Data storage security protection method and system based on solid state disk

Abstract

The invention relates to the field of computer security, and discloses a data storage security protection method and system based on a solid-state hard disk, which comprises the following steps: identifying the inherent physical difference of a flash memory chip in the solid-state hard disk so as to calculate the read-write time sequence micro-deviation of the solid-state hard disk; performing hash operation on the device root key and the unique identifier of the controller corresponding to the solid state disk to obtain a bound master control key; generating a temporary encryption key of the write-in operation to encrypt plaintext data of the host system to obtain ciphertext data; writing the ciphertext data into a flash memory physical page corresponding to the physical page address to obtain a ciphertext physical page address; when the access mode is a hostile attack mode, secretly updating the ciphertext physical page address into a new physical page address, and deleting the ciphertext physical page address; and when the access mode is a secure access mode and the verification key is consistent with the bound master control key, normal loading and data access requests of the solid state disk are allowed. According to the invention, the security and integrity of data storage can be improved.

Description

TECHNICAL FIELD

[0001] The application relates to a solid state disk-based data storage security protection method and system, and belongs to the computer security field. BACKGROUND

[0002] Data storage security protection refers to ensuring that the data stored on a solid state disk always meets the three basic security attributes of confidentiality, integrity and availability in the entire life cycle (from creation, storage, use, transmission to destruction) through a series of technical, management and physical means. Its significance lies in: for individuals, it is a basic guarantee for protecting privacy and property; for enterprises, it is a lifeline for maintaining core assets, business continuity and market competitiveness; for the country, it is a strategic cornerstone for protecting the safety of critical infrastructure, social stability and digital sovereignty.

[0003] The data storage security protection method mainly relies on the software and operating system level of the host end, and uses an operating system (such as BitLocker) or a third-party software to perform encryption and decryption on the host CPU to protect data storage. Since the key is generated by a software random number generator, there is a risk that the key will be copied, leaked or tampered with on the production line, thereby causing a high risk of data leakage. SUMMARY

[0004] The application provides a solid state disk-based data storage security protection method and system, which aims to improve the security and integrity of data storage.

[0005] To achieve the above purpose, the application provides a solid state disk-based data storage security protection method, which comprises: Identifying the inherent physical differences of the internal flash memory chip of the solid state disk to calculate the read-write timing micro-deviation of the solid state disk, generating a device root key of the solid state disk by using a preset physically unclonable function based on the read-write timing micro-deviation, performing a hash operation on the device root key and the unique identifier of the corresponding controller of the solid state disk to obtain a binding master key; In response to a write operation of the host system on the corresponding logical block address of the solid state disk, a temporary encryption key of the write operation is generated by combining the binding master key and the logical block address to encrypt the plaintext data of the host system to obtain ciphertext data; The logical block address is converted into a physical page address, and the ciphertext data is written into the flash memory physical page corresponding to the physical page address to obtain a ciphertext physical page address; Detecting the access mode of the logical block address, when the access mode is a malicious attack mode, the ciphertext physical page address is secretly updated to a new physical page address, and the ciphertext physical page address is deleted; When the access mode is a secure access mode, a PUF entropy source of the solid state disk is collected to calculate a verification key of the solid state disk, and when the verification key is consistent with the binding master key, normal loading and data access requests of the solid state disk are allowed.

[0006] Optionally, the device root key of the solid state disk is generated based on the read-write timing micro-deviation by using a preset physically unclonable function, including: The read-write timing micro-deviation is filtered to obtain a filtered read-write timing micro-deviation; The filtered read-write timing micro-deviation is standardized to obtain a standardized timing micro-deviation; The standardized timing micro-deviation is subjected to digital bit conversion to obtain a binary vector; Auxiliary data of the binary vector is generated by the physically unclonable function; Based on the auxiliary data, the binary vector is stabilized to obtain a stabilized binary vector; The stabilized binary vector is subjected to hash calculation to obtain a hash state; The hash state is subjected to bit length adjustment to obtain a device root key.

[0007] Optionally, the device root key and a unique identifier of a corresponding controller of the solid state disk are subjected to hash operation to obtain a binding master key, including: The unique identifier is subjected to hash calculation to obtain an identification binary sequence; A connection order of the device root key and the identification binary sequence is determined; Based on the connection order, the device root key and the identification binary sequence are subjected to data concatenation to obtain a combined binary sequence; The combined binary sequence is subjected to hash calculation to obtain a hash output digest; The hash output digest is subjected to bit length adjustment to obtain a binding master key.

[0008] Optionally, the inherent physical difference of the solid state disk internal flash memory chip is identified, including: A storage unit of the flash memory chip is determined; An intermediate voltage state of the storage unit is constructed; Based on the intermediate voltage state, current timing data of the storage unit is collected; Based on the current timing data, timing characteristics of the storage unit are extracted, wherein the timing characteristics include noise amplitude, jump frequency, and average switching time; Based on the timing characteristics, the inherent physical difference of the storage unit is determined.

[0009] Optionally, the calculating the read-write timing micro-deviation of the solid state disk comprises: dividing a timing measurement block of the solid state disk according to inherent physical differences corresponding to the solid state disk; determining a plurality of groups of address pairs of the timing measurement block; defining read-write competition rules of the plurality of groups of address pairs; collecting read-write time difference samples of the plurality of groups of address pairs based on the read-write competition rules; calculating read-write timing micro-deviations of the plurality of groups of address pairs according to the read-write time difference samples.

