JP3974158B2 - Event place search method, communication system, entity device, computer program thereof, and recording medium thereof - Google Patents

Description

The present invention relates to a communication network.
More particularly, the present invention relates to an entity advertisement method in a semantic information network based on a communication model in which events are exchanged peer-to-peer between entities based on matching of semantic information.

  First, in order to explain the semantic information network, the transition of communication forms and business models on the Internet will be described. FIG. 33 is a schematic diagram showing the transition of the business model on the Internet.

  The Internet, based on the Experimental Network (ARPANET, Advanced Research Projects Agency Network) launched by the US Department of Defense in 1969, is now indispensable for our lives. The Internet, which interconnects computers all over the world, has undergone several technological innovations, and has provided us with a network environment where anyone can easily access information on the network anytime, anywhere.

  In the Internet in the first generation, information distribution based on a “broker type distribution model” is common. In this business model, an information distributor (for example, a personal computer communication company) positioned as a broker centrally manages information of an information provider and distributes information to users as necessary. That is, the broker played a role of connecting the user and the information provider. For this reason, brokers can store large amounts of information (storage such as hard disks), high-performance computing devices (such as high-performance workstations) for processing information, and ensuring sufficient communication bandwidth with users. Therefore, a huge capital investment was necessary.

In contrast, the WWW (World Wide Web) announced in 1992 is an innovative technology that dramatically improves the convenience of the Internet and expands the base of users. This opened the door to the second generation of the Internet. While users can use all information (contents) around the world via a browser, information providers can now send information without going through a broker (information distributor).
In the WWW, each information provider manages information in a super-distributed manner and hyperlinks the information, so that a super-distributed database of information can be constructed only by the information provider. This has made it possible to “construct an information distribution network” that does not require an information distributor. This is the true form of the Internet, and the reason why WWW is said to be the second generation of innovation. In this specification, a new business model made possible by a distribution technology such as WWW is referred to as a “brokerless distribution model”.

While information distribution based on a brokerless distribution model, that is, information distribution led by information providers, is possible, information that users want can be easily identified from information flooding around the world and information provision It has become an important issue to be able to specify a user who is suitable for distributing information to a person. Unfortunately, the WWW did not provide a solution for this issue. Therefore, new brokers called portal sites, search services, recommendation services, directory services, and traders have appeared. This business model is called a “broker search model”.
In the broker-type search model, information (source) search or information distribution destination search is performed through the broker, but information distribution with the discovered partner is end-to-end (end-to-end). -end), in other words peer-to-peer, directly, ie based on a brokerless delivery model.

  Therefore, Napster, which was announced in January 1999 and became a hot topic, can also be positioned as a broker-type search model. As an aside, the inventor has some doubt that the majority of the world has positioned Napster as P2P (peer-to-peer, brokerless distribution model). The brokerless distribution model proposed for the first time by the WWW is that an information distribution network can be constructed by itself without the presence of a broker. However, Napster is not just a network, but a mere point. In other words, when Napster is defined as P2P, all forms of communication, including FTP (File Transfer Protocol) and thread telephone, are defined as P2P, but this completely reflects the essence of P2P. It is not done. This is because the origin of communication starts with peer-peer communication. This is a classic way of thinking called a P2P communication system (P2P interaction system), which is different from the topical P2P now. Looking at other current reviews, there are many misunderstood commentaries. Refer to the following documents 1-4 for detailed explanations about these.

− Takanari Hoshiai, Satoshi Kubota: “The New Trend of the Internet: Non-Broker Model Type Search Model and Autonomous Distributed Technology”, Denki Journal, Vol. 121, No. 3, pp.178-184 (2000.12 reception, published in 2001.3) (Reference 1) : Corresponding to Non-Patent Document 1 below)-Takanari Hoshiai: "Invited Talk: New Trend of Internet" Non-Broker Model "and its Realization Technology" SION: Semantic Information Network "", SSE2000-235 & IN2000-191, pp.65-72 (Received 2000.12, published on 2001.3) (Reference 2: Corresponding to Non-Patent Reference 2 below)-Takanari Hoshiai, Hiroshi Shibata: "Autonomous Distributed Collation Environment Architecture for Proposal of Proposal Information and its Performance Evaluation" Journal (DI), Vol.J83-DI, No.9, pp.1001-1012 (Received 1999.10, 2000-09) (Reference 3: Corresponding to Non-patent Reference 3 below)-Takanari Hoshiai, “Jnutella Workshop Invited lecture ”, 2001.5.22, http://www.jnutella.org/jnudev/jws-052201_2.shtml (Reference 4: Serial corresponding to the non-patent literature 4)

  However, in the broker-type search model, it is difficult to meet the various needs of users and information providers in real time because the broker manages the metadata of information collectively, and everything is included in the management policy of the broker. Because it has to be entrusted, all services are stopped when the system operated by the broker goes down, and it is difficult to protect the privacy of users and information providers, so it is more free and controlled by a specific broker. A new business model that has not been implemented has come to be demanded. In other words, the broker (central manager) that has been generally used in the business model so far is excluded, and a suitable partner is directly searched and discovered between entities (for example, between a user and an information provider). A business model that can do this is now required.

  From this problem recognition, a new concept proposed by the inventor in 1998 is the “brokerless search model”. This is directed to the same world as P2P (Peer-to-Peer), which has been attracting attention all over the world, with Gnutella announced in March 2000 as an opportunity. JXTA, announced in April 2001, is aiming for a similar world. P2P is an innovative technology since the WWW and is called the third generation.

  As an aside, the inventor of the present application is convinced that the brokerless policy model will be the fourth generation. The brokerless policy model is a business model for determining a unique policy for each entity by negotiating the policies of each entity between the entities, and there is no administrator involved in the decision. Say.

  It should be noted that the combination of the broker-type search model and the broker-less distribution model is a hybrid P2P, and the broker-less search model is a pure P2P. FIG. 34 is a schematic diagram showing the dimensions of the P2P model.

The brokerless search model (P2P model) is a vision, concept, and philosophy, and the world it aims for is very simple and clear. Specifically, the goal is to build the following world.
In other words, first, the roles that brokers have played in the past can be shared by each entity as volunteers, so that various network services can be constructed without assuming the existence of a specific operator or administrator. It can be operated.
Secondly, any entity participating in the operation as a volunteer is left behind so that it will not affect the entire network service even if the entity withdraws from the operation due to various reasons such as disability or withdrawal. An entity can continue services by self-organizing.

  Next, the semantic information network will be described in comparison with a conventional network.

  The conventional network is conceptually expressed by a network model shown in FIG. In FIG. 26, each terminal registers a terminal address (for example, a network address and a host address) in the network as its own identifier. On the other hand, the transmission terminal sends an event (packet) composed of the destination address (transmission destination terminal address) and data shown in FIG. 27 to the network. That is, the basic concept of a conventional information network is to correctly deliver an event to a destination based on a destination address.

Consider a case where a proposal-type service (a personalized information proposal service for proposing personalized information for each user and a yellow page service such as a search service) is realized using this network. Conventionally, a centralized server system (broker) as shown in FIG. 28 must always be interposed in order to realize a personalized information proposal service. Similarly, in the yellow page service, it is necessary to assume a broker called a search service. Such a centralized business model through a broker is called a broker model.
In this broker model, a broker (broker) plays a role of connecting an information provider and a user. That is, the user and the information provider can meet only through the broker.

  The broker model will be described in detail with reference to FIG. As shown in FIG. 29, in a business model composed of three business roles of a service provider (content information provider) 103 in addition to an end user (consumer 101 in the figure) and an information provider (content provider 102), The service provider 103 manages the content information registered from the information provider (content provider 102) and the end user preference information registered from the end user (consumer 101), and the registered content information and end user preference information. Thus, a service for proposing content suitable for the end user (consumer 101) is provided as a proposal-type service. Such a service provider corresponds to a broker.

Specific services include a search service such as Yahoo, a trading service, and a personalized information proposal service that proposes content that matches the end user's interest without direct intervention of the end user.
As shown in FIG. 30, these services are realized as an integrated service system (service application) in each service provider 103 existing on the network 109 or the content distribution network 110.

  Due to these forms, the broker model has the following problems (1) to (4).

  (1) The information provider (content provider 102) has no means for announcing content information to the end user (consumer 101) without a third party. An information provider (content provider) -led pursuit society that circulates content information itself could not be realized. That is, the information provider (content provider 102) that is the content provider cannot distribute the content information according to its own policy without intervening third parties.

  (2) In order to enjoy the service offered by the service provider 103, the end user (consumer 101) must explicitly recognize the presence, location, and access method of the service provider 103.

