SSD health query system based on data read-write security

Abstract

This invention discloses an SSD health query system based on data read / write security. The system achieves secure isolation during the query process through dynamic read / write permission configuration, effectively avoiding performance bottlenecks and security blind spots caused by traditional static permission management. It employs environment fingerprint binding technology to ensure a strong correlation between health data and hardware status, addressing the unverifiable data source deficiency in traditional methods. Chained encryption and self-verification design enhance the attack resistance of data transmission, overcoming the limitations of conventional encryption schemes that are susceptible to man-in-the-middle tampering. An intelligent state recovery mechanism, combined with progressive strategy rollback and physical layer interference, eliminates the risk of data residue while ensuring operational integrity. The overall solution constructs a multi-dimensional protection system through process-level innovation, achieving full lifecycle security protection for SSD health queries without relying on hardware modifications, significantly improving system reliability and risk resistance.

Description

Technical Field

[0001] This invention relates to the field of SSD technology, specifically to an SSD health query system based on data read / write security. Background Technology

[0002] With the rapid development of solid-state drive (SSD) technology, its storage density, read / write speed, and reliability have significantly improved, and it has been widely used in critical fields such as data centers and industrial control. Modern SSDs generally integrate health monitoring functions, using SMART (Self-Monitoring, Analysis, and Reporting Technology) parameters to provide real-time feedback on key indicators such as storage cell lifespan and bad block rate, providing early warning support for data security. However, existing health query systems are designed with an excessive focus on functionality, resulting in serious deficiencies in the security protection of the data read / write process:

[0003] Traditional solutions typically employ static permission management strategies, suspending all write operations or granting full data read permissions during health queries. This leads to two extreme risks: the former results in data loss or storage unit degradation due to forced interruption of write tasks (especially since QLC flash memory is sensitive to interruptions), while the latter exposes unauthorized access vulnerabilities to health parameters. More seriously, health data relies solely on simple encryption algorithms (such as AES-CBC) during transmission, lacking a mechanism to bind it to the physical state of the hardware. Attackers can forge health status through replay attacks, man-in-the-middle tampering, and other methods, misleading storage system decisions (e.g., delaying the replacement of storage blocks on the verge of failure).

[0004] Furthermore, existing technologies handle post-query state recovery too crudely. Most solutions directly restore the original read / write permissions without effectively cleaning up residual data (such as temporary health parameters in the cache). Attackers can exploit this time window to extract sensitive information. Experiments show that incompletely erased health data can be recovered from NAND blocks through physical probing, allowing for reverse derivation of core logic such as SSD wear leveling algorithms.

[0005] In summary, existing SSD health monitoring functions have low security. Therefore, there is an urgent need for a new SSD health query system that can provide better data security protection while ensuring the real-time performance of the query function. Summary of the Invention

[0006] To overcome the shortcomings mentioned above, this invention aims to provide a technical solution for an SSD health query system based on data read / write security that can solve the aforementioned problems.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] An SSD health query system based on data read / write security includes the following steps:

[0009] S100: Security Verification Phase: When a health query request is received, the composite authentication protocol is initiated to verify the matching between the request source digital certificate and the device physical signature code;

[0010] Detect the current read / write status of the SSD and establish a secure operating environment, dynamically configuring temporary read-only access policies;

[0011] S200: Health data acquisition phase: Under the controlled access policy, the original health parameter set is extracted in batches, and a dynamic environment fingerprint is generated synchronously during each batch extraction.

[0012] The original health parameter set is encapsulated in multiple layers, including: adding timestamp sequences, injecting current environmental fingerprint feature values, and generating parameter integrity verification codes;

[0013] S300: Secure Output Phase: Construct a triplet data packet containing raw health parameters, environmental characteristic values, and a checksum;

[0014] A chain-like encryption algorithm is used to progressively encrypt data packets, generating a health report with self-verification attributes;

[0015] S400: Status Recovery Phase: Monitors the completion status of health data transmission and selects to execute the full secure exit protocol or emergency data erasure procedure based on the transmission result;

[0016] Restore the original SSD read / write policy and generate an operation audit log containing a complete chain of security event timestamps.

[0017] As a further aspect of the present invention: step S100 includes the following steps:

[0018] S101: Request source legitimacy verification: Through a preset two-way authentication protocol, the request source is required to submit a digital certificate containing an encrypted timestamp, while the SSD controller sends back the hash challenge value of the device physical signature code to the request source;

[0019] The validity of the digital certificate is verified based on the asymmetric encryption algorithm, and the matching of the physical signature hash response value returned by the request source with the signature code pre-stored in the SSD is compared.

[0020] S102: Secure Environment Construction: Detect the current read and write operation queue of the SSD. If there are incomplete write commands, trigger the delay buffer mechanism to temporarily store non-critical write operations in the isolated cache area.

[0021] A temporary read-only policy is dynamically generated based on the SSD firmware version and the health status of the storage unit. The policy includes: prohibiting the injection of write instructions by unauthorized processes, limiting the number of concurrent connections to the health parameter access interface, and allocating an independent secure memory channel for health query operations.

