Cover Image

Not for Sale



View/Hide Left Panel
Click for next page ( 438


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 437
A.4 Communication System Support Section A.3 presents example urban, suburban, metropolitan, and rural ITS communication system designs and cost estimates to procure and install communication systems. This section win present methodologies to estimate the costs to operate the communication system and calculate the life cycle costs (LCC). Example L~CC estimates are presented Hat illustrate typical trade-offs encounters in ITS communication system design. This section addresses Me following: . . General Overview: Ma~nt~nabiiity of advanced communication systems Reliability, availability, and maintainability planning and estimating; Maintenance personal staff planning and es~nadng; Spare parts planning and estimating; Test equipment requirements; and Life cycle cost es~nadng methodologies and examples. A.4.1 General Overview: Maintainability of Advanced Communication Systems A major concern of jurisdictions relating to the deployment of advanced communications technology is ability to maintain new technology. This issue involves several sub-issues including: Ease of maintenance, a basic issue since it impacts training; New gaining of the maintenance staff; New test equipment requirements and availability; and Spares requirements and associated cost of availability. To provide a basis for We discussion of maintainability, it is important to understand Mat by advanced communications technology we mean that which is not in experimental form, has been L:\NCHIWha~` NCHRP3-51 Phase2FmalReport A4-1

OCR for page 437
proved by bow laboratory and field testing and has been productized. Produchzation means Mat We product is fully documented is in production with established manufacturer's product support. It is furler a technology which is emerging into operational use but which has not been historically applied to ITS applications on a widespread basis. HistoncaBy, copper tw~sted-pa~r coax and point-to-po~nt wireless communications have been applied to llS communications applications. Typically, proprietary communications network designs were used. Advanced communications technology is considered to be optical, digital networks, digital wireless, subband wireless, VSAT, and similar products, including advanced local area networks ~ANs), budges, routers and switches. Maintainability does not just happen; it must be designed into systems. Elements Mat make systems maintainable are: Use of open systems standards wig proven interfaces (bow hardware and software); Modular design where components are not agony coupled and are easily replaceable; Built-in test features web real-time tests conducted; Real-dme fault detection and reposing; Network management to support remote fault monitoring by Be maintenance organization; Levels of test (bu~It-in diagnostics) to isolate failures in specific replaceable modules to a high degree of confidence; Safe, hot, change-out of failed modules eliminating the need to deactivate power and reinitialize the system; Test bus to support use of test equipment; and Use of fault propagation prevention to contain a fault, preventing Be distribution of "garbaged" data throughout Be communications network. Modem communications equipment is evolving to incorporate Be of the Test Architecture and Bus of Be loins Test Action Group and EKE's Pll49.5. The EKE Pll49.5 Test Bus allows removable modules to be tested and results reported via a network ~nanagement channel. The Test Bus supports both a system-level test and a removable module-level test through use of test equipment, aver Be system level repair has been accomplished. t;WCNRP~ap ~NCHRP 3-51 Phase 2 final Report A4-2

OCR for page 437
Operational circuitry also monitors statistical data errors and reports these errors to a central network management monitoring function. A trend toward higher bit error rates can provide an indication of an upcoming failure In electronics or a marginal condition In the physical link. Modern communications networks typically have unintetruptable power supplies (UPS) which protect the communications network against over and under voltage conditions. Systems designed for maintainability include monitoring and reporting of battery conditions and the over/under voltage condition of prime power. Where a battery backup is used and battery charge is depleted, well designed systems support graced! shutdown and prewashing via network management. Loopback test capability is very beneficial in identifying physical link problems and also in assessing marginal conditions in signal modulation and demodulation. Bit pattern tests incorporating loopback can help identify a marginal communications condition in much He same manner as computer memory tests. Loopback tests are also beneficial in isolating multiplexer problems. A.4.~.1 Network Management Network management includes He hardware 'hooks" and associated software to facilitate management of a communications network Perhaps one of He most important aspects of modem commum cations system design is attention to including network management within the architecture. Network management generally involves: Infrastructure Management: . Representation and control of communications infrastructure resources; Network Element Management: ~ Individual network element management at the level of hardware and software support; Resource Management: L:\NCHRPPba~\ NCHRP 3-51 Phase 2 Fmal Report A4-3