  (3) It has not been easy to share content, content information, and end user preference information between different service providers 103. In the past, information was shared by exchanging information according to a protocol decided in advance between the individual service providers 103.

  (4) Due to the centralized management of content information and end user preference information in the service provider 103, it has not been easy to construct a highly scalable system. Until now, classical methods such as preparing a plurality of server systems for managing content information and end user preference information have been used.

  By the way, distributed processing technologies such as WWW (World Wide Web) and CORBA (Common Object Request Broker Architecture) provide an environment where content distributed and distributed on a heterogeneous distributed system can be easily obtained without leaving anyone. Makes it easy to become a content provider. On the other hand, it is not easy to find the content desired by the end user from the enormous amount of content flooding all over the world, and it is desired to quickly establish a technology for that purpose.

  In view of such circumstances and the problems of the broker model described above, the inventor of the present application directly sends content to an end user who is appropriate for his content without assuming the presence of a broker. A non-broker model that can be offered was proposed.

As a technology for realizing a non-broker model, an invention by Gnutella and SUN (US Patent 5,870,605: Document 5: corresponding to Patent Document 1 below), an invention by NEC (Japanese Patent Application No. 11-328654, Japanese Patent Application Laid-Open No. 2000-165449: Document 6) : Corresponding to Patent Document 2 below).
Gnutella is a network that searches and discovers a specific entity from an unspecified number of entities (information) distributed on the Internet based on a P2P (Peer-to-Peer) interaction model. In Gnutella, an entity is also called a servant. In other words, each information provider supervises and manages metadata (metafile), and by setting an arbitrary connection between metafiles (between hosts storing metafiles), only the information provider can Build a super-distributed database of data. In addition, by distributing and coordinating information providers, it is possible to construct a metadata search network that does not require a broker. However, a search request packet from a consumer is broadcast to all information providers based on the connection. In other words, since P2P interaction is repeated for all hosts based on connection information, unnecessary traffic increases in proportion to the increase in the number of hosts and the type of metadata to be handled, resulting in extremely low scalability. There's a problem. In addition to scalability, there are problems in terms of interoperability, information sharing, scalability, security, and privacy.

  Document 5 enables the request and use of information in the absence of a broker in a situation where an information provider (Publisher) and a user (Subscriber) do not have knowledge of each other by a network system that performs content-based routing. The method was shown. However, there is a problem in scalability because there is no routing mechanism that suppresses unnecessary event forwarding. Furthermore, since there is no ontology between territories (such as a mechanism for converting an event dictionary), there are problems in terms of interoperability, information sharing, and extensibility. In other words, Document 5 defines territories that share a common event dictionary, but the territory implementation method and the method for sharing information between different territories are not clarified. Therefore, there is little description of specific feasibility regarding territory.

  Document 6 proposes information routing that can route not only packets with a specific destination address, but also packets that only contain a region of interest designation. Information routing enables searches that do not require a broker. However, there is no concept of event place, gateway, or session that limits the uniqueness of the ontology system or the spread of the event. Therefore, there are problems in terms of scalability, security, privacy, interoperability, information sharing, and extensibility.

  As described above, the conventional technology for realizing the non-broker model has the following four problems.

The first challenge is the lack of security and privacy protection.
In the conventional non-broker model implementation technology, there is no mechanism for limiting the range in which information is distributed. In other words, the destination of the event cannot be limited. Therefore, anyone who is connected to the network can view the information.
Moreover, the terminal point of information becomes an end user's terminal, and this may be referred to from the outside. Therefore, even if a terminal has information that is not desired to be disclosed to the outside, the information may be disclosed when connected to the network.
In addition, since there is no concept of a session, a network can be constructed only with event transmission / reception terminals, but a network provider and an event transmission / reception party cannot be separated.

The second problem is lack of interoperability.
When areas with different semantic information systems (ontology systems) are created, there is no mechanism for connecting these areas, so events cannot be shared between them.
Also, only necessary events cannot be transferred.

The third problem is lack of extensibility.
That is, it is difficult to add / delete dynamic functions.
Also, it is not easy to expand the network scale.

The fourth problem is lack of scalability.
That is, the increase in the number of event senders / receivers cannot be dealt with.
Further, it is impossible to cope with an increase in semantic information type (metadata type).
Also, unnecessary event forwarding cannot be suppressed.

Therefore, the present inventor has proposed a Semantic Information-Oriented Network (SION) as a technology for realizing a non-broker model other than the non-broker model described above. In addition, the documents regarding the semantic information network by the inventors of the present application are the following documents 7 to 8.
− Takanari Hoshiai, Hiroshi Shibata: “Autonomous Distributed Collaborative Environment Architecture for Proposal Information Proposal and Its Performance Evaluation”, IEICE Transactions (DI), Vol.J83-DI, No.9, pp.1001 -1012 (2000-09). (Reference 7: Corresponding to Non-Patent Document 5 below)-Takanari Hoshiai, Hiroshi Shibata, Takamichi Sakai, Keiichi Koyanagi: "Semantic Information Network Architecture: SION Architecture", NTT R & D, Vol.50, No.3, pp.157-164 (2001.3). (Reference 8: Corresponding to Non-Patent Reference 6 below)

  The SION proposed in the above documents 7 to 8 is an autonomous distributed network that distributes packets based on semantic information (metadata), and a specific entity is dynamically selected from entities distributed on the network. Search and discover. The concept of SION is shown in FIG. As is clear from comparison between FIG. 31 and FIG. 26, SION is a metanetwork based on semantic information (what person to send) instead of the conventional destination address (to whom it is sent). .

Here, an outline of the semantic information network will be briefly described with reference to FIG.
The entities 22a, 22b, and 22c (actually terminals and the like) for each event receiver and the entities 21a and 21b (actually terminals and the like) for each event sender are connected via the SION1. Each entity 22 (a, b, c) previously registers semantic information indicating the characteristics of the event (information) desired to be received as a filter. Entity 21 (a, b) provides a “stimulus” to SION 1 by sending an event. Here, it is assumed that the entity 21a gives a stimulus. This event includes at least semantic information. The SION 1 collates the semantic information of the transmitted event with each registered filter.
If the semantic information of the transmitted event matches the registered filter, the filter “fires” and transmits the event to the corresponding entity 22. Here, it is assumed that the filter of the entity 22b has fired. When the filter fires, the corresponding entity 22b starts autonomously and receives the event. Thereby, only a specific entity can be searched and discovered.

  In SION, a network is constructed by autonomously coordinating all entities including SION components. The SION network components include a semantic information switch (SI-SW), a semantic information router (SI-R), a semantic information gateway (SI-GW), an event place, and a session.

The SI-SW collates the semantic information registered as a filter with the semantic information attached to the event, and activates the event receiving entity that has fired as a result. The SI-R selects a path between the SI-SWs, and transfers an event transmitted from the entity terminal to another SI-SW. The event place is a range of the event. The SI-GW transfers events between event places. A session is a connection between an SI-SW and an entity. By combining these as necessary, a secure and scalable P2P network can be constructed with a bottom-up approach.
US Pat. No. 5,870,605 JP 2000-165449 A Takanari Hoshiai, Satoshi Kubota: "New Trend of the Internet-Non-Broker Model Search Model and Autonomous Distributed Technology", The Journal of Electrical Engineering, Vol. 121, No. 3, pp.178-184 Takanari Hoshiai: “Invited Talk: New Trend of the Internet“ Non-Broker Model ”and its Realization Technology“ SION: Semantic Information Network ””, SSE2000-235 & IN2000-191, pp.65-72 (2000.12 reception, 2001.3 published) Takanari Hoshiai, Hiroshi Shibata: “Autonomous Distributed Collaborative Environment Architecture for Proposal Information Proposal and Its Performance Evaluation”, IEICE Transactions (DI), Vol.J83-DI, No.9, pp.1001- 1012 (Received 1999.10, 2000-09) Takanari Hoshiai, “Jnutella Workshop Invited Lecture”, 2001.5.22, http://www.jnutella.org/jnudev/jws-052201_2.shtml Takanari Hoshiai, Hiroshi Shibata: "Autonomous Distributed Collaborative Environment Architecture for Proposal Information Proposal and Its Performance Evaluation", IEICE Transactions (DI), Vol.J83-DI, No.9, pp.1001- 1012 (2000-09). Takanari Hoshiai, Hiroshi Shibata, Takamichi Sakai, Keiichi Koyanagi: “Semantic Information Network Architecture: SION Architecture”, NTT R & D, Vol.50, No.3, pp.157-164 (2001.3).

  Conventionally, the semantic information network (SION) described above has the following problems.

