CN1625121A - Hierarchical cooperated network virus and malice code recognition method - Google Patents

Abstract

A layered coordinate network virus and vicious code recognizing method. The characteristics are: use strong self-protection mechanism of biological immunity for reference, correspond the network virus and vicious code recognizing method to the multi-level protective mechanism of biological immunity system, judge the dangerous degree of the stand-by detected script through statistically analyzing the word frequency of the key words, based on the point view of 'self-collection' of register form operation analyze and judge the exceptional behavior of the register form written in the form route, and recognize non-self the executing sequence of the programming interface of the applied program, at last send all the exceptional behavior information to the network control station through the network. It well solves the problem of the identification of the unknown network virus and vicious code, and has good capacity of identification, realizes the monitoring and management of the network virus and vicious code of single system and the whole sub-network.

Description

A kind of internet worm of layered cooperative and malicious code recognition methods

Technical field:

The invention belongs to the computer network security technology field, particularly relate to the recognition technology of internet worm and malicious code.

Background technology:

According to Institute of Electrical and Electric Engineers " potentiality " magazine (the IEEE POTENTIALS that publishes in the U.S., October calendar year 2001 fourth phase 16-19 page or leaf) introduce, existing computer anti-virus recognition technology roughly can be divided into following several: (1) is primarily aimed at known viruse based on the scanning of condition code.(2) virtual machine technique, its basic thought are suspicious program to be placed to carry out under the virtual machine environment be used to judge whether be virus, but still face the effect of virtual machine and the problems such as self-security that how to guarantee virtual machine at present.(3) heuristic, its basic thought are to attempt to detect family viral and detect unknown virus by extensive condition code.This method usually depends on condition code technology and virtual machine technique, and the recognition effect to unknown virus also haves much room for improvement at present.(4) behavioural analysis method promptly utilizes the peculiar behavior that monitors virus to detect the method for virus.This method requires at first to summarize the general behavior pattern of virus, designs the finite state machine pattern of the corresponding behavior then, and state transition is corresponding to the behavior of program, and receive status is for detecting virus.The problem of this method is to the new virus that emerges in an endless stream, and is difficult to summarize a general behavior pattern.(5) verification and method.This method generates a check information and preservation in the initial condition of machine, (verification failure) reports to the police when check information generation ANOMALOUS VARIATIONS then, the subject matter of this method is that to implement expense too big, also faces problems such as the installation of new application program and edition upgrading simultaneously.Generally speaking, in existing computer anti-virus technology, the condition code scanning technique is mainly used in the identification known viruse, and all the other various recognition technologies that propose at unknown virus all also have shortcoming and limitation separately.

Because internet worm and malicious code are just to come into vogue in recent years and bringing the network safety event of serious harm, a kind of methods of computer virus infection that prevent that Chinese patent application numbers 96114050 proposes can only be taken precautions against the early stage computer virus of part, the present this anti-virus cards market of thoroughly having faded out; Chinese patent application number 96109573 firewall systems that propose are that connection or information into and out of internal network are carried out safety inspection, do not have the ability of recognition network virus and malicious code substantially.Therefore, these technology are not suitable for the identification of internet worm and malicious code.

Summary of the invention:

Deficiency at existing network virus and malicious code recognition technology, the present invention proposes a kind of internet worm and malicious code recognition methods of layered cooperative, to solve the abnormal behaviour identification problem of unknown network virus and malicious code, realize monitoring to virus of the unknown network in individual system and the whole subnet and malicious code abnormal behaviour.

The internet worm of layered cooperative of the present invention and malicious code recognition methods, comprise: from script file, isolate keyword, obtain by the method for injecting dynamic link library (Dynamic Linked Library: be called for short DLL) that application programming interface (Application Programming Interface: be called for short API) is carried out sequence and registration table writes the list item path, registration table is write the list item path and the API sequence is kept in hard disk or the internal memory; It is characterized in that:

To the keyword word frequency statistics analysis of script and make unusual judgement;

Registration table is write the list item path to carry out oneself's identification and makes unusual judgement;

The API sequence is carried out nonego identification and made unusual judgement;

Abnormal behaviour information is sent to net control station;

The script file that described script file is meant the script file write with the Javascript language, write with the VBScript language and embedded Javascript or the script file of VBScript code;

