned with the aforementioned instruction execution cycle time sequence, enabling a positional mapping of energy changes on the instruction execution time axis. When energy changes begin to show a significant deviation from the normal execution state at a certain time position, that time position is determined as the energy anomaly starting point; when the abnormal change returns to the normal fluctuation range after a period of time, the corresponding recovery position is determined as the energy anomaly ending point.

[0027] It should be noted that the energy anomaly start point refers to the time point at which a significant abnormal deviation in hardware energy changes first occurs during instruction execution, representing the time boundary at which abnormal activity begins; while the energy anomaly termination point refers to the time point at which the abnormal energy change ends and returns to a normal fluctuation state, representing the time boundary at which abnormal activity ends. By marking these two boundary positions on the timeline of the instruction execution cycle, a clear abnormal time segment can be formed, which can be used to locate the range of instructions where execution deviations may occur.

[0028] After determining the start and end points of the energy anomaly, these two time points are used as time boundaries to record and filter the instruction activities transmitted on the PlantDGP node's internal bus within that time period. Specifically: By reading the instruction transmission log or instruction cache record at the node bus interface, all instruction fragments that were transmitted, decoded, or entered the execution phase within the aforementioned time segment are retrieved and arranged according to their order of appearance on the bus. All instructions within this time range are extracted to form a continuous instruction set. Since the time interval covered by this instruction set corresponds to the energy anomaly change interval, it may contain relevant instruction sequences that cause deviations or anomalies in the execution state. This instruction set is referred to as the first offset instruction stream.

[0029] The first offset instruction stream refers to a set of continuous instruction segments that appear and participate in the execution activities in the PlantDGP node bus during the energy anomaly time period. Its characteristic is that there is a temporal correspondence between this instruction set and the energy anomaly change, so it is used to analyze the source of instructions that may have execution offset or abnormal behavior.

[0030] After obtaining the first offset instruction stream, each instruction within it is parsed one by one. First, the opcode portion of the instruction encoding is read to identify the specific operation type corresponding to the instruction, such as data read, data write, arithmetic processing, or jump control. Then, the operand information contained in the instruction is parsed to determine the registers, memory addresses, or input data sources involved in the instruction. After parsing, the data read and write behaviors involved in each instruction are sequentially traced. When the data read by an instruction originates from the data location written by one or more previous instructions, a data transfer relationship is recorded between the instructions, thus forming a data dependency relationship between instructions. Simultaneously, the target address is identified for instructions with jump or branch control attributes. When an instruction changes the execution order of subsequent instructions and transfers the execution path to other address locations, the control flow jump relationship formed by the instruction is recorded. Data dependency refers to the sequential dependence between multiple instructions due to shared data read / write operations, meaning that the execution of a later instruction depends on the data result produced by the previous instruction. Control flow jump relationship, on the other hand, refers to the execution path changes formed by jump, branch, or call instructions, reflecting the transfer of program control path during instruction execution. Identifying these two types of relationships can further clarify the execution association structure between instructions within the first offset instruction stream.

[0031] In this embodiment, based on data dependencies and control flow jump relationships, context instruction completion is performed on the first offset instruction stream to obtain the instruction stream to be verified, specifically as follows: In the first offset instruction stream, each instruction is used as the starting point to trace back to the previous write instruction corresponding to its source operand, forming a forward data dependency chain. Starting with each instruction, trace backwards to the next read instruction corresponding to its destination operand to form a backward data dependency chain; The forward data dependency chain and the backward data dependency chain are merged to obtain the first data dependency chain; Based on the control flow jump relationship, identify the instructions to be removed in the first data dependency chain; Remove the instructions to be removed from the first offset instruction stream, and extract the missing context instruction fragments from the historical instruction cache of the PlantDGP node according to the first data dependency chain to fill them in, so as to obtain the instruction stream to be verified.