  First, in a conventional semantic information network, an event path setting request for establishing an event path is propagated to all semantic information routers (SI-R) and semantic information switches (SI-SW) in an event place traced by a shared link. Therefore, there has been a problem that network traffic increases. Therefore, there is a need for a technique for reducing network traffic when transferring an event path setting request or transferring an event.

Further, in the conventional semantic information network, there is a problem that an entity has not established a method for disclosing its existence to other entities. Therefore, there is a need for a technique for an entity to disclose its existence to other entities without a broker.
Furthermore, a technique for realizing an open topology in a topology constituted by shared links between entities is required.

The present invention has been made in view of the above circumstances, and an object thereof is to provide a method for improving the performance of a semantic information network.
In particular, it aims to provide an advertisement method for an entity to disclose its existence.

In order to solve the above problems, the present invention provides an event place for advertising an entity in the communication system based on a shared link between entities in a communication system including entities including a semantic information switch. A search method, in which an entity searches for another entity in the vicinity by broadcasting, and the entity establishes a shared link with another entity discovered in the search process, A base event place participation process that participates in a base event place that is an event place in which another entity discovered in the search process participates, and a base event in which the entity participates in the base event place participation process. In the event place, the discovery process of finding an entity that is a member of the event place that meets its needs, and the entity establishes a shared link with the entity found in the base event place, found event place other than pre-Symbol-based event place the found entity in the base event place is that participating discovered as an entity is a member of the event place that matches their needs in the discovery event place By establishing a shared link with the discovered entity, an event play other than the discovered event place where the discovered entity participates in the discovered event place. An event place search process repeated at discovering the gist event place search method characterized in that it comprises a.

Further, the present invention is a communication system configured by an entity including a semantic information switch and advertising the entity based on a shared link between the entities, and an entity constituting the communication system includes: It is an event place where other entities discovered by participating in the discovery by searching for other entities in the vicinity and establishing a shared link with the other entities discovered in the search. Participating in the base event place, the entity discovers an entity that is a member of the event place meeting the needs of the participating base event place, and the entity detects the base event place. By establishing a shared link between the discovery entity in scan, the discovered entities in the base event place to discover the event place other than pre-Symbol-based event place that participates, is the discovery The discovered entity participates in the discovered event place by establishing a shared link with an entity discovered as an entity that is a member of the event place that meets its needs at the event place It is characterized by repeating the discovery of an event place other than the found event place .

Further, the present invention is an entity device comprising a semantic information switch that collates semantic information registered as a filter with semantic information given to an event and activates an event receiving entity that has fired as a result, and constitutes a semantic information network An event place where the other discovered entities participate by searching for other nearby entities by broadcasting and establishing a shared link with the other entities discovered in the search. Participate in the base event place, find an entity that is a member of the event place that meets your needs in the joined base event place, and establish a shared link with the entity found in the base event place This In, the base events found event place other than the discovered entity prior SL-based event place that participates in place, is a member of the event place that matches their needs in the discovery event place entities Repeatedly discovering an event place other than the discovered event place in which the discovered entity participates in the discovered event place by establishing a shared link with the discovered entity as It is characterized by that.

Further, the present invention is an entity device comprising a semantic information switch that collates semantic information registered as a filter with semantic information given to an event and activates an event receiving entity that has fired as a result, and constitutes a semantic information network The event place search method according to claim 1, wherein a discovery is performed in the search process by establishing a shared link between a search process for searching for another entity in the vicinity by broadcasting and another entity discovered in the search process The base event place participation process that participates in the base event place, which is the event place where other entities participated, and the base event place in which the entity participated in the base event place participation process meet their needs A discovery process to discover entity is a member of the event place that, the entity, the base events to establish a shared link between the discovered entities in place, the discovered entities in the base event place but I found the event a place other than the previous SL-based events place you are participating, the shared link between the discovery has been discovered entity as an entity that is a member of the event place to meet its own needs in the event place by establishing, event place repeating said finding entities in the discovery event place to discover events place other than the event place which is the discovery participating It is characterized in that it has a cord process, the.

Further, the present invention is an entity device comprising a semantic information switch that collates semantic information registered as a filter with semantic information given to an event and activates an event receiving entity that has fired as a result, and constitutes a semantic information network In addition, by establishing a shared link between a search process for searching for other nearby entities by broadcasting and another entity discovered in the search process, the other entities discovered in the search process participate. In the base event place participation process that participates in the base event place that is the event place to be participated, and in the base event place that participated in the base event place participation process, the entity that is a member of the event place that meets its needs A discovery process for finding the presentity, the entity, the base events to establish a shared link between the discovered entities in place, the base event the pre discovered entities that have participated SL in place By discovering an event place other than the base event place and establishing a shared link with an entity discovered as an entity that is a member of the event place that meets the needs of the found event place. is executed and the event place search process for the discovered entity in the event places that are repeated to find an event place other than the event place which is the discovery participating, the processing of each step of It is a computer program characterized.

  The recording medium of the present invention is a computer-readable recording medium that records the computer program.

  According to the advertisement method of the present invention, since there is no broker (administrator) that manages the public information of the entity, a self-organization network with high fault tolerance can be constructed. Moreover, it is not realistic to manage a huge number of public information (entity properties) with a broker, but such management is not necessary in the method of the present invention.

Furthermore, according to the advertisement method of the present invention, the event place suitable for the entity can be narrowed down according to the formula, so that a desired event place can be efficiently searched.
In the advertisement method of the present invention, network traffic (the number of event transfers and the number of event path setting request transfers) can be reduced by limiting the spread range of the chain reaction.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
Hereinafter, all computers such as personal computers, workstations, portable information terminals, mobile phones, and wearable computers are collectively referred to as “hosts”. Furthermore, a host on which SION software is installed is referred to as “SION entity” or simply “entity”.

  An entity is a unit of autonomous decentralization in SION, and is a virtualization of a host equipped with SION software as an autonomous operating entity in SION (see FIG. 1). SION software also provides various mechanisms for individual entities to perform autonomous distributed coordination.

Here, the entity will be described in detail. Entities are classified into three main types as shown in FIG.
(1) Entity acting as a service (service entity: hereinafter referred to as “SE”)
(2) Entity acting as a network component (network entity: hereinafter referred to as “NE”)
(3) Entities that act from the viewpoints of both services and network components (service network entity: hereinafter referred to as “SNE”)

  The service entity is a virtualized application program as an entity. With a plug-in mechanism provided by SION-MT (SION-Management Tool), a general application program can be made into an entity as a service entity. Note that SION-MT itself is also a service entity. The service entity can send events to the SION and receive events from the SION via the session. In addition, a plurality of sessions can be established for SION from one service entity. Service entities are basically not involved in network construction / operation.

The network entity is an entity for constructing SION. As described above, SI-SW (semantic information switch), SI-R (semantic information router), SI-GW (semantic information gateway), EP (event place) , Session, statistical information entity, fault handling entity, Alive entity, and the like. In the SION of the present embodiment, such a network component can be dynamically generated in the entity as many as necessary as necessary. For example, a plurality of SI-SWs can be dynamically generated in one entity. It is also possible for SI-SWs in the same entity to belong to different EPs.
These are used when building a backbone P2P network or a hybrid P2P network.
The SNE is an entity that behaves from the viewpoint of both a service and a network component, and is used when constructing a pure type P2P network.
This is a form in which each SNE helps each other to build a volunteer-type network, and the SNE distributes and cooperates to self-organize the network.

  The hybrid P2P network is a network configured by, for example, an operator such as a network provider generating an event place in advance on a host or the like, and a service entity such as a personal terminal connecting to the event place through a session. . The pure P2P network is a network configured by connecting entities as individual transmitting / receiving terminals via shared links, and the smallest unit for sharing an event is an event place. The backbone type P2P network is a connection form between networks, for example, a network in a state where pure type P2P networks in a plurality of regions are connected via a hybrid type P2P network.

FIG. 24 shows an internal configuration of an entity when a plurality of network entities are included in one entity.
In this figure, the entity control unit accepts various requests from the control panel.
The control panel provides a GUI (Graphical User Interface) to owners of entities such as applications and people. For example, an entity name can be given to the entity.
The network entity factory creates an NE (network entity) in the entity. Based on the generation request received by the entity control unit, the network entity factory generates the NE described above.