Described injection DLL acquisition API execution sequence and registration table write the list item path and are meant, by DLL is injected in the target program (being program to be monitored) as remote thread, adopt the API of the method intercepting target program of replacing Import Address Table (Import Address Table:IAT) to carry out sequence then, and write the list item path from the parameter acquisition registration table of registration table api function;

Described to script the keyword statistical analysis and make unusual judgement and be meant and from script file, isolate 29 keyword copyfile, Createobject, Delete, FolderDelete, RegWrite, Virus, .Write, GetSpecialFolder, keys, opentextfile, readall, .save, startup, execute, .add, buildpath, copyfolder, createfolder, createtextfile, deletefile, fileexists, folderexists, getfile, getfolder, getparentfolder, format, .run, do copy, document.write, and carry out following steps:

(1) 29 keywords are divided into three groups, first group for creating object keyword: Createobject; Second group is no risky operation keyword itself: Virus .Write, GetSpecialFolder, keys, opentextfile, readall, startup, execute .add, buildpath, fileexists, folderexists, getfile, getfolder, getparentfolder .run, document.write; The 3rd group is to have the keyword that possibility is destroyed operation: copyfile, Delete, FolderDelete, RegWrite .save, copyfolder, createfolder, createtextfile, deletefile, format, do copy;

(2) the desired value f of the word frequency that these 29 keywords occur in the normal script of statistics _i, the desired value f of the word frequency that these 29 keywords occur in the unusual script is added up in 1≤i≤29 _i', 1≤i≤29, it is poor to calculate the normalization word frequency of 29 keywords in normal and unusual script $e_i=(f_i-f_i^{\′})/\underset{i=1}{\overset{29}{\mathrm{\Σ}}}(f_i-f_i^{\′}),1\≤i\≤29;$

(3) the statistics word frequency m that keyword occurs in current script to be detected _i, the risk factor Risk of script to be detected is calculated in 1≤i≤29,

$\mathrm{Risk}=G\underset{i=1}{\overset{29}{\mathrm{\Σ}}}P\left(i\right)F\left(i\right)$

Wherein P (i), F (i) and G are respectively:

(4) risk factor threshold value TH is defined as:

$\mathrm{TH}=\underset{i=0}{\overset{29}{\mathrm{\Σ}}}P\left(i\right)/29$

When risk factor Risk surpasses threshold value TH, send early warning information to net control station;

Describedly registration table is write the list item path carry out oneself identification and make unusual judgement and take following steps:

(1) the normal registration table of target program (program to be monitored) writes the list item path under the collection normal condition, and deposits in the database, and each normal registration table writes the list item path and is called " oneself ", and its set is called " oneself's collection ";

(2) read current registration table and write the list item path, compare, if not in " oneself's collection ", then send abnormal behaviour information to net control station with original in the database " oneself " operation;

Described the API sequence carried out nonego identification and made unusual judgement and take following steps:

(1) API selection operation:

(a) the API sequence of target program under the intercepting normal condition, and be W with the sliding step ₀Mode it is cut into length is L ₀Trail S ₀

(b) the API sequence of target program under the intercepting operation with virus state, and be W with the sliding step ₀Mode it is cut into length is L ₀Trail R ₀

(c) compare trail S ₀And R ₀In different sequences, extract the api function that constitutes these sequences, with these api functions as api function collection to be monitored;

(2) according to selected api function, the API sequence of target program under the intercepting normal condition, and be that W is cut into the string that length is L with it with the sliding step, generate oneself's collection S;

(3) the current API that obtains target program carries out sequence, and is that W is cut into the string that length is L with it with the sliding step, reads N API sequence at every turn and carries out following testing process:

(a) produce initial detector collection D ₀: produce pre-detector at random according to the api function of selecting, filter oneself's (promptly deleting), and then obtain the initial detector collection with the API sequence of oneself's coupling; The matching strategy here is the part matching strategy, and promptly two sequences match and if only if these two character strings are in r position consistency continuously;

(b) more current AP carries out arbitrary detector that sequence and detector are concentrated: if find to mate then this sequence of mark and total matching number added 1, when the total matching number of API sequence to be detected that obtains in real time reaches threshold value G _nThe time, send out abnormal behaviour information to net control station;