[0032] It should be noted that after obtaining the first offset instruction stream, each instruction is analyzed one by one. First, the source operands contained in the instruction are parsed to determine the source location of the data read during execution, such as register number or memory address. Then, starting from the position of the instruction in the first offset instruction stream, the instructions are searched backward in the order of arrangement, looking for instructions that perform write operations on the same register or memory address. When a write instruction that can provide a data source for the current instruction is found, it is recorded as the previous write instruction of the current instruction, and the process continues to use this write instruction as a new trace node, further searching backward in the same way for the write source corresponding to its own source operand, thus forming a data source path extending forward from the current instruction.

[0033] After tracing the source operand, each instruction in the first offset instruction stream is further analyzed to determine the location of the data written after the instruction's execution, such as the register or memory address. Then, starting from this instruction, the process proceeds sequentially, searching for instructions that read from the same register or memory address. When an instruction that reads the written result is found, it is recorded as the next read instruction after the current instruction. This read instruction then becomes the new tracing node, and the search continues in the same manner to find subsequent instructions that read or depend on this data, thus forming a data usage path extending from the current instruction. Connecting these consecutive data usage paths forms a backward data dependency chain.

[0034] It should be noted that a forward data dependency chain refers to a sequence of instruction relationships that traces the data source backward from a certain instruction, and is used to describe the historical path of data generation; while a backward data dependency chain refers to a sequence of instruction relationships that traces the data usage process backward from the same instruction, and is used to describe the propagation and usage of the data generated by that instruction in subsequent execution processes.

[0035] After obtaining the forward and backward data dependency chains, the two dependency chains are integrated using the current instruction as the connection node, specifically as follows: The upstream data source instructions recorded in the forward data dependency chain are arranged in their execution order, and the downstream data usage instructions recorded in the backward data dependency chain are arranged in their appearance order. Then, using the current instruction as the intermediate connection point, the forward data dependency chain, the current instruction, and the backward data dependency chain are continuously concatenated to form a complete instruction association path covering the entire process of data generation, data transmission, and data usage. During the integration process, duplicate instructions are merged while maintaining the original execution order between instructions. The instruction sequence obtained through the above integration is the first data dependency chain.

[0036] Furthermore, after forming the first data dependency chain, each instruction in the dependency chain is further analyzed in conjunction with the previously identified control flow jump relationships, specifically as follows: First, identify whether there are any execution path changes in the first data dependency chain caused by jump or branch instructions, and determine the instruction sequence traversed by the actual execution path based on the jump target address. Then, compare the instructions in the first data dependency chain with the actual execution path. If some instructions are found to be connected in the data dependency chain but their location is not within the address range traversed by the actual execution path, these instructions are determined to be invalid related instructions due to the branch not being triggered or the path not being executed. These instructions that do not belong to the actual execution path are marked as instructions to be removed to avoid these invalid instructions interfering with the subsequent instruction flow reconstruction process.

[0037] Finally, after identifying the instructions to be removed, they are first removed from the original instruction sequence according to their position in the first offset instruction stream, leaving only the parts related to the actual execution path and valid data dependencies. Then, an integrity check is performed on the first data dependency chain. If some data source instructions or intermediate transmission instructions are found to be missing in the current instruction sequence, the instruction records executed within the corresponding time period are retrieved from the historical instruction cache stored in the PlantDGP node based on the instruction address information recorded in the dependency chain. Instruction fragments that can complete the data source or data transmission relationship are extracted from these fragments. The extracted context instruction fragments are then inserted into their corresponding positions according to their original execution order, ensuring the continuity of data dependencies between instructions. This forms a structurally complete instruction sequence that reflects the execution path, which is the instruction stream to be verified.

[0038] In this embodiment, the instruction stream to be verified is reconstructed in context to obtain a first trusted instruction set, specifically as follows: Extract the timestamp of each instruction in the instruction stream to be verified within a preset execution cycle, and sort the instructions according to the order of the timestamps to obtain the first instruction queue; In the first instruction queue, the time interval between adjacent instructions is identified. If the time interval is greater than the preset instruction gap threshold, a no-operation instruction is inserted between adjacent instructions to obtain the second instruction queue. Perform an instruction address space scan on the second instruction queue and identify abnormal address regions based on the scan results; The system reallocates a contiguous virtual address space to the address error region and migrates the instructions within the address error region to the newly allocated virtual address space, thus obtaining the migrated instructions. The migrated instructions are merged with the instructions in the first instruction queue that have not experienced any exceptions to obtain the first trusted instruction set.

