Method for erasing information from electronic scrap based on dual-security mechanism

Abstract

According to aspects of the inventive concepts, provided is a method for erasing information based on a dual-security mechanism. A storage medium feature database, an information erasure feature database, and a firmware system feature database are built to match cases for to-be-erased electronic scrap. An erasure solution and a native system data package are generated based on the matching results. The information is erased and an erasure result is evaluated; and the information is recovered on the erased electronic scrap, and a recovery result is evaluated, to implement comprehensive double security evaluation. The information erasure validity of the electronic scrap is checked based on the evaluation results. If an erasure result is invalid, erasure solutions are corrected online based on the evaluation result, until the erasure result is valid and the electronic scrap with a native system recovered is obtained.

Description

CROSS REFERENCE TO RELATED APPLICATIONS

The present application claims the benefit of priority under 35 USC 119 to Chinese patent application 201911116710.X, filed Nov. 15, 2019, the contents of which are incorporated herein by reference.

TECHNICAL FIELD

The present invention relates to the technical field of information erasure for electronic products, and in particular, to a method for erasing information from electronic scrap based on a dual-security mechanism.

BACKGROUND

In China, the update of electronic products such as smartphones and computers is accelerating with the continuous development of science and technology, resulting in an increasingly large amount of electronic scrap. Waste of resources, environmental pollution and information security problems due to improper treatment of the electronic scrap are getting worse. According to statistics, China's users change their mobile phones on average in 15 months, of which 20% will change their mobile phones within one year. According to statistics from the Ministry of Industry and Information Technology, as of October 2015, China has 1.302 billion mobile phone users, about 200 million mobile phones are discarded each year, but the recycling rate is less than 1% [1]. Computers are updated every 18 months, and 6 million scrap computers are produced each year in mainland China. This electronic scrap contains a lot of valuable metals. For example, a scrap computer contains 1.2% copper alloy, 1% precious metal alloy (such as gold, silver, and palladium), and each ton of scrap mobile phones contains about 280 g/t copper, 2 kg/t gold, 100 kg/t silver, and 100 g/t palladium [2]. Therefore, the improper treatment of electronic scrap will cause serious waste of resources and environmental pollution.

Under the nation's increasing emphasis on environmental protection and reuse of resources, the recycling of electronic scrap has received attention from the whole society [3]. At present, electronic scrap is mainly recycled by roadside mobile vendors and maintenance service providers, circulated in second-hand markets, received by regular enterprises for trade-in, and donated to underdeveloped areas. Due to the outdated technologies and equipment of recyclers and dealers, the second-hand electronic products obtained by the mobile traders are divided into non-reusables and reusables. The non-reusables are disassembled by small workshops and then subjected to strong acid or incineration to extract precious metals, with other parts discarded. The reusables are sold after simple repair, refurbishment, or assembly[1]. In recent years, some online and offline recycling platforms resell the scrap mobile phones after simple information erasure. However, the lack of relevant industry standards, laws and regulations, as well as the existence of a large amount of mobile traders, affect the development of formal recycling institutions. As a result, electronic scrap does not go through any or complete information erasure in the process from recycling to reselling. A large amount of user data is left in these products, which may involve the privacy and confidential information of individuals, enterprises, state agencies or even secrecy departments. This will seriously threaten user privacy and even national security [4].

Information erasure uses various technical means to completely erase the private data in a storage device, to avoid data disclosure [5]. Generally, the typical information deletion and formatting functions of an electronic device cannot really erase data on a storage medium of the device. Instead, the data can be recovered through simple processing of dedicated software[6]. Degaussing, thermal removal, physical destruction, and chemical corrosion on the storage medium can completely erase the information, but also destroy the storage medium. Therefore, such methods provide high security but compromise cost efficiency [7]. The preceding technologies cannot meet the technical requirements for erasing information from electronic scrap. There is still another technique: data overwriting, which uses meaningless and irregular data to overwrite raw privacy and confidential data. Theoretically, the original data information can be overwritten after a single overwrite. However, there may be some data residue, so overwriting needs to be performed several times in different ways to ensure the security. Currently, most overwriting software vendors use US DOD 5220.22M standard. According to this standard, random numbers are overwritten first, then complement numbers of the random numbers, and once again, random numbers [8]. In theory, data on the storage medium can be effectively destroyed as long as the number of overwrites is sufficient [9]. The BMB21-2007 standard issued in 2007, Security and Confidentiality Requirements for the Destruction of Carriers Involving State Secrets and Information Erasure, affirms the overwriting technology and specifies the overwrite times and data formats [10]. From this perspective, overwriting is the most cost-effective method for information erasure currently [11].

Information erasure and information recovery, as two important components of information security, are both complementary and contrary. They are complementary because they are indispensable different applications in the information security field. They are contrary in terms of techniques and applications [3]. The continuous improvement of the information recovery technology poses increasingly high requirements for the information erasure technology.