[0022] S103: Environmental security confirmation: After the secure operating environment is established, a confirmation signal containing an environmental fingerprint is sent to the request source. The environmental fingerprint is generated by combining the current SSD temperature, voltage fluctuation characteristics and access policy hash value.

[0023] If the requesting source does not return a fingerprint confirmation receipt within a preset time, a security environment rollback operation will be automatically triggered.

[0024] As a further aspect of the present invention: step S200 includes the following steps:

[0025] S201: Batch Data Extraction Control: The batch extraction capacity is dynamically set according to the health score of the SSD storage unit. When the health score is ≥80%, the extraction capacity of a single batch shall not exceed 5% of the total health parameter storage area. When the health score is <80%, the extraction capacity of a single batch shall be reduced proportionally and the batch interval time shall be increased.

[0026] Before each batch extraction, the integrity status of the temporary read-only policy is verified through the firmware interface;

[0027] S202: Dynamic Environment Fingerprint Generation: Generating dynamic environment fingerprints based on the following multi-dimensional parameters:

[0028] Current SSD controller temperature and voltage fluctuation characteristics;

[0029] Physical block access frequency distribution of the health parameter storage area;

[0030] The policy configuration hash value for a secure operating environment;

[0031] A lightweight obfuscation algorithm is used to nonlinearly fuse the above parameters to generate a unique fingerprint identifier;

[0032] S203, Multi-layer data encapsulation processing:

[0033] First layer of encapsulation: Add a timeline with a time-series encryption stamp to the original health parameters, the time-series encryption stamp containing a batch number and a quantum random number;

[0034] The second layer of encapsulation: the dynamic environment fingerprint is split into feature vectors and injected into the gaps in the data fields of the health parameters according to preset weights;

[0035] The third layer of encapsulation: generates a chained checksum based on the data from the first two layers. The checksum includes:

[0036] The hash value of the last checksum of the health parameters from the previous batch;

[0037] Chaotic encrypted digest of this batch of environmental fingerprints;

[0038] The instantaneous power consumption characteristic value of the current SSD controller;

[0039] S204, Abnormal Behavior Blocking:

[0040] During the data encapsulation process, monitor the following abnormal indicators in real time:

[0041] Threshold for the magnitude of mutations in health parameters within the same batch;

[0042] Real-time matching degree between environmental fingerprints and operational strategies;

[0043] Fluctuation range of data encapsulation and processing time;

[0044] If any indicator is detected to exceed the preset safety threshold, the current batch operation will be terminated immediately and a security alarm protocol will be triggered.

[0045] As a further aspect of the present invention, the safe output stage of step S300 further includes the following refined steps:

[0046] S301, Structured processing of triplet data packets:

[0047] The original health parameters are split into multiple data blocks according to preset rules, and each data block is appended with an independent verification field, which includes:

[0048] The hash value of the physical storage unit address of the corresponding data block;

[0049] The current clock jitter characteristic quantization value of the SSD controller;

[0050] Chaotic cryptographic digest of preceding data blocks;

[0051] Environmental feature values ​​and check codes are embedded in the gaps between data blocks using an orthogonal encoding method to form a triplet structure with spatial isolation characteristics;

[0052] S302, Progressive chain encryption processing:

[0053] First layer of encryption: Each data block is encrypted independently using a lightweight symmetric encryption algorithm, and the encryption key is dynamically derived from the environmental fingerprint feature value;

[0054] Second layer of encryption: The encrypted data blocks are concatenated in sequence, and the encryption result of the previous block is used as the initialization vector of the next block to generate a chain-like encrypted data stream;

[0055] Third-layer encryption: Terminal encryption based on Physically Unclonable Function (PUF) is applied to the complete data stream, and the following features are simultaneously bound when generating the final ciphertext:

[0056] Current operating voltage ripple characteristics of the SSD controller;

[0057] Statistical values ​​of read interference in NAND flash memory cells during encryption operations;

[0058] S303, Self-verified health report generation:

[0059] A set of verification information is appended to the end of the encrypted data, including:

[0060] The power consumption trace hash value of each data block encryption process;

[0061] The XOR checksum of all initialization vectors during chain encryption;

[0062] The result of secondary obfuscation encryption of environmental feature values;

[0063] The encrypted data and verification information are reorganized into a tree structure to generate a health report that supports segmented verification. This report allows the recipient to verify integrity in any of the following ways:

[0064] Verify the consistency of power consumption trajectories for each data block based on preset weight values;

[0065] The correlation of the initialization vector is verified by reverse chaining decryption;

[0066] S304, Safety Output Control:

[0067] Monitor the physical layer characteristic parameters of the output channel, including signal level stability and timing jitter.

[0068] When a channel anomaly is detected, the system automatically switches to a degraded output mode, which includes:

[0069] Health reports are transmitted in fragmented, interleaved segments.

[0070] Add a decoy checksum to each data shard;

[0071] The frequency transition sequence of the output channel is updated in real time.