FIG. 4 shows a network configuration example of SION. In SION, various forms of P2P networks can be constructed by arranging and combining entities according to the purpose. Also, different forms of P2P networks can be seamlessly linked. This includes: “All entities are autonomously distributed and collaborated by a simple and simple mechanism called“ chain reaction based on stimulation and firing ””, “Reducing P2P network form differences to entity placement problems”, etc. Is achieved.
Note that an entity advertises its existence based on a public protocol. This is called advertisement. The advertisement includes a method using an “advertisement base (origin entity, origin event place, etc.)” that acts as an advertisement transmission base, and a method based on broadcast cascade.
In the following, individual network entities are described in detail.

<SI-SW (SI-Switch)>
SI-SW (semantic information switch) provides a switching mechanism for switching events based on semantic information. In SION, this SI-SW is implemented as distributed software.
FIG. 5 shows the configuration of the SI-SW. In this figure, an event receiver 51 receives an event from an event sender. The collation unit 52 collates the semantic information given to the event received by the event reception unit with the filter. The event transmission unit 53 transmits the event ignited by the verification unit to the event receiver. The filter management unit 54 stores a filter and an event transmission destination when fired as a result of collation. The control request receiving unit 55 receives control requests from other switches, routers, and the like. The control unit 56 performs filter control (registration, release), session control (registration, release), shared link control (establishment, release, re-establishment, information response) and the like based on the request received by the control request reception unit. . The shared link setting information management unit 57 stores a shared link setting condition (setting partner SW ID, setting request / requested type).

  As shown in FIG. 6, SI-SW accommodates a service entity in a star shape via a session. Here, the session is a connection between the SI-SW and the service entity, and the service entity can send and receive events via the session. The service entity is an entity that transmits and receives events to and from the SI-SW. For convenience of explanation, the service entity may be referred to as an event transmission entity and an event reception entity. The former is an entity that transmits an event to the SI-SW, and the latter is an entity that receives an event from the SI-SW. A service entity can have both aspects.

The service entity is positioned as a distributed application on SION (an application program that transmits and receives events using SION), and the File Exchanger (file exchange application) and personal TV service shown in FIG. Has been implemented.
FIG. 7 shows the structure of an event (SION packet). The event includes a control information part (CI: Control Information) part, a semantic information (SI: Semantic Information) part, and a data part. Transmission data is stored in the data portion. Various data and programs such as text (document, telephone number, XML (Extensible Markup Language)), binary data, reference (URL (Uniform Resource Locator), Object Reference), proxy, agent, etc. can be stored as transmission data. it can.

  On the other hand, the semantic information section stores semantic information (vocabulary and its value) of transmission data and a vocabulary concept of semantic information (called event type). Here, the semantic information is metadata describing characteristics of transmission data, and belongs to any event type as shown in the semantic information system (ontology system) in FIG. That is, the semantic information is an event type instance, and the event type is a semantic information template. You can define inheritance relationships between event types. Note that some event type names are reserved for SION control (for SION network entities). For reference, an example of a semantic information system is shown in FIG. In this figure, the meaning information “title; yesterday”, “price; $ 30”, “artist name; beatles” belongs to the event type “popular” and also belongs to the event type “music”. Yes. In addition, the meaning information “title; night of Kiyoshi no Ko”, “price; $ 20”, “situation; Christmas” belongs to the event type “BGM” and also belongs to the event type “music”. .

The control information part is a data field used for SION execution control, and only this is not open to the service entity that is a SION user. The control information section includes matched filter identifiers, matched filter verification scores, synchronous statistical information collection flags, TTL values (within event places, between event places), hop counts (within event places, between event places), etc. Various control information is set.
As illustrated in FIG. 10, the event reception entity registers an event acquisition condition in the SI-SW via a session. This is called a filter.
In the filter, an event type of an event to be acquired and a matching condition with semantic information (for example, an object whose vocabulary “Price” is “$ 20 to $ 40” are acquired) are set. Note that the matching condition can also be selected in units of filters for complete matches with all vocabularies defined in the event type, partial matches with some vocabularies, weighted matches, and the like.

A plurality of filters can be registered from one session, but filters registered through the same session have an “or” relationship. That is, one session fires at most once for one event.
If a wild card is specified for the event type of the event to be acquired, all event types are subject to matching. Further, when 1 (logical value is true) is set as the matching condition with the semantic information, it means that the matching condition with the semantic information is satisfied unconditionally.

On the other hand, the event transmission entity transmits an event to the SI-SW via the session. At this time, the event receiving unit 51 of the SI-SW receives the event and passes it to the matching unit 52. The collation unit 52 collates the event semantic information part with the filter. Specifically, first, it is checked whether or not the event type is desired to be received. If the event type is satisfied, the semantic information and the matching condition are checked. As a result of the collation, when the condition is satisfied, the event receiving entity in which the matched filter is registered is activated and the event is notified. In SION, the transmission of an event is called “stimulus”, the fact that the filter matches the event is called “reaction”, and the activation of the entity that registered the matched filter and notification of the event is called “fire”.
In addition to event notification by firing, it also supports look-in type event notification. This is a reception mode in which an event spooled for each reception session is acquired by pulling. This look-in type event notification is effective for event reception across firewalls as shown in FIG. 35 and event reception when SE is not active.

The basic concept of SION is to make all the entities autonomously cooperate with each other by the chain reaction of stimulation and firing.
It is also possible to register semantic information in the filter and set distribution conditions (collation conditions) in the semantic information section of the event. In SION, such a form is called a patronage model. For example, “$ 20 <Price <$ 30” can be set in the event semantic information section. On the other hand, the form described in the text (a form in which a filter is registered in SI-SW and semantic information is registered in an event) is called an inquiry type model.

<EP (Event Place)>
The event place is a domain concept devised for the following purposes (see FIG. 11). The shared link and SI-R will be described later.
(1) Event transfer range: It is possible to limit the spread range of chain reactions based on stimulation and firing.
(2) Event routing target range: A set of SI-SWs connected by a shared link. Link topology and event routing method can be selected for each event place.
(3) Space in which the uniqueness of the semantic information system is guaranteed: An explosive increase in the event types handled is suppressed.

The following are examples of the main usage of event places with such features.
(1) Subnet unit (physical unit)
For example, it is used when building a local network of SION on a personal computer at work.
(2) Service unit (logical unit)
In general, the required semantic information system (vocabulary concept and vocabulary) differs for each service. For example, it is used when an event place is constructed for each service type such as a securities information providing service and a medical information service.
(3) Service operator unit Even when the same service type is used, a different event place is constructed for each service operator.

(4) Security Assurance Unit Only approved entities are allowed to participate in the event place. That is, it is allowed to establish a session with the SI-SW in the event place, or to establish a shared link with the SI-SW in the event place. Basically, information sharing between event places is not allowed.
(5) Unit of network operation / administrator For ease of operation and management, the event place may be further divided into a plurality of event places.
(6) Load distribution unit From the viewpoint of load distribution, an event place may be divided into a plurality of event places.

(7) Physical Unit Associated with Restriction of Session Establishment In SION, a connection between a service entity and SI-SW and a connection between SI-SW (shared link) are realized by a session. The session itself is a logical concept, but when physically implemented using a wireless LAN or the like, the grouping target is limited by the wireless reachable distance.
(8) Grouping of groups By using SI-GW, which will be described later, event places are linked in a hierarchical manner, a clustering configuration having a parent group / child group relationship can be constructed.

<SI-GW (SI-Gateway)>
The event place is for limiting the transfer range of the event, and basically the event cannot be shared with other event places. However, setting an SI-GW (semantic information gateway) enables event sharing between event places (see FIG. 11). Thereby, cooperation (federation) between event places constructed according to each use and purpose becomes possible.
In FIG. 11, for example, when event place A makes a cooperation request to event place B and this is approved by event place B, event place A generates SI-GW _{A, B.} Note that one-way or two-way cooperation can be specified as the cooperation request.

SI-GW _{A, B} establishes an event reception session for event place B, while establishing an event transmission session for its own event place. This makes it possible to transfer an event that has occurred in event place B to event place A.
SI-GW _{A, B} is a fine-grained firewall that makes all event types subject to forwarding or only specific event types subject to forwarding by setting the filter value registered through the event reception session. Control becomes possible.

In addition, when event transfer is performed between event places having different semantic information systems, it is necessary to convert event semantic information.
Therefore, SI-GW _{A, B} creates a semantic information conversion table between event place A and event place B, performs semantic information conversion based on the information, and then transfers the event to event place A. Do. The number of transfers (number of hops) between event places is limited by the TTL value.
Based on the above idea, interconnection with other P2P networks is also possible. In SION, the mechanism for this is called “S / x-GW”. For example, from SION, the Gnutella network is virtualized as one entity via S / G-GW.