In China, information erasure technologies are still evolving, with the focus on discussions about the importance of information erasure [4]. In recent years, China's research on technologies for erasing information from different storage media has made some progress. For magnetic storage media, literature [5] expounds different information erasure methods and their comparative research, and puts forwards the concepts of remote destruction and self-destruction to ensure information security. Literature [12] analyzes the advantages and disadvantages of various overwriting sequences in the overwriting technology, designs a new overwriting solution, and puts forward a data destruction software system for magnetic medium based on network security. Literature [13] designs an electronic file shredder and information erasure software for magnetic hard drives. And literature [14] provides a convenient, secure, and user-friendly magnetic medium information erasure solution based on Windows file systems. For flash-based storage media, literature [15] analyzes information erasure on solid state memory. Literature [16] designs information recovery and erasure software based on the study of Android storage system. Literature [4] designs a platform-independent fragment information erasure model for SQLite database, to erase fragment information on different platforms, where the model is built on a model-driven framework.

For the information erasure technology itself, literature [17] proposes a new method for erasing residual data. Literature [18] provides an erasure method with customizable overwrite times and literature [19,20] puts forward the idea of parallel data destruction to improve the information erasure efficiency. Literature [21] proposes security level-based information erasure, which is also the mainstream solution for existing information erasure equipment/software in the domestic market. Its overwriting policy is based on some foreign overwriting standards, with the security to be tested. In recent years, a variety of domestic patents for information erasure have emerged in China, but most of them are designed based on information erasure systems. Literature [22,23] designs a portable disk information erasure system and device. Literature [24] designs an overall information erasure system, including a front-end erasure device and a back-end server, and proposes global overwriting+local random overwriting to improve the erasure efficiency. Literature [25] provides a system that realizes real-time monitoring and effective log management of information erasure. Literature [26] designs a QR code-based security management solution for secret-related storage media. And literature [27] provides a process for erasing Linux-based disk files, and their traces and logs.

There are also a few patents for inventions for information erasure technologies. Literature [28,29,30] proposes different solutions for the failure to securely erase user data from the flash memory in time due to remote update, to eliminate common security risks while mitigating system burden. And literature [31] provides a method for erasing information based on a quantum random number generator, to effectively improve the overwriting efficiency and security. In summary, there still lacks theoretical support, technological innovation, and integrated system design for information erasure technologies in China. There also lacks information erasure specifications for civil industry.

Foreign countries give more emphasis on the study of information erasure technologies, and have developed corresponding specifications for each information erasure technology. For example, foreign countries have developed many standards for overwriting technologies, for different users and organizations to securely and efficiently erase information [4]. In the 1990s, foreign scholars had made many achievements in the study of flash-based information erasure. Gutmann [32] made in-depth analysis on the magnetic memory and random access memory, and proposed effective information recovery and information erasure methods. For the effect of applying the mature disk information erasure technologies to flash memory, the study made by Wei et al. [33] and Fundo et al. [34] shows that the complex structure of the latter poses higher requirements for information erasure solutions.

Literature [35] proposes an adaptive hybrid erasure solution based on the cost and benefit model, which can, in theory, securely and efficiently erase data in multiple flash storage modes. Literature [36, 37] provides encryption-based secure file erasure solutions to securely erase file keys and original data. Literature [37] also puts forward trusted information erasure verification solutions for governments, businesses, and individuals. For the issue that the service life of SSDs is damaged by data migration or block cleanup caused by purge- and encryption-based information erasure, Liu et al. proposes an ErasuCrypto solution. This solution not only addresses the said issue, but also erases invalid data [38].

Foreign countries have made deep research into information erasure of smart mobile terminals, and provide well-designed solutions for different physical storage interfaces to securely and effectively erase data from physical storage devices in different file systems [39]. For today's big data systems, Gnatyuk et al. [40] designs STM Shredder software based on a pseudo-random sequence generator to erase information securely and quickly. At present, most of the foreign patents for inventions related to information erasure are about the overwriting technologies. For example, literature [41] designs an overwriting solution based on multiple overwrites and verifications on generated non-compressed data blocks, which can effectively meet the overwriting requirement of “compression before storage” Literature [42] proposes a verification scheme of first performing consistency check and then comparing a known overwrite sequence with a sequence read out after erasure; and literature [43] provides an information erasure solution based on file overwriting.

In general, the information erasure process used by foreign patents is: generating an overwrite sequence based on the capacity of a to-be-erased area, obtaining an address of the area and performing overwriting cyclically, and verifying a result of each overwriting. In addition, the study, application, and industry development of foreign information erasure technologies are quite advanced. Blancco provides a systematic secure information erasure service, its products use more than 20 overwriting standards of different countries, and its customers spread around the globe, but its security is still in doubt. In short, information erasure with multiple-security guarantee that is customized based on features of the electronic scrap is still not available, and there is no research on online correction of erasure solutions. Therefore, it is difficult to balance the efficiency and security in the information erasure process.