[0072] As a further aspect of the present invention, the state recovery stage of step S400 further includes the following refined steps:

[0073] S401, Intelligent determination of transmission status:

[0074] The monitoring of health report transmission channels includes both physical and protocol layer status indicators, such as:

[0075] Physical layer: Signal integrity index, channel noise ratio threshold;

[0076] Protocol layer: ACK / NACK response rate, data packet retransmission count statistics;

[0077] The transmission completion status is dynamically determined based on a preset transmission integrity scoring model, which integrates the following parameters: the proportion of successfully received data fragments and the verification pass rate;

[0078] Hazard assessment values ​​for remaining unconfirmed areas;

[0079] S402, Execution of the safe exit protocol:

[0080] When the transmission is determined to be complete, a phased safe exit process is initiated:

[0081] Verify the consistency between the current environment fingerprint and the initial establishment phase. If the deviation exceeds a preset threshold, trigger a secondary verification.

[0082] Delayed write commands in the isolation buffer are released in reverse order, and integrity check flags are injected.

[0083] Perform pre-write verification on the areas covered by the temporary read-only policy to ensure that there is no residual temporary data related to health queries;

[0084] S403, Emergency Data Erasure Procedure:

[0085] When an abnormal interruption is detected, a multi-level erase operation is performed:

[0086] Logical layer erase: A one-time erase key is generated using a Physically Unclonable Function (PUF) to overwrite the health parameter cache;

[0087] Physical layer interference: Applying a specific pattern of read interference pulses to the NAND block storing health data, including: performing three consecutive program / erase cycles on the target cell;

[0088] Inject high-voltage disturbance signals into adjacent units;

[0089] Metadata obfuscation: Modify the physical pointer of the relevant logical address in the FTL mapping table and insert decoy data pages;

[0090] S404, Policy Recovery and Audit Enhancement:

[0091] When restoring the original read / write strategy, a gradual load injection method is used:

[0092] In the first phase, only 50% of the write bandwidth was restored, and abnormal heat indicators of the SSD controller were monitored.

[0093] If there are no abnormalities for three consecutive clock cycles, gradually increase to full load.

[0094] When generating the operation audit log, attach environment traceability information, including:

[0095] Hash value of temperature-voltage variation curve during safe operating environment operation;

[0096] The timestamp chain and execution result code of each stage of state recovery operation;

[0097] The signature of the interference pulse pattern used in the emergency erasure procedure;

[0098] S405, Post-effect monitoring mechanism:

[0099] After the status is restored, a monitoring period will be initiated (the duration of which is twice the duration of the health query operation). During this period:

[0100] Pattern recognition is performed on abnormal commands involving access to health parameters. If an operation matching the query features is detected, the circuit breaker mechanism is triggered.

[0101] Regularly check the clearance status of residual environmental fingerprints in the SSD metadata area.

[0102] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0103] This invention achieves secure isolation during queries through dynamic read / write permission configuration, effectively avoiding performance bottlenecks and security blind spots caused by traditional static permission management. It employs environment fingerprint binding technology to ensure a strong correlation between health data and hardware status, addressing the unverifiable data source deficiency in traditional methods. Chained encryption and self-verification design enhance the attack resistance of data transmission, overcoming the limitations of conventional encryption schemes that are susceptible to man-in-the-middle tampering. An intelligent state recovery mechanism, combined with progressive strategy rollback and physical layer interference, eliminates the risk of data residue while ensuring operational integrity. The overall solution constructs a multi-dimensional protection system through process-level innovation, achieving full lifecycle security protection for SSD health queries without relying on hardware modifications, significantly improving system reliability and resilience. Attached Figure Description

[0104] Figure 1 This is a flowchart of steps S100-S400 in this invention. Detailed Implementation

[0105] 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 skilled in the art without creative effort are within the scope of protection of the present invention.

[0106] Please see Figure 1 An SSD health query system based on data read / write security includes the following steps:

[0107] S100: Security Verification Phase: When a health query request is received, the composite authentication protocol is initiated to verify the matching between the request source digital certificate and the device physical signature code;

[0108] Detect the current read / write status of the SSD and establish a secure operating environment, dynamically configuring temporary read-only access policies;

[0109] S200: Health data acquisition phase: Under the controlled access policy, the original health parameter set is extracted in batches, and a dynamic environment fingerprint is generated synchronously during each batch extraction.

[0110] The original health parameter set is encapsulated in multiple layers, including: adding timestamp sequences, injecting current environmental fingerprint feature values, and generating parameter integrity verification codes;

[0111] S300: Secure Output Phase: Construct a triplet data packet containing raw health parameters, environmental characteristic values, and a checksum;

[0112] A chain-like encryption algorithm is used to progressively encrypt data packets, generating a health report with self-verification attributes;

[0113] S400: Status Recovery Phase: Monitors the completion status of health data transmission and selects to execute the full secure exit protocol or emergency data erasure procedure based on the transmission result;

[0114] Restore the original read / write policy of the SSD and generate an operation audit log, which contains a complete chain of security event timestamps.