(c) if evolutionary generation t surpasses threshold value G _eOr all the API sequences are labeled, continue to read next group API sequence and detect; Otherwise, for unmatched API sequence, then according to the variation of affinity degree, gene library evolution, three subset D producing at random _A, D _G, D _RWith memory collection D _MCommon composition detector collection D of future generation _t=D _A+ D _G+ D _R+ D _M, and D _A, D _G, D _RSubclass satisfies $\frac{D_A}{1}\≈\frac{D_G}{2}\≈\frac{D_M}{1};$

Produce the detector subset D by the variation of affinity degree _A, affinity degree variation is meant that the matching degree of arbitrary detector of concentrating when API sequence and detector is above affinity degree threshold value G _fThe time, produce N by variation _c(N _c〉=1) individual filial generation individuality;

Produce the detector subset D by gene library evolution _G, gene library evolution is meant the selection probability that improves the API that forms valid detector, i.e. P _Api=P _Api+ Δ P; And when reality generates detector, select probability to generate pre-detector by the roulette wheel method according to API, filter the oneself at last and generate the detector subset D _G

By producing the detector subset D at random _R

The existing detector that can mate unusual sequence is formed memory collection D _M

Described net control station is meant to be used for receiving script, registration table is write the network program that list item path and API sequence are carried out the abnormal information that analyzing and processing obtained.

Compared with prior art, the invention has the advantages that:

1,29 selected keyword word frequency obtain the normalization word frequency in normal script and the unusual script by adding up in the present invention, and provide the risk factor that risk factor and risk factor threshold value calculation method are judged script to be detected based on this, solved the identification problem of malicious script.

2, the angle that the present invention is based on registry operations " oneself's collection " comes the discriminatory analysis registration table to write the abnormal behaviour in list item path, is applicable to all types of target program.

3, the present invention will comprise gene library evolution, produce at random, four learning and Memory modules of the variation of affinity degree and memory collection and the abnormality detection that API carries out sequence combine, make that the API sequence is carried out the abnormality detection effect of nonego identification is better, and be applicable to all types of target program.

4, the present invention uses for reference the powerful self-protective mechanism of biological immune; to carry out the keyword statistical analysis to script first, registration table is write the list item path carry out oneself identification, API is carried out sequence carry out nonego and discern the abnormal behaviour of uniting to target program these three aspects and monitor, make to the recognition effect of unknown network virus and malicious code better.

5, adopt registration table that the present invention can automatic full and accurate ground logging program to write the list item path and API carries out sequence, viral and malicious code provides the firsthand information for further phase-split network.

In sum; the present invention uses for reference the powerful self-protective mechanism of biological immune; the multilayer protection mechanism of internet worm and malicious code recognition technology and Immune System is mapped; respectively from script is carried out the keyword statistical analysis; registration table is write the list item path carry out oneself's identification; API execution sequence is carried out nonego discern the abnormal behaviour identification problem that these three aspects have solved unknown network virus and malicious code preferably; and then solved the problem that prior art is difficult to discern to virus mutation and unknown virus; not only realized monitoring, and made that the keeper can be by security situation in real time monitoring and the management of net control station to whole subnet internet worm in the individual system and malicious code abnormal behaviour.

Description of drawings:

Fig. 1 is that the present invention carries out the internet worm of layered cooperative and the workflow diagram of malicious code identification.

Embodiment:

Below in conjunction with accompanying drawing and example the inventive method is done further concrete description.

Embodiment 1:

1, utilizes several universal miniature personal computers, be linked to be a network environment by switch

Concrete employing is three Pentium IV microcomputers in the present embodiment, with a Dell notebook, and enterprise servers, add a Great Wall 24 port one 0M/100M self adaptation Ethernet switch GES-1125 switches, three Pentium IV of several microcomputers microcomputer, a Dell notebook and enterprise servers are linked to be a network by switch.

Fig. 1 has provided present embodiment and has carried out the internet worm of layered cooperative and the workflow of malicious code identification.The direction of arrow has indicated workflow to order, and the arrow afterbody is next step input, and the arrow end is next step operation of carrying out.Wherein a Pentium sequence microcomputer is used for operational network control desk 1, remaining two Pentium IV microcomputer, Dell notebook and enterprise servers all are used for carrying out script are carried out keyword word frequency statistics analysis 2, registration table is write the list item path carry out oneself identification 3 and API is carried out sequence carry out nonego and discern 4, and the analysis result of these three aspects is all sent to net control station 1.

