CN105245594A - A Naming Method for ICN Network Information - Google Patents

Abstract

The invention provides an ICN network information naming method. The ICN network information naming method comprises the following steps: 1), searching a domain name of a master server where to-be-named information is located and a main body providing information to generate a provider segment; 2), according to the category of the master server where the to-be-named information is located, generating a classification segment; 3), according to the situation of the to-be-named information, selecting a direct naming method or attribute value pair naming method to name the information, thereby generating a name segment; 4), utilizing the to-be-named information as an input according to a provider/category/name to generate a verification code segment by utilizing a digital signature algorithm; connecting the provider segment, the classification segment, the name segment and the verification code segment together by utilizing field separators to form the name of the information. The invention fuses advantages of layering naming and planar naming and can improve the ICN network information routing efficiency.

Description

A Naming Method for ICN Network Information

technical field

The invention relates to the technical field of information center network, in particular to an ICN network information naming method.

Background technique

With the continuous development and changes of applications on the Internet, the existing Internet based on TCP/IP has gradually exposed many inadaptations, such as insecurity, poor mobility, poor reliability, and poor flexibility. When users access the network, they are more concerned about "what is needed" rather than "where is the needed thing". However, the existing Internet is a host-based communication model, and attention must be paid to the "where" issue. This host-based communication model is no longer suitable for the needs of current network development.

Therefore, how to "pull" back the information required by users from the network has undoubtedly become a key core issue that the ICN network needs to solve. Research on this issue must change the concept, transform from the traditional host-centered communication model to an information-centered communication model, and establish an ICN network intelligent routing mechanism that supports information "pull" access.

The so-called information center network means that everything in the network can be regarded as information. It can be said that it is a network of information interconnection rather than host interconnection. Its core object is information, and each information is identified by the name of the information. The ICN network adopts an information-oriented communication model to replace the traditional host-oriented communication model, replaces traditional host-to-host "push" information access with host-to-network "pull" information access, and replaces traditional store-and-forward routing with cache-and-forward routing. It can fundamentally solve the problems existing in the current network.

For the ICN network, all the information flowing in it is named information, and the entire network and its terminals are driven by various information. There is a huge amount of information in the network, and the network system needs to distinguish each information. Therefore, how to name the information so that the network can run efficiently is the key basic issue of the ICN network.

The existing ICN network naming schemes can be classified into two different types: hierarchical and planar. The hierarchical method means that information is named according to a hierarchical relationship, and the information is defined through the hierarchical relationship to which it belongs, which facilitates the aggregation of information names. The flat method means that the naming of information is not hierarchical, and the unique identification code is used to define the information, which is easy to expand and add additional information. A good naming scheme should satisfy the four characteristics of persistence, aggregation, global uniqueness and self-validation. Generally speaking, persistence requires information names to be flat, and aggregability requires naming to be hierarchical. Persistence and aggregability are a pair of contradictions.

Contents of the invention

In order to solve the above technical problems, the present invention provides an ICN network information naming method.

The technical scheme adopted in the present invention is: a kind of ICN network information naming method, comprises the following steps:

Step 1. Search for the domain name of the main server where the information needs to be named and the subject of the information, and connect them with subordinate delimiters or escape characters to generate the provider segment;

Step 2, according to the need to name the classification of the master server where the information is located, the domains in the classification are connected with sub-separators to generate classification segments;

Step 3. Choose the direct nomenclature or attribute value naming method according to the situation where the information needs to be named, and generate a name segment. If the information to be named has a clear meaning or does not need a name field to indicate the meaning of the information, use the direct nomenclature. Otherwise, use the attribute Pair-value nomenclature, domains in the name segment are joined by subordinate delimiters;

Step 3: Input the information that needs to be named according to the provider/category/name, and use the digital signature algorithm to generate a verification code segment; connect the provider, classification, name and verification code 4 segments through segment separators to form information naming ;

Step 4: According to the completeness of the name of the requested information, the method of precise search, semi-precise search or fuzzy search is used to search the field.