OCR for page 437
Management of physical and logical elements of Me network including topology and configuration related to end-to-end connectivity; Service Management: . Management of services supported by He network; Maintenance Management: . Support for performance mon~tonng, fault detection, fault isolation, fault reporting and control for failure recovery; and Security Management: . Control and monitoring of network secunty. Network management technology is in a standards evolution phase. Open Systems Institute (OSI), International Standards Organization DSO), and He Intemational Telecommun~cabons Union DID) are supporting standards. ISO/lTU Recommendation X.701; "Open Systems Interconnection System Management Overview," and ITU Recommendation M.3000; "Overview of Telecornmun~cadons Management Network (IMN) Standards." From these standards efforts, Common Management Formation Services (CMIS) and Common Management Formation Protocol (CMIP) are emerging. CMIP supports CMIS by transporting management commands and information from one system (or subsystem) to another. Simple Network Management Protocol (SNMP) seems to be He current "default" standard in the communications network industry. Openv~ew (Hewlett Packard, ~c.) Netv~ew (DIM) and Sunnet (Sunsoft, Inc.) support SNMP. None of the SNMP-supported commercial management software packages provide a complete system management capability related to maintenance. fact, evaluations of He Free commercial SNMP software packages rated them: ~;\NC~Phase2~p`\ NCHRP3-51 Phase2F'nalReport A44

OCR for page 437
Alarm Monitoring: Alarm Filter~ng/Automation: C Alarm Correlation: Ease of Problem Identification: B Network Topology Viewing: B+ Ability to Identify a Failing Component: A B C SNMP is being improved and wiB most likely converge vv~ ~e CMIP standard. What is important in die system design process is to assure that ad components of He communications network support a common network management protocol (whether SUMP or the final approved standard). Without a common network management protocol supported by multiplexers, snatches, bridges, routers, etc., complete system-we maintainability Trough network management is not achievable. The information network management win provide, when properly implemented, is '~hat has failed," "where it has failed," and "impact on system performance." This can be extended Trough applications software to include inventory control of spares, replaceable modules, plus scheduling and mon~tonng of repair activity. A4.~.2 Factory Backup For large communications network subsystems such as SONET node equipment, ATM switches, bndge-routers, and over tokens equipment is available with a remote diagnostics port. This port is usually an RS-232 compatible port supported by a Public Telephone Network (PTN) compatible modem. The communications port can be used by Be manufacturer's maintenance support services to provide "expert" assistance to a user's maintenance organization in diagnosing a problem. If equipment win remote diagnostics ports is selected for a system, this feature provides a second tier maintenance backup to a jurisdictional maintenance organization. Where many subsystems include factory diagnostics communications ports, these may be incorporated into a common communications network,~w~ a single point interface to Be PrN at the Traffic Operations Center (TOC). This reduces the service cost of dial-up circuit access at L;WCHRP - warp NCHRP3-51 Phase2FmalReport A4-5

OCR for page 437
field locations. Typically an overlay RS-232 maintenance circuit would be integrated win the communications network design. This provides a means of communicating with the failed network subsystem. Where fault tolerant communications devices are used providing very high network availability, a multiplexed channel may be dedicated to remote diagnostics support. The factory maintenance port is usually a higher level maintenance port compared win the multifunctional network management link which supports failure reporting and maintenance activities. The factory maintenance port is designed for wide area communications interface supporting a communications link between equipment and factory maintenance. The port is further accessible for local, plugger test equnpment and may also be used by field maintenance. A.4.~.3 Contract Services for Network Management and Equipment Maintenance Support Larger networks with a network management protocol and remote diagnostics capability can be designed wad contract maintenance. To be responsive, He Maintenance Organization must be within a suitable maintenance response distance from Me communications network. This typically means Mat Be Maintenance Organization is within Me jurisdictional area The type of contract service is a function of: Use of fault tolerant equipment in Me field; Any maintenance responsibility desired to be maintained by He jurisdiction such as: Traffic controller maintenance, and Modem maintenance within traffic controllers; Degree of intelligence in the field and ability of He distributed intelligence to safely support traffic control in case of a communications failure; General jurisdicdonal traffic conditions; and Responsibility for maintenance service to He electronics and cable infrastructure. ~;\NCHRP`Phase~pr\ NCHRP 3-51 Phase 2 Final Report A4-6

