MXPA95003852A - Method and system for updating replicated databases in foreign and home telecommunication network systems for supporting global mobility of network customers - Google Patents

Abstract

A method and system of the present invention updates a replicated database in a foreign telecommunication network system situated in a foreign region and a home country telecommunication network system situated in a home country. The method and system improves call setup time and system availability in a foreign region for customers roaming outside their home country. Customer records are updated in a visitor database contained in a foreign region within the signaling system of foreign telecommunications network. Information concerning the updated database record is transmitted from the visitor database through the signaling network of a home country to a home database which contains the replicated version of the visitor database record in the foreign region. A field is added to the database record corresponding to a version number of the database record so as to identify the record referenced by queries of calls in progress. The home database record at the home country is changed to reflect changes in the visitor database at the foreign region by using the version number sequence while maintaining at all database sites the older version of the database record for query access until after completing the processing of all previous call querying the database record.

Description

"METHOD AND SYSTEM FOR UPDATING DATABASES REPLICATED IN FOREIGN TELECOMMUNICATIONS AND RESIDENCE NETWORK SYSTEMS TO SUSTAIN OR SUPPORT THE GLOBAL MOBILITY OF THE CUSTOMER NETWORK" Inventor (s): KIN K. LEUNG, North American, domiciled at 1 Rainford Road, Edison, New Jersey 08820, E.U.A.

• Causaire: AT &T CORP., Corporation of the State of Ne York, E.U.A., domiciled at 32 Avenue of the Americas 10013-2412 New York, New York, E.U.A.

FIELD OF THE INVENTION This invention relates to a method and system for updating replicated databases in a foreign telecommunications network and residence system to improve the establishment time and availability capacity of the system to sustain or support the overall mobility of the network of clients in foreign regions.

BACKGROUND OF THE INVENTION In the increasingly smaller world because of the telecommunications networks that join it, it is advantageous to globalize a telecommunications network without building expensive transport networks in foreign countries. In fact, some countries do not allow local companies from the United States to enter foreign countries to build such networks. Other expensive proposals to support global mobility include the use of satellites to locate mobile customers around the world. This design would be expensive. It does not matter what type of global system is used, however for certain mobile clients, the number of queries and updates that corresponds to a database record would increase dramatically. Thus, It would be advantageous to distribute and replicate the client's records in multiple geographic locations, that is, sites of different signaling networks of foreign countries for easy access, achieving a high level of efficiency and capacity of system availability, as well as the time of establishment of the improved call.

BRIEF DESCRIPTION OF THE INVENTION * 10 The features and advantages of the present invention overcome the disadvantage of the construction of expensive call transport systems in foreign countries. The system can be used with networks wired and wireless from foreign countries. In accordance with the present invention, a method and system for updating databases is described * replicated in a foreign telecommunications network and residence system to improve the time of call establishment and availability capability of the system in a foreign region for the transfer of calls from the clients of the residence network. A foreign region has a system for transporting calls and switches to send calls from a station telecommunications through the transportation system of calls to a destination. The signaling system or # Signal transmission is connected to the call transport system and includes a primary site database that contains customer records to send the call and other signaling functions. A residence signage system in a country of residence has a residence database. The visitor database is connected to the residence signaling system. A customer record is updated in the visitor database. The information that is related to the registration of the updated database is transmitted through the residence signaling network to the residence database that contains the previous version of the database registry. The medium is sensitive to the term of all the consultation calls of the previous version of both of the registers of the visitor and residence database to delay the previous version of the registry of the # database . Queries are sent to the database of residence, when calls originate from the country of residence to the customer network in the foreign region. Queries are sent to the visitor database when the calls originate from the foreign region to a customer network. In an aspect of the invention, the information of the current location in the visitor database is maintained for very mobile customers traveling in a region - »foreign The identifiers contained within the residence database that direct calls to the visitor database for inquiries. The use of identifiers contained in the residence database decreases the amount of constant updating to the residence database. The activity of the record update is recorded after a database record updated has been processed in the visitor or residence databases. An acknowledgment is sent from the residence database back to the visitor database after the updated database registration in the residence database has been processed. 15 Multiple versions of customer records can be maintained in the visitor and residence databases, so that queries for a call can have access to the same version of the associated record for consistent sending and signaling information. Each The updated record may include a version number of the record, to identify the record reported by the calls in progress queries. The records in the residence database are updated by the sequence of the version number of the records in the database updated. Each version of a record includes a field of call counter to indicate the number of calls in progress whose applications have been previously accessed to that version of the registry. The field of the call counter is incremented if a new call refers to it and the call counter field decreases when the call set-up is completed. The previous version of the data base record is deleted when the call counter is zero.

BRIEF DESCRIPTION OF THE DRAWINGS The above features and advantages of the present invention will be more fully appreciated from the following description, with reference to the accompanying drawings, in which: Figure 1 is a block diagram of an illustrative embodiment of the invention showing a service of telecommunications using the method and system of the present invention in a wired and wireless network. Figure 2 is a timing diagram for the main registration protocol according to the present invention. Figure 3 is a synchronization diagram for a primary site retention protocol.

Figure 4 is a synchronization diagram for a basic clock-date protocol. Figure 5 is a graph showing a mis-sent call for two-site replication, according to the present invention. Figure 6 is a graph showing a mis-sent call for replication at five sites, according to the present invention. Figure 7 is a graph showing a call misdirected for non-uniform traffic loading, according to the present invention. Figure 8 is a block diagram showing a centralized database for global mobility using the protocol according to the present invention. Figure 9 is a block diagram showing the use of residence databases, distributed for global mobility. Figure 10 is a block diagram showing an extended signaling network with replicated databases for global mobility. Figures llA and 11B are flow charts showing the general method of the invention.