[0010] Optionally, the generating the temporary encryption key of the write operation in combination with the binding master key and the logical block address comprises: loading the binding master key in a corresponding secure storage area of the binding master key to obtain input key material; mixing the input key material and the logical block address to obtain a pseudo-random key; constructing context information of the logical block address; defining an encryption key length of the write operation; performing expansion calculation on the pseudo-random key based on the context information and the encryption key length to obtain a temporary encryption key.

[0011] Optionally, the encrypting the plaintext data of the host system to obtain ciphertext data comprises: splitting a temporary encryption key corresponding to the host system to obtain a data key and an adjustment key; generating adjustment value plaintext of a logical block address corresponding to the write of the host system; encrypting the adjustment value plaintext based on the adjustment key to obtain a position adjustment value; determining a data key length of the data key to block the plaintext data to obtain a plurality of plaintext data blocks; performing block encryption on the plurality of plaintext data blocks through the data key to obtain a plurality of encrypted data blocks; sequentially assembling the position adjustment value and the plurality of encrypted data blocks to obtain ciphertext data.

[0012] Optionally, the detecting the access mode of the logical block address comprises: obtaining an I / O request sequence of a host system corresponding to the logical block address; extracting a key metadata sequence of the I / O request sequence; calculate an I / O request feature of the key metadata sequence, wherein the I / O request feature comprises a frequency feature, a space feature, an operation feature, and a time feature; calculate a threat score of the logical block address according to the I / O request feature; determine an access mode of the logical block address based on the threat score.

[0013] Optionally, when the verification key and the binding master key are consistent, the normal loading and data access request of the solid state disk are allowed, comprising: generate a binding verification token of the solid state disk according to the binding master key; generate a dynamic verification token of the solid state disk based on the verification key; calculate a Boolean value of the binding verification token and the dynamic verification token; when the Boolean value is false, the step of secretly updating the ciphertext physical page address to a new physical page address and deleting the ciphertext physical page address is executed; when the Boolean value is true, the normal loading and data access request of the solid state disk are allowed.

[0014] In order to solve the above problems, the application also provides a data storage security protection system based on a solid state disk, the system comprising: a binding master key generation module, configured to identify the inherent physical difference of the internal flash chip of the solid state disk, to calculate the read-write timing micro-deviation of the solid state disk, to generate the device root key of the solid state disk by using a preset physically unclonable function based on the read-write timing micro-deviation, and to obtain the binding master key by performing a hash operation on the device root key and the unique identifier of the corresponding controller of the solid state disk; a ciphertext data encryption module, configured to, in response to the write operation of the host system on the corresponding logical block address of the solid state disk, generate a temporary encryption key of the write operation in combination with the binding master key and the logical block address, to encrypt the plaintext data of the host system and obtain ciphertext data; a ciphertext data writing module, configured to convert the logical block address into a physical page address, and write the ciphertext data into the flash physical page corresponding to the physical page address to obtain a ciphertext physical page address; an abnormal defense module, configured to detect the access mode of the logical block address, and to secretly update the ciphertext physical page address to a new physical page address and delete the ciphertext physical page address when the access mode is a malicious attack mode; The normal access module is configured to, when the access mode is a secure access mode, collect a PUF entropy source of the solid state disk to calculate a verification key of the solid state disk, and allow normal loading and data access requests of the solid state disk when the verification key is consistent with the binding master key.

[0015] Compared with the problems described in the background art, the present application significantly improves the data security of the solid state disk by constructing a multi-level, dynamic and deeply bound active defense system with hardware, generating a device root key by identifying the inherent physical differences of the flash memory chip, and further generating a binding master key by combining the controller unique identifier, establishing a unique and unclonable hardware identity of the device, which fundamentally eliminates the possibility of cracking the security mechanism by copying firmware or cloning hardware, and realizes strong identity authentication at the device level. Secondly, during data writing, a temporary encryption key is dynamically generated by combining the binding master key and the logical block address, realizing the fine-grained data encryption, which makes each data block use an independent key, so that even if the key of a single data block is leaked, it will not affect the security of other data blocks, effectively containing the horizontal spread of attacks. The present application introduces an active defense mechanism based on access mode, which can secretly migrate data to a new physical address and destroy the old address when malicious attack mode is detected. This data migration strategy makes it difficult for attackers to locate and continuously attack target data, greatly increasing the difficulty and cost of attacks. Finally, by performing PUF-based verification key checking in the secure access mode, it is ensured that only a legitimate host can perform normal loading and data access after the device identity verification is passed. This mechanism strongly binds the data access permission with the physical identity of the device and the current session state, forming a closed-loop trust. The present application changes static passive defense to dynamic active defense, deeply integrates software-level security with hardware-level physical characteristics, and provides full-link and high-strength security protection for solid state disks from device identity, data encryption to access control. Therefore, the data storage security protection method based on the solid state disk provided by the embodiment of the present application can improve the security and integrity of data storage. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 The flowchart of the data storage security protection method based on the solid state disk provided by an embodiment of the present application is shown.

[0017] Figure 2 The module diagram of the data storage security protection system based on the solid state disk provided by an embodiment of the present application is shown.

[0018] The object implementation, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0019] It is to be understood that the specific embodiments described herein are merely illustrative of the present application and do not limit the scope of the present application.

