Centralizing Storage and Verification Element for Telephone Network

The present application relates to co-pending applications: Ser. No. 08/426,256, filed Apr. 21, 1995 entitled "Network Information Concentrator"; and Ser. No. 08/438,931, filed Jul. 13, 1995 entitled "Network Information Architecture Having Centralizing Storage and Verification Element".

The present invention relates to phone networks, and more particularly to a system for storing call event records and verifying requests for retransmission of missing records.

FIG. 1 is a generalized prior art elementary telecommunications network showing the path between a call originating point, for example, origination phone 6, and the network such as the MCI network, which will be the assumed network for this disclosure. Typically, a call from an origination phone 6 is handled by a local exchange carrier LEC 8. If, for example, an 800 toll-free number is dialed by the origination phone 6, the LEC 8 recognizes that MCI is the carrier for this particular 800 number. Consequently, the call is routed via line 9 to the MCI origination switch 10. This is the entry point of the call into the MCI network, and further routing may be necessary. However, in order to simplify the discussion of the present invention, simply routing through the switch 10 is presumed. Typically, switches of the type discussed are well-known in the art and identified, by way of example, as a DEX 600E switch, manufactured by Digital Switch Corporation, or alternatively a DMS 250 manufactured by Northern Telecom Corporation. International switches are provided by Ericcson Limited. The switch 10 has an adjunct processor (AP) 14 associated with it. The purpose of the adjunct processor is to receive call detail records (CDR) from the switch and collect them in buffers. Adjunct processors are of the type manufactured by Stratus, once marketed by IBM, and identified as a System 88 processor. This system later became a Stratus XA 2000, which may be used. Each switch in the network, such as switch 11, has an adjunct processor associated with it.

The call detail records are employed by various applications run by the network. For example, cloned data is provided along line 16 to a billing application 18 which posts the normal billing information for calls made by customers. Additional conventional applications include the acquisition of traffic statistics 24, as well as network management 28. Each of these applications utilizes specific data fields within the CDR data cloned by the involved adjunct processor 14. As will be appreciated, in a complicated network, each adjunct processor must communicate with each application so that the number of outputs from each AP, as well as the total number of inputs to the various applications, becomes unwieldy and difficult to manage from a hardware and software point of view.

Users of the various applications 18, 24 and 28 are schematically indicated by box 22. These may be databases of the MCI network or databases of customers who selectively load data from one or more applications. For example, a large corporate customer (user) may wish to monitor statistics regarding toll-free 800 number calls for which it is being billed. The traffic statistics application 24 includes such data for the customer which is extracted from the call detail records (CDR) provided by the various adjunct processors in the system.

The conventional data access point (DAP) 32 controls the routing of calls between network switches. The DAP also may block a call if it is unauthorized, or if network conditions warrant such blocking. The DAP is basically a large processor based database that includes customer profile information. The DAP stores call origination information, including such data as origination phone number, time of day a call was made, etc.

The present invention is directed to a Storage And Verification Element (SAVE) that replaces adjunct processors 14. A SAVE is connected at its input to a number of switches and network elements for collecting network event information into a logically centralized location. A network information concentrator (NIC) has its inputs connected to the outputs of each SAVE. An important function of the NIC is generating retransmission of data from SAVEs in the event that data gaps or duplicates are detected. The result is assurance of data integrity as early in the call record process as possible.

The above-mentioned objects and advantages of the present invention will be more clearly understood when considered in conjunction with the accompanying drawings, in which:

FIG. 1 is a block diagram of prior art network information architecture indicating the connections between an adjunct processor and various data applications.

FIG. 2 is a block diagram of the present invention indicating the inclusion of a network information concentrator between SAVEs and data applications.

FIG. 3A is a data flow diagram of the SAVE connected at its input to a DAP.

FIG. 3B is a data flow diagram of the SAVE connected at its input to a network switch.

FIG. 4 is a glossary of abbreviations included in the following description of the invention.

In describing preferred embodiments of the present invention illustrated in the drawings, specific terminology is employed for the sake of clarity. However, the invention is not intended to be limited to the specific terminology so selected, and it is to be understood that each specific element includes all technical equivalents which operate in a similar maimer to accomplish a similar purpose.

It is pointed out that FIG. 4 is a glossary of abbreviations used in this description.

