Printer Friendly

Alascom Is Served by a Centralized Network Alarm and Control System.

The Alascon communications network handles over 48 million telephone calls a year, and provides such services as WATS, telgerams, telex, marine radio, television and dedicated as well as dial-up data service.

These services are provided to all Alaskans, who are widely scattered over a sparsely populated state that encompasses an area more than two and a half times the size of Texas. Couple this with rugged terrain--mountains, glaciers and arctic tundra--plus adverse weather conditions that include winds of 90 to 100 mph, whiteouts, temperatures of 50 to 80 degrees below zero, and wind-chill factors of 150 below, and you have just a sampling of some of the concerns that Alascom, the state's telecommunications company, had to consider when designing its communication network.

The Alascon communications system is owned and operated by Pacific Telecom of Vancouver, Washington. The network consists of its own satellite (Aurora) equipped with 24 transponders, over 200 earth stations, three major satellite gateway terminals, three major switching centers, and over 6,000 miles of high-density (1200 to 2400-channel) terrestrial microwave radio systems. Net Is Technologically Complex

The technological complexities of the Alascom network vary from small VHF radio systems to highly complex broad-band transmission systems such as fiber-optic and coaxial-cable systems, analog and digital microwave systems, single channel per carrier (SCPC) and time-division multiple-access (TDMA) satellite systems, as well as newly established electronic switching systems.

As early as 1976, Alascon realized that maintaining the massive flow of performance data in a rapidly expanding communications environment such as its telecommunication network required some new and innovative methods to maintain and administer the telephone plant.

So that Alascom's ability to provide reliable, high-quality and cost-effective communication service to Alaskans was maintained, now and into the future, a centralized network management system was designed and implemented in 1981. The goals of this project were to improve service and increase profits: Goal of the Project Is Twofold

To improve service (a) by providing increased circuit quality through centralized surveillance of long-haul transmission facilities, and (b) by providing increased circuit reliability through trend analysis--that is, detection of trouble conditions or circuit deteriorations before customer complaints.

To increase profits (a) by improving the integrity of network information through centralized reporting (since otherwise, measured data may not always be reported in a timely fashion, measured results may be reported incorrectly through human error, and some data may not be reported at all), (b) by increasing the productivity of highly skilled technical personnel through better control methods (such as minimizing the number of trips to remote sites for preventive and corrective maintenance) and (c) by improving the efficiency of operating personnel and resources through better diagnosis of transmission problems and supervision of corrective actions. Remote Control Is Big Aspect

An important aspect of a centralized network management facility is the ability to collect information on the health of the telecommunications system and to remotely control the state of this equipment. This requires an alarm, surveillance and control function. There are numerous facets the problem of providing this function for a complex network such as Alascom's. These include:

* The system must have the ability to monitor and control not only central office switching equipment but all types of telecommunications equipment, from small VHF facilities to major satellite earth-station terminals.

* The system must provide the ability on a regional basis to perform central offive through microwave radio remote testing.

* The system must provide maintenance personnel with control features to make trunks busy, transfer radio channels, and start or transfer engine generators. These are a small sampling of control features that may be required by the operations department.

* The alarm and surveillance system should also have the ability to acquire analog telemetry information from each remotely monitored station. This telemetry information will be used to generate summary reports containing trend analysis. These reports will also supply operations and maintenance personnel with the necessary information to predict network faults long before alarm conditions or system outages occur. A sampling of the types of reports that could be generated includes network realiability, network availability, network major and minor failures, network preventive maintenance, network circuit outrages (trouble ticketing), network performance trends and network inventory (such as spares, test equipment and fuel).

The Alascom Network Alarm and Control System (ANACS) has been designed to solve these problems and to make availabel system that will provide alarm, control and monitoring capabilities to the entire Alascom network as an entity.

Configures by Pulsecom division, Harvey Hubbell Incorporated, ANACS is a network-wide Datalok 10 control and supervisory system that monitors network performance and activity, reports alarm and fault occurrences, provides data for systemic preventive maintenance, predicts trends and determines downtime probabilities. The regionally located Datalok master stations, using Hewlett-Packard HP-1000 computers under the control of technician/operators, collect alarm information and analog data from the remote sites within the region's maintenance territory. These area locations are called "regional control centers.) The system presently incorporates three of them, one each at the Fairbanks, Anchorage and Lena Point toll centers. Additional control centers are planned for the Bartlett and Eagle River earth stations and for Pipeline Operations.