[0020] The embodiment of the present application provides a data storage security protection method based on a solid state disk. The execution subject of the data storage security protection method based on the solid state disk includes but is not limited to at least one of electronic devices such as a server and a terminal which can be configured to execute the method provided by the embodiment of the present application. In other words, the data storage security protection method based on the solid state disk can be executed by software or hardware installed in a terminal device or a server device. The server includes but is not limited to a single server, a server cluster, a cloud server or a cloud server cluster.

[0021] Referring to Figure 1 Fig. 1 is a flowchart of a data storage security protection method based on a solid state disk provided by an embodiment of the present application. In the embodiment, the data storage security protection method based on the solid state disk includes: S1, identifying inherent physical differences of a flash memory chip inside a solid state disk to calculate read-write timing micro-deviation of the solid state disk, generating a device root key of the solid state disk by using a preset physically unclonable function based on the read-write timing micro-deviation, and performing a hash operation on the device root key and a unique identifier of a corresponding controller of the solid state disk to obtain a binding master key.

[0022] In the embodiment of the present application, the inherent physical differences of the flash memory chip inside the solid state disk can be quantified to calculate the read-write timing micro-deviation, and a digital fingerprint unique to the device is generated based on the read-write timing micro-deviation. The flash memory chip refers to a NAND Flash memory wafer and its package which constitute a storage core of the solid state disk. The inherent physical differences refer to microscopic structural non-uniformity which cannot be avoided and eliminated in the physical aspect due to atomic-level process fluctuations in the semiconductor manufacturing process.

[0023] As an embodiment of the present application, the identification of the inherent physical differences of the flash memory chip inside the solid state disk includes: determining a storage unit of the flash memory chip; constructing an intermediate voltage state of the storage unit; collecting current timing data of the storage unit based on the intermediate voltage state; extracting timing features of the storage unit according to the current timing data, wherein the timing features include noise amplitude, jump frequency and average switching time; determining the inherent physical differences of the storage unit based on the timing features.

[0024] The storage unit refers to the most basic data storage unit of a flash memory chip. The intermediate voltage state refers to a non-stable, sensitive balance point between the "programmed" and "erased" stable states by precise control to adjust the floating gate charge of the storage unit. The current time series data refers to a series of time-sequentially arranged numerical values obtained by continuously and high-speed current sampling of a storage unit in an intermediate voltage state within a period of time. The time sequence feature refers to a key numerical index extracted from the original current time series data, which can quantitatively describe the data pattern. The noise amplitude refers to the difference between the high current state and the low current state of the current time series data when jumping. The jump frequency refers to the average number of times of jumping between the high and low states of the current per unit time. The average switching time refers to the average time of keeping the current at the high level and the average time of keeping the current at the low level.

[0025] Optionally, the intermediate voltage state of the storage unit can be constructed by a machine learning algorithm, such as linear regression, neural network, etc.

[0026] Optionally, the current time series data of the storage unit can be collected by a high-precision current sensor.

[0027] Optionally, the time sequence feature of the storage unit can be extracted by statistical and signal processing methods, such as mean and variance, jump event analysis, etc.

[0028] The embodiment of the present application can quantify the physical uniqueness of the flash memory chip into a measurable digital parameter by calculating the read-write time sequence micro-deviation of the solid state disk, and provide a stable and reliable response basis for the physically unclonable function. The read-write time sequence micro-deviation refers to the small, repeatable but unpredictable deviation in operation completion time caused by the inherent, random physical differences within the chip when performing standardized read-write operations on the flash memory chip of the solid state disk.

[0029] As an embodiment of the present application, the calculation of the read-write time sequence micro-deviation of the solid state disk comprises: According to the inherent physical differences corresponding to the solid state disk, the time sequence measurement block of the solid state disk is divided; A plurality of groups of address pairs of the time sequence measurement block are determined; The read-write competition rules of the plurality of groups of address pairs are defined; Based on the read-write competition rules, read-write time difference samples of the plurality of groups of address pairs are collected; According to the read-write time difference samples, the read-write time sequence micro-deviation of the plurality of groups of address pairs is calculated.

[0030] The timing measurement block refers to a plurality of physical storage areas specially divided in the reserved space of the solid state disk for performing timing measurement. The plurality of groups of address pairs refers to physical address combinations selected from the timing measurement block according to a specific strategy for two-by-two testing. The read-write competition rule refers to a standardized protocol defining how to fairly and controllably let two addresses in an address pair start at the same time and record their arrival times. The read-write time difference sample refers to the difference between the times taken by two addresses in an address pair to complete an operation measured according to the read-write competition rule in one test.

[0031] Optionally, the read-write competition rule of the plurality of groups of address pairs can be defined by open standards such as NVMe protocol, PCIe bus, and CXL.

[0032] Optionally, the read-write timing micro-deviation of the plurality of groups of address pairs can be calculated by robust statistics such as interquartile range method, median instead of mean, and truncated mean.

[0033] The device root key of the solid state disk generated by the preset physically unclonable function based on the read-write timing micro-deviation can achieve inseparable binding of the key and the physical carrier, eliminate the key static storage link, and fundamentally avoid the risk of direct extraction of the key. The hardware root of trust established thereby provides a native trusted basis for key functions such as secure boot and data encryption. The preset physically unclonable function refers to a mathematical function that maps a read-write timing micro-deviation of a solid state disk as a “seed” into a series of seemingly random, but unique and repeatable numerical sequences for a specific device through a public and standardized algorithm. The device root key refers to a high-entropy, stable, and unique encryption key finally derived from the inherent physical differences of a solid state disk through a preset physically unclonable function.