To sum up, the complexity of flash memory and the diversity of existing storage media in the market make a generic and effective information erasure solution impossible.

SUMMARY

Recycling the increasing large amount of electronic scrap has become a trend to avoid waste of resources and environmental pollution. However, the difficulty in secure erasure of information in the electronic scrap badly limits the recycling rate. The existing information erasure devices and software in the market mostly use security level-based overwriting technologies, and the overwriting policies are also based on foreign standards, making it hard to ensure a secure and efficient information erasure process for the electronic scrap. To resolve the foregoing problems, this application proposes an information erasure method based on a dual-security mechanism, including a feature data case matching module, an information erasure and recovery evaluation module, an erasure solution adjustment and native system recovery module. Firstly, use a feature data case matching module to match feature data cases by: identifying a storage medium feature of to-be-erased electronic scrap and identifying an information erasure feature of the electronic scrap, then combining with a storage medium feature database and an information erasure feature database to match and determine an erasure solution; and identifying a firmware system feature of the electronic scrap and matching a similar case in a firmware system feature database to determine a native system data package for rewriting after the information is securely erased. Then, use an information erasure and recovery evaluation module to execute the erasure solution for the electronic scrap, extract a feature of the erased electronic scrap in this step, and build an information erasure evaluation model to evaluate an information erasure effect; use an information recovery mechanism to recover the information on the erased electronic scrap, extract an information recovery feature, and build an information recovery evaluation model to evaluate an information recovery effect; and build a comprehensive dual-security evaluation model based on field-specific expert knowledge and the preceding evaluation results, to obtain a comprehensive dual-security evaluation value. Lastly, determine whether the information erasure meets a set standard; and if the information erasure does not meet the set standard, correct an erasure solution according to the evaluation result and evaluating an information erasure and recovery effect, until the information erasure result meets the standard; or if the information erasure meets the set standard, perform a last erasure operation on the erased electronic scrap to erase the recovered information, format the electronic scrap, and write a matching native system into the electronic scrap to obtain the electronic scrap with the native system recovered.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 shows an embodiment of a policy for erasing information from electronic scrap based on a dual-security mechanism.

FIG. 2 is a flowchart of an embodiment of a method for erasing information from electronic scrap based on a dual-security mechanism.

FIG. 3 is a flowchart of an embodiment of a method for matching feature data cases.

FIG. 4 is a flowchart of an embodiment of a method for evaluating information erasure and recovery effects.

FIG. 5 is a flowchart of an embodiment of a method for adjusting an information erasure solution and recovering a native system.

DETAILED DESCRIPTION

In accordance with the inventive concepts, provided is a method for erasing information from electronic scrap based on a dual-security mechanism. The method is implemented using a feature data case matching module, an information erasure and recovery evaluation module, and an erasure solution adjustment and native system recovery module, as shown in FIG. 1 and FIG. 2.

Feature Data Case Matching Module

Inputs of the module are to-be-erased electronic scrap X and field-specific expert knowledge K_now. Outputs are an erasure solution{λ_pass,S,λ_size,λ_order, . . . }, and a native system data package u_data for the to-be-erased electronic scrap. The module is used to obtain the erasure solution, and the native system data package u_data for the to-be-erased electronic scrap, based on the to-be-erased electronic scrap, the field-specific expert knowledge K_now, a storage medium feature database D_medium^data, an information erasure feature database D_delete^data, and a firmware system feature database D_firmware^data. A mapping relationship is as follows:{{λ_pass,S,λ_size,λ_order, . . . },u_data}=f_match(X,D_medium^data,D_delete^data,D_firmware^data)  (1)where f_match(⋅) (represents a mapping model for feature extraction and case matching of the to-be-erased electronic scrap. λ_pass represents overwrite times. S={s₁,L,s_λpass} represents a set of λ_pass overwrite sequences. s₁,L,s_λpass may be all-0s sequences, all-1s sequences, random sequences, uncompressed sequences or fixed sequences. λ_size represents the size of the overwrite packet; λ_order represent the arrangement order of different overwrite sequences in a overwrite packet.Information Erasure and Recovery Evaluation Module

Inputs of the module are the erasure solution {λ_pass,S,λ_size,λ_order, . . . }, and the field-specific expert knowledge K_now. Outputs are the information erasure feature{η_delete^null,η_delete^damage, . . . }, the information recovery feature {η_recover^iden,η_recover^back, . . . }, and a comprehensive dual-security evaluation value ξ_combine. The module is used to perform, based on the erasure solution and field-specific expert knowledge, information erasure and evaluation on the to-be-erased electronic scrap, perform information recovery and evaluation, and obtain the comprehensive dual-security evaluation value. A mapping relationship is as follows:{{η_delete^null,η_delete^damage, . . . },{η_recover^iden,η_recover^back, . . . },ξ_combine}=f_estimate({λ_pass,S,λ_size,λ_order, . . . },X,K_now)  (2)where f_estimate(⋅) represents a mapping model for implementing the foregoing process.Erasure Solution Adjustment and Native System Recovery Module