<Reference model>
SION uses the concept of a reference model. The lowest layer is SION and is an event transmission layer. Provides a network interface for higher layers. The middleware layer is a community network and corresponds to the intelligence layer of SION. The highest layer is the application layer (service entity).
SION-Management TOOL (SION-MT) is also positioned as one of the service entities. This is a GUI (Graphical User Interface) tool for issuing a SION network interface and does not have service-specific logic. However, a plug-in mechanism capable of incorporating another application program (an application having service logic) into SION-MT is provided. This makes it easy to build a service entity customized for each user.

<Online increase / decrease method>
Here, the network entity increase / decrease method will be described particularly from the viewpoint of increasing / decreasing SI-SW.
<Purpose of SI-SW increase / decrease>
First, the purpose of SI-SW increase / decrease in SION will be described.
(1) From the viewpoint of improving the total processing capacity of the event place, SI-SWs in the event place are added and the event filtering process is load-balanced. Moreover, SI-SW is reduced from the opposite viewpoint.
(2) A flexible and global network is constructed in a bottom-up manner by flexibly establishing a shared link for dynamically generated SI-SW.

<Type of increase / decrease>
SION provides several types of increase / decrease forms, but here, a representative form is shown in FIG.
(1) Combining and separating event places As shown in FIG. 13A, a plurality of event places can be combined. Here, the synthesis of event places refers to expanding the spillover range of an event by making a plurality of event places one in an event place, which is a unit that limits the spillover range of an event.
This is because a shared link is established between event places by making a composite request (shared link establishment request) to any SI-SW or event place belonging to the event place, and the combination of both is realized. The
The request source and the request destination of synthesis (shared link establishment) may be either SI-SW or event place. On the other hand, in the case of division, the established shared link is released and separated into event places.

(2) Participating in and leaving the event place As shown in FIG. 13B, by making a participation request (shared link establishment request) to the SI-SW in the event place or the event place, A shared link is established with the existing SI-SW, and it is possible to participate in the event place.
When the SI-SW leaves the event place, the shared link with the other SI-SW is released, the shared link is reconstructed, and the event place is moved away. At this time, the state of the SI-SW that has moved out transitions to the suspended state.
The SI-SW can be taken into the event place by requesting the SI-SW to establish a shared link from the event place side. This is called absorption. The opposite is called splitting.

(3) Increasing / decreasing SI-SW As shown in FIG. 13 (c), when an SI-SW addition request is made to an SI-SW in an event place or an event place, the designated host The number of SI-SWs specified for (entity) is newly generated, and a shared link is established with the existing SI-SW. On the other hand, when a SI-SW reduction request is made to the event place, the shared link of the designated SI-SW is released, and then the shared link is reestablished, and the SI-SW is deleted.
At this time, the state of the SI-SW transitions to “Non-Existent”.

(4) Federation between event places As shown in FIG. 13 (d), an SI-GW is generated by making a federation (cooperation) request to the SI-SW in the event place or to the event place. Both event places cooperate through the session.
The federation request source and the request destination may be either SI-SW or event place.
FIG. 14 shows the state transition of SI-SW in the above embodiment. Since the state of the SI-SW when the SI-SW is generated is the “suspend” state, the SI-SW does not become “active” unless it belongs to any event place.
Moreover, SI-SW with which a shared link is established cannot be deleted.

<Entity arrangement and business model>
As described above, a hybrid business model and a pure business model can be constructed according to the arrangement of network entities.
The former is a form in which an event place is constructed by arranging network entities in advance on a host provided by a network provider or service provider, and a user (event sender / receiver) participates in the event place via a session.

On the other hand, the latter is a form in which a pure P2P network is constructed by arranging network entities on individual hosts of event senders and receivers and establishing a shared link for each.
By using SION, both forms can be realized with the same network architecture and control method. That is, it can be reduced to the problem of the arrangement method of the network entity. Moreover, it is also possible to cooperate both via SI-GW. The advantage of SI-GW is that when attention is paid to a certain event place A, event place B (SI-GW) linked to the event place can be seen as one service entity of event place A. That is, general service entities and event places can be handled in the same manner. By developing this idea and deploying SION / Gnutella-GW, it can be easily linked with other P2P networks such as Gnutella. From the SION (event place) side, the entire Gnutella network appears as one service entity participating in the SION. Such an SX-GW (X = I, G) can be arranged by an event place operator, but the SX-GW is incorporated into the SION-MT by using a SION-MT plug-in mechanism. Therefore, it is possible for the end user to arrange the service.
As a result, a seamless network can be built up, and an adhesive for a business model (broker model, other brokerless model) having a different SION can be obtained.

<Event routing method>
The shared link (shared link) is a concept for performing bidirectional event sharing between a plurality of different SI-SWs. For example, the SI-SW can establish a shared link between SI-SWs by making a shared link (SL) establishment request to another SI-SW. Note that various event routing methods can be dynamically selected between SI and SW by setting filter values to be described later.
FIG. 15 shows a state in which SI-SW2 and SI-SW3 are each making a shared link establishment request to SI-SW1.

<Establishment of shared link>
Here, in FIG. 15, a mechanism from the establishment request to establishment of the shared link to the SI-SW 1 by the SI-SW 2 will be described.
As shown in FIG. 16, SI-R is installed between SI-SW and SI-SW, and SI-R performs event routing (event transfer) between SI-SW based on semantic information. . The configuration of SI-R will be described with reference to FIG.
The event transfer unit 71 receives an event from a switch and transmits the event to another switch. The control request receiving unit 72 receives control requests such as shared link control (establishment / cancellation) and filter registration control from other switches and routers. The control unit 73 performs the control received by the control request receiving unit. The filter registration information management unit 74 records filter registration information (registration counter for each event type).

Hereinafter, an operation for establishing a shared link will be described.
(1) SI-SW2 issues a shared link establishment request including the identifier of SI-SW2 to SI-SW1.
(2) At this time, as shown in FIG. 16, the SI-SW ₁ dynamically generates SI-R _1,2 by the network entity factory in order to establish SL _1,2 , and also shares shared link setting information. The management unit 57 stores that the SI-SW 2 is a shared link establishment request source. SI-SW1 passes the identifier of SI-SW2 to SI-R _1,2 .

(3) The control unit 73 of SI-R _1,2 establishes a session for event reception with respect to SI-SW2. Furthermore, the session for event transmission is established with respect to SI-SW1 to which self belongs. A combination of transmission and reception sessions established by SI-R can be said to be a shared link.
(4) Thereby, SL _1,2 is established, and for example, an event sent by SE 4 to SI-SW ₂ is also sent to SI-SW ₁ via SI-R _1,2. .
(5) At this time, the SI-R _1,2 can select the event transfer method by setting the filter registered for the SI-SW2. An example of the event transfer method will be described below with reference to (a) and (b).

(A) The event sent to SI-SW2 is unconditionally transferred to SI-SW1. That is, SI-R _1,2 fires for all events.
This is possible by setting a wildcard as the event type that you want to receive for the filter that SI-R _1,2 registers through the event reception session and setting 1 (logical value true) as the matching condition. Become. This is called “unconditional routing”. This is mainly used for broadcast communication.

(B) Only necessary events for SI-SW1 are transferred from SI-SW2 to SI-SW1 (unnecessary event transfer is not performed). That is, SI-R _1,2 fires only for a specific event. For example, SE2 is a filter registered against SI-SW1, by SI-R _{1, 2} is filter registered in the SI-SW2, this method can be realized.
(I) Of the filter values registered by SE2 for SI-SW1, SI-R _1,2 registers only the event type for SI-SW2 (the matching condition is always true). Thus, only the event type is collated in SI-SW2, and as a result, when SI-R _1,2 fires, SI-R _1,2 sends the event to SI-SW ₁ and collates with semantic information. SI-SW1, the event type registered in SI-SW1, and (ii) the event type registered by SE2 for SI-SW1 and the matching condition are registered in SI-SW2 by SI-R _1,2 However, it is roughly divided into forms (event routing by vocabulary) in which complete filtering is performed in SI-SW2. The former is mainly used for attributed multicast communication, and the latter is mainly used for unicast communication and multicast communication. This is used in failure processing notifications, statistical information notifications, alive notifications, reply notifications, advertisement notifications, and the like.
A set of filters selected by the SI-R for event sharing (event path) is route selection information for event forwarding (event routing).
In addition to multi-hop communication (unicast, broadcast, multicast, attributed multicast) using a shared link and an event path, direct communication with a communication destination entity is also possible. Thus, in SION, the method of the chain reaction between entities can be dynamically controlled only by the difference in filter values set by SI-R. That is, the various communication modes described above can be realized. Such a simple and single mechanism is one of the major features of SION.
Further, when generating an EP (event path) or registering a filter from a service entity, an event transfer method (filter value at the time of registering a filter in SI-SW by SI-R) can be selected.