[0034] As an embodiment of the application, the device root key of the solid state disk generated by the preset physically unclonable function based on the read-write timing micro-deviation includes: Filtering the read-write timing micro-deviation to obtain a filtered read-write timing micro-deviation; Standardizing the filtered read-write timing micro-deviation to obtain a standardized timing micro-deviation; Converting the standardized timing micro-deviation into a digital bit to obtain a binary vector; Generating auxiliary data of the binary vector through the physically unclonable function; Stabilizing the binary vector based on the auxiliary data to obtain a stabilized binary vector; Performing hash calculation on the stabilized binary vector to obtain a hash state; The hash state is subjected to bit length adjustment to obtain a device root key.

[0035] The filtered read-write timing micro-deviation refers to mathematical processing of original, noise-filled read-write timing micro-deviation samples to suppress random noise and eliminate abnormal values, thereby obtaining a more smooth and more close-to-true physical nature deviation value. The standardized timing micro-deviation refers to mapping of the filtered micro-deviation values in different dimensions and distribution ranges to a standard and comparable numerical interval through a unified mathematical transformation, thereby obtaining a standardized micro-deviation value. The binary vector refers to conversion of the standardized continuous timing micro-deviation value into a digital sequence composed of 0 and 1. The auxiliary data refers to a piece of data generated from the binary vector by a physically unclonable function, which does not contain key information but contains sufficient error correction clues. The stabilized binary vector refers to the binary vector that is completely consistent with the registration time and stable and reliable after error correction of the newly generated binary vector that may have errors by using the stored auxiliary data. The hash state refers to a fixed-length hash value finally output by the internal calculation of a standard cryptographic hash function after the stabilized binary vector is input into the standard cryptographic hash function.

[0036] Optionally, the filtered read-write timing micro-deviation can be obtained by a filtering algorithm, such as median filtering, mean filtering, weighted filtering, etc.

[0037] Optionally, the standardized timing micro-deviation can be obtained by Z-Score standardization technology.

[0038] Optionally, the hash state can be calculated by a cryptographic hash function, such as SHA-256, SHA-3, etc.

[0039] The embodiment of the application realizes the two-factor hardware binding of the key and the flash memory chip and the main controller by performing hash operation on the device root key and the unique identifier of the corresponding controller of the solid state disk to obtain the binding master key, establishes a cryptographic association between the flash memory and the main controller, ensures that the key is valid only in a specific hardware combination environment, and any single replacement of the flash memory or the main controller will cause the key to be invalid, thereby preventing component disassembly and reuse and significantly improving the physical attack threshold. The binding master key refers to a synthetic key obtained by irreversible mixing operation of the device root key derived from the physically unclonable feature of the flash memory chip and the unique identifier of the solid state disk main controller by a cryptographic hash function.

[0040] As an embodiment of the application, the hash operation on the device root key and the unique identifier of the corresponding controller of the solid state disk to obtain the binding master key comprises: hashing the unique identifier to obtain an identification binary sequence; determining a concatenation order of the device root key and the identification binary sequence; based on the concatenation order, concatenating the device root key and the identification binary sequence to obtain a combined binary sequence; hashing the combined binary sequence to obtain a hash output digest; adjusting the bit length of the hash output digest to obtain a binding master key.

[0041] The identification binary sequence refers to a fixed-length digital sequence composed of 0 and 1 obtained by converting the unique identifier of the solid state disk controller through a hash function. The concatenation order refers to the rule of defining which one is in front and which one is behind when concatenating the device root key and the identification binary sequence into a longer binary sequence. The combined binary sequence refers to a longer, single continuous binary sequence formed by concatenating the binary representation of the device root key and the identification binary sequence according to the preset concatenation order. The hash output digest refers to the fixed-length hash value output by the function after inputting the combined binary sequence into a standard cryptographic hash function.

[0042] Optionally, the concatenation order of the device root key and the identification binary sequence can be determined by derivation rules such as parity rules, threshold rules, etc.

[0043] S2, in response to the write operation of the host system to the logical block address corresponding to the solid state disk, combining the binding master key and the logical block address to generate a temporary encryption key of the write operation to encrypt the plaintext data of the host system to obtain ciphertext data.

[0044] The present application implements non-volatile storage by permanently writing the data provided by the host system into the physical block of the flash memory chip in response to the write operation of the host system to the logical block address corresponding to the solid state disk. The logical block address refers to a virtual address used when the host operating system exchanges data with the solid state disk. The write operation refers to the complete process of safely, efficiently and persistently storing the data provided by the host system into the flash memory medium according to the internal rules of the solid state disk.

[0045] The temporary encryption key of the write operation is generated by combining the binding master key and the logical block address, so that each data block or region is protected by an independent key, thereby greatly limiting the scope of potential attacks, and even if a single key is cracked, it only affects local data and cannot endanger the security of the entire disk.

[0046] As an embodiment of the present application, the combination of the binding master key and the logical block address to generate the temporary encryption key of the write operation includes: The binding master key is loaded in the corresponding secure storage area to obtain input key material; Mixing the input key material and the logical block address to obtain a pseudo-random key; Context information of the logical block address is constructed; The encryption key length of the write operation is determined; Based on the context information and the encryption key length, the pseudo-random key is expanded to obtain a temporary encryption key.