[0039] It should be noted that, firstly, the execution time identifier recorded for each instruction in the instruction stream to be verified when it enters the processing unit's execution phase is read. This time identifier is usually automatically generated by the node's internal clock counter during the instruction decoding or scheduling phase, and is used to indicate the actual entry time of the instruction within the execution cycle. By extracting the above time identifiers one by one from the instruction stream to be verified, the timestamp information corresponding to each instruction can be obtained. Then, using a preset execution cycle as the time range, the timestamps of all instructions are uniformly mapped to the same time interval, and the instructions are rearranged in ascending order of timestamps to ensure that the instruction order is consistent with their actual execution time. The instruction sequence formed after the above sorting is the first instruction queue.

[0040] After obtaining the first instruction queue, the timestamps between adjacent instructions are compared one by one according to the instruction order to determine the time interval between the two instructions during execution. When the time interval between a pair of adjacent instructions exceeds a preset instruction gap threshold, it indicates that there is an unrecorded or uncaptured execution cycle within that time period. To maintain the continuity of the instruction execution time structure, a placeholder no-operation instruction is inserted between the two instructions to fill the gap on the timeline. The no-operation instruction is only used to maintain the integrity of the execution time structure and does not participate in actual data processing. After checking and supplementing all time intervals in the first instruction queue in the above manner, a new instruction sequence is formed, which is the second instruction queue.

[0041] It should be noted that the first instruction queue refers to the original sequence of executed instructions arranged only in timestamp order; while the second instruction queue is a sequence of instructions with a continuous time structure obtained by adding no-operation instructions to eliminate time gaps based on the first instruction queue. The main difference is that the second instruction queue maintains the integrity of the execution cycle on the timeline.

[0042] After obtaining the second instruction queue, the instruction addresses corresponding to each instruction are read one by one, and a continuity check is performed according to the address order from low to high to determine the distribution of instructions in the address space. During the scan, if some instruction addresses are found to jump, be repeatedly occupied, or not match the predetermined instruction storage range, then the address distribution of that part is determined to be abnormal. At the same time, when multiple abnormal addresses appear consecutively or are concentrated in a certain address range, that address range is designated as an address abnormal region. By checking all instruction addresses in the second instruction queue in the above manner, regions in the address space where instructions are abnormally arranged or abnormally occupied can be identified.

[0043] After identifying the address anomaly region, a contiguous, unoccupied address segment is first selected within the available virtual address range of the node, and this segment is used as the new instruction storage location. Then, the instructions within the address anomaly region are read one by one in their original order in the second instruction queue, and each instruction is reassigned the aforementioned contiguous virtual address, thus restoring the previously scattered or abnormally arranged addresses to a continuous structure. After address allocation, the address identifier of the corresponding instruction is updated to the newly allocated virtual address, while maintaining the execution order and data association between instructions. Through this process, the instructions from the original address anomaly region are migrated as a whole to the new contiguous address segment, forming the migrated instruction set.

[0044] After migrating instructions from address-abnormal regions, the migrated instructions are integrated with the instructions in the first instruction queue that do not exhibit address abnormalities. Specifically, firstly, instructions in the first instruction queue with normal addresses that have not been marked as abnormal are retained and arranged according to their original execution order; then, the migrated instructions that have undergone address reallocation are inserted into their corresponding positions according to their original time positions in the second instruction queue, ensuring that all instructions maintain consistency in both time order and address structure. The instruction set formed after this merging and organization is the first trusted instruction set.