* DETAILED DESCRIPTION OF THE INVENTION The present invention now allows the replicated databases to be updated in a telecommunications network system to improve the cost of set-up time and the availability capacity of the system without the overhead typically associated with such help control protocols as the protocols for retention of the primary site and the basic timekeeper clock. In accordance with the present invention, the registration of a client is updated in a database of the primary site within the signaling or signal transmission system of a telecommunications network, which includes a call transport system with switches to send calls from a local telephone station through the call transport system to a destination. The information that is related to the updated database record is transmitted through the signaling network to at least one database of the secondary site, the which contains the previous replicated version of the primary site database record. The customer record is changed to the secondary site to reflect changes in the database at the primary site. Older versions of the registry are maintained in both of the bases of primary and secondary site data for access by ? query until after the processing of all previous calls that consult the versions of the record in the database site is completed. The design of the replicated database in accordance with the present invention can also be used to support global mobility. It can be applied to the services of Universal Personal Telecommunications, Personal Communication (wired and wireless) and future global mobility services offered by the Networks Intelligent The design will provide significant savings in transmission media and operating costs and minimize the delays associated with the call establishment time. There is a shorter implementation time, since the design with the global implementation is based on the current network infrastructure, requiring the deployment of only a small amount of equipment. The design is also consistent with the existing protocol standards. The proposed design for global mobility divides the world (in addition to the United States) in various regions that normally cover several countries. To sustain global mobility, a single local or resident database or local multiple databases are installed in the United States and a visitor database is installed in each region. A record of the data of signage of each client is stored in the base of residence data. If a client is a visitor from a foreign country, the client's registration is also replicated in the visitor database of the visited region. Calls for the transfer of customer calls originating from the United States and the visited region can be processed by the residence and visitor database, respectively. Due to the local availability capacity of the signaling information in the visited region, this design can reduce the means or accessories of transmission and cost of the operation. The call establishment time and response time of other signaling functions are also decreased. Figure 1 illustrates an intelligent network, shown generally in 10, which has a transport network, generally indicated at 12 and the signaling network such as the common SS7 network 14. The transport network 12 includes a network 20"Wired and could include a wireless network 16 for mobile units 22. The transport network 12 is connected to the central office 23, _ local and 24 local telephone station. The signaling network includes at least one primary site 26, which contains database records for sending purposes and other signaling functions. A secondary site 28 is separate geographically of the primary site 26 and includes records of the database replicated from the primary site 26, such as the illustrated A Record that corresponds to the signaling and general delivery information for a client. For purposes of understanding, a general, overall description of the foundations of the system is established, followed by more detailed specifications. Since the outputs of the distributed database are common for PCN services, UPT services and possibly other services involved in the use of the 10 databases ff (eg, Network Control Points, NCPs) in the IN , the general terminology for distributed database systems is adopted in the following specification. The computer system 30, 30a for the respective primary and secondary sites, where parts of the database are located, referred to hereinafter as sites. The customer records to send the call, features, service profiles and other Signaling information are called records. The readings and writings to the database are also mentioned as queries and updates, respectively. To allow the recovery of system failures (except broken disks), each site maintains a log (also known as a log) of all the updates made to the local database in a database. 36, 36a stable memory, which can survive the failures In accordance with the present invention, used with telecommunications services, the database 26a consists of a collection of individual records, which are distributed and replicated in multiple sites, such as the database 28a of the secondary site. Queries associated with a call and updates for a client, have access to a particular record in the base of data. In this way, the reading set and the writing set of a request and an update are a record. As a result, if queries and updates are processed atomically, unnecessary, inconsistent data can be avoided. 15 In addition, several queries can be sent for access to the associated register during the call set-up time, which usually takes several seconds. Although the record may have been updated in the average time, it is advantageous not to eliminate immediately an obsolete copy of the record, in such a way that the subsequent consultations of the calls in progress can be processed in a consistent manner according to the previous record. This eliminates the need to retain the records for the processing of the query without relate about the inconsistency of the data.

Additionally, calls will not be mishandled while the customer record is being updated. This would be more difficult if the reading set and the query and update writing set consisted of many records and files as in the database systems of other applications. Since the reading set and the writing set consists of a record, it is advantageous to treat the records as the "unit" for distribution and replication in different places. Since the read set (or write set) of each request (or update) is a record, the help control protocol must only maintain the internal and mutual help at the record level. With the internal help, groups of data in each record in each site are always consistent. For example, the sending data in a record must be valid in such a way that the calls can be sent properly. Mutual cooperation is preserved if copies of a record replicated in different sites become identical in a finite amount of time after many updates have been sent to the registry. The mandatory protocol is designed to allow the system to recover from system failures, that is, those updates that have been sent will not be made. As illustrated in Next, the present invention provides a protocol individual, which performs control of cooperation or assistance * and mandatory functions. The telecommunications system of the invention has the basic operating parameters: a. All sites are connected by a signaling network and messages sent from one site to another may be lost in the network. Even if the messages reach 2? - their destination sites, they can not necessarily arrive in the order in which they are sent. b. The records of the 26a database are distributed and replicated in the sites. For example, a record (record A) is replicated at sites N, indexed by i = 1,2, ..., N. To take advantage of the balance of In this case, the initial queries of the calls that require access to the A record are sent to those N sites according to some known call distribution algorithm for * those with skill in the art, such as the generalized wheel assignment algorithm or a static scheme.