2, to the keyword statistical analysis of script and make malicious code and judge unusually

As script is carried out the keyword word frequency statistics analyze 2 among Fig. 1, specifically take following operating procedure:

(1) collects a large amount of normal foot presents and malicious script file, suggestion normal foot presents and malicious script file all are no less than 50, isolate 29 keyword copyfile from script file, Createobject, Delete, FolderDelete, RegWrite, Virus, .Write, GetSpecialFolder, keys, opentextfile, readall, .save, startup, execute, .add, buildpath, copyfolder, createfolder, createtextfile, deletefile, fileexists, folderexists, getfile, getfolder, getparentfolder, format, .run, do copy, document.write:

(2) 29 keywords are divided into three groups, first group for creating object keyword: Createobject, second group is no risky operation keyword: Virus itself, .Write, GetSpecialFolder, keys, opentextfile, readall, startup, execute, .add, buildpath, fileexists, folderexists, getfile, getfolder, getparentfolder, .run, document.write, the 3rd group is to have the keyword that possibility is destroyed operation: copyfile, Delete, FolderDelete, RegWrite, .save, copyfolder, createfolder, createtextfile, deletefile, format, do copy;

(3) as the normal script keyword word frequency statistics A1 among Fig. 1: the desired value f that adds up the word frequency that these 29 keywords occur in the normal script _i(1≤i≤29);

(4) as the unusual script keyword word frequency statistics A2 among Fig. 1: the desired value f of the word frequency that these 29 keywords occur in the statistics malicious script _i' (1≤i≤29);

(5) as the calculating normalization word frequency A3 among Fig. 1: it is poor to calculate the normalization word frequency of 29 keywords in normal and unusual script $e_i=(f_i-f_i^{\′})/\underset{i=1}{\overset{29}{\mathrm{\Σ}}}(f_i-f_i^{\′})(1\≤i\≤29);$

(6) as the script A4 to be detected of the analysis among Fig. 1: from hard disk, read the script file of appointment or from the temporary file catalogue of browser (as IExplore.exe), read the script file that browser is being visited, the statistics word frequency m that these 29 keywords occur in this script _i

(7) calculate A5 as the risk factor among Fig. 1: calculate the risk factor Risk of script to be detected,

$\mathrm{Risk}=G\underset{i=0}{\overset{29}{\mathrm{\Σ}}}P\left(i\right)F\left(i\right)$

Wherein P (i), F (i) and G are respectively:

(8) calculate the risk factor threshold value, the computational methods of risk factor threshold value TH are:

$\mathrm{TH}=\underset{i=0}{\overset{29}{\mathrm{\Σ}}}P\left(i\right)/29$

(9) as the transmission early warning information A6 among Fig. 1: when risk factor Risk surpasses threshold value TH, early warning information is sent to net control station 1 (Socket according to Windows operating system works out corresponding transmitting/receiving program) by network.

3,, registration table is write the list item path carries out oneself and discern and make unusual judgement and can take following implementation step as registration table is write the list item path carry out oneself identification 3 among Fig. 1:

(1) write list item path B1 as the intercepting registration table among Fig. 1: the DLL that injects intercepting registration table api function is to target program, as IExplore.exe and Outlook.exe, obtain registration table api function implementation status and parameter, and write the list item path from the parameter acquisition registration table of registration table api function.Inject the method for DLL and can use the remote thread method for implanting, the remote thread function can be referring to the CreateRemoteThread among the MSDN, (ImportAddress Table: the API that method Import Address Table) can intercept target program carries out sequence injecting DLL employing substitute I AT, attention will be to GetProcAddress and LoadLibraryA, LoadLibraryExA, LoadLibraryW, LoadLibraryExW does special processing, specifically can publish referring to Microsoft, " Windows kernel programming " (Programming Applications for Windows) of cut in the Jeffree (Jeffrey Ritcher) work;

(2) as the oneself of the collection among Fig. 1 B2: operational objective program under normal condition, as not containing the webpage of malicious code with IExplore.exe visit or collecting the mail that do not contain internet worm and malicious code etc. with Outlook.exe, the normal registration table of collecting target program (is IExplore.exe or Outlook.exe at this) under the normal condition writes the list item path, and deposit in the database, each normal registration table writes the list item path and is called " oneself ", and its set is called " oneself's collection ";

(3) write list item path B3 as the current registration table to be detected of the collection among Fig. 1: in the target program running, the registration table that obtains target program by the DLL that injects in real time writes the list item path, as the operation of the registry writes of IExplore.exe or Outlook.exe, and registration table is write the list item path be kept in the shared drive; Meanwhile, the registration table detection module reads current registration table and writes the list item path from shared drive, compares with original in the database " oneself " operation, discerns B4 as the oneself among Fig. 1; If in " oneself's collection ", then do not send abnormal behaviour information to net control station, as the transmission abnormal behaviour information B5 among Fig. 1.