OCR for page 437
Where fault tolerance and intelligent distnbuted control are used in a system design, the need for 2=hour per day, 7 days per week monitoring and guaranteed maintenance corrective action response gene~y is not consider. Monitoring and maintenance are conducted during stan~d work hours at a cost reduction which can be as much as 50%. Where failure response time guarantees are included in the maintenance contract, We response dine period directly relates to service cost. Typically a 2, 4, 6, or 8-hour response time selection is available with 2-hour response tune costing 30% more Wan S-hour response time. Again, with fault tolerance in the field, response time is not critical. With fault tolerant systems We maintenance contract usually guarantees corrective action within the next ~ hour work day following a failure. Where a cable infrastructure maintenance contract is in use, typically co~rec~ve action guarantees and even cost of maintenance action are not included in the contract. Cost of monitoring, detecting, and isolating a cable break is covered in the connect. The reason for this is that the maintenance contractor does not know the degree of difficulty associated with accessing the break, nor the length of time that it will take to access the cable. A cable break under a Dive or street requires: Coordination and approval of the junsdiction; Traffic management and safety considerations; Concrete (or other surface) penetration; Conduit location, access penetration; and Break repay, including restoration of conduit and cable integrity ninth environment (such as installing a splice closure). Thus, cable infrastructure repair is typically accomplished on a billable time and materials contract basis. With automated built-~n test capability and network management, contract maintenance may not be justified based on network complexity. With automated fault detection, isolation, and repordug through a properly designed communications network, the jurisdictional maintenance ~:`NCHR~\ N~3-51 IF A4-7

OCR for page 437
technician is told what is broken, what module(s) to change, and where. Thus, the need for contract maintenance service is perhaps not justified. For networks Mat do not stress fault detection, isolation, and fault propagation prevention and reporting, network ~oubleshoodug becomes a complex task requ~nng detailed knowledge of network hardware and software, necessitating consideration of contract maintenance support, even with experts providing the maintenance service, due to We level of maintenance complexity. Contract maintenance may optionally relieve the jurisdiction of the requirement to purchase spares. Typically, the contract maintenance organization provides the spares and is responsible for failed module repair at the component level. While this can save a jurisdiction perhaps 15% of the cost of con~mun~cadons electronics Hardware only, not including installation and test and not including cable infrastructure) the disadvantages are: Funding is usually available to procure required operational spares as part of the initial system procurement, but maintenance funds are usually more difficult to obtain for He purpose of purchasing spares after the system has been in operation. If the maintenance contact becomes unsatisfactory Cause of poor maintenance management, spares would have to be procured then, if not initially procured; Generally, not all of He equipment in He system is provided by He provider of maintenance services; therefore, He quality of maintenance service may be degraded; and . Generally, it is beneficial for He jurisdiction to have a trained maintenance staff which can support technical planning for modification and expansion as well as being responsible for maintaining He system to meet required availability objectives of the junsdiction. In summary, contract maintenance is available to jurisdictions, wig a savings in personnel, test equipment, and spares cost. By electing contract maintenance, a jurisdiction becomes vulnerable to increased maintenance service cost with few alternatives because He maintenance organizadon owns spares, test equipment, and maintenance knowledge. The better solution is to use He hardware manufacturers as a backup to jurisdictional maintenance and to repair modules at He component level, with He jurisdiction purchasing operational spares and integrating system-level test equipment Into He operational system. Thus, He jur~sdicHon is in control of its L::U OCR for page 437
operations, system availability objectives, and resources for technical planning and growth support. A.4.~.4 Impact of Fault Tolerance on Maintainability Fault tolerance directly impacts system reliability and also impacts maintenance. Through use of the fault tolerant designs available in advanced communications products, a failure can occur without ~ntern~pting communications. The network management software reports We occurrence of We failure and the fact that We backup module has been committed to operation or that a counter-rotat~ng optical nag is operating in a failure infonnation routing mode. Maintenance can be scheduled, bred on optimized use of maintenance resources. Thus, crisis maintenance is eliminated. Where large distributed systems are involved, use of fault tolerance can be justified based on maintenance cost savings. Group more efficient use of maintenance time, the system can be mainlined with fewer repair personnel and fewer trips to the field. Un~nterruptable Power Supplies (UPS) also reduce maintenance costs and can be justified. Powering on and off electronics induces the highest stress on semiconductors, which results in failures. By eliminating power cycling during loss of prime power, failures of electronic equipment are reduced. Maintenance time to restart He system and synchronize databases after a power failure is also eliminated. A.4.~.5 Reliability of Advanced Communications Technology as Related to Maintainability Older communications equipment used discrete components, each having a discrete failure rate. Advance communications equipment typically uses highly integrated circuitry which supports reduced component count Similarly, advanced communications equipment typically uses low power semiconductor components with low heat dissipation, unlike older technology. Lower power and heat dissipation supports higher reliability since semiconductor failure rate increases with junction temperature. L:`NCE~\ NCHRP3-51 IF A4-9