[0045] S3, extract the backtracking features of each instruction in the first trusted instruction set, and verify the preset instruction execution path based on the backtracking features to obtain the second trusted instruction set; In this embodiment, the backtracking features of each instruction in the first trusted instruction set are extracted, and the preset instruction execution path is verified based on the backtracking features to obtain the second trusted instruction set, specifically: Traverse each instruction in the first trusted instruction set, extract the source register set and destination register set corresponding to the instruction, and record the changes in register state before and after instruction execution; The reverse instruction sequence of each instruction is constructed based on the register state change, and the reverse instruction sequence is compared bit by bit with the instruction template library preset by the PlantDGP node. Based on the comparison results, the first path instruction and the second path instruction are identified, and the second path instruction is moved into the pending instruction buffer pool. Perform a second backtracking on the second path instruction in the pending instruction buffer pool to obtain the second path instruction verified by the second backtracking. The second path instruction for secondary backtracking verification is merged with the first path instruction, and then reorganized according to the preset arrangement order of the first trusted instruction set to obtain the second trusted instruction set.

[0046] It should be noted that after obtaining the first trusted instruction set, the instructions are traversed and parsed one by one according to their order in the instruction set. For each instruction, the operand information in its encoding structure is first read to identify the register locations where data needs to be read during execution and the register locations where data needs to be written after execution. The set of registers whose data is read is recorded as the source register set, and the set of registers whose data is written is recorded as the destination register set. After determining the above register sets, the data state records in the registers before and after instruction execution are read. The data state stored in the registers before instruction execution and the updated data state in the registers after instruction execution are recorded respectively, and the changes between the two are marked to form a register state change record.

[0047] The so-called register state change refers to the degree and direction of the change in the data state stored in the same register before and after the execution of a certain instruction. It is used to characterize the actual impact of the instruction on the register contents during the execution process, thereby reflecting the modification behavior of the instruction on the data state.

[0048] After obtaining the register state change record for each instruction, the execution process of the instruction is reversed based on the state changes it causes to the registers to determine the reverse operation that can produce the same register change result. Specifically, based on the register state changes, combined with the instruction's operation type and operand structure, a set of instruction sequences that can recover or explain the source of the register change is constructed in the opposite direction to the original execution order, thus forming a reverse instruction sequence.

[0049] The reverse instruction sequence refers to a set of instruction structures used to explain the source of data changes by reverse deducing the instruction execution logic based on register changes. The resulting reverse instruction sequence is then compared bit-by-bit with a pre-built instruction template library in the PlantDGP node. During the comparison, the instruction operation type, operand structure, and register usage are checked sequentially to ensure consistency with the standard instruction patterns recorded in the template library, thereby determining the rationality of the current instruction in the execution path.

[0050] It should be noted that when the reverse instruction sequence corresponding to a certain instruction can find a template record in the instruction template library that has the same structure or behavior, the instruction is determined to conform to the preset execution path rules and is marked as a first path instruction. When the reverse instruction sequence corresponding to a certain instruction cannot find a matching structure in the template library, or only partially has the same structure but not completely the same behavior, the instruction is determined to have uncertainty in the execution path and is marked as a second path instruction. For the parts identified as second path instructions, in order to avoid directly removing them and causing execution path breaks, they are uniformly transferred to the pending instruction buffer pool for temporary storage, so that their execution paths can be further backtracked and verified in subsequent steps.

[0051] After storing the second path instruction into the pending instruction buffer pool, each instruction in the buffer pool is subjected to execution path backtracking analysis again, specifically as follows: Starting from the instruction's position in the first trusted instruction set, and considering the data dependencies and control flow jump relationships between its preceding and following instructions, the upstream and downstream execution environments of the instruction are re-traced to confirm whether the instruction constitutes reasonable execution behavior under certain specific execution conditions or branch paths. When backtracking analysis reveals that although the instruction does not form a complete match in the template library, its execution logic has a reasonable data transfer relationship or control path relationship with the preceding and following instructions, it is determined to be a second-path instruction that has passed secondary backtracking verification. A first-path instruction refers to an instruction that can be confirmed as belonging to the standard execution path during the initial template comparison stage; while a second-path instruction refers to an instruction whose legality cannot be fully confirmed during the initial comparison stage and requires further backtracking analysis for confirmation. The difference between the two lies in the different stages of determining the legality of the path.