Inputs of the module are the information erasure feature {η_delete^null,η_delete^damage, . . . }, the information recovery feature {η_recover^iden,η_recover^back, . . . }, the comprehensive dual-security evaluation value ξ_combine, solution {λ_pass,S,λ_size,λ_order, . . . }, the native system data package u_data, and the field-specific expert knowledge K_now. Outputs are a corrected erasure solution {λ_pass^new,S^new,λ_size^new,λ_order^new, . . . }, and the electronic scrap with the native system recovered Z. The module is used to completely erase the information based on corrected information erasure solutions, to obtain the electronic scrap with the native system recovered. A mapping relationship is as follows:{{λ_pass^new,S^new,λ_size^new,λ_order^new, . . . },Z}=f_UpWr({η_delete^null,η_delete^damage, . . . },{η_recover^iden,η_recover^back, . . . },ξ_combine,{λ_pass,S,λ_size,λ_order, . . . },u_data,K_now)  (3)where f_UpWr(⋅) represents a mapping model that can represent the erasure solution adjustment and native system recovery process.

As shown in FIG. 3, in a preferred embodiment, the workflow for the feature data case matching module is as follows. The inputs of the module are the to-be-erased electronic scrap X and the field-specific expert knowledge K_now. The outputs are the erasure solution {λ_pass,S,λ_size,λ_order, . . . }, and the native system data package u_data for the to-be-erased electronic scrap. The intermediate processing process includes sub-modules such as hardware scan, temporary information backup, storage medium feature identification, information erasure feature identification, erasure solution matching, firmware system feature identification, and firmware system feature matching.

A basic structure of the designed storage medium feature database D_medium^data is as follows:[{MediumType,Capacity,AdressMode,WRmode, . . . },{λ_pass,λ_size,λ_order, . . . }]  (4)where {MediumType,Capacity,AdressMode,WRmode,L} represents features of a storage medium such as a type, a capacity, an addressing mode, and a read/write mode.

A basic structure of the designed information erasure feature database D_delete^data is as follows:[{ComStorage,ComFormat,ComCode,PriName,PriLang,PriFormat,PriPath, . . . },{S, . . . }]  (5)where {ComStorage,ComFormat,ComCode,PriName,PriLang,PriFormat,PriPath,L} represents information erasure features such as a common data storage location, a common data format, a common data encoding rule, an independent software name, an independent software development language, an independent software data format, and an independent software installation path.

A basic structure of the designed firmware system feature database D_firmware^data is as follows:[{SysType,SysVersion,DevLang,RunMech,FileDir,L},u_data]  (6)where {SysType,SysVersion,DevLang,RunMech,FileDir,L} represents firmware system features such as a system type, a system version, a development language, an operating mechanism, and a standard file directory.

The following process is performed for the to-be-erased electronic scrap X. Firstly, connect to the to-be-erased electronic scrap through a data interface, scan the to-be-erased electronic scrap in a hardware scan module, temporarily back up the information contained in the to-be-erased electronic scrap, and identify the storage medium feature, the information erasure feature, and the firmware system feature based on the scanned information. This process is expressed as follows:

where f_scan(⋅) represents the process of scanning the to-be-erased electronic scrap and f_iden(⋅) represents the process of identifying the storage medium feature d_medium^fea, the information erasure feature d_delete^fea, and the firmware system feature d_firmware^fea of the to-be-erased electronic scrap.

Next, combine with the identified storage medium feature d_medium^fea, the identified information erasure feature d_delete^fea, the storage medium feature database D_medium^data and the information erasure feature database D_delete^data to match and obtain the erasure solution {λ_pass,S,λ_size,λ_order, . . . }. This process is expressed as follows:{λ_pass,S,λ_size,λ_order, . . . }=f_match^solution(d_medium^fea,d_delete^fea,D_medium^data,D_delete^data)  (8)where f_match^solution is a matching model based on a case reasoning algorithm.

Reference for the case-based reasoning algorithm: Yan Aijun, Qian Limin, Wang Pu: A Comparative Study on a Weight Allocation Model for Case Reasoning Attributes, ACTA Automatica Sinica, 2014, 40(09):1896-1902.

Then, match the identified firmware system feature d_firmware^fea with the firmware system feature database D_firmware^data to obtain the native system data package u_data. This process is expressed as follows:u_data=f_match^firmware(d_firmware^fea,D_firmware^data)  (9)where f_match^firmware(⋅) is a matching model based on a case reasoning algorithm.

Reference for the case-based reasoning algorithm: Yan Aijun, Qian Limin, Wang Pu: A Comparative Study on a Weight Allocation Model for Case Reasoning Attributes, ACTA Automatica Sinica, 2014, 40(09):1896-1902.