To avoid the inconsistency of unnecessary data, if a call establishment involves multiple queries to the same record, the subsequent queries of a call are sent to the same place where the initial query of the call is processed. c. Each site 26, 28 can maintain versions # Multiple of a record, such that queries for a call can access the same version of the associated record for consistent sending and other signaling information. An obsolete version of the record is deleted after the processing of all query calls has been completed. d. For the purposes of recovery, each site records the update activities in a log in stable storage, which can survive the system failures considered in this invention. For each A record, a site with a copy of the record is selected as its primary site (PS) 26 and all other secondary sites (SS) 28. (Note that the PS can also serve as the PS for many other records stored on this site. For simple implementations, this is probably the case). The queries for the A record are sent to the Primary Site 26 or any of the Secondary Sites 28 according to the algorithm of call distribution in use. However, all updates for the A record are sent first to and processed by the primary site 26. If the processing of an update is completed without causing inconsistency of the data, the update is deposited in the primary site. So, the information that is related to the update is sent from primary site 26 to all child sites to update all replicated copies of the original database record. This cooperation control and the mandatory protocol is mentioned as the Primary Registration Protocol (PWP). To ensure the correct operations of the protocol, two fields are added to each record: 1) a version number (VN) and 2) a call counter (CC). He version number is used to identify the version of a record reported by the calls in progress queries, while the call counter indicates the number of calls in progress that have been previously accessed in that version of the record. If the call counter is zero and the record has been updated, the oldest version of the record is deleted. At any given time, there are probably very few versions (for example, two) of the same record that exists in a site, because the entire establishment time lasts only a few seconds and the time of interaction is typically much longer than that. For the purpose of understanding the details of the present invention, the following notations apply.