[0052] After completing the second backtracking verification of the second path instructions, the verified second path instructions are retrieved from the pending instruction buffer pool and integrated with the previously confirmed first path instructions. During integration, the first trusted instruction set is used as the original arrangement basis, maintaining the original structure of each instruction in terms of time and execution order. Then, the second path instructions, verified through the second backtracking, are re-inserted into their corresponding positions according to their original positions in the first trusted instruction set, allowing all confirmed valid instructions to re-form a complete execution sequence. The instruction set obtained after this process is the second trusted instruction set. The first trusted instruction set is mainly used for time structure correction and address structure adjustment, focusing on ensuring the integrity of the instruction sequence in terms of execution structure. The second trusted instruction set, on the other hand, is formed by further verifying the execution path validity based on the first trusted instruction set. Its focus is on confirming the trustworthiness of the instruction execution path, thereby providing a more reliable instruction foundation for subsequent trusted execution.

[0053] S4, perform mirror execution of the second trusted instruction set, and generate an execution license signal based on the mirror execution result; In this embodiment, the second trusted instruction set is mirrored and run, and an execution license signal is generated based on the mirroring result, specifically as follows: A second trusted instruction set is loaded in a pre-defined isolated image execution environment to simulate the actual node hardware state and input data; The image execution log is obtained by collecting the instruction execution sequence, resource access records, and output responses during the image's operation in real time. Compare the image execution log with the expected standard output instruction by instruction to identify any abnormal jumps or illegal accesses. If the image execution result matches the expected standard output, a positive execution permission signal is generated; otherwise, an execution rejection signal is generated and the abnormal instruction is recorded.

[0054] It should be noted that the preset isolated image execution environment refers to a virtualized execution space independent of the actual PlantDGP node control hardware. This environment can completely reproduce the node hardware's processing logic, register states, bus access characteristics, and input / output interfaces, while maintaining physical isolation from the real execution environment to ensure that it does not affect the actual hardware operation during execution. After loading the second trusted instruction set, the hardware state of the actual node is simulated in this environment, including the processor's operating state, register initial values, memory unit data, and bus load conditions. Input data corresponding to the real node is provided to the environment so that the instruction set can execute under conditions consistent with the actual operating logic.

[0055] After the second trusted instruction set is loaded and execution begins in the image environment, the entire instruction execution process is continuously monitored. Specifically, this involves recording the execution order of each instruction, register reads and writes during execution, memory and bus access operations, and the output response of the execution results, while also marking abnormal events or delays. The collected data is then organized chronologically to form the image execution log.

[0056] After obtaining the image execution log, the execution result, register changes, and resource access information of each instruction in the log are compared line by line with the predefined standard output. The comparison includes whether the instruction execution order matches the expectation, whether the register write values ​​conform to the standard results, whether memory or bus resource accesses exceed the allowed range, and whether there are any unexpected jumps in the control flow. Through this instruction-by-instruction comparison, abnormal jumps or illegal access behaviors that occur during image execution can be identified—that is, instruction behaviors that deviate from the standard execution path or violate resource access rules.

[0057] Finally, after completing the instruction-by-instruction comparison, if the mirrored execution log is completely consistent with the expected standard output—that is, the execution order, register state changes, and resource accesses of all instructions conform to the standard—a positive execution permission signal is generated, allowing subsequent instructions to be executed on the node. If there is a deviation in the comparison result—that is, abnormal jumps, illegal accesses, or output results that do not conform to the standard are found—an execution denial signal is generated, and the abnormal instruction is recorded in the log for subsequent analysis and processing. Through the above processing, it can be ensured that an instruction is granted execution permission only when its execution behavior is completely consistent with the preset standard, thereby achieving reliable control over node execution.

[0058] S5 binds the execution permission signal to the second trusted instruction set and sends it to the PlantDGP node for instruction execution.