  The lexical event routing method does not cause any unnecessary event forwarding, but the filter registration processing overhead for routing is enormous, so this method is effective when the number of event transmissions is sufficiently larger than the number of filter registrations. .

On the other hand, the routing method based on the vocabulary concept is positioned as a compromise between the two methods described above. In other words, it is effective when the number of filter registrations and the number of event transmissions are about the same, or when the ratio between the number of filter registrations and the number of event transmissions cannot be predicted.
With these mechanisms, event multi-hop and multicast based on semantic information are realized.

(6) The above-described procedures (2) to (5) are similarly performed in the SI-SW 2 that is the shared link establishment request source. That is, SI-SW2 stores SI-SW1 as an establishment request destination, and dynamically generates SI-R (SI- _R2,1 ) in order to establish _SL2,1 , Do the same. Thereby, a bidirectional shared link is established between SI-SW1 and SI-SW2, and events can be shared with each other. By not registering a filter, a one-way shared link can be logically established.

(7) The number of event transfers, that is, the number of hops between SI and SW can be limited by the TTL value. As a result, multi-hop between SI and SW can be realized.
As is clear from the above description, SI-R is an entity having aspects of both event transmission and event reception, and is not essentially different from a service entity that is a general service application. In SION, an entity used for SION control such as SI-R is called a network entity, particularly distinguished from a service entity. In SION, all service entities and network entities are treated as a common entity, and autonomously operated according to a simple and consistent common logic of event transmission and event reception, that is, chain reaction of stimulus and firing. , Provide a super-distributed and super-loosely coupled architecture in which all entities can cooperate autonomously and distributedly.

For example, consider the case in FIG. 18 where SE3 registers a filter for SI-SW2. At this time, SI-R _2,1 receives the notification from SI-SW2, and registers the filter registered by SE3 in SI-SW1 via the reception session. Similarly, SI-R _1,3 registers the filter in SI-SW 3 via the reception session. Thus, the event path is established by the filters registered by the SI-R sequentially spreading to adjacent SI-Rs like a message game based on the shared link. This is called an “event path setting request”. This setting request is propagated to all SI-Rs in the event place, but the propagation range can be limited by the TTL value.

  That is, the event path spreading method according to the present embodiment is a method of sequentially spreading the event path setting to the entities in the communication system based on the shared link between the entities. The spread range between the entities is limited by the TTL value included in the request.

  Specifically, for example, in the present embodiment, the entity (device) sets an event path that sequentially propagates the registration of the filter to the entity device based on the shared link between the entity devices in the semantic information network. When there is a request, the semantic information router compares the TTL value included in the request with the number of stages following the shared link from the entity device that generated the request, and the number of stages is equal to or less than the TTL value. Only in some cases, a filter is registered in the adjacent semantic information switch based on a shared link.

  As another specific example, the TTL value included in the request is decreased every time the request is spread, and a filter is registered in the semantic information switch when the TTL value is greater than zero.

The restriction on the spread range will be briefly described with reference to FIG. Here, for convenience of explanation, it is assumed that an event path is established for lexical concept routing.
Each entity shall advertise the same vocabulary concept. In FIG. 19A, the entity 2 (strictly speaking, the SI-R belonging to the entity 2) has started an event path setting request. Here, since the TTL value is 2, the event path setting request is propagated only to entity 1, entity 3, and entity 4, and as a result, the event path is established as shown in FIG. The Similarly, when the entity 11 starts an event path setting request, an event path is established for the entity 10 and the entity 9. Therefore, event sharing is not performed between the entity 2 and the entity 11.

  However, as shown in FIG. 19B, when the entity 5 initiates an event path setting request, an event path is established for the entity 4, the entity 5, the entity 6, the entity 7, and the entity 9. The As a result, as shown in FIG. 19C, event paths are established for all entities in the event place except the entity 8.

  Thus, if the vocabulary concept is frequently used (reputed or popular) in the event place, the event path will eventually be established in the event place. Become. On the other hand, vocabulary concepts that are not so popular will eventually be deceived. The numbers on the event path in FIG. 19 indicate the multiplicity.

  From the viewpoint of distributed object technology, the event path can also be interpreted as follows. Entities that are super-distributed around the world have some kind of “correlation” between the entities. There are various properties that give correlations, such as entity names, group names, and attributes (location, interest, reputation, fashion, service). Entities have a connection between entities based on the correlation, and cooperate in a distributed manner. In SION, this correlation is expressed by a vocabulary concept and a vocabulary and managed as an event path. The strength of the correlation corresponds to generalization / specialization of vocabulary concepts, multiplicity of event paths, and the like.

  That is, in SION, correlation between entities is dynamically controlled and managed by an event path. This is the control of the chain reaction by the filter. Therefore, the entity does not have a fixed entity identifier. For example, an IP (Internet Protocol) address is a fixed identifier of an entity based on location. In SION, instead of this, entity properties, event place properties and the like described as vocabulary concepts / vocabularies are used as entity identifiers. The entity registers these as filters in the SI-SW, thereby declaring the property of the entity (entity identifier) and advertising its own property based on the shared link. This corresponds to the propagation of the event path, and thereby the event path is established.

  As described above, it is possible to reduce network traffic (the number of transfer of event path setting request) by limiting the spread range of chain reaction. Nevertheless, for event paths for popular vocabulary concepts and popular vocabulary concepts, the event paths can be established for all entities in the event place.

<Release and re-establish shared link>
When a specific SI-SW is moved away or removed, the shared link established between the SI-SWs must be released and the shared link reestablished. The outline is introduced below.
In FIG. 20, the first number of the shared link establishment request indicates the order of those requests. First, a shared link establishment request is made from SI-SW2 to SI-SW1. Next, a shared link establishment request is made from SI-SW3 to SI-SW2. Finally, a shared link establishment request is made from SI-SW 4 to SI-SW 2. Each SI-SW can make a shared link establishment request only once for each of the other SI-SWs, but establishes a shared link based on a link topology that can accept the establishment request without limitation. . This guarantees that no loop will occur between entities joined by the shared link. That is, an open topology can be realized by a simple method.

When this establishment request is successful, a shared link is established between the SI-SWs by the procedure described above. At this time, for example, SI-SW 2 shares a list of SI-SW (SI-SW 1) for which the establishment request has been made and SI-SW (SI-SW 3, SI-SW 4) for which establishment has been accepted. It is held in the link setting information management unit 57.
In this situation, for example, when the SI-SW 2 needs to be removed or removed, a shared link release request is issued to the SI-SW 2 and a new shared link needs to be established after the shared link is released. . The procedure is shown below.
For convenience of explanation, each part of SI-SW1 is a (for example, control part 56a) at the end, each part of SI-SW2 is at the end (for example, control part 56b), and each part of SI-SW3 is the end. C (for example, the control unit 56c) and each unit of the SI-SW 4 will be described with d (for example, the control unit 56d) at the end.

(1) A shared link release request is notified to SI-SW2. The control request receiving unit 55b of the SI-SW2 receives the request and controls the control unit 56b.
(2) The control unit 56b of the SI-SW2 notifies the 1-hop SI-SW (SI-SW1, SI-SW3, SI-SW4) that the shared link is released. This is made possible by sending an event with the TTL value of the event set to 1. At this time, for the SI-SW (SI-SW3, SI-SW4) that accepted the establishment of the shared link to itself, the establishment request destination of the shared link on behalf of itself, that is, the SI that has made the establishment request. -Teach SW1. If there is no entity that has made an establishment request, an arbitrary entity is selected from the establishment requesting entities to itself, and this is set as the establishment request destination of the shared link on behalf of itself. Thereafter, the control unit 56b of the SI-SW2 notifies the SI-R _{2,1, SI} to cancel the shared link (SL _2,1 , SL _2,3 , SL _2,4 ) established by the SI-SW ₂ itself. _{-R2,3, SI-R2,4} (however, these SI-Rs are not shown). Each control unit 73 of these SI-Rs releases a session with another SI-SW in response to this.

(3) On the other hand, the control units 56c and 56d of the SI-SW3 and SI-SW4 release all the shared links (SL _{3,2, SL4,2} ) that they have established for the SI-SW _2, and newly A request for establishing a shared link is made to the SI-SW (SI-SW1), and the shared link is reestablished. The control unit 56a of SI-SW1 cancels SL _1,2 and waits for an establishment request from SI-SW3, SI-SW4.