4,, the API sequence is carried out nonego identification and made unusual judgement and can take following implementation step as API is carried out sequence carry out nonego identification 4 among Fig. 1.

Need to prove: if do not consider speed, can not move (1) step and (2) step, directly use all api functions; Perhaps do not move for (1) step, directly in all api functions, choose.

(1) at first whole api functions are renumberd, and the api function general collection used of definite target program, as the API set C1 of the use among Fig. 1:

(a) because all api function is too much, about 3000, api function can be divided into 20 groups, every group about 150, and at the corresponding DLL that injects of each group api function generation;

(b) these DLL are injected target program respectively,,, and from the file of record, obtain the api function tabulation that target program uses normal and be with operational objective program under the malicious situation as IExplore.exe or Outlook.exe;

(2) API selection operation, choose C2 as the API among Fig. 1:

(a) the API sequence of target program under the intercepting normal condition, and be W with the sliding step ₀Mode it is cut into length is L ₀Trail S ₀, W wherein ₀Value can be for 1 to L ₀Between arbitrary integer, suggestion is got

L ₀Value can be for greater than 8 integer, suggestion gets 8,16,32 or 64;

(b) the API sequence of target program under the intercepting operation with virus state, and be W with the sliding step ₀Mode it is cut into length is L ₀Trail R ₀

(c) compare trail S ₀And R ₀In different sequences, extract the api function that constitutes these sequences, with these api functions as api function collection to be monitored;

(3) renumber C3 as the API among Fig. 1: selected api function is renumberd, so that expression API sequence;

(4) as the oneself of the collection among Fig. 1 C4: according to selected api function, intercept the API sequence of target program under the normal condition, and it is cut into the string that length is L, generate oneself's collection S, wherein W with sliding step W ₀Value can be for 1 to L ₀Between arbitrary integer, suggestion is got L ₀Value can be for greater than 8 integer, suggestion gets 8,16,32 or 64;

(5) the current API that obtains target program carries out sequence, reads N API sequence at every turn and carries out following testing process, and as IExplore.exe or Outlook.exe, suggestion N value is 128, carries out sequence C 5 as the current API that obtains target program among Fig. 1:

(a), produce initial detector collection D as the start detection among Fig. 1 and judge whether termination condition satisfies C7 ₀: produce pre-detector at random according to the api function of selecting, filter oneself's (promptly deleting), and then obtain the initial detector collection with the API sequence of oneself's coupling; The matching strategy here is the part matching strategy, and promptly two sequences match and if only if these two character strings are in r position consistency continuously;

(b) as the coupling C6 among Fig. 1, more current API carries out arbitrary detector that sequence and detector are concentrated: if find to mate then this sequence of mark and total matching number added 1, when the total matching number of API sequence to be detected that obtains in real time reaches threshold value G _nThe time, send out abnormal behaviour information to net control station, as the transmission abnormal behaviour information C8 among Fig. 1;

(c) as the start detection among Fig. 1 and judge whether termination condition satisfies C7, if evolutionary generation t surpasses threshold value G _eOr all the API sequence is labeled, continues next group API sequence is detected;

(d) for unmatched API sequence, then according to the variation of affinity degree, gene library evolution, three subset D producing at random _A, D _G, D _RWith memory collection D _MCommon composition detector collection D of future generation _t=D _A+ D _G+ D _R+ D _M, and D _A, D _G, D _RSubclass satisfies $\frac{D_A}{1}\≈\frac{D_G}{2}\≈\frac{D_M}{1};$

(e) as the variation of the affinity degree among Fig. 1 C9, detector subset D _AProduced by affinity degree variation, the variation of affinity degree is meant that the matching degree of arbitrary detector of concentrating when API sequence and detector is above affinity degree threshold value G _fThe time, produce N by variation _c(N _c〉=1) individual filial generation individuality;

A kind of advise the concrete variation method that adopts can for: surpass the affinity threshold value that makes a variation if current API carries out sequence and arbitrary detector match bit array, generate several a of [1, L] at random, morphed in this detector a position, obtain a filial generation detector; So circulation is 4 times, and each detector that need make a variation is generated 4 filial generation detectors.