[0059] In this embodiment, the execution permission signal is bound to the second trusted instruction set and sent to the PlantDGP node for instruction execution, specifically as follows: Obtain the timestamp and node identifier from the execution permission signal, and append the timestamp and node identifier to the instruction sequence header of the second trusted instruction set to form an instruction packet to be issued; The instruction packet to be sent is encrypted in segments, and a cyclic redundancy check code is added to each segment to obtain the encrypted instruction segment sequence. The encrypted instruction segment sequence is divided into packets and encapsulated according to the preset network transmission protocol between PlantDGP nodes, and the sequence number of the current segment and the total number of segments are embedded in each data packet; Data packets are sent sequentially, and the system waits for a reception confirmation signal from the node during the sending process. After all data packets have received a confirmation signal, record the completion log of this instruction issuance.

[0060] It should be noted that after generating the execution permission signal, the timestamp information and the corresponding PlantDGP node identifier contained in the signal are first extracted. The timestamp is used to identify the generation time of the permission signal, and the node identifier is used to indicate the target node for the instruction. Subsequently, the extracted timestamp and node identifier are appended to the beginning of the instruction sequence in the second trusted instruction set, so that the instruction sequence contains verification information for the node's execution permission before transmission, thus forming a complete instruction packet to be sent. By embedding the timestamp and node identifier at the beginning of the instruction sequence, it can be ensured that the node can verify the legality of the instruction and confirm its target execution environment when receiving the instruction, thereby achieving the binding of the instruction and the permission signal.

[0061] After the instruction packet to be sent is generated, the entire instruction packet is divided into several instruction segments according to a preset length. Each segment is encrypted to ensure that the instruction content is not read or tampered with by unauthorized nodes during transmission. Simultaneously, a cyclic redundancy check (CRC) code is attached to each segment to verify data integrity and detect errors during transmission at the instruction receiving end. After encryption and checksum processing, a complete sequence of encrypted instruction segments is obtained, ensuring that each segment possesses both data confidentiality and verifiability.

[0062] After generating the encrypted instruction segment sequence, each segment is encapsulated according to the preset network transmission protocol between PlantDGP nodes. During encapsulation, the sequence number of the current segment and the total number of instruction segments are embedded in each data packet, so that the receiving node can reassemble the instruction segments in order and confirm whether all have been received. Through packet encapsulation and sequence number labeling, the instructions can be accurately reassembled during network transmission, while also facilitating the nodes' verification and management of data packet integrity.

[0063] After packet encapsulation, each data packet is sent to the target node sequentially according to its sequence number. During transmission, after each data packet is sent, further transmission is paused, awaiting a reception acknowledgment signal from the node to confirm successful and error-free reception of the segment. The next segment is only sent after receiving the node's acknowledgment signal, ensuring the reliability and traceability of the entire command delivery process and preventing packet loss or out-of-order delivery.

[0064] Finally, after confirming that all segments in the instruction packet have received the reception confirmation signal from the node, the instruction issuance process is recorded to generate a completion log. The completion log includes information such as the instruction packet identifier, node identifier, issuance time, number of segments, and reception status of each segment, which is used for subsequent auditing, tracking, and anomaly troubleshooting. By recording the completion log, the traceability and integrity of the instruction issuance process can be ensured, providing a reliable operational basis for node execution.

[0065] Example 2, Figure 2 This invention presents a secure authentication and trusted execution system for the PlantDGP cooperative control instruction set, comprising a data acquisition module, an instruction reconstruction module, an instruction verification module, an image execution module, and an instruction execution module. The data acquisition module is used to acquire the interaction control commands between PlantDGP nodes and extract the energy fluctuation characteristics of the interaction control commands based on the hardware execution status parameters. The instruction reconstruction module is used to identify the phase offset region of the interactive control instruction based on the energy fluctuation characteristics, obtain the instruction stream to be verified, and reconstruct the context of the instruction stream to be verified to obtain the first trusted instruction set. The instruction verification module is used to extract the backtracking features of each instruction in the first trusted instruction set, and to verify the preset instruction execution path based on the backtracking features to obtain the second trusted instruction set. The image execution module is used to perform image execution of the second trusted instruction set and generate an execution license signal based on the image execution result; The instruction execution module is used to bind the execution permission signal with the second trusted instruction set and send it to the PlantDGP node for instruction execution.