The digital signature algorithm adopts a digest algorithm or a key cryptosystem RSA algorithm.

Beneficial effects of the present invention:

The present invention integrates the advantages of hierarchical naming and planar naming, and divides the name of an information into four parts, which are provider, category, name, and verification code. The provider indicates who issued the information, and the category indicates the information. Which type of information does the information belong to (which can include subcategories), the name indicates the name of the information, and the verification code is similar to a digital signature to ensure the integrity, non-impersonation, non-tampering and non-repudiation of the information. By adopting the naming scheme, the efficiency of ICN network information routing can be improved.

Description of drawings

Fig. 1 is a format diagram of information naming in the present invention;

Fig. 2 is the format diagram of the provider in the present invention;

Fig. 3 is the format diagram of classification section among the present invention;

Fig. 4 is the hierarchical classification structure example of information among the present invention;

Fig. 5 is the format chart of direct nomenclature among the present invention;

Fig. 6 is a format diagram of attribute-value pair nomenclature in the present invention.

detailed description

As shown in the figure, a method for naming ICN network information includes the following steps:

Step 1. Search for the domain name of the main server where the information needs to be named and the subject of the information, and connect them with subordinate delimiters or escape characters to generate the provider segment;

Step 2, according to the need to name the classification of the master server where the information is located, the domains in the classification are connected with sub-separators to generate classification segments;

Step 3. Choose the direct nomenclature or attribute value naming method according to the situation where the information needs to be named, and generate a name segment. If the information to be named has a clear meaning or does not need a name field to indicate the meaning of the information, use the direct nomenclature. Otherwise, use the attribute Pair-value nomenclature, domains in the name segment are joined by subordinate delimiters;

Step 3: Input the information that needs to be named according to the provider/category/name, and use the digital signature algorithm to generate a verification code segment; connect the provider, classification, name and verification code 4 segments through segment separators to form information naming ;

The digital signature algorithm adopts a digest algorithm or a key cryptosystem RSA algorithm.

The present invention is further described below in conjunction with specific examples.

According to the requirements of network information naming and the characteristics of information access, we found that the naming of information includes four components, and each part is called a segment. That is to say, the name of each piece of information in the ICN network consists of four segments, as shown in FIG. 1 .

1. Provider

When we obtain a specific kind of information from the Internet, we are more concerned about who provided the information, such as Sohu, Sina, Netease, etc., because the source of information can represent the credibility of the information to a certain extent. Therefore, the provider needs to include who provided the information.

In addition, in order to make the naming of information conform to the current Internet usage habits, it is also necessary to include the provided top-level domain name information in the provider, which can also provide better aggregation for routing lookups.

Therefore, the provider is a fixed-length segment (such as 8 bytes), which is subdivided into two domains: domain name and provider, which are used to represent the provider domain name and provider name of the information, as shown in Figure 2 .

(1) Domain name: The current domain name is divided into two types: "generic top-level domain" (Generictop-level domain, gTLD) and "country code top-level domains, ccTLD". gTLDs are top-level domains used by some specific organizations, such as ".com", ".net", ".org", ".travel", etc.; ccTLDs are top-level domains assigned to countries or regions, such as ".cn", ".cc", ".us", ".tk", etc. At present, more than 200 countries have allocated top-level domain names according to the ISO3166 country code. A domain name is a fixed-length domain (such as 4 bytes), which consists of one of the above top-level domain names, such as ".org", ".com", ".cn" and so on.

(2) Provider: Indicates the subject that provides the information, generally the main part of the domain name used in the current network, such as sohu, sina, google, baidu, etc. Provider is also a fixed-length field (eg 4 bytes).

It is stipulated that the domains are connected by a subordinate separator (here, "."), so that the provider segment can be expressed as ""domain name.provider", such as "com.baidu", "com.sina" and so on.