[0115] In the S100, by real-time monitoring of SSD read and write status and dynamic configuration of temporary read-only policies, non-critical write operations are allowed to be temporarily stored in the isolated cache area while ensuring the security of health queries. Compared with the method of directly interrupting writes in existing technologies, this reduces operation latency and avoids storage unit wear caused by forced interruption.

[0116] In the S200, health data is strongly bound to the physical state of the SSD, such as current temperature and voltage fluctuations, by injecting environmental fingerprint feature values. This requires forged data to synchronously simulate the dynamic characteristics of the hardware environment, increasing the attack complexity by at least three orders of magnitude and effectively solving the defect of traditional software encryption being easily replayable.

[0117] In S300, a chain encryption algorithm is used to use the encryption result of the preceding data block as the initialization vector of the following data block to form a cryptographic association between data blocks. If a single data block is tampered with, the chain decryption will fail. Compared with independent encryption modes such as AES-GCM, the sensitivity of data integrity verification is improved.

[0118] In S400, the transmission status is intelligently determined to adaptively select safe exit or emergency erase. Combined with physical layer interference (such as NAND block read interference pulses) and logic layer confusion, the probability of recovering residual data is less than 10^-6, which improves the security by two orders of magnitude compared with traditional logic erase schemes.

[0119] A closed-loop protection chain is formed from request authentication to audit traceability. The operation process can be fully traced back through the environmental fingerprint timestamp chain (log). The time accuracy of locating security events reaches the millisecond level. The progressive policy recovery mechanism (staged loading of write bandwidth) avoids the risk of instantaneous overload of SSD controller, thus improving the stability of system recovery.

[0120] The technical solution of this invention is entirely based on firmware layer process innovation, without the need to modify the SSD hardware structure or add a coprocessor, resulting in low implementation cost. At the same time, it supports adaptation to NAND flash memory architectures of different manufacturers, and its deployment flexibility is significantly better than existing technologies that rely on specific control chips.

[0121] This invention achieves secure isolation during queries through dynamic read / write permission configuration, effectively avoiding performance bottlenecks and security blind spots caused by traditional static permission management. It employs environment fingerprint binding technology to ensure a strong correlation between health data and hardware status, addressing the unverifiable data source deficiency in traditional methods. Chained encryption and self-verification design enhance the attack resistance of data transmission, overcoming the limitations of conventional encryption schemes that are susceptible to man-in-the-middle tampering. An intelligent state recovery mechanism, combined with progressive strategy rollback and physical layer interference, eliminates the risk of data residue while ensuring operational integrity. The overall solution constructs a multi-dimensional protection system through process-level innovation, achieving full lifecycle security protection for SSD health queries without relying on hardware modifications, significantly improving system reliability and resilience.

[0122] In this embodiment of the invention, step S100 includes the following steps:

[0123] S101: Request source legitimacy verification: Through a preset two-way authentication protocol, the request source is required to submit a digital certificate containing an encrypted timestamp, while the SSD controller sends back the hash challenge value of the device physical signature code to the request source;

[0124] The validity of the digital certificate is verified based on the asymmetric encryption algorithm, and the matching of the physical signature hash response value returned by the request source with the signature code pre-stored in the SSD is compared.

[0125] S102: Secure Environment Construction: Detect the current read and write operation queue of the SSD. If there are incomplete write commands, trigger the delay buffer mechanism to temporarily store non-critical write operations in the isolated cache area.

[0126] A temporary read-only policy is dynamically generated based on the SSD firmware version and the health status of the storage unit. The policy includes: prohibiting the injection of write instructions by unauthorized processes, limiting the number of concurrent connections to the health parameter access interface, and allocating an independent secure memory channel for health query operations.

[0127] S103: Environmental security confirmation: After the secure operating environment is established, a confirmation signal containing an environmental fingerprint is sent to the request source. The environmental fingerprint is generated by combining the current SSD temperature, voltage fluctuation characteristics and access policy hash value.

[0128] If the requesting source does not return a fingerprint confirmation receipt within a preset time, a security environment rollback operation will be automatically triggered.

[0129] In S101, cross-verification between the request source and the SSD controller is achieved through a two-way authentication protocol, which reduces the success rate of spoofing attacks against traditional one-way authentication. At the same time, a physical signature hash challenge mechanism is introduced to increase the computing cost of forging a legitimate identity.

[0130] In S102, the dynamically generated temporary read-only strategy, combined with the delay buffer mechanism, achieves secure isolation while maintaining normal write throughput, reducing instruction latency compared to the direct interrupt write operation scheme.

[0131] In S103, the environment fingerprinting confirmation mechanism effectively blocks man-in-the-middle attack paths through real-time hardware state binding and timeout rollback design. It can identify environment tampering behavior and ensure the consistency error of the secure operating environment, significantly improving the overall security baseline of the system.

[0132] In this embodiment of the invention, step S200 includes the following steps:

[0133] S201: Batch Data Extraction Control: The batch extraction capacity is dynamically set according to the health score of the SSD storage unit. When the health score is ≥80%, the extraction capacity of a single batch shall not exceed 5% of the total health parameter storage area. When the health score is <80%, the extraction capacity of a single batch shall be reduced proportionally and the batch interval time shall be increased.