As shown in FIG. 4, the workflow for the information erasure and recovery evaluation module is as follows. The inputs of the module are the erasure solution {λ_pass,S,λ_size,λ_order, . . . }, and the field-specific expert knowledge K_now. The outputs are the information erasure feature {η_delete^null,η_delete^damage, . . . }, the information recovery feature {η_recover^iden,η_recover^back, . . . }, and the comprehensive dual-security evaluation value ξ_combine. The intermediate process includes sub-modules such as information erasure, information erasure feature extraction, erasure effect evaluation, information recovery, information recovery feature extraction, information recovery effect evaluation, and comprehensive dual-security evaluation.

The specific implementation process is as follows: erase the information of the to-be-erased electronic scrap X based on the given erasure solution {λ_pass,S,λ_size,λ_order, . . . }, extract an information erasure feature of the erased electronic scrap, and evaluate an information erasure effect. This process is expressed as follows:

where X_delete represents the erased electronic scrap; ξ_delete represents an information erasure evaluation value obtained by an information erasure model f_delete^estimate(⋅); f_delete(⋅) represents the process of erasing the information; f_delete^extra(⋅) represents the process of extracting the information erasure feature of X_delete; and f_delete^estimate(⋅) represents an erasure evaluation model that is implemented using a neural network algorithm. Reference for the neural network algorithm: Li Long, Wei Jing, Li Canbing, Cao Yijia, Song Junying, Fang Baling: Load Model Prediction Based on an Artificial Neural Network, Transactions of China Electrical Technology, 2015, 30(08):225-230.Here {η_delete^null,η_delete^damage,L} represents features such as an available space percentage and a damaged space percentage of X_delete, and calculation formulas are as follows:

$\begin{array}{cc}{\eta }_{\mathrm{delete}}^{\mathrm{null}}=\frac{\mathrm{Available}\mathrm{storage}\mathrm{capacity}}{\mathrm{Rated}\mathrm{storage}\mathrm{capacity}-\mathrm{Used}\mathrm{storage}\mathrm{capacity}}\times 100\%& \left(11\right)\end{array}$

$\begin{array}{cc}{\eta }_{\mathrm{delete}}^{\mathrm{damage}}=\frac{\mathrm{Damaged}\mathrm{storage}\mathrm{capacity}}{\mathrm{Rated}\mathrm{storage}\mathrm{capacity}-\mathrm{Used}\mathrm{storage}\mathrm{capacity}}\times 100\%& \left(12\right)\end{array}$

Use the information recovery mechanism to recover the information of X_delete, extract an information recovery feature of the recovered electronic scrap X_recover, and evaluate an information recovery effect. This process is expressed as follows:

where X_recover represents the electronic scrap obtained after information recovery is performed on X_delete; ξ_recover represents an information recovery evaluation value obtained by an information recovery evaluation model f_recover^estimate(⋅); f_recover(⋅) represents the process of recovering the information; f_recover^extra(⋅) represents the process of extracting the information recovery feature of X_recover; and f_recover^estimate(⋅) represents an information recovery evaluation model that is implemented using the neural network algorithm.

Reference for the neural network algorithm: Li Long, Wei Jing, Li Canbing, Cao Yijia, Song Junying, Fang Baling: Load Model Prediction Based on an Artificial Neural Network, Transactions of China Electrical Technology, 2015, 30(08):225-230.

{η_recover^iden,η_recover^back,L} represents features such as a recoverable data identification rate and a recovery degree of X_recover, and calculation formulas are as follows:

$\begin{array}{cc}{\eta }_{recover}^{\mathrm{iden}}=\frac{\mathrm{Identifiable}\mathrm{data}\mathrm{fragment}\mathrm{bytes}}{\mathrm{Total}\mathrm{original}\mathrm{data}\mathrm{file}\mathrm{bytes}}\times 100\%& \left(14\right)\end{array}$ $\begin{array}{cc}{\eta }_{recover}^{back}=\frac{\mathrm{Recovered}\mathrm{data}\mathrm{units}}{\mathrm{Original}\mathrm{data}\mathrm{file}\mathrm{units}}\times 100\%& \left(15\right)\end{array}$

Evaluate the information erasure and recovery processes based on the field-specific expert knowledge K_now and the preceding information to obtain the comprehensive dual-security evaluation value ξ_combine. This process is expressed as follows:ξ_combine=f_{del_recover}^estimate({η_delete^null,η_delete^damage,L},{η_recover^iden,η_recover^back,L},ξ_delete,ξ_recover,K_now)  (16)where f_{del_recover}^estimate(⋅) represents a comprehensive dual-security evaluation model that is implemented using a case reasoning algorithm.

Reference for the case-based reasoning algorithm: Yan Aijun, Qian Limin, Wang Pu: A Comparative Study on a Weight Allocation Model for Case Reasoning Attributes, ACTA Automatica Sinica, 2014, 40(09):1896-1902.