OCR for page 437
New integrated circuits are being designed wig test interfaces that integrate with board level test buses. Thus, maintainability at He system, board, and chip levels is enhanced. A Bird trend, use of Digital Signal Processors (DSP) to perform communications functions, is also improving advanced communications reliability and further supports maintainability. The results are fewer components, intelligence to support self diagnostics, and generally more communications functionality. Use of DSPs, Programmable Logic Arrays (PLAs), Applications Specific Integrated Circuits (ASIC) and other large-scale integration techniques in modern communications: Reduces component count; Reduces mechanical connections; and . Reduces problems protecting circuitry against noise (coupled and caused by ground planes). Thus, reliability of Me communications device is increased. This, combined w~ff1 the ability to Impose built-in testing down to the component level, the need for maintenance operations for advanced communications devices is decreased, and the ease of conducting a maintenance operation when needed is increased. A.4.~.6 Built-in Test Equipment As opposed to built-in tests, which are test features incorporated into the design of operational communications electronic equipments, test equipment may be integrated into the system. Test equipment supplements those devices which do not have built-~n tests or are incapable of ~nco~poradog built-in test features. A fiber cable, for instance, requires special test equipment to determine We location of a broken connection. Prudent system design integrates test equipment wig the operational system using test data buses, switches, and patch panel capability. Integrating test equipment wig the operational system has He following benefits: . Easy access and use for testing; L.\NC~RP\~.spt\ NCHRP 3-51 Phase 2 Fmal Report A4-10

OCR for page 437
Availability when needed (not in the back of a maintenance van out in the field); Highly complimentary to bu~It-in test features of Be system; and it Can be budgeted and justified as part of the operational system. Test equipment supporting advanced communications technology is being designed to be: Compatible wad recognized test bus standards; Able to be integrated into a common system test capability; Able to be controlled and test results read by system bu~It-in test software supporting central network management capability; and Generally modularly expandable and tailorable Trough software (example is protocol analyzer). Thus, it is prudent to incorporate test equipment in Me system design as needed to fib the test gaps. A.4.~.7 Jurisdictional Maintenance Skills and Compatibility with Advanced Communications Technology There is no question Hat advanced communications technology is more complex Han conventional technology; however, complexity and understandability are not Incompatible In advanced systems. There are many reasons for this including: High degree of functional integration places complexity below the replaceable module level, making it only necessary to understand the higher level function at Be module level; Use of built-in test (Brie), fault isolation, and fault reporting. Requirements for detailed analysis of bit patterns, voltages, and waveforms are minimized because the '~IT' report defines He problem and its location; Advanced communications use self calibration m~nuniz~ng the need to manually adjust circuit parameters. The trend is approaching total automatic adjustment, even for radio frequency commun~cabons equipment; ~WCHR~se2~t\ NCHRP3-51 Phase2FmalReport Af11