(f) as the gene library evolution C10 among Fig. 1: the detector subset D _GProduced by gene library evolution, gene library evolution is meant the selection probability that improves the API that forms valid detector, makes that this API has higher selection probability, i.e. P when generating pre-detector by the roulette wheel method _Api=P _Api+ Δ P.The selection probability that it is pointed out that all API is consistent when beginning, has identical selected probability P _ApiAnd for avoiding local optimum, the step-length of gene library evolution is very little each time, and promptly API selects the incremental change Δ P of probability very little, and for all API, Δ P is identical here;

API selects the code of probability lift portion to be abbreviated as in the gene library evolution:

For (each gene Gene of valid detector)

Begin

Selection probability P [Gene]=P[Gene of this gene Gene]+Δ P.

End

Wherein Δ P is less usually counting.If for any Gene, initial p [Gene] is 100, and Δ P can be made as 0.1 or 0.01.

(g) as the C11 of generation at random among Fig. 1, detector subset D _RBy producing at random, produce detector at random and be meant to concentrate for detector and keep a certain proportion of detector to come from the mode that produces at random that this is in order to keep the diversity of detector at each;

(h) as the memory collection C12 among Fig. 1: memory collection D _MBe made up of the detector that can mate unusual sequence, it both can generate by off-line before beginning detects in real time, and the detector that also can detect unusual sequence in the actual monitoring process joins memory and concentrates;

5, net control station 1 is the program with network data newspaper receiving function, can write with visual programming tools, writes as VC++ or Delphi, has visualization interface and also can report and the storehouse that reads and writes data by receiving network data; Database can use Microsoft SQL Server database.The keeper can obtain by net control station and script, registration table are write list item path and API sequence carry out the abnormal behaviour information that analyzing and processing obtains.

6, according to the method described above, comprise script is carried out keyword word frequency statistics analysis 2, registration table is write the list item path carries out oneself identification 3 and API is carried out sequence carry out nonego identification 4, listed the testing result at 75 kinds of Email viruses, Email worm-type virus and malicious code below, the result shows that the present invention has good effect to internet worm and malicious code.

Sequence number	Title	Kind	Whether report virus to be
				????1	????Bloodhound.vbs.worm	??Email，worm
????2	The Bloodhound.vbs.worm mutation	??Email，worm					Be
			????3	????vbs.mesut	??email	Be
????4	????Jesus	??Email，worm					Be
			????5	????Vbs.jadra	??email	Be
????6	????Vbs.infi	??email					Be
			????7	????Vbs.hatred.b	??email	Be
????8	????Vbs.godog	??email					Be
			????9	????Vbs.hard	??Email，worm	Be
????10	????Vbs.gascript	??Email，Trojan					Be
			????11	????I-Worm.CIAN	??email	Be
????12	????Vbs.vbswg.qen	??Email，worm					Be
			????13	????I-Worm.doublet	??Email，worm	Be
????14	????White?house	??Email，worm					Be
			????15	????I-Worm.chu	??email	Be
????16	????Loveletter	??Email，worm					Be
			????17	????freelink	??Email，worm	Be
????18	????Mbop.d	??Email，worm					Be
			????19	????Kounikewa	??Email，worm	Be
????20	????json888	Malicious code					Be
			????21	????gator[1]	Malicious code	Be
????22	????overkill2	Malicious code					Be
			????23	????redlof	Malicious code	Be

????24	????script.unrealer	Malicious code	Be
				????25	????vbs.both	Malicious code	Be
????26	????VBS.kremp	Malicious code	Be
				????27	????script.exploit	Malicious code	Not
????28	????script.happytime	Malicious code	Be
				????29	????vbs.godog	Malicious code	Be
????30	????I-worm.doublet	Malicious code	Be
				????31	????I-worm.chu	Malicious code	Be
????32	????vbs.baby	Malicious code	Be
				????33	????vbs.gascript	Malicious code	Be
????34	????vbs.jesus	Malicious code	Be
				????35	????vbs.mbop.d	Malicious code	Be
????36	????vbs.fasan	Malicious code	Be
				????37	????vbs.hard.vbs	Malicious code	Be
????38	????vbs.infi	Malicious code	Be
				????39	????vbs.jadra	Malicious code	Be
????40	????LOVE-LETTER-FOR-YOU	Malicious code	Be
				????41	????vbs.mesut	Malicious code	Be
????42	????JS.Exception.Exploit1	Malicious code	Be
				????43	????JS.Exception.Exploit2	Malicious code	Be
????44	Self-editing Writefile	Malicious code	Be
				????45	The Writefile mutation	Malicious code	Be
????46	????IRC.salim	Malicious code	Be
				????47	????Vbs.vbswg.qen	Malicious code	Be
????48	????Bloodhound.vbs.3	Malicious code	Be
				????49	Bloodhound.vbs.3 mutation 1	Malicious code	Be
????50	Bloodhound.vbs.3 mutation 2	Malicious code	Be
				????51	Bloodhound.vbs.3 mutation 3	Malicious code	Be