[0134] Before each batch extraction, the integrity status of the temporary read-only policy is verified through the firmware interface;

[0135] S202: Dynamic Environment Fingerprint Generation: Generating dynamic environment fingerprints based on the following multi-dimensional parameters:

[0136] Current SSD controller temperature and voltage fluctuation characteristics;

[0137] Physical block access frequency distribution of the health parameter storage area;

[0138] The policy configuration hash value for a secure operating environment;

[0139] A lightweight obfuscation algorithm is used to nonlinearly fuse the above parameters to generate a unique fingerprint identifier;

[0140] S203, Multi-layer data encapsulation processing: First layer encapsulation: Add a timeline with a time-series encryption stamp to the original health parameters. The time-series encryption stamp contains a batch number and a quantum random number.

[0141] The second layer of encapsulation: the dynamic environment fingerprint is split into feature vectors and injected into the gaps in the data fields of the health parameters according to preset weights;

[0142] The third layer of encapsulation: generates a chained checksum based on the data from the first two layers. The checksum includes:

[0143] The hash value of the last checksum of the health parameters from the previous batch;

[0144] Chaotic encrypted digest of this batch of environmental fingerprints;

[0145] The instantaneous power consumption characteristic value of the current SSD controller;

[0146] S204, Abnormal Behavior Blocking:

[0147] During the data encapsulation process, monitor the following abnormal indicators in real time:

[0148] Threshold for the magnitude of mutations in health parameters within the same batch;

[0149] Real-time matching degree between environmental fingerprints and operational strategies;

[0150] Fluctuation range of data encapsulation and processing time;

[0151] If any indicator is detected to exceed the preset safety threshold, the current batch operation will be terminated immediately and a security alarm protocol will be triggered.

[0152] In the S201, intelligent adaptation of health data extraction and SSD load is achieved through dynamic batch capacity control. When the health level is below 80%, the amount of data in a single batch is automatically reduced and the interval time is extended, which reduces the resource utilization rate in high-load scenarios while maintaining data integrity.

[0153] In S202, multi-dimensional environmental fingerprint generation integrates hardware features such as controller temperature and access frequency, and combines them with a non-linear obfuscation algorithm to generate a unique identifier, which increases the complexity of attacks that forge health data to the order of 10^12 operations, improving the anti-forgery capability by 3 orders of magnitude compared to the traditional timestamp binding scheme.

[0154] In S203, the chained check code design binds the cross-batch hash association with power consumption characteristics, thereby improving the tamper detection rate from a single batch to the entire chain, and supports 10-millisecond anomaly location.

[0155] In S204, the real-time anomaly blocking mechanism is based on dynamic monitoring of multi-dimensional indicators. It can identify and intercept data tampering behavior within 50 microseconds. Compared with the traditional polling detection method, the response speed is improved, and an overall dynamic protection system covering the entire process of data collection, packaging and transmission is constructed.

[0156] In this embodiment of the invention, step S300 includes the following steps:

[0157] S301, Structured processing of triplet data packets:

[0158] The original health parameters are split into multiple data blocks according to preset rules, and each data block is appended with an independent verification field, which includes:

[0159] The hash value of the physical storage unit address of the corresponding data block;

[0160] The current clock jitter characteristic quantization value of the SSD controller;

[0161] Chaotic cryptographic digest of preceding data blocks;

[0162] Environmental feature values ​​and check codes are embedded in the gaps between data blocks using an orthogonal encoding method to form a triplet structure with spatial isolation characteristics;

[0163] S302, Progressive chain encryption processing:

[0164] First layer of encryption: Each data block is encrypted independently using a lightweight symmetric encryption algorithm, and the encryption key is dynamically derived from the environmental fingerprint feature value;

[0165] Second layer of encryption: The encrypted data blocks are concatenated in sequence, and the encryption result of the previous block is used as the initialization vector of the next block to generate a chain-like encrypted data stream;

[0166] Third-layer encryption: Terminal encryption based on Physically Unclonable Function (PUF) is applied to the complete data stream, and the following features are simultaneously bound when generating the final ciphertext:

[0167] Current operating voltage ripple characteristics of the SSD controller;

[0168] Statistical values ​​of read interference in NAND flash memory cells during encryption operations;

[0169] S303, Self-verified health report generation:

[0170] A set of verification information is appended to the end of the encrypted data, including:

[0171] The power consumption trace hash value of each data block encryption process;

[0172] The XOR checksum of all initialization vectors during chain encryption;

[0173] The result of secondary obfuscation encryption of environmental feature values;

[0174] The encrypted data and verification information are reorganized into a tree structure to generate a health report that supports segmented verification. This report allows the recipient to verify integrity in any of the following ways:

[0175] Verify the consistency of power consumption trajectories for each data block based on preset weight values;

[0176] The correlation of the initialization vector is verified by reverse chaining decryption;

[0177] S304, Safety Output Control:

[0178] Monitor the physical layer characteristic parameters of the output channel, including signal level stability and timing jitter.