RA (n) is the version of the A record with the version number (VN) where n is for n = 0,1,2, ..., and the call counter, exists initially in the system. Also, let v_late be the VN of the latest version of the existing A record in a site. One represents an update for record A, which successfully creates R ^ (n). The primary registration protocol of the present invention can be further described as follows. If the query for record A is the first query of a call, then start Access RA (vn_late); (* Access to the latest version of registration A on the site *) Register the VN, vn_late, for the call; CC (vn_late) «* - CC (vn_late) +1 then terminate in any other way start RA Access (n) that has been accessed by the previous consultation of the call; If the query is the last of the call, then start CC (n) «* - CC (n) - 1; If CC (n) = 0 and vn_late > n, then RA (n) is removed end finalize It is assumed that the site is capable of doing determination, if the query is the last of the call. If the site does not have this capability, it is assumed that the site is notified by a separate message at the end of a call processing. In this case, the following operations are performed, when such message is received: if CC (n) = 0 and vn_late > n, then R ^ (n) is eliminated. § 10 Verify if CC (n) < 0 before the CC (n) CC (n) ^ - 1 operation can serve as a sanitation check since CC (n) _ > 0 by definition. The details of processing an update for record A in the PS are as follows: 15 If the update does not cause the data to be inconsistent (for example, due to incomplete shipping data supplied by customers), then start (* Update deposited *) Create RA (vn_late + l) and set vn__late ^ - 20 vn_late + l; Register the deposited update, Uvn .ate 'in ^ a blog; If CC (vn_late-l) = 0, remove R ^ (vn_late-1); Send an update message, (vn? Ate ' vn beats) to all secondary sites; Begin a waiting period for the update U ± ate finish in any other way, start (* Update aborted *! Abort the update and inform the source end The details of processing an update for the A record in a secondary site i are as follows. 0 For an update message (Un, n), compare n with vn_late in the site; If n > vn late + 1 then start (The update arrives from the sequence *) 15 Add the update out of sequence (Un, n) to an update list for the A record, sorted in ascending order of the VNs for the * future processing finish 20 in any other way, if n = vn__late + l, then start (* Update with the correct VN *) Process U ^ -, create R ^ (n), and set vn_late vn_late 1 + 1; Register the delivered update, Unv j_ate, in the binnacle; If CC (vn_late-l) = 0, delete RA (vn_late-l); Send an acknowledgment with the VN vn_late and the i number of the site to the PS; Repeat this procedure to process the first update in the update list for record A if your VN equals vn_late + l end in any other way, if n < vn_late then start (* Duplicate update which has been '# 10 previously processed *) Send an acknowledgment with the VN n and the site number i to the PS end The details of the processing recognition for Registration A in the PS are as follows. If the acknowledgment has been received before, it is discarded; in any other way, recognition for the * Update A from site i is recorded in the blog; if all acknowledgments associated with Un have been received of all secondary sites 28, the waiting time is deactivated and a "finish" message is sent to inform the source that all copies of record A have been successfully updated. The Processing at the End of the Wait Time for the A Record in the PS is as follows: The One update is sent back to those sites from which the associated acknowledgments have not been received. As noted in the above, all updates for the A record are processed first by the primary 26 sites. If an update causes any inconsistency of the data, it will be rejected (aborted) by the primary site 26; in any other way, the update is deposited. The protocol of the present invention, also preserves the internal consistency for all replicated copies of the AI record. If the copies of RA (0) replicated in the primary site 26 and all the secondary sites 28 are identical and internally consistent, the protocol of the present The invention maintains internal consistency for all replicated copies of record A. For example, consider a set of • updates. { A; n = l, 2, 3, • • •} deposited in the PS where A maps RA (n-l) to RA (n). The mapping is represented by RA (n) = Un (RA (n-l)). Because RA (0) is consistent and Un are deposited by PS only if they do not cause inconsistency, the R ^ (n) = Un (Un_1 (• • -U-j_ (RA (0)) • • •)) for all n = l, 2,3, ... it is always internally consistent in the primary site 26. By the use of VNs, the protocol forces the Un to be placed in each secondary site 28 in the same sequence, in which they are deposited by the site 26 primary. Therefore, the internal consistency at the primary site 26 implies the internal consistency for the A record at all the secondary sites 28. The mechanism by which the protocol of the present invention maintains the consistency of the record, is similar to that for an exclusive registration protocol proposed in. . Chu and J. Hellerstein, "The Exclusive- riter Approach To Updating Replicated File in Distributed Processing Systems, "IEEE Transactions on Computers, Volume C-34, No. 6, June 1985, pp. 489-500, although this last protocol does not consider the delayed processing of out-of-order update messages, the recovery aspects of the failure and the characteristics of telecommunications applications It should be noted that, once the update is deposited by the primary site 26, it can always be deposited in the secondary sites 28. Thus, there is no need to repeat the base The method and system of the present invention also provide mutual consistency.If no additional updates are sent to the system after many updates have been sent for record A to the primary site 26, all copies replicated from record A in the sites, they will eventually become identical in a period of finite time, with the condition that all sites are * operational, perhaps after a failure for a finite amount of time. II. If the copies of RA (0) replicated in the primary site and all secondary sites are identical, the protocol of the present invention maintains the mutual consistency for all replicated copies of record A. Assume the updates. { A; n = l, 2, ..., M} have been deposited by the primary site and no longer they will send additional updates. The use of the VNs by the system and the method of the present invention, forces the updates U to be placed in all the secondary sites in the sequence, in which they are deposited in the primary site. Since RdO) is identical in all the sites, the copies of RA (M), which is equal to UM (UM_1 (• • -U-_ (RA (0)) •••)), all sites are identical after the updates . { Un; n = l, 2, ..., M.}. They are # processed by all sites. The system allows non-identical copies of the records replicated in different sites are available for the processing of the call, simultaneously. In this way, queries can access obsolete information in some secondary sites, while updates are pending for processing, thus causing misdirected calls. How I know - discussed in the above, if the probability of each case is satisfactorily small, exclusive access to the records during the updates is not necessary. Without exclusive access, system 10 is freely interlocked. In addition, the system 10 does not require extensive message exchanges between the sites, which in any other way might be needed for cooperation control protocols such as retention to provide exclusive access. Since the record is replicated at the N sites, ^ 10 The total number of messages exchanged between the primary site and the secondary sites for each update is 2 (N-l) in normal situations. This is the minimum number of message exchanges to support the degree of mutual and internal consistency necessary for sending and receiving signaling of the call. The system and method of the present invention can also be used in other applications as long as they can tolerate inconsistency # of the record and the probability of obsolete reference data is kept at an acceptable level. The system of the present invention also has recovery capability, such as for the following situations. to. Loss of Messages and Link Failures: A loss of update and recognition messages, will cause the waiting time to end on the site primary. When this happens, the primary site returns to * send the update message to the sites associated with the lost messages. b. Site Failures: If the primary site fails before an update is posted to the blog, the update is simply lost. The source discovers that it does not receive a complete message from the primary site after a certain period of time. In the case where the primary site fails after it has been written in the If the update is posted, the primary site can resume its operations after the failure as if the update was only written in the log. As a result, the update is sent to all child sites again. For those sites secondary users who have already placed the update successfully, simply discard the duplicate update and send an acknowledgment back to the primary site. Yes # a secondary site fails, the primary site resends the update, if there is one to the site fails when the time or wait interval associated with the update ends. After the update is sent back to a secondary site for a number of times, the primary site can declare to the secondary site that it is no longer accessible. In this case, other procedures may be taken of recovery when secondary sites resume operations. When the failed secondary site is recovered, it can also resume its operations as if the last update was only written in the log and the PS request to forward all those updates completed during the failure. The system and method of the present invention is compared with two commonly used protocols: Primary Site Lockout Protocols (PSL) (Figure 3) and Clock 0 Basic Dater (BTS) (Figure 4). In addition, several numerical examples are presented to represent the typical performance characteristics of the system of the present invention (Figures 5-7). A fraction of the calls can be misdirected by the method of the present invention, due to the access of obsolete information in some databases of the secondary site, while the updates are pending. The fraction of misdirected calls is a measurement of key performance to determine whether or not the The method of the present invention is applicable to a particular application. As noted in the above, a record is updated first at its primary site in the method. A vulnerable period is defined by the registration in a site secondary, as the time interval from the end of • f an update for registration at the primary site until the time when the update is placed on the secondary site (Figure 2). Since most of the updated record has been available at the primary site at the beginning of the vulnerable period, queries that access the registry at the secondary site during the vulnerable period refer to the obsolete data, thus causing misdirected calls. Certainly, such bad addresses will not occur, there should be no record replication. The Figure 10 represents a timing diagram for the method of the invention. For example, the vulnerable period for the update A in the secondary site i is from time A to B. The duration of a vulnerable period depends mainly of two factors: 1) the delay of the network incurred by an update message sent from the pry site to the secondary site and 2) the update response time (ie, the time of the arrival of an update until your processing is completed) in the secondary site. These factors in turn depend on the characteristics of the signaling network, the design of each database, the traffic load, the ratio of query and update speeds and the discipline of the service program for the query processing and update.

Based on an operating study with * Typical parameters in today's telecommunication networks, the results in Figures 5-7 show that the fraction of misdirected calls in the method of this invention is satisfactorily small (for example less than 10 ~ 4) for a broad range range of expected client behavior (in terms of call production and read-write frequency ratio of the R-register) in Personal Communication Networks (PCN), wireless networks, m * «1Q Universal Personal Telecommunications (UPT) services and other advanced services that are offered by Intelligent Networks (IN). In addition, when the method of this invention is compared with other aid control protocols, such as pry site lock protocols and basic date clock, Figures 3 and 4 show that since the method of this invention avoids extensive message exchanges # required by the existing protocols between the database sites, reduces vulnerable periods, reducing well misdirected calls. Now with reference to Figures HA and 11B, a high level flow chart representing the method of the present invention using the pry registration protocol of the present invention is illustrated.