????52	Bloodhound.vbs.3 mutation 4	Malicious code	Be
				????53	Bloodhound.vbs.3 mutation 5	Malicious code	Be
????54	Bloodhound.vbs.3 mutation 6	Malicious code	Be
				????55	Bloodhound.vbs.3 mutation 7	Malicious code	Be
????56	Bloodhound.vbs.3 mutation 8	Malicious code	Be
				????57	Bloodhound.vbs.3 mutation 9	Malicious code	Be
????58	????Vbs.bound	Malicious code	Be
				????59	????Vbs.charl	Malicious code	Be
????60	????VBS.Phram.D(vbs.cheese)	Malicious code	Be
				????61	????Vbs.entice	Malicious code	Be
????62	????Vbs.ave.a	Malicious code	Be
				????63	????Vbs.exposed	Malicious code	Be
????64	????Vbs.annod(vbs.jadra)	Malicious code	Be
				????65	????Vbs.nomekop	Malicious code	Be
????66	????Html.reality(vbs.reality)	Malicious code	Be
				????67	????Bloodhound.vbs.3	Malicious code	Be
????68	Bloodhound.vbs.3 mutation 1	Malicious code	Be
				????69	Bloodhound.vbs.3 mutation 2	Malicious code	Be
????70	Bloodhound.vbs.3 mutation 3	Malicious code	Be
				????71	Bloodhound.vbs.3 mutation 4	Malicious code	Be
????72	Bloodhound.vbs.3 mutation 5	Malicious code	Be
				????73	Bloodhound.vbs.3 mutation 6	Malicious code	Be
????74	Bloodhound.vbs.3 mutation 7	Malicious code	Be
				????75	Bloodhound.vbs.3 mutation 8	Malicious code	Be

Claims (1)

1, a kind of internet worm of layered cooperative and malicious code recognition methods comprise:

From script file, isolate keyword, obtain by the method for injecting dynamic link library (being called for short DLL) that application programming interface (being called for short API) is carried out sequence and registration table writes the list item path, registration table is write the list item path and the API sequence is kept in hard disk or the internal memory; It is characterized in that:

To the keyword word frequency statistics analysis of script and make unusual judgement;

Registration table is write the list item path to carry out oneself's identification and makes unusual judgement;

The API sequence is carried out nonego identification and made unusual judgement;

Abnormal behaviour information is sent to net control station;

The script file that described script file is meant the script file write with the Javascript language, write with the VBScript language and embedded Javascript or the script file of VBScript code;

Described injection DLL acquisition API execution sequence and registration table write the list item path and are meant, by DLL is injected in the target program as remote thread, the API execution sequence of the method intercepting target program of Import Address Table (being called for short IAT) is replaced in employing then, and writes the list item path from the parameter acquisition registration table of registration table api function;

Described to script the keyword statistical analysis and make unusual judgement and be meant and from script file, isolate 29 keyword copyfile, Createobject, Delete, FolderDelete, RegWrite, Virus, .Write, GetSpecialFolder, keys, opentextfile, readall, .save, startup, execute, .add, buildpath, copyfolder, createfolder, createtextfile, deletefile, fileexists, folderexists, getfile, getfolder, getparentfolder, format, .run, do copy, document.write, and carry out following steps:

(1) 29 keywords are divided into three groups, first group for creating object keyword: Createobject; Second group is no risky operation keyword itself: Virus .Write, GetSpecialFolder, keys, opentextfile, readall, startup, execute .add, buildpath, fileexists, folderexists, getfile, getfolder, getparentfolder .run, document.write; The 3rd group is to have the keyword that possibility is destroyed operation: copyfile, Delete, FolderDelete, RegWrite .save, copyfolder, createfolder, createtextfile, deletefile, format, do copy;