In some cases, the sub-delimiter may also be used in the name of the information (such as the "." used here). At this time, the escape character "\" commonly used in the computer field is used to distinguish, that is, add the escape character before the delimiter. The free character "\". For example, "com.\.baidu", where the first "." indicates a sub-separator, and the second "." is not a sub-separator. Among the sub-delimiters, parallel delimiters, and segment delimiters described below, the escape character "\" is used for processing when the corresponding delimiter is actually required as part of the name.

2. Classification

A good naming design should improve the efficiency of name lookups. And the hierarchical naming scheme has good query efficiency. Combining that the classification of physical objects in reality is hierarchical, the present invention adopts a hierarchical scheme in information classification to improve naming query efficiency.

Since the level of information classification is uncertain, the classification is a segment with an indeterminate length consisting of multiple domains. The number of domains is not fixed, but the length of each domain is fixed. Its format is shown in Figure 3.

Each subcategory in Figure 3 has a fixed length (for example, 4 bytes), and is used to represent a subcategory of information. The number of sub-categories in the classification section is uncertain, that is, some information classification levels are many, and some information classification levels are few. Therefore, the length of the entire classification segment is not fixed.

The specific information classification method is determined by different information provider subjects, for example, the information classification method of sohu is determined by sohu, and the information classification method of sina is determined by sina. However, the information classification structure should follow the pattern shown in Figure 4, that is, the classification of information follows the hierarchical structure.

It is stipulated that the domains in the category are connected by a subcategory separator ("." is used here). In this way, a category segment of information can be expressed as ""Subcategory 1. Subcategory 2. Subcategory 3.... Subcategory n", such as "entertainment.video.teleplay.history", "entertainment.video.teleplay.military", "entertainment.viedo.teleplay.history", etc.

3. name

The name is the main body of information naming, which is used to indicate the main meaning and content of the information, etc., and provide convenience for users who need the information to retrieve the information. Because the information in the network may exist in the network for a long time, this requires the naming of the information to be persistent, so the name segment adopts a flat naming method, that is, it does not have a hierarchy.

In reality, some information can use a name to clearly express the meaning of the information, such as "the final ranking of the 2015 World Women's Volleyball Grand Prix, "2014 top ten people who moved China", etc. But some information needs more qualifiers to satisfy The determination of the information, such as "26, red, variable speed, mountain bike", "14 inches, silver gray, I5-1.8GHCPU, 4G memory, 120G hard disk, ... laptop". Therefore, the name is There are two different naming methods for a segment with an unfixed length, which can be determined according to the actual situation of the required naming information.

(1) Direct nomenclature

The direct nomenclature is mainly for situations where the meaning of the information is relatively clear or the name field is not required to indicate the meaning of the information. Its format is shown in Figure 5.

Naming type: It is a fixed-length field (for example, 1 byte), indicating whether the name segment is named using the "direct naming method" or the "attribute-value pair naming method". Among them, 1 indicates that the direct naming method is adopted, and 2 indicates that the attribute value pair naming method is adopted.

It is stipulated that the domains in the name segment are connected by a subordinate separator (here, "."), so that the name segment can be expressed as "name type. name", such as "1.2014 Top Ten People in China", "1.afsdfasl\. jpg" etc. Here afsdfasl.jpg means a jpg picture, and afsdfasl only means the name of the picture file, not the specific meaning of the picture. The "\" here is an escape character, which is used to limit that "." is not a subordinate delimiter.

(2) Attribute value pair nomenclature

The nomenclature of attribute-value pairs is mainly aimed at situations where more attribute descriptions are required to clearly express the meaning of information, and its format is shown in Figure 6.

Naming type: It is a fixed-length field (for example, 1 byte), indicating whether the name segment is named using the "direct naming method" or the "attribute-value pair naming method". Among them, 1 indicates that the direct naming method is adopted, and 2 indicates that the attribute value pair naming method is adopted.