(4) Not only is it requested to release a shared link that has undergone legitimate procedures such as SI-SW removal or removal, but also due to SI-SW failure, power failure, session (communication path) failure, etc. In some cases, the shared link must be re-established. However, in such a case, there is a possibility that an alternative SI-SW necessary for reestablishing the shared link cannot be taught. In SION, in order to cope with such a situation, each SI-SW transmits an event of n hops (n is arbitrary) to grasp an alternative establishment request destination SI-SW. In SION, finer hop control is possible by using the TTL value and the hop attribute included in the event semantic information. Typical hop attributes include:
(I) Only SI-SWs that are requesting establishment of a shared link are targeted for hopping.
(Ii) Only SI-SWs that accept shared link establishment requests are targeted for hopping.
(Iii) All SI-SWs are hop targets.

  For example, the event flow when the hop attribute (i) is designated is indicated by a broken line in FIG. In FIG. 21, “(i / j)” appended to each SI-SW is shared link establishment information stored in each SI-SW. i indicates the SI-SW to which the SI-SW has made a shared link establishment request, while j indicates the SI-SW in which the SI-SW has accepted the shared link establishment request. For example, “(2/5, 6)” added to SI-SW3 indicates that SI-SW3 makes an establishment request to SI-SW2 and accepts establishment requests from SI-SW5 and SI-SW6. Represents. Each entity holds such shared link establishment information, so that, for example, when entity 3 makes a request to establish another shared link to another entity, the request can be rejected as an error. . In this situation, it is possible to know the presence of SI-SW2 by sending an event with a hop attribute of (i) with 2 hops in SI-SW6 and receiving a return event from SI-SW2. This is an alternative for SI-SW3 failure.

<Event path management method>
FIG. 22 shows how an event path (event routing information) set by SI-R is updated and managed when a service entity registers a filter in a one-way shared link. .
FIG. 22B shows an example of SI-R counter value transition when the SE registers / releases a filter for the event type “BGM” in the network configuration of FIG. 22A. is there.

In FIG. 22B, “SE _{i +} ” means that service entity i has registered BGM as an event type for the event to be acquired. On the other hand, “SE _i− ” means that the service entity has canceled the filter. The numbers in the table are counter values for event path management held by the SI-R filter registration information management unit 74, and SI-R is for setting an event path with respect to SI-SW. Indicates the number of times the filter is registered. Such management is performed for each event type. However, the actual filter registration from SI-R to SI-SW is controlled by the control unit 73 so as to be performed only when the counter value is incremented from 0 to 1. For example, when the SI-R transitions from 1 to 2, the filter registration is not performed, and only the fact that the filter registration is requested is stored in the counter of the filter registration information management unit 74. Similarly, when the service entity performs the filter cancellation, the counter is decremented to update the event path. However, the actual filter cancellation from SI-R to SI-SW has a counter value of 0. Sometimes it happens for the first time. This makes it possible to suppress the number of times an event path setting request or cancellation request is sequentially propagated (propagated) autonomously to the minimum necessary, thereby reducing event path control overhead. .

In order to simplify the description, only a one-way shared link is used for the description, but actually, the same event path management is performed in the bidirectional shared link.
Below, (b) of FIG. 22 is demonstrated easily.
(1) The initial value indicates that an event path for the event type “BGM” is not established.
(2) When the SE3 registers the filter, the event path setting request sequentially and autonomously spreads. As a result, the SI-R _3,2 registers the filter in the SI-SW2, and similarly, the SI- Event paths are set by R _2,1 and SI-R _2,4 registering with SI-SW as filters. At this time, each SI-R increments the counter.
(3) Furthermore, when SE5 registers a filter, through the SI-R _5,3, although setting request event path is spread against _{SI-R 3,2, SI-R} 3,2 is already Since a filter for transferring an event related to the event type BGM has already been registered (counter value is 1), the counter is simply incremented to 2, and no further filter registration is performed. Therefore, since the event path setting request does not spread to SI-R _2,1 and SI-R _2,4 , both counters are not changed.
(4) With the same concept, the counter values in the table are updated as shown in FIG.

  Next, the purpose and mechanism of entity advertisement (publication) will be described with reference to FIG. 3 and FIGS. There are two aspects of entity advertisement in SION: entity entry point advertisement (viewpoint 1) and entity property advertisement (viewpoint 2).

<Viewpoint 1-Advertisement of entity entry point>
An entity joins a base event place in order to become a component of the network. That is, a shared link is established with an arbitrary entity in the base event place. Thereby, the entity group connected by the shared link is self-formed. The base event place is an event place that serves as a base (starting point) for searching for (discovering) an optimum event place for an entity. That is, the base event place is an entrance to SION. Therefore, an entity in the base event place advertises its entry point to which a shared link is established for an entity that wants to join the base event place.

<Viewpoint 2-Advertisement of entity property>
An entity group connected by a shared link can advertise its own property. This corresponds to the event path described above.

  FIG. 23 is a schematic diagram showing an advertisement method. The flow until the entity is self-organized will be described below with reference to FIG. The entity makes a state transition according to the state transition diagram shown in FIG.

  (1) First, SION software is installed. The state of the entity at this point is “Non-Existent” (see FIG. 3).

  (2) By executing the SION software, the “Non-Existent” state is changed to the “Suspend” state (see FIG. 3). At this time, internal modules such as an entity control unit, a network entity factory, and a control panel are generated in the entity (see FIG. 24). An entity in this state is not known by SION.

(3) An entity in the “Suspend” state searches for an entry point for participating in the network as a component of SION. Specifically, an entity in the “Active” state in the vicinity is searched by broadcasting. The broadcast method depends on the implementation. For example, when implemented in a wireless network, all entities within the wireless reach are targeted for search. For example, when implemented in an IP network, IP broadcast is performed.
FIG. 23 shows that entity 1 has been found by a search by broadcast. If a nearby entity cannot be found by broadcasting, a well-known base entity can be used.

(4) By joining the discovered entity 1, a shared link is established between the entity (for convenience of explanation, hereinafter referred to as entity Y) and entity 1. Thereby, the entity Y is self-organized as a network component in the base event place (see FIG. 25). At this time, the entity Y transitions from the “Suspend” state to the “Active” state (see FIG. 3).
A set of such entities (a set of entry points for participating in the network) is called a “base event place”. The base event place is an event place serving as a base for searching for an optimum event place for the entity Y.

  An entity in the “Active” state can advertise (publish) its existence (entity property). Entity properties include entity names, nicknames, group names, attributes, and so on. For example, the entity 3 can publish an event place (event place α) in which the entity 3 participates as an attribute of the entity property. In addition, there is XML or the like as a description language for entity properties. By registering this as a filter in the semantic information switch in the entity, the entity property is propagated to other entities based on the shared link. Basically, it propagates to all entities in the event place, but the propagation range can be limited by the TTL value. In the entity property, the entity name is used as an identifier in multihop unicast communication, and the attribute is used as an identifier in multicast communication with a multihop attribute.

  (5) The entity Y that participated in the base event place releases an “entity property that meets its own needs” by sending an event (discovery) for searching for the “event place that it wants to participate in” Discover the entities that Here, it is assumed that the entity 3 is found as a result. Further, it is assumed that the entity 3 has already participated in the event place α.

  (6) By joining the entity 3, the entity Y can participate in the event place α. Thereby, the existence of the event place α can be known.

  (7) Furthermore, the entity Y can discover a new entity that meets its needs by sending an event for the same search in the event place α. For example, it is assumed here that the entity 12 has been discovered. This means that the entity 12 that could not be found in the base event place can be found via the entity 3. By repeating such an operation, it is possible to gradually reach an event place that meets its needs.

The advertisement method shown here described above has the following effects.
First, since there is no broker (administrator) for managing public information of entities, a self-organization network with strong fault tolerance can be constructed. In addition, it is not realistic to manage a huge number of public information (entity properties) with a broker, but such management is not necessary in this method.

  Secondly, the event place suitable for the entity can be narrowed down to the following formula, so that a desired event place can be efficiently searched. For example, even if the spillover range of the chain reaction (the number of discovery event hops and the spillover range of the event path setting request for the entity property) is limited by the TTL value, the entity 12 is discovered via the entity 3, As a result, another event place (event place β) to which the entity 12 belongs can be found. It should be noted that network traffic (event transfer count, event path setup request transfer count) can be reduced by limiting the spread range of the chain reaction.

Next, a hardware configuration for realizing this embodiment will be described.
The shared link setting information management unit 57 and the filter registration information management unit 74 in FIGS. 5 and 17 can only read out a nonvolatile memory such as a hard disk device, a magneto-optical disk device, a flash memory, or a CD-ROM. It is based on a storage medium, a volatile memory such as a RAM (Random Access Memory), or a combination thereof.