[0047] The input key material refers to the original binding master key loaded from the secure storage area of the solid state disk without any processing. The pseudo-random key is a fixed-length binary sequence that looks completely random but is deterministically generated after mixing the input key material and the logical block address. The context information refers to a piece of data that explicitly describes the purpose, source and operating environment of the key. The encryption key length refers to the number of binary bits contained in the finally generated temporary encryption key.

[0048] Optionally, the pseudo-random key can be obtained through a cryptographic hash function, such as HMAC.

[0049] Optionally, the temporary encryption key can be obtained through a key derivation function, such as HKDF, AES-CMAC KDF, etc.

[0050] The embodiment of the present application encrypts the plaintext data of the host system to obtain ciphertext data, which can ensure that the information on the storage medium is unreadable to any unauthorized party, thereby effectively preventing data leakage due to physical theft or loss of equipment, and resisting offline attacks by directly reading the chip.

[0051] As an embodiment of the present application, the encryption of the plaintext data of the host system to obtain ciphertext data comprises: Splitting the temporary encryption key corresponding to the host system to obtain a data key and an adjustment key; Generating an adjustment value plaintext of the logical block address corresponding to the writing of the host system; Encrypting the adjustment value plaintext based on the adjustment key to obtain a position adjustment value; Determining the data key length of the data key to block the plaintext data to obtain a plurality of plaintext data blocks; Encrypting the plurality of plaintext data blocks by the data key to obtain a plurality of encrypted data blocks; Assembling the position adjustment value and the plurality of encrypted data blocks in sequence to obtain ciphertext data.

[0052] The data key is a symmetric key that is split from the temporary encryption key and is used for core encryption operation of the host plaintext data block. The adjustment key is a symmetric key that is split from the temporary encryption key and is used for generating the position adjustment value bound to the logical block address. The adjustment value plaintext is a standardized and fixed-length data structure, and its only purpose is to carry the logical block address information in a clear format for encryption. The position adjustment value is a 128-bit binary sequence that is strongly related to a specific LBA and appears random after AES encryption of the adjustment value plaintext by the adjustment key. The data key length refers to the number of binary bits contained in the data key. The plurality of plaintext data blocks refers to a series of continuous and equal-length data segments obtained by dividing the complete plaintext data sent by the host system according to the fixed block size required by the encryption algorithm. The plurality of encrypted data blocks refers to a series of continuous and meaningless ciphertext segments obtained by encrypting each block in the plurality of plaintext data blocks by the data key and its corresponding position adjustment value.

[0053] Optionally, the adjustment value plaintext of the logical block address can be generated by dynamic entropy injection coding of the hardware trusted execution environment, such as TEE in ARM TrustZone.

[0054] Optionally, the plurality of plaintext data blocks can be obtained by an encryption algorithm, such as AES-128 algorithm, AES-256 algorithm, etc.

[0055] S3, converting the logical block address into a physical page address, and writing the ciphertext data into the flash physical page corresponding to the physical page address to obtain a ciphertext physical page address.

[0056] The embodiment of the present application can uniformly distribute the write operation to all physical blocks by converting the logical block address into a physical page address, avoid local over-wear, greatly prolong the overall service life of the fixed hard disk, support power-off protection, ensure the atomicity of address mapping table update, and prevent data loss caused by accidental power-off. The physical page address refers to complete and multi-level address information of a smallest programmable unit on a flash memory chip.

[0057] Optionally, the physical page address can be converted by a flash translation layer in the solid state disk.

[0058] The embodiment of the present application can complete data persistence by writing the ciphertext data into the flash memory physical page corresponding to the physical page address, obtaining a ciphertext physical page address, safely storing the encrypted information in the non-volatile flash memory medium, realizing physical isolation of the ciphertext data, and ensuring the final landing of the encrypted storage. The ciphertext physical page address refers to the final and determined physical location of the physical page on the flash memory after the ciphertext data is successfully written.

[0059] S4, detecting the access mode of the logical block address, when the access mode is a malicious attack mode, secretly updating the ciphertext physical page address to a new physical page address, and deleting the ciphertext physical page address.

[0060] The embodiment of the present application can effectively identify and distinguish normal user access and malicious attack behaviors such as data theft and ransomware by detecting the access mode of the logical block address, thereby providing a decision basis and trigger signal for subsequent active defense measures. The access mode refers to the statistical characteristics and behavior sequence of input / output requests initiated to multiple logical block addresses within a specific time window.

[0061] As an embodiment of the present application, the detection of the access mode of the logical block address comprises: Obtaining the I / O request sequence of the host system corresponding to the logical block address; Extracting the key metadata sequence of the I / O request sequence; Calculating the I / O request characteristics of the key metadata sequence, wherein the I / O request characteristics include frequency characteristics, spatial characteristics, operation characteristics and time characteristics; According to the I / O request characteristics, calculating the threat score of the logical block address; Based on the threat score, determining the access mode of the logical block address.