As shown in FIG. 5, the workflow for the erasure solution adjustment and native system recovery module is as follows. The inputs of the module are the information erasure feature {η_delete^null,η_delete^damage, . . . }, the information recovery feature {η_recover^iden,η_recover^back, . . . }, the comprehensive dual-security evaluation value ξ_combine, erasure solution {λ_pass,S,λ_size,λ_order, . . . }, the native system data package u_data, and the field-specific expert knowledge K_now. The outputs are the corrected erasure solution {λ_pass,S^new,λ_size^new,λ_order^new, . . . }, and the electronic scrap with the native system recovered Z. The intermediate processing process includes sub-modules such as determining of a dual-security evaluation result, erasure solution correction, information erasure, and native system recovery.

The specific implementation process is as follows. Firstly, compare the obtained comprehensive dual-security evaluation value λ_combine with a threshold ξ_set specified based on the expert knowledge. If ξ_combine^new≤ξ_set, perform a last erasure operation, and then recover the native system. This process is expressed as follows:

where f_delete^last(⋅) represents the process of the last erasure operation; and X_delete^last represents the electronic scrap obtained after the f_delete^last(⋅) operation is performed on X_recover. If ξ_combine>ξ_set, correct the information erasure solution based on a storage medium feature d_medium^fea, an information erasure feature d_delete^fea, a firmware system feature d_firmware^fea, and the native system data package u_data of the to-be-erased electronic scrap, an erasure solution {λ_pass,S,λ_size,λ_order, . . . } used by the information erasure and recovery evaluation module, a storage medium feature database D_medium^data, an information erasure feature database D_delete^data, a firmware system feature database D_firmware^data, and the field-specific expert knowledge K_now. This process is expressed as follows:{λ_pass^new,S^new,λ_size^new,λ_order^new, . . . }=f_update(ξ_combine,d_medium^fea,d_delete^fea,d_firmware^fea,{λ_pass,S,λ_size,λ_order, . . . },D_medium^data,D_delete^data,D_firmware^data,K_now,X,{η_delete^null,η_delete^damage, . . . },{η_recover^iden,η_recover^back, . . . })  (18)where f_update(⋅) represents a mapping model for updating the erasure solution, which is implemented using a case reasoning algorithm.

Reference for the case-based reasoning algorithm: Yan Aijun, Qian Limin, Wang Pu: A Comparative Study on a Weight Allocation Model for Case Reasoning Attributes, ACTA Automatica Sinica, 2014, 40(09):1896-1902.

Above d_medium^fea, d_delete^fea, and d_firmware^fea represent the storage medium feature, the information erasure feature, and the firmware system feature of the to-be-erased electronic scrap; {η_delete^null,η_delete^damage, . . . } represents the information erasure feature obtained after information erasure is performed on X; and {η_recover^iden,η_recover^back, . . . } represents the information recovery feature obtained after information recovery is performed on X_delete.

The original information erasure solutions are replaced with new information erasure solution:{λ_pass,S,λ_size,λ_order, . . . }←{λ_pass^new,S^new,λ_size^new,λ_order^new, . . . }  (19)

Then, use the new information erasure solution to evaluate an information erasure and new recovery effect until anew comprehensive dual-security evaluation value ξ_combine^new≤ξ_set is obtained.

To sum up, this application proposes a method for erasing information based on a dual-security mechanism. Its novelty lies in intelligently generating, based on features of to-be-erased electronic scrap, an erasure solution that can achieve both efficiency and security. Its main idea is to: build a storage medium feature database, an information erasure feature database, and a firmware system feature database to match cases for the to-be-erased electronic scrap; generate an erasure solution and a native system data package based on the matching results; erase the information, evaluate an erasure result, recover the information on the erased electronic scrap, and evaluate a recovery result, to implement comprehensive double security evaluation; check the information erasure validity of the electronic scrap based on the evaluation results; and if an erasure result is invalid, correct erasure solutions online based on the evaluation result, until the erasure result is valid and the electronic scrap with a native system recovered is obtained. The method according to this application provides information erasure solutions for different systems. Case matching improves the information erasure efficiency. Information erasure and recovery evaluation ensures secure erasure. Erasure solution correction provides a unified solution for different electronic scrap and achieves both efficiency and security in the information erasure process.