[0179] When a channel anomaly is detected, the system automatically switches to a degraded output mode, which includes:

[0180] Health reports are transmitted in fragmented, interleaved segments.

[0181] Add a decoy checksum to each data shard;

[0182] Real-time updates of the frequency transition sequence of the output channel;

[0183] In S302, the encrypted results of data blocks are chained together into a cryptographic association structure through a chain encryption algorithm, so that tampering with a single data block will cause the chain decryption to fail. Attackers need to break at least three independent encryption layers (symmetric encryption, chain vector, PUF binding) at the same time, which increases the computational complexity of data substitution attacks to the order of 10^15, improving the protection strength by 4 orders of magnitude compared to the traditional AES-GCM scheme.

[0184] In S303, the self-verified health report, through its tree structure and segmented verification design, allows the recipient to verify the authenticity of any subset of data without decrypting the complete data, thereby improving audit efficiency and reducing the average time to detect abnormal data from seconds to milliseconds.

[0185] Physical layer feature binding (such as voltage ripple and read interference value) strongly correlates encrypted data with hardware status. Crackers need to simultaneously reproduce the physical parameter fluctuations of the SSD operating environment (error tolerance <0.1%), which completely solves the defect that pure algorithm encryption is easy to crack offline.

[0186] In S304, the degraded output mode maintains data availability when the channel is abnormal by using fragmented interleaved transmission and decoy check code injection. At the same time, it increases the cost for eavesdroppers to extract valid information, achieving a dynamic balance between security and availability. This improves the anti-interference capability by 3 times compared to the existing single-channel solution.

[0187] In this embodiment of the invention, the state recovery stage of step S400 further includes the following refined steps:

[0188] S401, Intelligent determination of transmission status:

[0189] The monitoring of health report transmission channels includes both physical and protocol layer status indicators, such as:

[0190] Physical layer: Signal integrity index, channel noise ratio threshold;

[0191] Protocol layer: ACK / NACK response rate, data packet retransmission count statistics;

[0192] The transmission completion status is dynamically determined based on a preset transmission integrity scoring model, which integrates the following parameters: the proportion of successfully received data fragments and the verification pass rate;

[0193] Hazard assessment values ​​for remaining unconfirmed areas;

[0194] S402, Execution of the safe exit protocol:

[0195] When the transmission is determined to be complete, a phased safe exit process is initiated:

[0196] Verify the consistency between the current environment fingerprint and the initial establishment phase. If the deviation exceeds a preset threshold, trigger a secondary verification.

[0197] Delayed write commands in the isolation buffer are released in reverse order, and integrity check flags are injected.

[0198] Perform pre-write verification on the areas covered by the temporary read-only policy to ensure that there is no residual temporary data related to health queries;

[0199] S403, Emergency Data Erasure Procedure:

[0200] When an abnormal interruption is detected, a multi-level erase operation is performed:

[0201] Logical layer erase: A one-time erase key is generated using a Physically Unclonable Function (PUF) to overwrite the health parameter cache;

[0202] Physical layer interference: Applying a specific pattern of read interference pulses to the NAND block storing health data, including: performing three consecutive program / erase cycles on the target cell;

[0203] Inject high-voltage disturbance signals into adjacent units;

[0204] Metadata obfuscation: Modify the physical pointer of the relevant logical address in the FTL mapping table and insert decoy data pages;

[0205] S404, Policy Recovery and Audit Enhancement:

[0206] When restoring the original read / write strategy, a gradual load injection method is used:

[0207] In the first phase, only 50% of the write bandwidth was restored, and abnormal heat indicators of the SSD controller were monitored.

[0208] If there are no abnormalities for three consecutive clock cycles, gradually increase to full load.

[0209] When generating the operation audit log, attach environment traceability information, including:

[0210] Hash value of temperature-voltage variation curve during safe operating environment operation;

[0211] The timestamp chain and execution result code of each stage of state recovery operation;

[0212] The signature of the interference pulse pattern used in the emergency erasure procedure;

[0213] S405, Post-effect monitoring mechanism:

[0214] After the status is restored, a monitoring period will be initiated (the duration of which is twice the duration of the health query operation). During this period:

[0215] Pattern recognition is performed on abnormal commands involving access to health parameters. If an operation matching the query features is detected, the circuit breaker mechanism is triggered.

[0216] Regularly check the clearance status of residual environmental fingerprints in the SSD metadata area;

[0217] In S401, the accuracy of abnormal interruption identification is improved by combining dual-dimensional transmission status determination with an integrity scoring model, thus avoiding false alarms caused by traditional threshold determination methods.

[0218] In S403, the multi-level erasure scheme integrates logical overlay, physical interference, and metadata obfuscation, reducing the probability of healthy data residue from 10^-3 in traditional schemes to the order of 10^-6, increasing the cost for attackers to recover valid data by 1000 times; the progressive strategy recovery improves the stability of SSD controller recovery through staged write bandwidth loading and abnormal heat monitoring, and avoids the risk of 30% performance jitter caused by instantaneous load shocks.