[0062] The I / O request sequence refers to a set of I / O commands arranged in chronological order in a specific time window sent by a host system to a target logical block address. The key metadata sequence refers to a sequence of core description information extracted from the I / O request sequence for each I / O command, such as timestamp, logical block address, request length, operation type, etc. The I / O request feature refers to a set of quantitative indicators obtained by statistical calculation of the key metadata sequence. The frequency feature refers to a quantitative indicator describing the frequency of I / O request occurrence, such as IOPS, bandwidth, access frequency, etc. The spatial feature refers to a quantitative indicator describing the distribution characteristics of I / O request in the logical address space, such as LBA range, order degree, randomness, etc. The operation feature refers to a quantitative indicator describing the type and composition of I / O request, such as read / write ratio, average request size, command type distribution, etc. The time feature refers to a quantitative indicator describing the distribution law and rhythm characteristics of I / O request in the time dimension, such as access interval distribution, burstiness, access time point, etc. The threat score refers to a comprehensive quantitative value for evaluating the matching degree of current access behavior with known malicious attack patterns and the potential danger level.

[0063] Optionally, the I / O request feature of the key metadata sequence can be calculated by sequence deep learning, such as Transformer model, graph neural network, etc.

[0064] Optionally, the threat score of the logical block address can be calculated by supervised learning technology, such as SVM, decision tree, neural network, etc.

[0065] The embodiment of the application can directly interrupt data stealing by updating the ciphertext physical page address to a new physical page address when the access mode is a malicious attack mode, making the physical address obtained by the attacker through physical or underlying logic vulnerabilities invalid instantly, and unable to read the target data, ensuring that sensitive data is not obtained by the enemy. The new physical page address refers to a completely new physical storage location for storing a copy of the original ciphertext data dynamically allocated in the flash translation layer of the solid state disk in response to security threats.

[0066] The embodiment of the application can effectively prevent data residual leakage by deleting the ciphertext physical page address, eliminating the risk of attackers stealing information from abandoned areas through advanced means such as chip-level physical detection.

[0067] S5, when the access mode is a secure access mode, collect the PUF entropy source of the solid state disk to calculate the verification key of the solid state disk, and when the verification key and the binding master key are consistent, allow normal loading and data access request of the solid state disk.

[0068] The embodiment of the application can generate a unique and unclonable root key for the device by collecting the PUF entropy source of the solid state disk when the access mode is a secure access mode, the key can be used as a key source for data encryption and for realizing identity authentication and anti-counterfeiting of the device, wherein the PUF entropy source refers to a unique and uncopyable hardware fingerprint generated due to microscopic physical differences in the manufacturing process of the solid state disk chip.

[0069] The embodiment of the application can be used to verify the authenticity of the device identity by calculating the verification key of the solid state disk, so as to ensure that the communication object is a trusted and unaltered solid state disk, and to be able to establish an end-to-end encrypted communication channel to protect the confidentiality of data in the transmission process, wherein the verification key refers to a temporary key generated based on the unique hardware fingerprint (PUF entropy source) of the device and used to establish a trust relationship between the two communication parties.

[0070] Optionally, the verification key of the solid state disk can be calculated by a hash function, such as SHA-3.

[0071] The embodiment of the application can successfully establish a trust chain between the device and the host by allowing normal loading and data access requests of the solid state disk when the verification key and the binding master key are consistent, which lays a foundation for all subsequent security operations, realizes secure startup of the firmware, and ensures that only authorized and unaltered firmware code can run, thereby ensuring the integrity of the device from the source, wherein the normal loading refers to the whole process of the solid state disk performing complete and uninterrupted internal software and hardware initialization according to a predetermined process after passing the security authentication, and entering an operational state. The data access request refers to a set of instructions sent by the host system to the solid state disk through a standard interface, aiming to read or write user data.

[0072] As an embodiment of the application, when the verification key and the binding master key are consistent, the normal loading and data access requests of the solid state disk are allowed, including: generating a binding verification token of the solid state disk according to the binding master key; generating a dynamic verification token of the solid state disk based on the verification key; calculating the Boolean value of the binding verification token and the dynamic verification token; when the Boolean value is false, performing the step of updating the ciphertext physical page address to a new physical page address and deleting the ciphertext physical page address; when the Boolean value is true, allowing the normal loading and data access requests of the solid state disk.

[0073] The binding verification token refers to a standard reference token generated by the internal security module of the solid state disk using a pre-stored, protected binding master key, combined with a dynamic, one-time random challenge parameter, through a cryptographic operation (such as HMAC) in the verification process. The dynamic verification token refers to an instant token generated by the security module of the solid state disk based on the verification key calculated in real time at the moment, combined with the same dynamic random challenge parameter as when the binding period verification token is generated, through the same cryptographic operation in the verification process. The Boolean value refers to a logical data type, which specifically refers to the binary result obtained by comparing the binding verification token and the dynamic verification token through a secure comparison function (such as a constant-time comparison algorithm).

[0074] Optionally, the generating the binding verification token of the solid state disk according to the binding master key comprises: generating a random challenge value and a verification security timestamp of the solid state disk; calculating the binding verification token of the solid state disk based on the random challenge value, the verification security timestamp and the binding master key through the following formula:

[0075] wherein, the binding verification token, the hash-based message authentication code function, the binding master key, the random challenge value, the verification security timestamp, the bitwise XOR operation.

[0076] It needs to be explained that in this application, the formula represents the bitwise XOR operation, which is to mix the random challenge value and the verification security timestamp , which is a lightweight and efficient information mixing method, ensuring that the second input of the function is dynamic and unpredictable.

[0077] As another embodiment, the Boolean value is calculated through the following formula:

[0078] wherein, the Boolean value, the constant-time comparison function, the binding verification token, the dynamic verification token.