REFERENCES

• 1. Zhou Jing: Study on Promoting the Development of China's Scrap Mobile Phone Recycling Industry Through Standardization[J], Quality and Standardization, 2018(01):41-43.
• 2. Ye Diqing: On Recycling of Computers, Mobile Phones, and Other Electronic Scrap[J], Resource Conservation and Environmental Protection, 2014(01):144.
• 3. Wang Jianfeng: Data destruction: an Important Branch in Data Security[J], Computer Security, 2006(08):53-54.
• 4. Ye Qingming: Research on MDA-based SQLite Fragment Data Erasure Technology[D], Kunming University of Science and Technology, 2016.
• 5. Xu Jing, Zhu Youdian, Lai Fan: Data Destruction Technology for Magnetic Storage Media[J], Journal of Southwest China Normal University (Natural Science Edition), 2007(04):107-110.
• 6. Yin Yanbin, Wen Weiping: Secure Computer Data Erasure and Privacy Protection[J], Information Network Security, 2009(05):55-58.
• 7. Chen En, Liu Xiaojie, Li Tao, Lu Zhengtian, Hu Xiaoqin, Dong Chengxi: Design and Implementation of a Database Destruction System[J], Computer Engineering and Design, 2008(10):2499-2501.
• 8. Wang Ben, Zhu Dali, Sun Degang: Research and Verification of Overwriting-based Information Erasure Technology[C], Proceedings of the 18th National Conference on Information Confidentiality, Beijing: Jincheng Publishing House, 2008.
• 9. Wei Chengwei, Liu Liwei, Wang Nan, Li Yong: Application and Research of Data Destruction Technology[J], Computer and Modernization, 2010(10):180-182.
• 10. Li Tao: Information Storage and Information Destruction Technology[J], Information Security and Technology, 2010(06):45-48.
• 11. Shen HeLei, On Information Erasure Technology[J], Information Security and Communication Confidentiality, 2009(08):92-94.
• 12. Lu Zhengtian, Li Tao, Hu Xiaoqin, Zhao Kui, Zeng Jinquan, Peng Lingxi: A Data Self-Destruction Method[J], Application Research of Computers, 2009, 26(01):350-351+355.
• 13. Wang Ben: Research and Implementation of Storage Media Information Erasure Technology[D], Beijing Jiaotong University, 2009.
• 14. Bai Yang: Research and Implementation of Windows-based Magnetic Medium Data Erasure Technology[D], Hubei University of Technology, 2010.
• 15. Qi Feng, Gao Qi, He Peng: Research on Method for Erasing Information from Solid State Storage and Devices[J], Secrecy Science and Technology, 2011(06):54-59.
• 16. Sun Dian: Design and Implementation of Android-based Data Recovery and Erasure Software[D], Beijing University of Posts and Telecommunications, 2014.
• 17. Zhou Kaimin, Zhao Qiang, Zhang Xiao, Deng Gaoming: Research on Residual Data Erasure and Security[J], 2006(17):2769-2771.
• 18. Wu Shasha, Wang Minshen, Wu Yiping, Xiong Jinbo: Secure Erasure of Data on Storage Media, Computer Systems & Applications, 2017, 26(11):36-44.
• 19. Shang Fang, Wang Xiaoyu, Li Sheng, Zhang Tong: Development and Application of a Parallel High-Speed Electronic Data Destruction Platform[J], Heilongjiang Electric Power, 2017, 39(06):547-550+556.
• 20. Huang Wuming, Ma Zheng, Ji Xiaojiang: Method and System for Performing ROM Flashing or Data Erasure on Multiple Mobile Concurrently[P], Chinese Patent: CN108271150A, 2018 Jul. 10.
• 21. Du Yinxia: Study on Securely Erasing Residual Data from Hard Drives[D], Hebei University of Science and Technology, 2012.
• 22. Han Bing, Zhang Tong, You Junsheng, Wang Xiaoyu, Shang Fang, Liu Sheng, Zhong Zhichen: Portable Electronic Data Erasure Apparatus Based on USB Flash Drive[P], Chinese Patent: CN103294960A, 2013 Sep. 11.
• 23. Han Bing, Zhang Tong, You Junsheng, Wang Xiaoyu, Shang Fang, Liu Sheng, Zhong Zhichen: Portal Apparatus for Erasing Data from Electronic Storage Medium and Data Erasure Method Based on Said Apparatus[P], Chinese Patent: CN103295638A, 2013 Sep. 11.
• 24. Liu Jun, Xu Changle, Geng Lida: Method for Completely Erasing Original Information on Smartphone[P], Chinese Patent: CN107277089A, 2017 Oct. 20.
• 25. Yang Guodong, Kang Hongjuan, Guo Xin, Tan Yuanquan: Information Erasure System and Method for Android Device[P], Chinese Patent: CN109683918A, 2019 Apr. 26.
• 26. Jing Yixin, Han Min, Tang Wei, Yu Pengfei, Liao Wei: Apparatus and Method for Erasing Information from Secret-related Storage Medium[P], Chinese Patent: CN103117084A, 2016 Apr. 13.
• 27. Wang Xinxin, Wang Jinguo, Cui Xin'an: Method and Apparatus for Completely Erasing Files Based on Linux File System[P], Chinese Patent: CN109656888A, 2019 Apr. 19.
• 28. Shi Jia, Zhang Hengliang, He Xiaoyan: Method and Apparatus for Erasing Data from Storage System[P], Chinese Patent: CN103902468A, 2014 Jul. 2.
• 29. Wang Yuan, Xia Fan, Chen Ningyi, Zhou Yunfeng: Method and Apparatus for Erasing Information on Terminal Device[P], Chinese Patent: CN106909542A, 2017 Jun. 3.
• 30. Li Minghao: Data Erasure Method and Apparatus[P], Chinese Patent: CN107562376A, 2018 Jan. 9.
• 31. He Yuanhang, Yang Jie, Zhang Liangliang, Liu Jinlu, Liu Jia, Xu Bingjie: Data Erasure Method Based on Quantum Random Number[P], Chinese Patent: CN107562384A, 2018 Jan. 9.
• 32. Gutmann P. Secure deletion of data from magnetic and solid-state memory[C]//Proceedings of the Sixth USENIX Security Symposium, San Jose, Calif. 1996, 14: 77-89.
• 33. Wei M Y C, Grupp L M, Spada F E, et al. Reliably Erasing Data from Flash-based Solid State Drives[C]//FAST. 2011, 11: 8-8.
• 34. Fundo A, Hysi A, Tafa I. Secure Deletion of Data from SSD[J]. 2014.
• 35. Sun K, Choi J, Lee D, et al. Models and design of an adaptive hybrid scheme for secure deletion of data in consumer electronics[J]. IEEE Transactions on Consumer Electronics, 2008, 54(1): 100-104.
• 36. Lee J, Heo J, Cho Y, et al. Secure deletion for NAND flash file system[C]//Proceedings of the 2008 ACM symposium on Applied computing. ACM, 2008: 1710-1714.
• 37. Swanson S, Wei M. Safe: Fast, verifiable sanitization for ssds[J]. University of California, San Diego, Tech. Rep, 2010.
• 38. Liu C, Khouzani H A, Yang C. Erasucrypto: A light-weight secure data deletion scheme for solid state drives[J]. Proceedings on Privacy Enhancing Technologies, 2017, 2017(1): 132-148.
• 39. Reardon J, Basin D, Capkun S. Sok: Secure data deletion[C]//2013 IEEE symposium on security and privacy. IEEE, 2013: 301-315.
• 40. Gnatyuk S, Kinzeryavyy V, Sapozhnik T, et al. Modern Method and Software Tool for Guaranteed Data Deletion in Advanced Big Data Systems[C]//International Conference of Artificial Intelligence, Medical Engineering, Education. Springer, Chain, 2018: 581-590.
• 41. Vaisanen K, Lalli L, Brew J. Apparatus, a system, a method and a computer program for erasing data stored on a storage device: U.S. Pat. No. 9,286,231[P]. 2016 Mar. 15.
• 42. LeGargean B, Gonzalo V. Method and system for verifying a data erasure process: WO. Patent 18/052,703[P]. 2018 Feb. 2.
• 43. Kellokoski P, Torma M, Nurminen P, et al. Data Erasure Method and Apparatus: U.S. patent application Ser. No. 16/334,344[P]. 2019 Jul. 11.