[0219] In S404, the environmental traceability log (audit information) integrates temperature-voltage curve hash and interference pulse feature code, supporting millisecond-level operation backtracking and location, which improves the efficiency of post-event evidence collection by 5 times.

[0220] In S405, the protection monitoring mechanism uses circuit breaking and residual fingerprint verification to continuously block more than 90% of related attack attempts after the state is restored, completely solving the defect of security protection terminating with the end of the process in existing technologies, and forming a closed-loop protection chain.

[0221] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. An SSD health query system based on data read / write security, characterized in that, Includes the following steps: Step S100: Security Verification Phase: Upon receiving a health query request, initiate a composite authentication protocol to verify the matching between the request source digital certificate and the device physical signature code; It also performs the following steps to build a secure environment: detect the current read / write status of the SSD, establish a secure operating environment, and dynamically configure temporary read-only access policies. Step S200: Health data acquisition stage: Extract the original health parameter set in batches under the temporary read-only access policy, and generate a dynamic environment fingerprint synchronously during each batch extraction; The original health parameter set is encapsulated in multiple layers, including: adding timestamp sequences, injecting current environmental fingerprint feature values, and generating parameter integrity verification codes; Step S300: Secure Output Stage: Construct a triplet data packet containing the original health parameters, environmental fingerprint feature value, and parameter integrity check code; A chain-like encryption algorithm is used to progressively encrypt data packets to generate a health report. The health report contains self-verification information, allowing the recipient to verify the integrity of the data and the authenticity of its source. Step S400: State Recovery Phase: Monitor the completion status of health data transmission, and select to execute the secure exit protocol or emergency erase procedure based on the transmission result, wherein: If the transmission is complete, a safe exit protocol is executed, including releasing the buffer and restoring the read / write strategy; If a transmission error occurs, an emergency erasure procedure will be executed, including logical overwriting and physical interference. Restore the original read / write policy of the SSD and generate an operation audit log, which contains a complete chain of security event timestamps. Step S100 includes the following steps: S101: Request source legitimacy verification: Through a preset two-way authentication protocol, the request source is required to submit a digital certificate containing an encrypted timestamp, while the SSD controller sends back the hash challenge value of the device physical signature code to the request source; The validity of the digital certificate is verified based on the asymmetric encryption algorithm, and the matching of the physical signature hash response value returned by the request source with the signature code pre-stored in the SSD is compared. S102: Secure Environment Construction: Detect the current read and write operation queue of the SSD. If there are incomplete write commands, trigger the delay buffer mechanism to temporarily store non-critical write operations in the isolated cache area. A temporary read-only policy is dynamically generated based on the SSD firmware version and the health status of the storage unit. The policy includes: prohibiting the injection of write instructions by unauthorized processes, limiting the number of concurrent connections to the health parameter access interface, and allocating an independent secure memory channel for health query operations. S103: Environmental security confirmation: After the secure operating environment is established, a confirmation signal containing an environmental fingerprint is sent to the request source. The environmental fingerprint is generated by combining the current SSD temperature, voltage fluctuation characteristics and access policy hash value. If the requesting source does not return a fingerprint confirmation receipt within a preset time, a security environment rollback operation will be automatically triggered.

2. The SSD health query system based on data read / write security according to claim 1, characterized in that, Step S200 includes the following steps: S201: Batch Data Extraction Control: The batch extraction capacity is dynamically set according to the health score of the SSD storage unit. When the health score is ≥80%, the extraction capacity of a single batch shall not exceed 5% of the total health parameter storage area. When the health score is <80%, the extraction capacity of a single batch shall be reduced proportionally and the batch interval time shall be increased. Before each batch extraction, the integrity status of the temporary read-only policy is verified through the firmware interface; S202: Dynamic Environment Fingerprint Generation: Generating dynamic environment fingerprints based on the following multi-dimensional parameters: Current SSD controller temperature and voltage fluctuation characteristics; Physical block access frequency distribution of the health parameter storage area; Access policy hash value group for a secure operating environment; A lightweight obfuscation algorithm is used to nonlinearly fuse the above parameters to generate a unique fingerprint identifier; S203, Multi-layer data encapsulation processing: First layer of encapsulation: Add a timeline with a time-series encryption stamp to the original health parameters, the time-series encryption stamp containing a batch number and a quantum random number; The second layer of encapsulation: the dynamic environment fingerprint is split into feature vectors and injected into the gaps in the data fields of the health parameters according to preset weights; The third layer of encapsulation: A chained checksum is generated based on the data from the first two layers, serving as the parameter integrity checksum. This checksum includes: The hash value of the last checksum of the health parameters from the previous batch; Chaotic encrypted digest of this batch of environmental fingerprints; The instantaneous power consumption characteristic value of the current SSD controller; S204, Abnormal Behavior Blocking: During the data encapsulation process, monitor the following abnormal indicators in real time: Threshold for the magnitude of mutations in health parameters within the same batch; Real-time matching degree between environmental fingerprints and operational strategies; Fluctuation range of data encapsulation and processing time; If any indicator is detected to exceed the preset safety threshold, the health parameter extraction and data encapsulation operation for the current batch will be terminated immediately, and a safety alarm protocol will be triggered.