[0066] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0067] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0068] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0069] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0070] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A secure authentication and trusted execution method for PlantDGP cooperative control instruction set, characterized in that, Includes the following steps: Obtain the interaction control commands between PlantDGP nodes, and extract the energy fluctuation characteristics of the interaction control commands based on the hardware execution status parameters; Based on the energy fluctuation characteristics, the phase offset region of the interactive control command is identified, the command stream to be verified is obtained, and the context of the command stream to be verified is reconstructed to obtain the first trusted command set. Extract the backtracking features of each instruction in the first trusted instruction set, and verify the preset instruction execution path based on the backtracking features to obtain the second trusted instruction set; The second trusted instruction set is mirrored and run, and an execution license signal is generated based on the mirroring results. The execution permission signal is bound to the second trusted instruction set and sent to the PlantDGP node for instruction execution.

2. The secure authentication and trusted execution method for the PlantDGP cooperative control instruction set according to claim 1, characterized in that, The extraction of energy fluctuation characteristics of interactive control commands based on hardware execution state parameters specifically includes: Collect hardware execution status parameters of PlantDGP nodes when executing historical instructions, and align the hardware execution status parameters by time to obtain a hardware status timing sequence. The interactive control commands are matched with the hardware state timing sequence on the time axis to identify the hardware state change range corresponding to each interactive control command. Calculate the cumulative change in voltage fluctuation amplitude and the instantaneous rate of change in current fluctuation frequency within the range of hardware state changes; The time distribution curve of energy fluctuation is constructed based on the cumulative change and the instantaneous rate of change, and the energy fluctuation characteristics are extracted from the time distribution curve.

3. The secure authentication and trusted execution method for the PlantDGP cooperative control instruction set according to claim 2, characterized in that, The phase offset region of the interactive control command is identified based on energy fluctuation characteristics to obtain the command stream to be verified, specifically as follows: The energy fluctuation characteristics are mapped to the instruction execution cycle of the PlantDGP node, and the start point and end point of the energy anomaly are marked on the time axis of the instruction execution cycle. Using the energy anomaly start point and energy anomaly end point as boundaries, extract all instruction segments flowing through the PlantDGP node bus within the corresponding time period to obtain the first offset instruction stream. For each instruction in the first offset instruction stream, perform opcode parsing and operand tracing to obtain the data dependencies and control flow jump relationships between instructions; Based on data dependencies and control flow jump relationships, context instruction completion is performed on the first offset instruction stream to obtain the instruction stream to be verified.

4. The secure authentication and trusted execution method for the PlantDGP cooperative control instruction set according to claim 3, characterized in that, The step of performing context instruction completion on the first offset instruction stream based on data dependencies and control flow jump relationships to obtain the instruction stream to be verified is as follows: In the first offset instruction stream, each instruction is used as the starting point to trace back to the previous write instruction corresponding to its source operand, forming a forward data dependency chain. Starting with each instruction, trace backwards to the next read instruction corresponding to its destination operand to form a backward data dependency chain; The forward data dependency chain and the backward data dependency chain are merged to obtain the first data dependency chain; Based on the control flow jump relationship, identify the instructions to be removed in the first data dependency chain; Remove the instructions to be removed from the first offset instruction stream, and extract the missing context instruction fragments from the historical instruction cache of the PlantDGP node according to the first data dependency chain to fill them in, so as to obtain the instruction stream to be verified.

5. The secure authentication and trusted execution method for the PlantDGP cooperative control instruction set according to claim 4, characterized in that, The context reconstruction of the instruction stream to be verified to obtain the first trusted instruction set is specifically as follows: Extract the timestamp of each instruction in the instruction stream to be verified within a preset execution cycle, and sort the instructions according to the order of the timestamps to obtain the first instruction queue; In the first instruction queue, the time interval between adjacent instructions is identified. If the time interval is greater than the preset instruction gap threshold, a no-operation instruction is inserted between adjacent instructions to obtain the second instruction queue. Perform an instruction address space scan on the second instruction queue and identify abnormal address regions based on the scan results; The system reallocates a contiguous virtual address space to the address error region and migrates the instructions within the address error region to the newly allocated virtual address space to obtain the migrated instructions. The migrated instructions are merged with the instructions in the first instruction queue that have not experienced any exceptions to obtain the first trusted instruction set.