OCR for page 437
Automated module test sets and electronic modules with test buses support automated component failure isolation. Maintenance sums needed relate to how to use the removable module test set and how to change large-scale integration components. ~ addition, many manufacturers offer board-level repair service as an option; Use of bus standards at module and system levels means basic standards are He same for signals and control protocols between equipment (OST layers ~ and 2~. . Use of the same standard link and network protocols and network management standards between network elements; and Emphasis on ergonomics In systems operation and maintenance simplifies understanding of system '~ealth.~, With advanced communications, when systems are properly designed, troubleshooting is automated and maintenance is generally accomplished at a higher level of understanding (more low-level functions are bundled together to provide a higher level function). Standards are used for interfaces and ergonomics are considers in presentation of bu~It-in test results to maintenance personnel. Emphasis on braining is at Be subsystem and systems level wig less emphasis on module-level maintenance. Because of their fi~ncdonal understanding and basic communication knowledge, jurisdictional maintenance personnel are generally capable of making Be transition from conventional to advanced communications support Generally, maintenance personnel Avid support Be opportunity to move forward wig technology, as long as they are provided the 'fools" to accomplish progress. This means that Be system is supplied ninth modern BIT features and integrated test equipment to support maintenance operations and that cen~ized communications network management and monitoring are included In the system. Without attention to automated maintenance assistance in new systems, jurisdictional maintenance support for advanced technology implementation we most likely be limited. The task of troubleshooting and maintaining a complex system, without BIT and control maintenance monitoring features, would be difficult even wig trained maintenance personnel. L;`NCHRPPhase~p ~NCHRP 3-51 Phase 2 Fm~1 Report A~12

OCR for page 437
The transition from copper infrastructure to wireless infrastructure has proved easily accommodatable by jurisdictional maintenance personnel. Communications principles apply, such as signal attenuation loss versus distance, and ban~vv~ as a function of distance. Concepts of frequency multiplexing apply in He sense of wave division multiplexing. Cables are color coded in Be standard manner. The difference is in configuration, splicing, and test equipment to measure parameters. With automated splicing tools which splice and verify the connection bow mechanically and opthcally, splicing of optical cable is simplified. For a cable comn~un~cabons maintenance technician to become a wireless technician requires training; however, Be transition of a jurisdictional mobile radio maintenance technician to support a modern digital wireless communications network is achievable without major difficulty. The reason is Mat Be basic theory of wireless communications is already understood by Be maintenance technician. Thus, the functions of the advanced wireless digital transceiver and associated BIT are more easily understood as an extension of current knowledge. In summary, wad a basic knowledge of communications technology, advanced commun~cabons applying communications principles, built-in test features, and central network monitoring, a jurisdictional maintenance technician is capable of transition~ng from conventional to advanced communications network maintenance. A.4.~.S Software Maintainability In communications equipment employing software, software ma~ntainabilit~,r is a concern. Older communications equipment did not employ structured, modular software developed under recognized procedures for quality software products. The Institute of Electrical and Electronic Engineers (' ~ ~E) has prepamcl standards for software development, documentation, and testing. LIFE Standard 1219-1992, entitled "Standard for Software Maintenance" applies. EKE document 89-1983, entitled "Measures to Produce Reliable Software" and WEE Standard 10008-1987, entitled "Standard for Software Unit Testing" are important in achieving software maintainability. Attention to software documentation as defined in ISLE Standard 1063-1987 (Software User Documentation) and 101~1993 (Guide to Software Design Description) furler supports software maintainability. L:;wC~rpt\ NCHRP 3-51 Me 2 Flu Rein A=13