As shown in block 100, the record of the client in the pry site is updated first. Then the update is checked for any inconsistency of the data in block 102. If there is an inconsistency, the update problem is terminated in block 104. If there is no inconsistency in the data, then the update is processed in block 106 and a new version of the record is created in block 108. The update is posted in the binnacle in block 110. The information that is related with the update it is then transmitted to the databases in the secondary sites in block 112. The secondary site updates using the version number in block 114. Then the secondary site creates a new version of the record in block 116. The The update is stored in the bin on the secondary site in block 118. If the call counter of the last version of the record is zero in block 120, then the registration of the old database is deleted in block 122. Regardless of whether the call counter is zero or not, the secondary site sends an acknowledgment to the pry site as the update has been processed successfully in block 124. The process continues with the next update in block 126.

Design of the Replicated Database for Global Mobility The protocol of the present invention is also adapted to be used with a replicated database design to support global mobility with wired and wireless networks or a combination thereof. To sustain the service, personal mobility and terminal, the communications network must store, maintain and retrieve the information of the mobility of the clients for signaling purposes. This information includes that specified in the customer's environment reference (for example, the characteristics of the terminal in use, the services subscribed, the location data and the logic of sending the calls). In this way, a key burden for mobility management is to develop an efficient database architecture, in such a way that mobility data can be easily available # for signaling functions, such as call setup and sending. 20 The current approach to support terminal mobility < requires a resident database (or Registries of Local or Resident Location, HLR) and a visitor database (or Visitor Location Records, VLR). This HLR-VLR architecture has really been established as a industry standard in the Global System for Mobile Telecommunications (GSM) for Europe and the < # IS-41 recommendations for North America. The resident database can be accessed by the wired or fixed wireless network, while the visitor database is connected to a switch (referred to as the mobile switching center, MSC) in the wireless network. The sending and other signaling functions of each call initiated or destined for a mobile client, requires the use of the information of the location stored in the databases. # 10 The protocol and associated architecture to support terminal mobility are well defined and understood by those skilled in the art. The notation of personal mobility and service is relatively new. As a result, the protocols and the architecture for personal mobility and service are still going to be standardized. The present invention relates to a network-signaling architecture based on the basis of # Replicated data to provide personal and service mobility for customers traveling around the world. 20 In the following description, two alternative database architectural designs are examined, which must be adopted for the management of mobility. Due to the unique characteristics of global mobility, an extended base-based signaling network is proposed of replicated data to sustain global mobility. The ability to apply architecture to mobility personnel and service and its compatibility with industrial standards for establishing terminal mobility. The operation of the system and cost considerations for the proposed architecture have also been discussed. The system and method of the present invention can be applied to customer location information (i.e., shipping data) of personal mobility. As explained in the following, the present invention can easily cover the mobility of the service and reference environments such as Universal Personal Telecommunications (UPT) services. In order for the IN to support the services, it must be based on the extensive use of databases to direct the call and other signaling functions.

Each of the origin and reception ends of a UPT call may be a wireless terminal or a wire terminal. # An example of an existing database architectural design to support UPT services for clients traveling abroad are shown in Figure 8. The customer's location information is stored in a centralized database 200 located physically in the United States. To make this design feasible, it is required: 1. Customers traveling abroad have a registration of the visited country, in such a way that the network is informed of your current locations (for example in ^ terms of POTS numbers in wired networks or mobile station call transfer numbers in wireless networks). This location information for all clients that subscribe to UPT services is stored in the centralized database, which can be maintained by a Network Control Point, NCP. The signal transfer points 204 in the United States network communicate with the signal transfer points 206 in country 1 and any other country in 210. Country 1 includes a VLR 212 connected to an MSC 214 in the wireless 216 network. Country 1 also includes a wire 218 network. Country n also includes similar elements mentioned with the premium notation. 15 2. Each time a customer changes location, location data (for example, a POTS number on a wired network or an indicator at point HLR to VLR 212 that # serves in the location area, a group of cells where the user is located in a wireless 216 network) so that the client in the database was updated. Such database updates can be initiated: a) explicitly by the clients connected to a wired network as it is similar to the current 500 services or b) automatically by the wireless network in which the clients are located. 3. When a call is destined for a ^ UPT client located either in the United States or abroad, the signaling network consults the centralized 200 database for location information for call setup and other signaling functions. The signaling network that uses the distributed databases for mobility management is shown s > at in Figure 9. In the description that follows, the elements common to those described in Figure 8 are given the same reference numbers. In this distributed database design, customer records are divided and stored (but not replicated) in different physical locations, which can be interpreted as locations 230 of residence of the clients. For this reason, the distributed database design is also mentioned as the residence database design. Currently, the centralized database design can be seen as a special case of distributed design where all the data is placed in an individual location. On the other hand, distributed databases offer potential advantages over centralized design, such as improved availability capacity, reliability and the flexibility of load balancing. f This distributed database design can be used extensively for many services, for example Advanced Network 800 services and Software Defined Network (SDN), where customer data for the same service is distributed and stored in databases Multiple NCP Although each client record is stored in a primary and a secondary NCP for reliability reasons, only the primary copy is used for signaling functions under normal conditions and the copy secondary is treated only as a backup in the case of failures. In this way, the systems in use must be classified as the distributed design. The main motivation for such wide use of distributed databases is to overcome the limitation of the energy of the processing and restriction of the memory of each NCP. An individual database or at most a few databases (ie, NCPs) are needed in the initial deployment of a new service. As the demand for the service grows, databases are necessary additional, doing in this way that the design of the centralized database evolves in the distributed design. The assumptions and operations for the establishment of the call, dispatch and other functions for the design of the distributed database are identical for those of the design of centralized database, as discussed in the above.

