Remote test that QoS.

A real-life circuit test can solve problems before the customer picks up the phone or switches on the modern.

Teleport Communications Group, Inc. (TCG) provides local telephone service in 65 major markets nationwide. When the company began implementing ISDN services in 1997, it wanted a solution to remotely monitor and test ISDN circuits before connecting customers to the network. The company also wanted the capability of remotely duplicating customer applications to rapidly resolve problems.

To meet these objectives, TCG had to be able to test a circuit using real traffic before it went on line. Sergio Santiago, the company's ISDN product manager, says, "Other local carriers dispatch technicians first and troubleshoot after. TCG wanted a way to test the circuit with real digital signals and the ability to accurately simulate an individual customer's traffic. We needed to be able to do complete end-to-end testing from the network management center (NMC). We wanted to be sure that when we connected ISDN customers, they had a working line. Also, if they had problems, we wanted to rapidly isolate the cause and get them back on line quickly without the delay of dispatching service personnel to the site."

TCG selected the TSI-1567 ISDN Test Unit (ITU), from Telesync, Inc., of Norcross, Ga., to handle the testing requirements. The ITU can place and receive calls and combines the features of a protocol analyzer and multichannel bit error rate test set (BERTS) in a single unit. It provides a dial-up means of automatically looping back or generating BERT patterns on any of the 23 DS-0 B channels in an ISDN PRI DS-1 stream. The TSI- 1567 ITU is used to troubleshoot BONDed calls, switched digital services, and translation tables in switches.

One ITU is located in every switch location. As new switches are brought on line, they are immediately outfitted with an ITU. In this way, the ITU can be used to turn up PRI service to the customer, as well as provide a maintenance and troubleshooting tool when required.

A dial-up modem is used to access the RS-232 port of the ITU. Local operations personnel, as well as the NMC personnel, can dial into any ITU in the network of switches to monitor or test any PRI application.

Accessing the ITU, the operation and maintenance personnel can emulate the end user's applications and determine in a matter of minutes whether the problem lies within the network or at the customer' s equipment. "Although located at the switch, the ITU can test both the local loop and the customer interface," Santiago says.

The network switch is translated to route the incoming calls assigned to the ITU into the DS- 1 applied to the unit. The switch translations that affect the ITU can be changed to mirror a customer's PRI application, giving the operations personnel the capability to exactly emulate the customer's application at the Layer 2 and 3 level, namely, call setup and network control.

Upon receiving a trouble call from a customer, the NMC personnel dial up the switch ITU in the city where the trouble is reported. The test format is built on the display screen of the workstation and a call is placed to the customer's application. Layer 1, 2, and 3 information is scanned using the ITU's overview, L1, L2, and L3 screens. The log screen will show the messages transmitted and received by the switch from the customer's application on the D channel.

This method of sectionalization--the determination of "trouble in or out" takes approximately 20 seconds, the length of time it takes to place a call. If the call is successful, the speed of the call (56K or 64K) is determined, as well as whether it is voice or data, and, most importantly, the destination of the call. By changing the translations of the ITU to match those of the customer, including trunk routing and long-distance carrier selection and routing index, the technician can place the same call as the customer into the network and to the far-end destination. In this instance, the overview and log formats will be most helpful in analyzing a failure, if one occurs.

To date, network call failures are typically of two types. One is calls made utilizing a long-distance carrier whose networks are not end-to-end 64K. The other is calls routed through tandem switching arrangements and meet-point trunking locations that do' not have sufficient trunking capacity to accept the ISDN traffic loads. A call can be placed into the network at 64K, but there is no guarantee that it will complete at 64K.

As more and more carriers upgrade their networks to 64K clear-channel capability, the problem of placing and receiving 64K calls should disappear.

TCG's monitoring/testing strategy was designed in cooperation with Odyssey Solutions, Inc. "TCG has taken a proactive approach to providing quality customer service by implementing a maintenance strategy before it introduced its primary rate ISDN service offerings," says Warren Schmitt, Odyssey's chief technology officer. "This will provide significant benefits for its ISDN customers."

Circle 269 for more information from Telesync, Inc.