Attribute-value pair: each attribute-value pair is a fixed-length field (such as 8 bytes), which is used to represent an attribute of information and its value, and describe the information from one side. The format is "attribute name.attribute value ", such as "memory.4G". Attribute-value pairs can provide information support for information name retrieval. Since there is no subordinate relationship between attribute value pairs, but a parallel relationship, multiple attribute value pairs are connected with a parallel delimiter (":" is used here), so that the format of the attribute value pair of information is "attribute 1 .attribute value 1:attribute 2.attribute value 2:...:attribute n.attribute value n"", such as "size.14 inches: color.silver gray:CPU.I5-1\.8GH:memory.4G:hard disk size. 120G"".

Core name: The core name segment is a field with an indeterminate length, which is used to represent the subject modified by the attribute-value pair.

It is stipulated that the named type, attribute value pair and core name are connected by a subordinate separator (here, "."), so that the name segment can be expressed as "named type. attribute value pair. core name", and further expressed as " Naming type.attribute 1.attribute value 1:attribute 2.attribute value 2:...:attribute n.attribute value n.core name", such as "2.size.14 inches:color.silver gray:CPU.I5-1\. 8GH: Memory. 4G: Hard disk size. 120G. Notebook".

4. Verification code

The verification code is mainly used to provide protection for the integrity, non-impersonation, non-repudiation of the provider, and non-tampering of information in the three segments of provider, category, and name.

The verification code can use public digital signature algorithms such as digest algorithm and public key cryptosystem RSA, and the information consisting of three segments of information provider, classification, and name is used as input to generate a verification code sequence. Users can use different signature algorithms when publishing information. Regardless of the length of the information composed of the three segments of provider, category, and name, the length of the verification code generated by the same algorithm is the same, and the length of the verification code generated by different algorithms may be different. Therefore, the captcha is a segment of variable length.

The segments of the specified information are connected by a segment separator ("/" is used here), so that a complete ICN network information name is:

"domain.provider/subcategory1.subcategory2.subcategory3....subcategoryn/namingtype.name/captcha"

Such as "com.sohu/News.Domestic.People/1.2014 Top Ten People in China/8D34D566EF23EA09"

Such as "com.Baidu/picture.cartoon picture/1.99bOOPIC77.jpg/6D24C521BC21D402"

or "domain.provider/subcategory1.subcategory2.subcategory3....subcategoryn/nametype.property1.propertyvalue1:property2.propertyvalue2:...:propertyn.propertyvalue n. Core Name/Verification Code"

Such as "com.jd/Computer, Office.Computer machine.Notebook/2.Size.14 inches:Color.Silver gray:CPU.I5-1\.8GH:Memory.4G:Hard disk size.120G.MacBook/3FECC54B709AC476".

The name generated according to this method satisfies the four characteristics of persistence, aggregation, global uniqueness and self-verification.

Persistence requires that the information name has a flat feature, so that the change of the name will not be affected by the change of the level, so that the name can persist. In the present invention, the name segment is named in a flat method, which can maintain the persistence of the name. Although the classification section adopts a hierarchical naming method, for a provider, its classification is relatively fixed, such as JD.com's classification of "computer, office.Computer machine.Notebook". Therefore, the persistence of the ranking can also be guaranteed.

Aggregability requires naming to be hierarchical in order to speed up the efficiency of information retrieval. In the present invention, a hierarchical naming method is adopted in the two sections of provider and category, "domain name.provider/subcategory 1.subcategory 2.subcategory 3....subcategory n/...". It can achieve a high degree of clustering of names and provide the efficiency of information retrieval by name.