For a given customer, the signaling networks or Switches have the computational logic to determine which database among many, contains the signaling data for the client. Because a current signaling network provides a very high degree of connectivity, it consults the databases distributed throughout the United States, does not cause unacceptable delays, or incurs high costs for call establishment and other signaling functions. The use of distributed databases will continue to be a viable approach. If high connectivity is maintained in the future, there will be no need for replicated customer records instead of multiple ones to reduce query delays and associated costs. Without However, this is not the case for signage in other global areas, as anticipated to sustain global mobility in the future. In this situation, customer records replicated in a visitor database located in foreign countries will reduce operating costs and will shorten the query delays as explained in the following. A signaling network based on replicated databases is shown in Figure 10, which illustrates the basic use of the replicated database of this invention, to be used in global mobility. HE used as reference numbers for common elements to # other units of Figures 8 and 9. 1. Depending on the anticipated traffic load and other engineering considerations, the world (in addition to the United States) is divided into several regions, each of which covers a number of countries aliens, such as Region A 246 illustrated and Region B 248. 2. A pair of databases 250 viewers are physically installed in each region and connected to the network 0 signaling in the United States, for example through two additional STP 252, physically located in this region. The number of STPs installed in a region and the number of signaling links that connect STPs to the signaling network in the United States can be adjusted and determined in accordance with traffic, performance and reliability requirements. 3. In the process of registering a client that $ visit a foreign country, a location record and other related signaling information for the client, establishes 250 visitor database for the region. This new information will also be sent back to update the associated residence database 230 in the United States. As a result, each call transfer client has two replicated copies of data from signage: one in the residence database in the United States and the other in the visitor database * located physically in or near the visited country. The registration in the visitor database will be withdrawn when the client leaves the region. The replicated database can be accessed for the establishment of the call and other signaling purposes. For the call that originates from the country or region visited and destined for the transfer of calls from the clients, the queries can be sent to the base visitor data. Calls initiated in the United States will consult the residence database. There are three main advantages for the use of replicated databases: 1. Substantial Cost Savings 15 Due to the local availability of necessary customer data, the transmission cost for signaling messages for calls originated from the ? country o. visited region can be saved. The savings will be substantial for: a) customers who transfer the calls to half of the way around the world, b) calls require multiple queries in the database, which will probably be the case for advanced services in the future and c) a majority of calls originate from the country or region visited, which also will be the case for some business clients. Also, as the number of messages sent long distance is decreased, the cost of accessories can also be reduced as less transmission accessories (for example, signaling links) will be necessary to support the same amount of call traffic. 2. Increased Delay Efficiency The call establishment time and response times for other signaling functions can be reduced. ? 10 3. Compatibility with Existing Wireless Standards The design of the replicated database is compatible with industry standards such as IS-41 and GSM specifications for wireless networks. the basis of visitor data in a region can be treated as the HLR by the wireless network in the visiting country. The rest of the standardized protocols for wireless networks continue being applicable. When receiving updates from a local wireless network, the visitor database is responsible to send the updates to the residence database in the United States to ensure the consistency of the client's data in the residence and visitor databases. Hence, the design of the replicated database does indeed apply not only to wireline services, but also services wireless In accordance with the present invention, the consistency must be maintained between the replicated copies of the customer data stored in the residence and visitor databases. The system and method of the present invention using the Primary Registry Protocol (P P) can be applied to the residence and visitor databases for global mobility. The Registration Protocol The primary of the present invention is advantageous since telecommunication applications can tolerate the inconsistency between replicated records over a short period of time. As noted in the above, the main consequence of giving access to obsolete data is a misdirected call. The analysis of operation or effectiveness of the Primary Registry Protocol of the present invention establishes that the probability of misdirection in the present invention is small for the expected client behavior in wireless networks and the IN. The present The invention conserves both of the internal and mutual data consistencies and also allows the system to recover from certain failures. The main advantage of the Primary Registry Protocol is its low overhead, especially when compared to existing protocols for systems computers (for example, bank systems), which are intended to maintain data consistency more strong every time Using the Primary Registry Protocol of the present invention to support overall mobility, the visitor database and the resident database is designated as the Primary Site and the Secondary Site, respectively for updating purposes. As noted in the above, this provision is appropriate because the location of the updates and other changes in the W 10 signaling information, are most likely generated from customers, who are transferring calls in foreign countries. Although the above discussion focuses on location information for UPT clients, it must It should be noted that the design of the replicated database currently applies well to the mobility of the service and reference environments. The location information, the subscribed services and the reference environment are specified on a per client basis. In this way, the The design of the replicated database and the associated protocols can treat the information per client related to the subscribed services and other data groups included in the reference environment, in the same way as the location information. As for what related to the efficiency performance of the design of the replicated database, the only difference * among the various data groups will be your query relations to update (for example, the speed ratio, query at the location update speed for a UPT client). However, it is expected that the most frequently updated data group will probably be the location information for a mobile client; the query relationships to update for other data groups of - 10 signaling will be greater than that for the location data. As a result, the update algorithm of the present invention will also apply to the customer records containing the subscribed services, reference environment and other signaling information. 15 One factor that can degrade the effectiveness of the replicated design is the query-to-update relationship. If the relationship is high (that is, the customer's information is rarely changed), the overload in updating the replicated records will be minimal. In any other way (for example, for a mobile client moving from location to location), the update overhead may be important. The following strategy will increase the effectiveness of the replicated design in such situations. A wireless, mobile client is transferring calls in a foreign country. As a client gets moves from one location area to another, information from The location of the client in the visitor database is updated frequently. The corresponding customer record in the residence database is also updated. In this way, the transmission and processing costs to maintain the current location information in the residence database will be important, especially if the client is located far away from the United States. The situation will become even worse if the "10 client moves from location to location without making or receiving calls, that is, no revenue is received to cover the cost incurred in managing the database updates, a way to reduce the overhead for the location updates, is to keep the current location information for certain very mobile customers only in the visitor database. The residence database, on the other hand, contains an identifier that indicates to the visitor database in which current location information is stored, (ie, the visitor database has the complete record of the information associated with this client that transfers calls and the residence database has only partial customer information).