The regional control centers will provide 24-hour surveillance of the radio, carrier and switching systems, plus site support equipment under their control. In case of failure or deterioration of any communications link, the fault will be pinpointed rapidly so that maintenance crews are dispatched to the correct site and equipped with the correct test equipment and spares. Host Computer Is in Anchorage

The control center computers communicate with a "host" computer, which is collocated with the Anchorage control center computer. The purpose of the host is to obtain selected alarm and analog information from the regional control centers for storage in a data base. The data base will provide a source from which reports about network performance may be derived. The host computer is also used by the system manager to update the software used by the control centers. The host is capable of servicing up to 16 regional centers in this manner.

The host is also capable of providing remote-control operation of the Hewlett--Packard Selective Level Measuring Set (SLMS) and the Transmission Impairment Measuring Set (TIMS). Under control of the stored program, carrier test points and circuit parameters may be measured automatically at predetermined intervals and compared with information reported earlier. This trend information will provide circuit quality and reliability to operations managers, and will furnish feedback to technician personnel on the results of their maintenance programs.

The original placement of remote auto-surveillance SLMS units at Anchorage, Glennallen, Delta Junction, Pedro Dome, Beacer Creek, Ripinski, Ketchikan and Hemlock Valley provides complete in-service peformance monitoring of the Alaska Interstate Microwave System (AIMS). Circuit Mileage Is 5.4 million

The initial ANACS network primarily monitors the major terrestrial interstate traffic between Anchorage (Alascom) and Port Angeles, Washington (AT&T). The total circuit mileage in this route segment is approximately 5.4 million miles. A conservative estimate of cost savings to Alascom through activation of ANACS auto-surveillance is $442,000 per year. The cost savings are being realized through fewer circuit outages and via better utilization of operations maintenance personnel.

ANACS employs a distributed processing concept of intelligence and controlling power. This method greatly reduces the amount of time to acquire information and also allows each regional control center to be completely self-contained and able to operate on its own, yet be an integral part of the overall system.

The "network host" facility and each control center are linked via a high-speed 2400 to 4800-b/s data link. The main purpose of this communication link is for the transmission of bulk data to and from the host computer for diagnosis and trend analysis of the communications system resulting from alarm conditions or telemetry information. Host Is USed to Compile Data

The network host facility will compile all data from the regional centers and prepare statistical reports for Operations Management. The host also maintains the necessary hardware and software to provide complete updated operating and trend-analysis information to the regional control centers, as well as a mass-storage capability. The CRT display formats are designed to present current data to the system maintenance personnel in a meaningful and convenient manner. The flash and/or variable-intensity feature is used to draw the operator's attention to conditions requiring prompt action. The CRT, in addition to serving as the primary display device, may also be used as an interactive display input medium. The system operator will use the capabilities of the CRT to control the communications system.

The system will normally function in the following manner: The regional controller will poll each remote station in the region, sequentially. The sequence in which the controller polls the stations will be defined by a polling list. When pooled, each station will respond with the current station status. Remote Sends Acknowledgement

If no alarm point has changed state, the remote will respond with a four-character acknowledgement. Otherwise, the status of all alarm points will be transmitted to the controller. The regional controller will compare the status of each point with that received on the previous poll and generate an alarm for each point that has changed state. For points that were already in alarm, a continuing alarm will be generated. These alarms will be sorted and sent to the CRT and line printer at the regional controller. In addition, the alarm data is transmitted to the host computer.

Periodically, the regional controller will request the analog data from all stations. This data is out of limits, it will be displayed and printed at the controller.

The regional controller can also exercise control over the output relays at the remote stations. Polling will be suspended while executing controls. In response to operator command, the controller will send a sequence of relay commands to a remote, which will perform the required operation and send a response back to the controller.

Alarm and other data transmitted to the host will be logged on magnetic tape to provide a system history trail. Alarms will also be printed on the line printers at the host. Operator Has Total Control

The operator will control and monitor the operation of the system through commands entered at the regional-controller or host consoles. At the regional console, the operator can control the polling operation and exercise control over the relay outputs. He can also print and display the status of the system and certain system tables. From the host console, the operator can perform all the functions that he can perform from the individual controller consoles. In addition, he can create and modify regional and station definitions. All regional-controller operator commands that affect system operations will be logged on magnetic tape at the host.

The regional-controller and host computers can suppress specific data from display and/or printout, depending on the various functions to be performed at each level. For instance, at the host level, specific alarms can be suppressed from being listed on the serial printer. However, at the regional level, because operators are interested in individual site performance and must respond immediately to high-priority alarms, these same alarms would be printed.

During normal polling, the regional console display shows alarm-and-control status data to the operator, suppressing only normal analog data. (An analog point that is out of limits is processed as an alarm.) Data Is Displayed Real-Time

At the network host, network data is displayed on a real-time basis only as requested by the system manager. The normal display at a host CRT is blank. Data is displayed, upon request, when building the data base, or when desired, to display the data from a selected regional center.