Global uniqueness requires information to be unique across the entire network, that is, no two pieces of information in the network will have the same name. The global uniqueness here actually refers to the uniqueness of the first three paragraphs of information, namely "domain name.provider/subcategory1.subcategory2.subcategory3....subcategoryn/namingtype.name/" or "domainname.provider by/subcategory1.subcategory2.subcategory3....subcategoryn/namingtype.property1.propertyvalue1:property2.propertyvalue2:...:propertyn.propertyvaluen.corename/ "only. Because as long as the first three names are the same, the generated verification codes are also the same. Intuitively, this feature is difficult to implement. However, the present invention adopts the limitation of the provider segment, because the provider segment in the network is unique (for example, two providers cannot apply for the domain name of sohu), that is, there are no two information providers in the network using the same provider segment. The global uniqueness of information is thus reduced to global uniqueness within the same provider. This requires only that the two segments "category/name" be globally unique within a provider. This is a simple name check problem.

Self-verification is to ensure the integrity of information naming, non-impersonation, non-repudiation of providers, and non-tampering of information. In the present invention, public digital signature algorithms such as digest algorithm and public key cryptosystem RSA are adopted, and the information composed of the information provider, classification, and name is used as input to generate a verification code sequence as a named verification code segment, It can satisfy the self-verification of information naming.

2. Information requests

In the ICN network, when users need to obtain information from the network, they need to send an interest request packet. The first field of the interest packet is the name of the requested information, which is used to retrieve the required information from the network.

Under the naming system of the present invention, when the user needs to request the required information from the network, there are three ways to fill in the content name field of the interest packet.

(1) Accurate search

Precise search means that the user clearly knows the details of the requested information, that is, the name of the requested information. In this way, the complete information naming content can be filled in according to the information naming format to form a content name field, such as "com.sohu/news.domestic.people/1.2014 top ten people who moved China".

The content name field in the interest packet does not require a verification code.

(2) Semi-precise search

Semi-precise search means that the user knows the provider of the requested information, and can further know some more detailed information, such as classification, name and so on. In this way, part of the information naming content can be filled in according to the information naming format to form a content name field, such as "com.sohu/news/2014 top ten people who moved China", "com.jd//laptop".

In semi-precise search, if the requester only knows part of the information of a certain segment, he can only fill in the known part; if all the information of a certain segment is unclear, he can not fill in the information of this segment, but indicate the segment separator of this segment cannot be missing.

(3) Fuzzy search

A fuzzy lookup is a situation where the user knows only part of the requested information, but not the provider of the requested information. In this way, part of the information naming content can be filled in according to the information naming format to form a content name field, such as "com/news/2014 top ten people who moved China", such as "com//Lenovo Notebook".

In fuzzy search, if the requester knows the top-level domain name of the provider segment, it needs to fill in the top-level domain name; if the requester only knows part of the information of a segment, it can only fill in the known part; if all the information of a segment is unclear, It is not necessary to fill in the information of this segment, but the segment delimiter indicating this segment cannot be missing. The difference between fuzzy lookup and semi-precise lookup is whether you know the subject of the provider domain name, such as "sohu", "sina", etc.

Claims (2)

1. A method for naming ICN network information, characterized in that: comprising the following steps: Step 1. Search for the domain name of the main server where the information needs to be named and the subject of the information, and connect them with subordinate delimiters or escape characters to generate the provider segment; Step 2, according to the need to name the classification of the master server where the information is located, the domains in the classification are connected with sub-separators to generate classification segments; Step 3. Choose the direct nomenclature or attribute value naming method according to the situation where the information needs to be named, and generate a name segment. If the information to be named has a clear meaning or does not need a name field to indicate the meaning of the information, use the direct nomenclature. Otherwise, use the attribute Pair-value nomenclature, domains in the name segment are joined by subordinate delimiters; Step 3: Input the information that needs to be named according to the provider/category/name, and use the digital signature algorithm to generate a verification code segment; connect the provider, classification, name and verification code 4 segments through segment separators to form information naming . 2. A kind of ICN network information method as claimed in claim 1, it is characterized in that: described digital signature algorithm adopts digest algorithm or key encryption system RSA algorithm.