As a result, the residence database does not You need to update the location information every time a customer changes their location abroad. In addition, calls originating from the United States, destined for a client traveling abroad, first retrieve the identifier from the residence database and then give access to the information of the current location in the visitor database for the establishment of the call and other signaling functions. ? 10 The increased focus on keeping client data complete, only in the visitor database, applies not only to location information, but also applies to other data groups with frequent updates. In addition, the improvement in focus must be used on a per-client basis because different clients have different mobility and call patterns. As discussed in the foregoing, the key parameters to identify these clients for whom the improvement scheme can be applied include: 20 1. The relationship of calls to the client originating from the visited region to all calls destined for that client, 2 The database of the query-to-update relationship (for example, an indication of mobility) and 3. The cost relationship to update the base of residence data (with updated requests sent from the visited region) and that for remote consultation of the visitor database for calls originated from the United States. Based on these parameters of each traveling client, the signaling network can determine whether the improvement in focus should be applied to the client (ie, only the visitor database contains the complete record of the client and the residence database). stores only parts of the customer's data, which are updated less frequently). In fact, the approach can be applied to a client dynamically on a regular basis, ie the network can turn the focus on or off periodically (for example, every night) for a traveling client, depending on the customer's call production and statistics on the frequency of update during the previous period. In accordance with the present invention, the network of extended signage based on the replicated database, has many benefits and can be seen as a means to globalize the telephone network without actually building transport networks in foreign countries, which telephone companies in the United States frequently do not are allowed or prefer not to do so. Still, the design of the replicated database in effect makes use of networks foreign companies to support global mobility for the customer network. The replicated design costs less to implement than other methods of the system, such as the use of satellites to locate mobile clients around the world (ie, global mobility, terminal). Most importantly, the system of the present invention is based on the current infrastructure of the wired and wireless networks and complies with the existing rules of cellular networks. In this way, the replicated design not only reduces the costs of operation and delay of call establishment, when compared to other alternative database designs, but also allows the United States telephone network to offer global mobility services to customers sooner The proposed design also helps the US network to compete with some foreign telephone companies. The network may choose to allow access to the visitor database only by the networks (wired and wireless) of existing business partners; the competing networks will not be provided with the signaling information stored and used in the visitor database.

As a result, for calls originating in a foreign country and destined for a client of transfer of calls of the network, only the networks of the company in question are equipped to transport the calls. Such a provision can also be used to improve the position of the network in the United States in association with foreign service providers. In another business arrangement, the telephone network may also choose to provide, for a tax, some foreign carriers with certain Signaling information to transport calls destined for the clients of the network that travel in the country. In addition, if the telephone network can work with its foreign business partners, the design of the replicated database can be implemented in an alternative way.

If an arrangement can be made between the telephone network and your business partner abroad, instead of installing the remote database of the network and the PTS in a foreign country, the records of the clients of the network that travel in the country they can be downloaded directly from the residence database to a database in the partner's network. Therefore, as in the design of the original replicated database, each traveling client continues to have two copies of their information for signaling purposes. The protocols and operations for the design replicated remain applicable to this new framework, although a telecommunications network in the United States does not have physically own visitor databases and STP in that country. It should be understood that the above description is only a preferred embodiment of the invention. Numerous other provisions may be contemplated by one skilled in the art, without departing from the spirit and scope of the invention. It is noted that in relation to this date, the The best method known to the applicant for carrying out the aforementioned invention is that which is clear from the present description of the invention. Having described the invention as above, property is claimed as contained in the following: #

Claims (29)

1. A method for updating the replicated database in foreign and residency telecommunications network systems to improve the call set-up time and the availability capability of the system in a foreign region to transfer calls to clients, the method is characterized in that it comprises The stages of: updating the customer record on a base of visitor data, contained in a foreign region within the signaling system of a foreign telecommunications network, in which the telecommunications network includes a call transport system with transmitters to send calls from a local foreign station through the system of transport of call to a destination, transmit the information that is related to the registration of the updated database of the visitor database through the signaling network of a residence telecommunications system to a residence database , which contains the replicated version of the visitor database record in the foreign region, add a field to the database record that corresponds to a version number of the database record in such a way that it identifies the versions of V ~ the records mentioned by the consultations of calls in progress and change the registration of the residence database in the country of residence, to reflect the 5 changes in the visitor database in the foreign region using the sequence of the version number, while maintaining all the database sites, the previous version of the database record for the access of / Í? L queries, until after completing the processing of 10 all previous calls that consult the version of the database record.

2. The method in accordance with the claim 1, characterized in that it includes sending queries to the base of 15 residence data, when the calls are originated from the country of residence to the clients of the network in the foreign region. #

3. The method according to claim 20 1, characterized in that it includes sending requests to the visitor database, when the calls originate from the foreign region to a client.

4. The method according to claim 25 3, characterized in that it includes directing the query to the base of resident data for access to the client's registry, if the client does not make a call transfer in a foreign region.