[0079] Compared with the problems described in the background art, the present application significantly improves the data security of the solid state disk by constructing a multi-level, dynamic and deeply bound hardware active defense system, generates a device root key by identifying the inherent physical differences of the flash memory chip, and further generates a bound master key by combining the controller unique identifier, establishes a unique and unclonable hardware identity of the device, which fundamentally eliminates the possibility of cracking the security mechanism by copying firmware or cloning hardware, and realizes strong identity authentication at the device level. Secondly, during data writing, a temporary encryption key is dynamically generated by combining the bound master key and the logical block address, realizing the fine-grained data encryption, which makes each data block use an independent key, so that even if the key of a single data block is leaked, it will not affect the security of other data blocks, effectively containing the horizontal spread of attacks. The present application introduces an active defense mechanism based on access mode, which can secretly migrate data to a new physical address and destroy the old address when a malicious attack mode is detected. This data migration strategy makes it difficult for attackers to locate and continuously attack target data, greatly increasing the difficulty and cost of attacks. Finally, by performing PUF-based verification key checking in the secure access mode, it is ensured that only the legal host can perform normal loading and data access after the device identity verification, which strongly binds the data access permission with the physical identity of the device and the current session state, forming a closed-loop trust. The present application changes the static passive defense to dynamic active defense, deeply integrates the software layer security with the hardware layer physical characteristics, and provides full-link and high-strength security protection for the solid state disk from device identity, data encryption to access control. Therefore, the data storage security protection method based on the solid state disk provided by the embodiment of the present application can improve the security and integrity of data storage.

[0080] As shown in Figure 2 , it is a function module diagram of a data storage security protection system based on a solid state disk.

[0081] The data storage security protection system based on a solid state disk 200 can be installed in an electronic device. According to the functions realized, the data storage security protection system based on a solid state disk includes a bound master key generation module 201, a ciphertext data encryption module 202, a ciphertext data writing module 203, an abnormal defense module 204 and a normal access module 205. The modules of the present application can also be called units, which refer to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0082] In the embodiment of the present application, the functions of each module / unit are as follows: The binding master key generation module 201 is configured to identify inherent physical differences of internal flash memory chips of the solid state disk, calculate a read-write timing micro-deviation of the solid state disk, generate a root key of the solid state disk by using a preset physically unclonable function based on the read-write timing micro-deviation, and perform hash operation on the root key and a unique identifier of a corresponding controller of the solid state disk to obtain a binding master key. The ciphertext data encryption module 202 is configured to, in response to a write operation of a host system on a corresponding logical block address of the solid state disk, generate a temporary encryption key of the write operation in combination with the binding master key and the logical block address, encrypt plaintext data of the host system to obtain ciphertext data. The ciphertext data writing module 203 is configured to convert the logical block address into a physical page address, and write the ciphertext data into a flash memory physical page corresponding to the physical page address to obtain a ciphertext physical page address. The anomaly defense module 204 is configured to detect an access mode of the logical block address, and when the access mode is a malicious attack mode, secretly update the ciphertext physical page address to a new physical page address and delete the ciphertext physical page address. The normal access module 205 is configured to, when the access mode is a secure access mode, collect a PUF entropy source of the solid state disk to calculate a verification key of the solid state disk, and when the verification key is consistent with the binding master key, allow normal loading and data access request of the solid state disk.

[0083] In detail, the modules in the data storage security protection system 200 based on the solid state disk in the embodiments of the present application use the same technical means as the data storage security protection method based on the solid state disk in the above-mentioned Figure 1 , and can produce the same technical effects, which will not be described here.

[0084] It is obvious for those skilled in the art that the present application is not limited to the details of the above-mentioned exemplary embodiments, and the present application can be implemented in other specific forms without departing from the spirit or essential characteristics of the present application.

[0085] Finally, it should be noted that in the above-mentioned embodiments, each embodiment can be combined or independent, and deleting any one of them does not affect the technical implementation of other embodiments. The above embodiments are only used to illustrate the technical solutions of the present application but not to limit it. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced equivalently without departing from the spirit and scope of the present application.

Claims

1. A data storage security protection method based on a solid state disk, characterized in that, The method comprises: Identify the inherent physical differences of the internal flash chip of the solid state disk to calculate the read-write timing micro-deviation of the solid state disk, based on the read-write timing micro-deviation, using a preset physically unclonable function, generate the root key of the device of the solid state disk, hash operation is carried out on the root key of the device and the unique identifier of the corresponding controller of the solid state disk to obtain the binding master key; In response to the write operation of the host system to the corresponding logical block address of the solid state disk, the binding master key and the logical block address are combined to generate the temporary encryption key of the write operation, so as to encrypt the plaintext data of the host system to obtain the ciphertext data; Convert the logical block address to a physical page address, and write the ciphertext data into the flash physical page corresponding to the physical page address to obtain the ciphertext physical page address; Detect the access mode of the logical block address, and when the access mode is a malicious attack mode, the ciphertext physical page address is secretly updated to a new physical page address, and the ciphertext physical page address is deleted; When the access mode is a secure access mode, collect the PUF entropy source of the solid state disk to calculate the verification key of the solid state disk, and when the verification key and the binding master key are consistent, allow the normal loading and data access request of the solid state disk.