6. The secure authentication and trusted execution method for the PlantDGP cooperative control instruction set according to claim 5, characterized in that, The process involves extracting backtracking features from each instruction in the first trusted instruction set and verifying the preset instruction execution path based on these backtracking features to obtain the second trusted instruction set. Specifically: Traverse each instruction in the first trusted instruction set, extract the source register set and destination register set corresponding to the instruction, and record the changes in register state before and after instruction execution; The reverse instruction sequence of each instruction is constructed based on the register state change, and the reverse instruction sequence is compared bit by bit with the instruction template library preset by the PlantDGP node. Based on the comparison results, the first path instruction and the second path instruction are identified, and the second path instruction is moved into the pending instruction buffer pool. Perform a second backtracking on the second path instruction in the pending instruction buffer pool to obtain the second path instruction verified by the second backtracking. The second path instruction for secondary backtracking verification is merged with the first path instruction, and then reorganized according to the preset arrangement order of the first trusted instruction set to obtain the second trusted instruction set.

7. The secure authentication and trusted execution method for the PlantDGP cooperative control instruction set according to claim 6, characterized in that, The step of mirroring the second trusted instruction set and generating an execution license signal based on the mirroring result is as follows: A second trusted instruction set is loaded in a pre-defined isolated image execution environment to simulate the actual node hardware state and input data; The image execution log is obtained by collecting the instruction execution sequence, resource access records, and output responses during the image's operation in real time. Compare the image execution log with the expected standard output instruction by instruction to identify any abnormal jumps or illegal accesses. If the image execution result matches the expected standard output, a positive execution permission signal is generated; otherwise, an execution rejection signal is generated and the abnormal instruction is recorded.

8. The secure authentication and trusted execution method for the PlantDGP cooperative control instruction set according to claim 7, characterized in that, The process of binding the execution permission signal with the second trusted instruction set and sending it to the PlantDGP node for instruction execution is as follows: Obtain the timestamp and node identifier from the execution permission signal, and append the timestamp and node identifier to the instruction sequence header of the second trusted instruction set to form an instruction packet to be issued; The instruction packet to be sent is encrypted in segments, and a cyclic redundancy check code is added to each segment to obtain the encrypted instruction segment sequence. The encrypted instruction segment sequence is divided into packets and encapsulated according to the preset network transmission protocol between PlantDGP nodes, and the sequence number of the current segment and the total number of segments are embedded in each data packet; Data packets are sent sequentially, and the system waits for a reception confirmation signal from the node during the sending process. After all data packets have received a confirmation signal, record the completion log of this instruction issuance.

9. A secure authentication and trusted execution system for the PlantDGP collaborative control instruction set, applied to the secure authentication and trusted execution method for the PlantDGP collaborative control instruction set as described in any one of claims 1-8, characterized in that, It includes a data acquisition module, an instruction reconstruction module, an instruction verification module, an image execution module, and an instruction execution module: The data acquisition module is used to acquire the interaction control commands between PlantDGP nodes and extract the energy fluctuation characteristics of the interaction control commands based on the hardware execution status parameters. The instruction reconstruction module is used to identify the phase offset region of the interactive control instruction based on the energy fluctuation characteristics, obtain the instruction stream to be verified, and reconstruct the context of the instruction stream to be verified to obtain the first trusted instruction set. The instruction verification module is used to extract the backtracking features of each instruction in the first trusted instruction set, and to verify the preset instruction execution path based on the backtracking features to obtain the second trusted instruction set. The image execution module is used to perform image execution of the second trusted instruction set and generate an execution license signal based on the image execution result; The instruction execution module is used to bind the execution permission signal with the second trusted instruction set and send it to the PlantDGP node for instruction execution.

Images