At the regional control centers (and at the host, when displaying regional-controller data), alarm data is displayed or printed out only when it is equal to or above a certain severity level. (Operators can request display of alarm data of any severity level.) Severity levels for individual alarms are set at the host during the data-base building process. Severity levels can range from zero (highest) to seven (lowest).

All network CRT terminals, host (on request) and regional, can display suppressed alarms in another form. As an example, the number of remote stations with certain active alarm conditions are listed in the top portion of the terminal display. This information includes the number of remotes within a region that have such conditions as: no response, equipment operating on auxiliary power, technician at site, open doors, low waveguide pressure, and AC power failure. The display also indicates the number of remotes with any critical alarm condition. Operators Can Request Data

As mentioned, operators can request, through terminal keyboard entry, data that is normally suppressed from display or printout. This capability reflects the basic data-base management philosophy of the ANACS--that is, no data will be lost.

The autonomous regional controller computers (up to 16 of them; see Figure 1) report all system actions and occurrences to the network host computer in Anchorage. One task of the host is to analyze network-wide problems and network trends by monitoring alarm, analog and control information from the regional control centers.

For trend analysis, an 800-bpi magnetic tape drive and a serial line printer are interfaced to the host computer for recording all network alarms and control actions reported by the regionals. The HP-1000 host computer (see Figure 2) with a 1,024-kilobyte memory, collects the regional data over full-duplex channels employing 4800-baud asynchronous modems.

At the network host, one of the tasks performed by the system manager is to build and maintain a data base of the system. These definition tables are used to instruct the regional control center on polling sequences and content. This is done by building the data base through the host CRT terminal onto a disk drive. When requested by the system manager, the host computer downloads the application software to the designated regional computer.

The system is designed to have network operations information available at headquarters through a CRT terminal display and a printer interfaced directly into the host computer at the Anchorage Toll Center. This information is a summary of alarm data reported earlier by the regional control centers. It provides a means for upper-level management to receive periodic reports about network performance.

Each regional control center is tasked to monitor and control up to 64 remote stations within its assigned region, place network data into priorities, and send specified data to the network host. In normal operation, the regional computer polls the remotes under its jurisdiction to detect change-of-status (COS) conditions.

If a remote sends alarm data, the regional computer compares the past state of that alarm input to the present state. The regional-controller CRT display indicates the state of an alarm as "new" for new alarms, "continuing" for old alarms, and "clear" for alarms that have returned to the normal state. "Derived" alarms are logical combinations of other status information from a single station.

Derived alarms result from a combination of alarms, which the regional-controller computer processes by taking a Boolean sum of up to five alarm conditions. For example, during a commercial power failure, the ENGINE RPM HIGH alarm might be reported with no immediate action required. During normal operation, a DR POWER BOARD HIGH VOLTAGE might be reported, and the maintenance center notified to investigate the condition on the next visit to the site. If however, both alarms were reported simultaneously, along with a generator frequency alarm, a derived major alarm would be output indicating that generator overspeed is causing equipment problems, and that immediate action is required to correct the condition. Simultaneous Minor Alarms Create "Derived"

Similarly, a power equipment minor alarm, if reported simultaneously with one or more minor radio bay alarms, would cause a major alarm to be derived. The derived alarm table can be designed to meet the particular requirements of the regional control center.

The regional computer communicates with each remote station over a full-duplex, dedicated facility through a 1200-baud asynchronous, DC-powered, full-duplex modem. Polling sequences and contents are established by network host system definition tables that are downloaded to the regional controller.

On any terminal in the system, the regional-controller CRT display is separated into three sections. The top section (the status-display area) gives the time of day, the average poll time per remote and the accumulated alarm status in that region. The middle ares (the alarm-display area) lists alarms and, when requested, analog or control data. At the bottom of the display is the input-prompt area. This is where the computer gives feedback to the operator in the form of four lines of scrolled display.

For simple commands, the input-prompt display area will indicate command execution and any entry errors. For commands that require a series of keyboard entries, the operator is prompted by the display for each keyboard input. In the control modes, the display prompts the operator for individual control-point data in order to aid the operator in locating a specific control point and function. As operator/computer interaction continues, the prompt area will scroll to allow room for new entries.

Regional control center operators canalso request a printout or display of a variety of lists from the system data base to aid in troubleshooting problems. Operators can request region command lists, alarm message tables, the polling list of remote stations, the list of stations suspended from polling, severity level of each remote station, and others. There is no need to contact the host for a printout of the present configuration. All Regional Controls and Systems Are Alike

Controls and systems operations do not vary from regional control center to regional control cneter. Technicians can operate any CRT terminal in the network without re-training.