(2) the desired value f of the word frequency that these 29 keywords occur in the normal script of statistics _i, the desired value f of the word frequency that these 29 keywords occur in the unusual script is added up in 1≤i≤29 _i', 1≤i≤29, it is poor to calculate the normalization word frequency of 29 keywords in normal and unusual script $e_i=(f_i-f_i^{\′})/\underset{i=1}{\overset{29}{\mathrm{\Σ}}}(f_i-f_i^{\′}),$ 1≤i≤29：

(3) the statistics word frequency m that keyword occurs in current script to be detected _i, the risk factor Risk of script to be detected is calculated in 1≤i≤29,

$\mathrm{Risk}=G\underset{i=1}{\overset{29}{\mathrm{\Σ}}}P\left(i\right)F\left(i\right)$

Wherein P (i), F (i) and G are respectively:

(4) risk factor threshold value TH is defined as:

$\mathrm{TH}=\underset{i=0}{\overset{29}{\mathrm{\Σ}}}P\left(i\right)/29$

When risk factor Risk surpasses threshold value TH, send early warning information to net control station;

Describedly registration table is write the list item path carry out oneself identification and make unusual judgement and take following steps:

(1) the normal registration table of target program writes the list item path under the collection normal condition, and deposits in the database, and each normal registration table writes the list item path and is called " oneself ", and its set is called " oneself's collection ";

(2) read current registration table and write the list item path, compare, if not in " oneself's collection ", then send abnormal behaviour information to net control station with original in the database " oneself " operation;

Described the API sequence carried out nonego identification and made unusual judgement and take following steps:

(1) API selection operation:

(a) the API sequence of target program under the intercepting normal condition, and be W with the sliding step ₀Mode it is cut into length is L ₀Trail S ₀

(b) the API sequence of target program under the intercepting operation with virus state, and be W with the sliding step ₀Mode it is cut into length is L ₀Trail R ₀

(c) compare trail S ₀And R ₀In different sequences, extract the api function that constitutes these sequences, with these api functions as api function collection to be monitored;

(2) according to selected api function, the API sequence of target program under the intercepting normal condition, and be that W is cut into the string that length is L with it with the sliding step, generate oneself's collection S;

(3) the current API that obtains target program carries out sequence, and is that W is cut into the string that length is L with it with the sliding step, reads N API sequence at every turn and carries out following testing process:

(a) produce initial detector collection D ₀: produce pre-detector at random according to selected api function, filter the oneself, and then obtain the initial detector collection; The matching strategy here is the part matching strategy, and promptly two sequences match and if only if these two character strings are in r position consistency continuously;

(b) more current AP carries out arbitrary detector that sequence and detector are concentrated: if find to mate then this sequence of mark and total matching number added 1, when the total matching number of API sequence to be detected that obtains in real time reaches threshold value G _nThe time, send out abnormal behaviour information to net control station;

(c) if evolutionary generation t surpasses threshold value G _eOr all the API sequences are labeled, continue to read next group API sequence and detect; Otherwise, for unmatched API sequence, then according to the variation of affinity degree, gene library evolution, three subset D producing at random _A, D _G, D _RWith memory collection D _MCommon composition detector collection D of future generation _i=D _A+ D _G+ D _R+ D _M, and D _A, D _G, D _RSubclass satisfies $\frac{D_A}{1}\≈\frac{D_G}{2}\≈\frac{D_M}{1};$

Produce the detector subset D by the variation of affinity degree _A, affinity degree variation is meant that the matching degree of arbitrary detector of concentrating when API sequence and detector is above affinity degree threshold value G _fThe time, produce N by variation _c(N _c〉=1) individual filial generation individuality;

Produce the detector subset D by gene library evolution _G, gene library evolution is meant the selection probability that improves the API that forms valid detector, i.e. P _Api=P _Api+ Δ P; And when reality generates detector, select probability to generate pre-detector by the roulette wheel method according to API, filter the oneself at last and generate the detector subset D _G

By producing the detector subset D at random _R

The existing detector that can mate unusual sequence is formed memory collection D _M

Described net control station is meant to be used for receiving script, registration table is write the network program that list item path and API sequence are carried out the abnormal information that analyzing and processing obtained.