5. The method according to claim 1, characterized in that it includes the step of deleting the records from the visitor database, when the associated network clients leave the foreign region. * 10

6. The method of compliance with the claim 1, characterized in that it includes depositing the updated record of the visitor and residence databases registering the activity of the updated record in a log in a stable data storage area 15 after the registration of the updated database has been processed and deposited in the visitor and residence databases, respectively. *

7. The method according to the claim 20 1, characterized in that it includes sending an acknowledgment of the resident database back to the visitor database, when a record of the updated database has been processed within the residence database.

8. The method in accordance with the claim 1, characterized in that it includes the step of maintaining the current location information in the visitor database for very mobile clients traveling in the foreign region and maintaining an identifier in the residence database that directs the calls to the database of the visitor for consultations.

9. The method in accordance with the claim

10 8, characterized in that it includes the step of maintaining the identifier in the residence database by directing the calls to the visitor database for consultations on a per client basis. 15 10. The method according to the claim 1, characterized in that it includes maintaining multiple versions of the records in the residence and visitor databases, such that queries for a call can have access to the same version of the associated record 20 in the same database for consistent shipping and signaling information.

11. The method in accordance with the claim 1, characterized in that it includes eliminating the version of the 25 record of the previous database, in all the sites of the database after the processing of all the calls requesting the previous version of the database record is completed.

12. The method according to claim 1, characterized in that it includes adding a field to the register corresponding to a call counter to indicate the number of calls in progress, whose queries have been f previously accessed to that version of record. 10

13. The method according to the claim 12, characterized in that it includes increasing the call counter field in one, if a new call is informed and the call counter field in one decreases 15 when the establishment of a call is completed.

14. The method in accordance with the claim # 13, characterized in that it includes eliminating the registration of the obsolete, previous database, when the counter of the call 20 is zero.

15. The method according to claim 1, characterized in that it includes the step of verifying the registration of the updated database in the database 25 of the visitor site, for the inconsistency of the data.

16. The method according to claim 1, characterized in that the call transport system of the foreign region includes a wireless telephone network 5 and a wired telephone network.

17. A system for updating the replicated database in foreign and residency telecommunications network systems, to improve the call set-up time and system availability capacity in a foreign region for the transfer of customer calls, the system it is characterized in that it comprises: a foreign telecommunications system located in a foreign region and having a transport system 15 of calls and switches for sending calls from a telecommunications station through the call transport system to a destination, a signaling system operatively connected to the foreign call transport system and including 20 a visitor database within the signaling system that contains customer records to direct the call and other signaling functions, a residence telecommunications system that has a residence signaling system and residence data, the visitor database is connected to the residence signaling system, means to update a customer record in the visitor database, means to transmit the information that is related to the registration of the updated database to through the residence signaling network to the residence database, which contains the previous version of the database registry, and $ 10 responsible means to complete all the calls that consult the previous version of both of the records of the visitor and residence database to eliminate the previous version of the database registry.

18. The system according to claim 17, characterized in that it includes means for cleaning queries to the residence database, when the • calls originate from the country of residence to the customer network in the foreign region. 20

19. The system in accordance with the claim 17, characterized in that it includes means for sending requests to the visitor database, when the calls originate from the foreign region to a network of 25 customers

20. The system according to claim 17, characterized in that it includes means for maintaining the information of the current location in the visitor database for very mobile clients traveling in a foreign region and an identifier contained within the residence database that direct calls to the visitor database for inquiries.

21. The system of. compliance with the claim 17, characterized in that it includes storage means for recording the activity of the updated record in the visitor and residence databases, after the registration of the updated database has been processed and deposited in the visitor and residence databases , respectively.

22. The system according to claim 17, characterized by including means for sending an acknowledgment from the residence database back to the visitor database after the updated database record has been processed within the database. of residence data.

23. The system according to claim 17, characterized in that it includes means for maintaining multiple versions of the customer records in the visitor and residence databases in such a way that queries for a call can have access to the same version of the associated record in the same database for consistent shipping and signaling information. i

24. The system in accordance with the claim 10-17, characterized in that each updated record includes an identification field corresponding to a version number of the record, to identify the record informed by the calls in progress queries.

25. The system according to claim 24, characterized in that it includes means for updating the records in the residence database by the sequence * of the version number of the updated database records. 20

26. The system in accordance with the claim 17, characterized in that each updated record includes a call counter field to indicate the number of calls in progress, whose queries have been previously accessed to 25 this version of the registry.

27. The system in accordance with the claim 26, characterized in that it includes means for increasing the call counter field by one, if a new call is mentioned, and means for decreasing the call counter field by one, when the call establishment is completed.

28. The system according to claim 27, characterized in that it includes means for eliminating the registration of the previous database when the counter of the call is zero.

29. The system according to claim 17, characterized in that the call transport system of the foreign region includes a wireless telephone network and a wired telephone network. In testimony of which I sign the present in this City of Mexico, D.F., on September 8, 1995. Apo was or SUMMARY OF THE INVENTION A method and system for updating a replicated database in a foreign and residency telecommunications network system is described to improve the call establishment time and system availability capacity in a foreign region for the transfer of calls from the clients. The client's records are updated in a visitor database contained in a foreign region within the signaling system of the foreign telecommunications network. The information that is related to the registration of the updated database is transmitted from the visitor database through the signaling network of a country of residence to a residence database, which contains the replicated version of the registry of the visitor database in the foreign region. A field is added to the database record that corresponds to a version number of the database record to identify the record reported by calls in progress queries. The registration of the residence database in the country of residence is changed to reflect the changes in the visitor database in the foreign region, using the sequence of the version number, while keeping the database in all the sites, the oldest version of the registry of the database for the query access until after the processing of all previous calls that consult the database record is completed.