  24, 5, and 17 may be realized by dedicated hardware, and each unit in FIGS. 24, 5, and 17 may be a memory and a CPU (central processing unit). ), And a program for realizing the function of each unit in FIGS. 24, 5 and 17 may be loaded into a memory and executed to realize the function.

Also, a program for realizing a part or all of the functions of each unit in FIGS. 24, 5 and 17 is recorded on a computer-readable recording medium, and the program recorded on the recording medium is read into a computer system. The function may be realized by executing. Here, the “computer system” includes an OS and hardware such as peripheral devices.
Further, the “computer system” includes a homepage providing environment (or display environment) if a WWW system is used.

  The “computer-readable recording medium” refers to a storage device such as a flexible medium, a magneto-optical disk, a portable medium such as a ROM and a CD-ROM, and a hard disk incorporated in a computer system. Furthermore, the “computer-readable recording medium” dynamically holds a program for a short time like a communication line when transmitting a program via a network such as the Internet or a communication line such as a telephone line. In this case, a volatile memory in a computer system serving as a server or a client in that case, and a program that holds a program for a certain period of time are also included. The program may be a program for realizing a part of the functions described above, and may be a program capable of realizing the functions described above in combination with a program already recorded in a computer system.

  The embodiment of the present invention has been described in detail with reference to the drawings. However, the specific configuration is not limited to this embodiment, and includes design changes and the like within a scope not departing from the gist of the present invention.

In one Embodiment of this invention, it is a figure explaining the outline | summary of an entity. In the same embodiment, it is a figure explaining the type (kind) of an entity. In the same embodiment, it is a figure explaining the state transition of an entity. In the same embodiment, it is a figure explaining the structure of a semantic information switch, an entity, and a session. In the same embodiment, it is a figure explaining the structure of a semantic information switch. In the same embodiment, it is a figure which shows the structural example of a semantic information network. In the same embodiment, it is a figure explaining the structure of an event. In the embodiment, it is a figure explaining the outline | summary of a semantic information system (ontology system). In the embodiment, it is an example of a semantic information system (ontology system). In the same embodiment, it is a figure explaining registration of event acquisition conditions. In the same embodiment, it is a figure explaining the outline | summary of an event place. In the same embodiment, it is a figure explaining the view of a reference model. In the same embodiment, it is a figure explaining the increase / decrease form of a network entity. In the embodiment, it is a figure explaining the state transition of a semantic information switch. In the same embodiment, it is a figure explaining the state in which a semantic information switch performs the establishment request | requirement of a shared link. In the embodiment, it is a figure explaining the operation | movement in which a semantic information switch establishes a shared link. In the same embodiment, it is a figure explaining the structure of a semantic information router. In the embodiment, it is a figure explaining the operation | movement in which a semantic information switch establishes a shared link. It is a figure explaining the process of establishment of the event path and the fusion of an event path in the same embodiment. In the same embodiment, it is a figure explaining the operation | movement which a semantic information switch cancels | releases and re-establishes a shared link. In the same embodiment, it is a figure explaining the flow of the event which has a hop attribute. FIG. 6 is a diagram for explaining how an event path is managed and updated in a semantic information router when an arbitrary service entity registers a filter in the embodiment. In the same embodiment, it is a figure explaining the process until an entity performs advertisement and joins (joins) an event place. It is a figure explaining the module structure inside the entity in the embodiment. It is a figure explaining the process in which an entity self-organizes as a network component in an event place through the operation | movement of discovery (discovery) and joining (join) by the embodiment. It is a figure explaining the network by a prior art. It is a figure explaining the structure of the event (packet) in the network by a prior art. It is a figure explaining the network structure for implement | achieving a personalized information proposal service using the network by a prior art. It is a figure for demonstrating the broker model by a prior art. It is a figure explaining the realization method of a broker model by conventional technology. It is a figure explaining the concept of a semantic information network (prior art). It is a figure explaining the concept which shares an event in a semantic information network (prior art). It is the schematic which shows the transition of the business model from the former. It is the schematic which shows the dimension of a P2P (peer-to-peer) model. In one Embodiment of this invention, it is a figure explaining the federation over the firewall using SI-GW.

Explanation of symbols

  1: semantic information network (SION), 2: entity, 21 (a, b): entity, 22 (a, b, c): entity, 51: event receiving unit, 52: matching unit, 53: event transmitting unit, 54: Filter management unit, 55: Control request reception unit, 56: Control unit, 57: Shared link setting information management unit, 71: Event transfer unit, 72: Control request reception unit, 73: Control unit, 74: Filter registration information Management Department

Claims (6)

In a communication system configured by an entity including a semantic information switch, an event place search method for advertising an entity in the communication system based on a shared link between entities,
A search process in which an entity searches for other nearby entities by broadcast;
A base that participates in a base event place, which is an event place in which another entity discovered in the search process participates, by establishing a shared link with the other entity discovered in the search process. Event place participation process,
A discovery process in which the entity finds an entity that is a member of an event place that meets its needs in the base event place that participated in the base event place participation process,
The entity to establish a shared link between the discovery entity in the base event place, an event place other than pre-Symbol-based event place the found entity in the base event place is that participating Discover and establish the shared link with an entity discovered as an entity that is a member of the event place that meets its needs at the discovered event place, An event place search process that repeats discovering event places other than the found event place in which the entity is participating ,
An event place search method characterized by comprising: A communication system configured by an entity including a semantic information switch and performing advertisement of the entity based on a shared link between the entities,
An entity constituting the communication system searches for other nearby entities by broadcasting,
The entity participates in a base event place, which is an event place in which the other discovered entity participates, by establishing a shared link with the other entity discovered in the search,
In the participating base event place, the entity finds an entity that is a member of the event place that meets its needs,
The entity to establish a shared link between the discovery entity in the base event place, an event place other than pre-Symbol-based event place the found entity in the base event place is that participating Discover and establish the shared link with an entity discovered as an entity that is a member of the event place that meets its needs at the discovered event place, Repeated discovery of event places other than the discovered event place in which the entity has joined,
A communication system characterized by the above. An entity device comprising a semantic information switch that collates semantic information registered as a filter with semantic information attached to an event and activates an event receiving entity that has fired as a result, and constitutes a semantic information network,
Search other nearby entities by broadcast,
By establishing a shared link with other entities discovered in the search to participate in a base event place that is an event place in which the other discovered entities participate;
In the participating base event place, find an entity that is a member of the event place that meets your needs,
The base event to establish a shared link between the discovered entities in place, discover event place other than pre-Symbol-based event place the found entity in the base event place is that participating, the The discovered entity participates in the discovered event place by establishing a shared link with an entity that is discovered as an entity that is a member of the event place that meets its needs at the discovered event place Repeatedly discovering event places other than the found event place,
An entity device characterized by that. In the event place search method by the entity device comprising the semantic information switch that collates the semantic information registered as a filter with the semantic information given to the event and activates the event receiving entity fired as a result, and constitutes the semantic information network There,
A search process for searching for other entities in the vicinity by broadcasting;
A base event place participation process that participates in a base event place that is an event place in which another entity discovered in the search process participates by establishing a shared link with another entity discovered in the search process When,
A discovery process in which the entity finds an entity that is a member of an event place that meets its needs in the base event place that participated in the base event place participation process,
The entity to establish a shared link between the discovery entity in the base event place, an event place other than pre-Symbol-based event place the found entity in the base event place is that participating Discover and establish the shared link with an entity discovered as an entity that is a member of the event place that meets its needs at the discovered event place, An event place search process that repeats discovering event places other than the found event place in which the entity is participating ,
An event place search method characterized by comprising: An entity device that constitutes a semantic information network includes a semantic information switch that collates semantic information registered as a filter with semantic information assigned to an event, and activates an event receiving entity that has fired as a result.
A search process for searching for other entities in the vicinity by broadcasting;
A base event place participation process that participates in a base event place that is an event place in which another entity discovered in the search process participates by establishing a shared link with another entity discovered in the search process When,
A discovery process in which the entity finds an entity that is a member of an event place that meets its needs in the base event place that participated in the base event place participation process,
The entity to establish a shared link between the discovery entity in the base event place, an event place other than pre-Symbol-based event place the found entity in the base event place is that participating Discover and establish the shared link with an entity discovered as an entity that is a member of the event place that meets its needs at the discovered event place, An event place search process that repeats discovering event places other than the found event place in which the entity is participating ,
A computer program for executing the process of each process. The computer-readable recording medium which recorded the computer program of Claim 5.

Images