Claims

1. A computer-implemented method for erasing information based on a dual-security mechanism, comprising: a) identifying a storage medium feature of an electronic scrap X and identifying an information erasure feature of the electronic scrap X, and matching identified storage medium feature and identified information erasure feature in a storage medium feature database and an information erasure feature database to determine an erasure solution; andidentifying a firmware system feature of the electronic scrap X and matching identified firmware system feature in a firmware system feature database to determine a native system data package for rewriting after the information is securely erased;wherein step a) is expressed as follows: {λpass,S,λsize,λorder, . . . }=fmatchsolution(dmediumfea,ddeletefea,Dmediumdata,Ddeletedata), andudata=fmatchfirmware(dfirmwarefea,Dfirmwaredata);wherein:fiden(⋅) represents the process of identifying the storage medium feature dmediumfea, the information feature ddeletefea and the firmware system feature dfirmwarefea of the electronic scrap X;fmatchsolution represents the process of matching the identified storage medium feature and the identified information feature in the storage medium feature database Dmediumdata and the information erasure feature database Ddeletedata to output the erasure solution {λpass,S,λsize,λorder, . . . },Ddeletedata represents information erasure features comprising a common data storage location, a common data format, a common data encoding rule, an independent software name, an independent software development language, an independent software data format and an independent software installation path;λpass represents overwrite times, S represents a set of overwrite sequences, λsize represents a size of an overwrite packet, andλorder represents an arrangement order of overwrite sequences in the overwrite packet;fmatchfirmware(⋅) represents the process of matching the identified firmware system feature in the firmware system feature database Dfirmwaredata to output the native system data package udata;b) executing the erasure solution for the electronic scrap X to obtain an electronic scrap Xdelete, extracting a feature of the electronic scrap Xdelete, and evaluating an erasure effect by using an information erasure evaluation model;wherein step b) is expressed as follows:

2. The method of claim 1, wherein step a) further comprises scanning the electronic scrap X and temporarily backing up the information contained in the electronic scrap X.

3. The method of claim 1, wherein the storage medium feature database Dmediumdata is expressed by [{MediumType,Capacity,AdressMode,WRmode, . . . },{λpass,λsize,λorder, . . . }],wherein {MediumType,Capacity,AdressMode,WRmode, . . . } represents storage medium features comprising a type, a capacity, an addressing mode and a read/write mode;