2. The solid state drive based data storage security protection method of claim 1, wherein, The method comprises: Filter the read-write timing micro-deviation to obtain a filtered read-write timing micro-deviation; Standardize the filtered read-write timing micro-deviation to obtain a standardized timing micro-deviation; Convert the standardized timing micro-deviation to a binary vector; Generate auxiliary data of the binary vector through the physically unclonable function; Stabilize the binary vector based on the auxiliary data to obtain a stabilized binary vector; Hash calculate the stabilized binary vector to obtain a hash state; Adjust the bit length of the hash state to obtain a root key of the device.

3. The solid state drive based data storage security protection method of claim 2, wherein, The method comprises: Hash calculate the unique identifier to obtain an identifier binary sequence; Determine the connection order of the root key of the device and the identifier binary sequence; Based on the connection order, concatenate the root key of the device and the identifier binary sequence to obtain a combined binary sequence; Hash calculate the combined binary sequence to obtain a hash output digest; Adjust the bit length of the hash output digest to obtain a binding master key.

4. The solid state drive based data storage security protection method of claim 1, wherein, The method comprises: Determine the storage unit of the flash chip; Construct the intermediate voltage state of the storage unit; Based on the intermediate voltage state, collect the current timing data of the storage unit; According to the current timing data, extract the timing characteristics of the storage unit, wherein the timing characteristics include noise amplitude, jump frequency and average switching time; Determine the inherent physical difference of the storage unit based on the timing characteristics.

5. The solid state drive based data storage security protection method of claim 1, wherein, The calculation of the read-write timing micro-deviation of the solid state disk includes: According to the inherent physical difference corresponding to the solid state disk, the timing measurement block of the solid state disk is divided; Determine a plurality of groups of address pairs of the timing measurement block; Define the read-write competition rules of the plurality of groups of address pairs; Based on the read-write competition rules, collect the read-write time difference samples of the plurality of groups of address pairs; According to the read-write time difference samples, calculate the read-write timing micro-deviation of the plurality of groups of address pairs.

6. The solid state drive based data storage security protection method of claim 1, wherein, The combination of the binding master key and the logical block address includes: Load the binding master key in the corresponding secure storage area of the binding master key to obtain the input key material; Mix the input key material and the logical block address to obtain a pseudo-random key; Construct the context information of the logical block address; Determine the encryption key length of the write operation; Based on the context information and the encryption key length, the pseudo-random key is expanded to obtain a temporary encryption key.

7. The solid state drive based data storage security protection method of claim 1, wherein, The encryption of the plaintext data of the host system to obtain the ciphertext data includes: Split the temporary encryption key corresponding to the host system to obtain a data key and an adjustment key; Generate the adjustment value plaintext of the logical block address corresponding to the write of the host system; Based on the adjustment key, encrypt the adjustment value plaintext to obtain the position adjustment value; Determine the data key length of the data key to block the plaintext data to obtain a plurality of plaintext data blocks; Block encrypt the plurality of plaintext data blocks through the data key to obtain a plurality of encrypted data blocks; Assemble the position adjustment value and the plurality of encrypted data blocks in sequence to obtain the ciphertext data.

8. The solid state drive based data storage security protection method of claim 1, wherein, The detection of the access mode of the logical block address includes: Obtain the I / O request sequence of the host system corresponding to the logical block address; Extract the key metadata sequence of the I / O request sequence; Calculate the I / O request characteristics of the key metadata sequence, wherein the I / O request characteristics include frequency characteristics, spatial characteristics, operation characteristics, and time characteristics; According to the I / O request characteristics, calculate the threat score of the logical block address; Based on the threat score, determine the access mode of the logical block address.

9. The solid state drive based data storage security protection method of claim 1, wherein, When the verification key and the binding master key are consistent, the normal loading and data access request of the solid state disk are allowed, including: According to the binding master key, generate the binding verification token of the solid state disk; Based on the verification key, generate the dynamic verification token of the solid state disk; Calculate the Boolean value of the binding verification token and the dynamic verification token; When the Boolean value is false, execute the steps of secretly updating the ciphertext physical page address to a new physical page address and deleting the ciphertext physical page address; When the Boolean value is true, the normal loading and data access request of the solid state disk is allowed.

10. A data storage security protection system based on a solid state drive, characterized by, The system includes: The binding master key generation module is used to identify inherent physical differences of the internal flash memory chip of the solid state disk, to calculate a read-write timing micro-deviation of the solid state disk, to generate a root-of-trust key of the solid state disk by using a preset physically unclonable function based on the read-write timing micro-deviation, to perform a hash operation on the root-of-trust key and a unique identifier of a corresponding controller of the solid state disk, and to obtain a binding master key; The ciphertext data encryption module is used to, in response to a write operation of a host system on a corresponding logical block address of the solid state disk, combine the binding master key and the logical block address to generate a temporary encryption key of the write operation, to encrypt plaintext data of the host system to obtain ciphertext data; The ciphertext data write module is used to convert the logical block address into a physical page address, to write the ciphertext data into a flash memory physical page corresponding to the physical page address, and to obtain a ciphertext physical page address; The anomaly defense module is used to detect an access mode of the logical block address, to secretly update the ciphertext physical page address to a new physical page address and to delete the ciphertext physical page address when the access mode is a malicious attack mode; The normal access module is used to, when the access mode is a secure access mode, collect a PUF entropy source of the solid state disk to calculate a verification key of the solid state disk, and to allow normal loading and data access requests of the solid state disk when the verification key is consistent with the binding master key.

Images