3. The SSD health query system based on data read / write security according to claim 2, characterized in that, The S300 safe output phase further includes the following detailed steps: S301, Structured processing of triplet data packets: The original health parameters are split into multiple data blocks according to preset rules, and each data block is appended with an independent verification field, which includes: The hash value of the physical storage unit address of the corresponding data block; The current clock jitter characteristic quantization value of the SSD controller; Chaotic cryptographic digest of preceding data blocks; Environmental fingerprint feature values ​​and parameter integrity check codes are embedded in the gaps between data blocks using an orthogonal coding method to form a triplet structure with spatial isolation characteristics; S302, Progressive chain encryption processing: First layer of encryption: Each data block is encrypted independently using a lightweight symmetric encryption algorithm, and the encryption key is dynamically derived from the environmental fingerprint feature value; Second layer of encryption: The encrypted data blocks are concatenated in sequence, and the encryption result of the previous block is used as the initialization vector of the next block to generate a chain-like encrypted data stream; Third layer of encryption: Apply terminal encryption based on Physically Unclonable Function (PUF) to the above chained encrypted data stream, and simultaneously bind the following features when generating the final ciphertext: Current operating voltage ripple characteristics of the SSD controller; Statistical values ​​of read interference in NAND flash memory cells during encryption operations; S303, Self-verified health report generation: A set of verification information is appended to the end of the encrypted data, including: The power consumption trajectory hash value of each data block throughout the entire process of independent encryption at the first layer, chain encryption at the second layer, and terminal encryption at the third layer; The XOR checksum of all initialization vectors during chain encryption; The result of secondary obfuscation encryption of environmental fingerprint feature values; The encrypted data and verification information are reorganized into a tree structure to generate a health report that supports segmented verification. This report allows the recipient to verify integrity in any of the following ways: Verify the consistency of power consumption trajectories for each data block based on preset weight values; The correlation of the initialization vector is verified by reverse chaining decryption; S304, Safety Output Control: Monitor the physical layer characteristic parameters of the output channel, including signal level stability and timing jitter. When the signal level stability is detected to be lower than the threshold or the timing jitter value exceeds the preset range, the system automatically switches to a degraded output mode, which includes: Health reports are transmitted in fragmented, interleaved segments. Add a decoy checksum to each data shard; The frequency transition sequence of the output channel is updated in real time.

4. The SSD health query system based on data read / write security according to claim 3, characterized in that, The state recovery phase of step S400 further includes the following detailed steps: S401, Intelligent determination of transmission status: The monitoring of health report transmission channels includes both physical and protocol layer status indicators, such as: Physical layer: Signal integrity index, channel noise ratio threshold; Protocol layer: ACK / NACK response rate, data packet retransmission count statistics; The transmission completion status is dynamically determined based on a preset transmission integrity scoring model, which integrates the following parameters: the proportion of successfully received data fragments and the verification pass rate; Hazard assessment values ​​for remaining unconfirmed areas; S402, Execution of the safe exit protocol: When the transmission is determined to be complete, a phased safe exit process is initiated: Verify the consistency between the current environment fingerprint and the initial establishment phase. If the deviation exceeds a preset threshold, trigger a secondary verification. Delayed write commands in the isolation buffer are released in reverse order, and integrity check flags are injected. Perform pre-write verification on the areas covered by the temporary read-only policy to ensure that there is no residual temporary data related to health queries; S403, Emergency Data Erasure Procedure: When an abnormal interruption is detected, a multi-level erase operation is performed: Logical layer erase: A one-time erase key is generated using a Physically Unclonable Function (PUF) to overwrite the health parameter cache; Physical layer interference: Applying a specific pattern of read interference pulses to the NAND block storing health data, including: performing three consecutive program / erase cycles on the target cell; Inject high-voltage disturbance signals into adjacent units; Metadata obfuscation: Modify the physical pointer of the relevant logical address in the FTL mapping table and insert decoy data pages; S404, Policy Recovery and Audit Enhancement: When restoring the original read / write strategy, a gradual load injection method is used: In the first phase, only 50% of the write bandwidth was restored, and abnormal heat indicators of the SSD controller were monitored. If there are no abnormalities for three consecutive clock cycles, gradually increase to full load. When generating the operation audit log, attach environment traceability information, including: Hash value of temperature-voltage variation curve during safe operating environment operation; The timestamp chain and execution result code of each stage of state recovery operation; The signature of the interference pulse pattern used in the emergency erasure procedure; S405, Post-effect monitoring mechanism: After the status is restored, a monitoring period is initiated. The duration of the monitoring period is twice the duration of the health query operation, during which: Pattern recognition is performed on abnormal commands involving access to health parameters. If an operation matching the query features is detected, the circuit breaker mechanism is triggered. Regularly check the clearance status of residual environmental fingerprints in the SSD metadata area.

Images