FIG. 2 illustrates the substitution of SAVEs for prior art APs in a network. As in the case of the prior art, an originating point, for example phone 6, communicates via a local exchange carrier LEC 8 to the network. If an 800 toll-free number is dialed by the origination phone 6, the LEC 8 recognizes the network to handle the call, for example, MCI. As a result, the call is routed to an MCI origination switch 10. This is the entry point of the call into the MCI system, and further routing is normally required. A data access point 32 provides this function. A number of switches and other network elements feed data (12) to a corresponding Storage and Verification Element (SAVE) 14' which stores and distributes the data (33) to a consolidation system, referred to as the network information concentrator 34 (NIC), which is the subject matter of the previously identified co-pending application find incorporated by reference herein. Other types of network elements may include an enhanced voice services platform or an intelligent services network platform. The latter subject is discussed in an article by Richard Robrock entitled "The Intelligent Network--Changing the Face of Communications" published in the Proceedings of the I.E.E.E., Vol. 79, No. 1, January 1991.

The output of each SAVE 14' is preferably connected to an input of a distributed group of processors constituting the network information concentrator. The purpose of the concentrator is to compress and filter data supplied thereto, on an applications basis. By that it is meant that the concentrator 34 selects certain types of data fields from all of the data supplied to it, as required by a particular application (e.g., billing 18), which may then be employed by a user (22) (customer). As will be observed from FIG. 2, a single point of access to the applications is provided to transfer data from all network elements served. For the most part, the data format for the various applications is uniform so that the NIC 34 deals with standardized communications for the applications.

FIG. 3A is a system block diagram of the SAVE system illustrating data flow. An input to the SAVE is, for illustrative purposes, indicated as network packets of data output from a DAP (see 32 in FIG. 2). A corresponding discussion appears later, in connection with inputs from a network switch (FIG. 3B) instead of a DAP.

A SAVE-to-DAP interface (SDI) introduces the packets into the SAVE, and more particularly to an ADF Record Processor (ARP) which extracts raw data from packets. Particularly, this subsystem is responsible for disassembling the input data packet into its constituent applied data field (ADF) message pairs. The message pairs are then stored in a data buffer along with other types of preselected messages, such as messages from a Fraud Management System (FMS), included in the data network but not discussed herein. After buffering, the message pairs and FMS message is input to an Applied Data Field Block Formatter (ABF) which reads the queued ADF message pairs and the FMS message and creates disk blocks of a preselected size, padding out the disk block as needed. This subsystem also sets a block sequence number by which the block will be referenced in the future. The resulting block is then queued in a data buffer which splits the disk blocks. One path for the disk blocks is a SAVE-to-NIC interface (SNI) which generates messages for the NIC corresponding to the disk blocks. The disk blocks are also input to an ADF Block Storage Manager (ABS) which is responsible for archiving, on the indicated ADF record archive files, the blocks containing ADF message pairs and FMS messages to an appropriate disk data file. This is indicative of the first major function, namely storage, for the SAVE. The ABS maintains information pertaining to the sequence numbers of the data blocks written to the disks to support block retransmission. The ABS has a second output, constituting a confirmation message that is input to the SNI which enables the transmission of NIC messages. There are instances in network operation when it is necessary to retransmit ADF message pair data to the NIC. The SNI would receive such retransmission requests from the NIC and process such requests to a Retransmission Request Manager (RRM), after buffering in the indicated message buffer. The RRM verifies the validity of the request. This is indicative of the second major function of the SAVE, namely verification of retransmission requests. If the request is for a valid file and the start and end block sequence numbers are also valid, the requested data for transmission is retrieved from the hard disk where the data was stored (ADF Record Archive Files) and, after data buffering, is input to a retransmission input of the SNI for transmission to the NIC. In the event the data, requested for transmission, is impossible to retrieve, a negative acknowledgement message (NAK) is generated by the RRM and handled by the SNI as a response to the NIC.

In order to orchestrate all system administration, alarm processing and resource management, a processor, such as, but not necessarily, the IBM Netview 6000 is employed. Functionally, this processor is indicated in FIG. 3B as the Administration Menu And System Administration And Maintenance (SAM) box. Typically, administration menus are generated for system operators which allow the monitoring of the system and detection of alarm conditions. These types of processes are well-known in the art and are currently used in conjunction with DAPs.

The SAVE-to-LSE Interface (SLI) subsystem retrieves system alerts sent to it by the SAM and sends alert messages to an appropriate network component (not shown) such as a local support element (LSE). Typically, such alerts are transmitted to an LSE via a TCP/IP protocol connection.

FIG. 3B is similar to that of FIG. 3A and indicates the SAVE processes when connected to a network switch 10 instead of a DAP, as indicated in FIG. 2. The input to the SAVE will be network packets relating to switch call detail record data (CDR) instead of ADF pair and FMS messages, as was the case for the system of FIG. 3A. Other than the type of data processed within the SAVE, the operation of this system remains identical.

It should be understood that the invention is not limited to the exact details of construction shown and described herein for obvious modifications will occur to persons skilled in the art.