At each communication station, a Datalok 10 remote station monitors up to 372 alarm points that, in turn, monitor the status of equipment such as microwave system radios, multiplexers, generators, tower lighting, waveguide pressure and building doors. Analog information such as battery voltages, fuel levels, weather, and microwave-radio AGC levels also are monitored, resulting in acquisition of data from up to 48 analog points per site.

To maintain network continuity and to supervise communications traffic, up to 95 controls (55 latching and 40 momentary) can be provided at each remote station for remote switching as required. Where low alarm activity is anticipated, such as small earth stations, application of Datalok 10D dial-up units provides for 36 alarms and 10 controls over a standard telephone line. The acquired data is collected and sent to each respective regional control center for fault reporting and control supervision, and then to the network host to be stored on tape for later use.

The Datalok remote stations are completely modular and have no practical limits to expansion capability. Because of a common data base, alarm points can be expanded in increments of 12, latching control points in increments of five, momentary controls in increments of 10, and analog measurement points in increments of 16. The network configuration or application software for expansion is built at the network host.

As the ANACS network expands into new areas, the system manager can add new remote stations into the system or expand the size of an existing remote station at the host control console. Using a regional polling list, he can build or change the network data base, even down to the message text for an individual alarm point.

System security is of great concern to Alascom management, since both the building of the data base and the execution of controls could affect the entire network operations and configuration. Protection against unauthorized manipulation was included as part of the design criteria.

Data-base access is allowed only through the host terminals. Critical data, such as analog input high and low limits, can be accessed only through the host system control console. The building of system definition tables is also protected at the keyboard.

A scheme of several passwords protects against inadvertent keyboard entry, control manipulation and unauthorized database changes. To enter any command, an operator must first enter the proper password for the host or regional controller to accept the entry. If there is no additional entry after 30 seconds, the system "times-out," and thekeyboard appears dead to the operator. Other keyboard passwords must first be entered before a control is pre-selected or executed and before an element of the data base is added or changed. System Keeps Downtime to a Minimum

ANACS was designed to keep downtime to a minimum through a number of fall-back mechanisms. On keyboard request, a regional control center that normally controls 64 remote stations can take over the monitoring and control function of three adjacent regionals. Using control relay facilities at selected remote stations, a regional can poll up to 256 remotes.

If power fails at a regional, back-up battery power keeps data input in the regional controller's computer memory. The regional-controller computer will restart normal system operation automatically when power is restored. If the duration of a power failure outlasts the battery life, the affected regional controller, upon restoration of power, will automatically request a download of the entire regional-controller software and data-base package from the host without operator intervention at either end.

Remote operation of the dedicated Hewlett-Packard SLMS and TIMS at the remote sites is under control of the stores program at the host via the vendor's interface bus.

At pre-determined intervals, the radio, carrier or circuit test point is connected to the test set and the desired measurements are transmitted to the host. An HP 3755A Access Switch Controller and one 3754A Broadband Switch for each 10 test points will provide access of up to 1,000 test points per site. At the host, measurements are compared against threshold values, and those that are out of limits are reported via teleprinter to the responsible maintenance group.

Maintaining the massive flow of performance data in a rapidly expanding environment such as the Alascom communications network was a prime factor in the ANACS surveillance and control system design. This expansion capability is accomplished through a modular hardware and software architecture.

In the past two years of operation, ANACS (Alascom Network Alarm and Control System) has proven to be a very effective tool in the management of Alascom's communications network. As new centralized surveillance and control requirements arise in the network, ANACS will continue to provide a flexible, expedient and cost-effective solution to meet these demands, now as we grow into the future.

Acknowledgement is made of the efforts of Dell Burch, Harry March and Chuck Peters of Alascom for their contribution to the final design of ANACS. A special thanks to Larrh Epp of Simpact Associates for his guidance and expertise in the development of the system software.
COPYRIGHT 1984 Nelson Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1984 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Riddle, C.
Publication:Communications News
Date:Jul 1, 1984
Previous Article:Micros and Workstations LAN-Linked at Dolby Labs.
Next Article:Meteor-Burst Link Supports Radar Net.

Related Articles
Satellite Data Modems Offer Greater Networking FLexibility.
Network Management Adjusts to the New Era.
Videoconferencing on Alaska's north slope.
Can anybody figure out Alaska's phone wars?
Waking from the paging alarm nightmare.
Contingency plans beyond the battery backup.
Long-distance service in Alaska: providers offer telecommunication packages to businesses that are competitive and tailored to each company.
AT&T operates TAPS communications.
AT&T Alascom.
Building Connect: Siemens Building Technologies.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters