Заглавная страница Избранные статьи Случайная статья Познавательные статьи Новые добавления Обратная связь FAQ Написать работу КАТЕГОРИИ: АрхеологияБиология Генетика География Информатика История Логика Маркетинг Математика Менеджмент Механика Педагогика Религия Социология Технологии Физика Философия Финансы Химия Экология ТОП 10 на сайте Приготовление дезинфицирующих растворов различной концентрацииТехника нижней прямой подачи мяча. Франко-прусская война (причины и последствия) Организация работы процедурного кабинета Смысловое и механическое запоминание, их место и роль в усвоении знаний Коммуникативные барьеры и пути их преодоления Обработка изделий медицинского назначения многократного применения Образцы текста публицистического стиля Четыре типа изменения баланса Задачи с ответами для Всероссийской олимпиады по праву Мы поможем в написании ваших работ! ЗНАЕТЕ ЛИ ВЫ?
Влияние общества на человека
Приготовление дезинфицирующих растворов различной концентрации Практические работы по географии для 6 класса Организация работы процедурного кабинета Изменения в неживой природе осенью Уборка процедурного кабинета Сольфеджио. Все правила по сольфеджио Балочные системы. Определение реакций опор и моментов защемления |
NNI Optical Interface standardisationСодержание книги Поиск на нашем сайте
Although recommendations G.707/8/9have done an excellent job in standardising the logical signal presented at an NNI, this aloneis not enough to ensure a Mid-FibreMeet between equipment from two different manufacturers. G.957recognises this problem and specifies the relevant physical parameters for arange of standard optical interfaces, at the preferred SDH rates of STM-1, STM-4 and STM-16. In addition, there is also an extensionto the existing G.703electrical interlace specification to accommodate an electrical presentation of an STM-1 signal Despite this, there is still no guarantee of anything beyond traffic path interworking, as the management channels that flow across the mid-fibreboundary maystill becarrying information which is unintelligible to the network managers onthe other side. This is an important area that the standards bodies are still addressing SDH network elements Because traffictouting flexibility is one of the reasons forthe existence of SDH, the simplest wayof considering any SDH network element is as a group of transporttermination functions (TTFs)surrounding a trafficpath connectivity function (see figure 42.21) The TTFs are essentially the conversion functions from the VC level to either the STM or plesiochronoussignal levels, while the connectivity function allows VCs to berouted between the various TTFs.More than anything else, it is the size and flexibility ofthe connectivity function that distinguishes one type of network element from another. For example, a crossconnect may have a large connectivity function, capable of simultaneously routing a large number of VCs of a given type between a large number of TTFs, with very few restrictions on this connectivity. This, of course, is consistent with the normal role of a crossconnect, as a major traffic routing node in a transport network, where electronically control-table traffic routing flexibility is its raison d'etre. At the other end ofthe scale come Line System Terminals (LTEs), which normally incorporate a multiplexer as well. The connectivity between tributaries and aggregate ports of such a multiplexer is normally hard wired, hence no flexibility whatsoever. Between these two extremes come drop and insert (Add/Drop) multiplexers, which attempt to strike a balance, that leads to adequate, rather than comprehensive routing flexibility, with an attendant reduction in equipment costs. All ofthe above mentioned types ofnetwork element normally assume that any interconnection will be based on optical fibres. It is, however, also possible to use other interconnection media, in particular radio transmission. The problem in using radio interconnectionis that, unlike optical fibres, the available spectrum is finite, and in short supply. The liberal use of overhead capacity in SDH only exacerbates the already difficult problem of squeezing the trafficinformation into the existing arrangement of radio channel bandwidths. These problems are by no means insuperable, but they do tend to restrict radio interconnection of NNI signals to the lower end ofthe SDH range e.g. up to STM-2 at present. This inability to match the transmission capacity ofoptical fibres will progressively force radio based systems out of the core areas of transmission networks, and into the more peripheral areas. In particular, it is likely that radio based SDH equipment will be heavily used for access duties The reduction in cost that accompanies a network element with restricted connectivity results not only from simpler hardware, but also from simpler control software. This is somewhat surprising at firstsight, as it has often been said that SDH does not really make economic sense without a large measure of software control over all the network element functionality and, in particular, that of routing flexibility.However, the complexity of even relatively simple SDH network elements leads to a situation where control of complete networksrapidly escalates unless some restrictions are put on the traffic routing complexity. In short, every additional piece of traffic routing flexibility is a potential network control headache. The general problem of control of a whole SDH network, notwithstanding the traffic routing problem, is almost impossible if it is composed of network elements from different manufacturers, which do not have some form of common control interface. Before we can arrive at such a universal interface, we must first standardise, to a large extent, on the functions which each of the network elements actually perform. This is where the network element functional standards, CCITT G.781/2/3, come in. These are the recommendations which describe how the functionality of a network element can be decomposed into the basic 'atomic' elements of functionality, together with how such atomic functions may be combined and thereafter exchange both traffic and control information. (See Figure 42.22.) The rules for combination of these atomic functions enable a variety of network elements to he synthesised, with differing traffic capacities and routing capabilities. Indeed, one of the current problems is the potentially large number of different ways a conformant network element can be constructed. It is rather difficult for software based control systems to cope with this variety and hence there are efforts in several standards fora (notably ETSI), to standardise on a more restricted range of network element functionalities. G.781/2/3 were created primarily to describe different types of drop and insert multiplexers, however, the general principles and indeed, much of the detail, are just as applicable to LTEs and large cross connects. Moreover, the formalism required to describe complete SDH networks was produced very much with this same functional decomposition in mind. We shall see, later, how these atomic functions may be paired between network elements on opposite sides of a network to produce a complete traffic path, that is sufficiently well defined for a computer to recognise.
Figure 42.21 Generalised representation of the traffic paths through an SDH network element
Figure 42.22 Fragment of an SDH network element showing how it is broken down into atomic functions acording to CCITT G.783
Figure 42.23 Network management layering for the control of telecoms networks and services
1. Learn the following technical words and word-combinations:
42.4.6
42.4.9
Exercise 2 Read the text 42.4.6- 42.4.9.
Exercise 3 Find the Russian equivalents for the following English technical word-combinations:
Exercise 4 Find the English equivalents for the following Russian technical word-combinations:
Exercise 5 Answer the following questions: 1.How have mappings into Synchronous Containers been defined so far? 2.What do CCITT G.708 and G.709 contain? 3.What signal is the only byte synchronous mapping defined for? 4.What is the advantage of the byte synchronous mapping? When does this mapping become apparent? 5.What were the SONET standards designed for? What new standard appeared later? 6.What other terminology exists within SONET? 7.What does the recommendation G.957 specify? 8.Why is there no guarantee of anything beyond traffic path interworking? 9.What is the simplest way of considering any SDH network element? 10.Is the size and flexibility of the connectivity function of great importance for distinguishing one type of network element from another? 11.Why is it likely that radio based SDH equipment will be used for access duties? Why? 12.Under what conditions is it impossible to control a whole SDH network?
Exercise 6 Make a short report on SDH network element (part 42.4.9).
Exercise 7 a) Translate into Russian in writing part 42.4.9 paragraph 1. b) Translate into Russian in writing part 42.4.9 paragraphs 2 (up to “The reduction in cost…”).
Part 5 (42.5- 42.6) 5. Control and management 1. SDH based networks
Control and management Control and management within any type of telecommunications network (not just an SDH one) can best be viewed in terms of a series of layers (See Figure 42.23). At the lowest level, there is internal control of an individual network element, which performs internal housekeeping functions and deals in alarm and control primitives. CCITT G.783 has rigorously defined a minimum set of control and alarm primitives for each of the SDH atomic functions, and these are the basis of all SDH management. However, beyond this CCITT G.784 describes how such primitive information should be processed and ordered to produce derived information such as error rates etc, which is stored in logs of defined duration, and reported at set intervals, etc. At the lowest level, some of this information may look rather different to that specified in G 783/4 etc, however, the internal network element control system takes these and from them, synthesises the information that is required by the next level in the management hierarchy i.e. the element manager. The element manager is a piece of software which can control many individual network elements (usually in the range 10-1000), but it can only control them as individual elements and does not have any view of the traffic relationships between them. Usually it is located remote from the elements it is controlling and more often than not, it runs on some form of workstation. The interface between the element manager and the network elements is an obvious area for standardisation, as without this there is little chance of a single element manager controlling network elements from more than one manufacturer. Despite the progress that has been made in rigorously describing the network element atomic functions, it has still proved very difficult to agree on a software description of them. The approach adopted in CCITT, ANSI, and ETSI has been to describe complete network elements as collections of 'objects' in line with the rules of 'object orientated' programming. The idea is that an object is a software entity that has both attributes and behaviour. External stimulate (i.e. information, commands, etc.) trigger an object to behave in a certain way e.g. change its attributes, transmit information, commands etc. The claim is that a standardised object could be looked upon as the software equivalent of a hardware integrated circuit. One of the biggest areas of disagreement in generating an agreed standard set of SDH objects, is the questions of whether an object should represent a pieceof functionality or whether an object should represent a (small) piece of hardware. Although CCITT G.783 functionally decomposes SDH equipment, its says almost nothing on the way such atomic functions are split or combined in any real hardware implementation. The current view within both CCITT and ETSI is that the set of objects (collectively known as the 'Information Model') should present both a functional and a physical view of the network element that they represent. ETSI, in particular, is making good progress in generating a model along these lines. Not only is it necessary to have a standardised information model, so that the network element and element manager can understand line another, it is also necessary to have an agreed message set to go with it. Fortunately, this flows reasonably easily from the definition of the objects themselves. However, the existence of such a message set, then leads to the requirement for an agreed information protocol stack with which to transport it. For information transferred between a network element direct to its element manager, CCITT G.773 details several allowed protocol stacks, which split into two groups, those which have a full 7 layer structure and those which have a 'short stack', where layers 4, 5, and 6 are absent. Most observers favour the use of the heavier, but more flexible, seven layer protocol stacks for SDH networks, while the 'short' stacks are more appropriate for PDH equipment for which it is more difficult to justify the burden of the additional layers. Not only are there defined protocol stacks for information transfer direct from element manager to network element, but there is also a defined protocol stack for information transfer between individual network elements. In general, the majority of information transfer between network elements is actually information arising from an element manager, which is being relayed by an intermediate network element to a more remote element. (See Figure 42.24). This flow of management information amongst network elements, element managers and yet more managers at the network and service levels, gives rise to the concept of a Telecommunicatioas Management Network (TMN). The main recommendation concerning the TMN is CCITT M.30, which attempts to define a series of interfaces between different management entities in such a network. It is not confined purely to SDH networks, but SDH networks will probably be the first to implement an M.30 style TMN. (See Chapter 16.)
Figure 42.24 Control channels and protocols within a SDH network
SDH based networks An SDH based network can be viewed as the transmission bedrock which supports all other terrestrial telecommunications services. (See Figure 42.25.) As already mentioned, the main advantages of such a network are the ease and precision with which the available network bandwidth can be partitioned amongst the higher layer services, together with accurate monitoring of the quality of the transmission links. Despite this, the control and management of complete networks of the size operated by BT, France Telecom, or DBP is a difficult problem, which requires some degree of standardisation in the way such networks are functionally decomposed, in much the same way as the individual network elements have already been functionally decomposed into their atomic elements by CCITT G.783. The basic idea behind the functional decomposition described in CCITT G.snal is that a transport network can be stratified into a number of layers. Each layer provides a service to the layer above it, and is, in turn, a client of the layer below it, in much the same way as the ISO seven layer information transfer model consists of layers which participate in client-server relationships with their vertical nearest neighbours. (See Figure 42.26.) This layering within an SDH network could be viewed as a subdivision of ISO layer 1. As an example of a non-SDH client-server relationship in a trans-port network, consider Figure 42.27, which shows a 64kbit/s circuit being a client of the 2Mbit/service layer i.e. the 2Mbit/s layer, which can be viewed as a 2Mbit/s network, transports the 64kbit/s circuit between its desired end points. In order to do this, the 2Mbit/s layer will probably call upon the services of the 8Mbit/s layer, and so on up to the 140Mbit/s layer. The SDH counterpart of this simple PDH example is slightly more complicated in that the transport layers are divided between those concerned with end to end networking (i.e. the Path Layers) and those concerned with transport between each pair of SDH network elements along the route (i.e. the SIM section layer). As an additional complication, there are two path layers, the lower order paths consisting of VC1 s, VC2s or VC3s, and the higher order paths, which are VC4s in CEPT countries, hut could also be VC3s. (See Figure 42.28.)
Figure 42.25 Managed transmission network; SDH network as a bearer for other services
Figure 42.26 Layering and client server relationships between the layers of an SDH network
Figure 42.27 64kbit/s circuit making use of a 2Mbit/s service in order to get from A to B
Usually when a client layer makes use of a server, it is necessary to adapt the client signal to a form suitable for transport by the server layer. (See Figure 42.29.) Examples of this 'adaptation' function are the plesiochronous 'stuffing' which occurs when a 2Mbit/s channel is multiplexed into an 8Mbit/s one, or the progressive change in a VC12 pointer value to accommodate a small frequency mismatch with the VC4 into which it is being loaded. The adaptation function is only one of the network atomic functions that have been described in G.snal, and it is indeed fortunate that G.snal has been developed in full recognition of the contents of G.782/783 because many of the equipment and network atomic functions are identical. G.snal, not only describes a series of vertical client-server relationships for an SDH transport network, it also gives a structure to each of the individual layers. Each layer can he viewed as a network in its own right, which can he partitioned into a series of subnetworks. (See Figure 42.30.) These subnetworks are connected together by "link connections', and can, if necessary, be further subdivided into yet smaller subnetworks. The logical place to stop this process is where a whole subnetwork is completely contained within one network element. As an example consider a VC12 cross connect. The switching matrix (connectivity function) within this element could be considered as a subnetwork, which is capable of making VC12 layer subnetwork connections from one port to another. (See Figure 42.31.) To connect one of these VC12s to another cross connect, a link connection is now required, which will need to make use of the VC4 and STM layers. This concatenation of subnetwork and link connections across the VC12 path layer network begins and ends at the points where the VC12 is created and destroyed, namely the path termination points. It is at these points that the VCI2 POH is added or removed, exposing only the C12 Synchronous Container. This same path termination function is recognised in both G.783 and G.snal and it is located immediately next to the adaptation function described above. On the assumption that the C12 in question is carrying a 2Mbit/s circuit, then the combination of adaptation functions, termination functions and the chain of subnetwork and link connections in the VC12 layer have succeeded in transporting this 2Mbit/s circuit between the two end points of a single 2Mbit/s link connection. (See Figure 42.31.) From this point, the above analysis can now be repeated in the 2Mbit/s layer, where the particular 2Mbit/s circuit cited above will probably he found to be serving as part of a link connection for a number of 64kbit/s circuit (see Figure 42.27). This formal description of an SDH network opens up the prospect of real control and management of large networks consisting of network elements from several manufacturers. The pairing of path termination functions on opposite sides of a network, together with a similar operation on the same functions in the STM layer, allows a PTO to accurately monitor the service which the SDH network is delivering, and to pin-point those network elements responsible for any poor performance. Beyond that, the SDH network model greatly facilitates management of the all important traffic flexibility points i.e. those subnetworks which consist of an electronically controllable connectivity function.
Figure 42.28 Layering in an SDH transport network
Figure 42.29 Handling a link connection in a client layer by a path in a server layer
Figure 42.30 Partitioning of a transport layer network into a series of subnetworks 1. Learn the following technical words and word-combinations:
42.4.5
42.4.6
Exercise 2 Read the text 42.5- 42.6.
Exercise 3 Find the Russian equivalents for the following English technical word- combinations:
Exercise 4 Find the English equivalents for the following Russian technical word- combinations:
Exercise 5 Make up 4-5 questions on part 42.5 “Control and Management” and be able to ask your group-mates to answer them.
Exercise 6 Answer the following questions on part 42.6: 1. What are the main advantages of SDH based network? 2. What is the basic idea behind the functional decomposition described in CCITT G.sna 1? 3.How could layering within an SDH be viewed? 4.When does the plesiochronous “stuffing” occur? 5. What is the role of G.sna 1? How can each layer be viewed?
Exercise 7 a) Translate into Russian in writing part 42.5 paragraph 1. b) Translate into Russian in writing part 42.5 paragraphs 2 (up to “The idea is that…”)
Exercise 8 Make a short report on SDH based networks (part 42.6) Part 6 (42.6.1-42.6.2) 2. SDH network topologies 1. Deployment strategies SDH network topologies Once again traffic routing flexibility, its control and physical distribution within an SDH network, is one of the most important influences on the topologies proposed for the deployment of SDH equipment. The most obvious manifestations of this are the drop and insert ring topologies that are finding favour in the former junction areas of PTO networks. (See Figure 42.32.) The idea behind a drop and insert ring is that the ring structure can give a high degree of protection against cable cuts etc. due to its potential for routing traffic either way round the ring. In fact, one of the biggest advantages, in control terms, of this topology is the limitation on the re-routeing possibilities for any traffic affected by a cable break. Anything more complicated than a simple clockwise/counter-clockwise routing decision requires up to date knowledge of a rather more extensive portion of an SDH network than just a simple ring. The nodes of such a ring are populated by drop and insert multiplexers which have a restricted traffic routing flexibility that is tailored to the requirements of a ring. This gives a relatively low cost ring implementation, which nevertheless, when viewed as a single entity, appears as a restricted form of a cross connect. The versatility of drop and insert rings also extends to drop and insert chains, which can be considered as a 'flattened' version of a conventional ring. (See Figure 42.33.) Beyond this, a ring can also be made from a chain of drop and insert multiplexers, whose ends have been joined by an SDH line system. The main problem with implementing this type of ring, at least in the junction area, is that the existing layout of cables and ducts usually takes the form of a star, which formerly linked the old analogue local exchanges to their district switching centre. In many cases, a partial physical ring can be created by a small extension of the existing cable pattern. This can be supplemented by creating a logical ring when existing cables are laid in a star arrangement, although this obviously affords less protection against cable breaks (See Figure 42.34.) Finally, it is sometimes possible to produce a physical ring by linking some of the ring nodes with microwave radio rather than optical fibre.
Figure 42.31 Use of two VC12 cross connects to produce a path within the VC12 layer
Figure 42.32 SDH multiplexers deployed in a drop and insert ring
Figure 42.33 Drop and insert chain produced by flattening' a conventional drop and insert ring
Outside of the former junction areas, i.e. within the transmission core of the average PTO network, the normal topology advocated is that of a mesh of cross connects, interconnected by point to point transmission systems. (See Figure 42.35.) For this application, the SDH transmission systems, like their PDH counterparts, require the cross connects, which provide complete traffic routing flexibility at the VC4 and VC12 (and possibly other VCs as well), although not usually in the same network element. Used in this way, cross connects can be viewed as electronic replacements for the present day digital distribution frames. The benefits of this deployment of flexibility are alleged to be those of easier traffic path provisioning and a fast, efficient scheme for restoring failed paths which requires the absolute minimum of standby transmission capacity. This last benefit is heavily dependent on the control that is exercised over the cross connects, and here there are some potential problems. The main problem is that of database integrity. With a network of meshed cross connects, when a transmission link fails, the path restoration action will usually involve re-routeing all the affected paths through several alternative cross connects. In order to do this efficiently the network control system must rapidly command simultaneous switching actions in all of these crossconnects, which usually implies that the control system has a pre-determined plan of action which is based on the spare bandwidth that is thought to be available on the relevant transmission links and cross connects. Unfortunately the integrity of the control systems database may have been compromised because of other recent reconfiguration. This problem escalates rapidly as the number of cross connects in a network increases. There are several potential solutions to this problem. One is to simply not use the cross connects for protection against transmission failures, and, instead rely on transmission systems having 1 + 1 or 1:N protection. In this case the cross connects are used solely for off-line management of the network's transmission capacity. This is often call “Facilities Management” in North America. An alternative is to deploy the cross connects in a rather more bounded topology than the completely free mesh topology assumed above. A limiting case of this idea is to deploy them in a ring. The restriction on routing choices imposed by such a topology greatly cases the control problem, albeit with some significant increase in the standby line transmission capacity that must be available to allow a traffic re-route. Although the present day cost of such standby capacity is considerable, it has already been observed that raw point to point transmission capacity is the one thing that will continue to become cheaper, in real terms, for some time to come, hence this option may look increasingly attractive. Finally, the most interesting possibility is to endow a network of cross connects with a signalling system that will allow it to dynamically set up and clear down paths, in much the same way as a PSTN sets up conventional 64kbit/s circuits. The idea is that the intelligence required for this type of operation resides, not in a single large network manager, but, like the PSTN, distributed throughout the whole network. This greatly reduces the data base integrity problem, and should operate quickly enough to meet the criteria for fast restoration of failed paths, as well as facilitating other operations such as rapid provision of bandwidth (i.e. bandwidth “on demand”) in response to requests from the client services of the SDH network. Figure 42.35 Example of core transmission network topology that relies on cross connects for flexibility
Deployment strategies There are three main strategies for the deployment of SDH equipment: 1. Synchronous islands. 2. 'Thin' overlay networks. 3. Ad hoc deployment dictated by traffic growth, etc. Many PTOs are currently using the ad hoc approach and deploying the latest STM-16 (2.5 Gbit/s) transmission systems, not so much because they conform to the SDH standards, but because they offer higher transmission capacities than any available PDH systems. Even so, most PTOs which are doing this, are intending to eventually fill in the gaps between these ad hoc deployments so as to create an SDH overlay network. The idea behind the 'thin overlay' strategy is to rapidly deploy a limited SDH capability across the whole of a PTO network so that a small nucleus of key business customers can be offered the benefits of SDH as soon as possible. This type of deployment is most appropriate to networks where the important communities of interest are geographically widely dispersed. The synchronous 'island' deployment strategy assumes that the most important communities of interest are geographically concentrated, and that each one can start to benefit from SDH without necessarily having full SDH connectivity with similar communities of interest. The classic example here is that of the financial community in the City of London. With the passage of time, a synchronous island would normally increase in geographical size, so as to eventually coalesce with other islands. Long before this, they would probably be interconnected by an emerging SDH trunk network.
1. Learn the following technical words and word-combinations:
42.6.1
Exercise 2 Read the text 42.6.1- 42.6.2
Exercise 3 Find the Russian equivalents for the following English technical word- combinations:
Exercise 4 Find the English equivalents for the following Russian technical word-combinations:
Exercise 5 Answer the following questions: 1.What are the most important influences on the topologies proposed for the deployment of SDH equipment? 2.What is one of the biggest advantages of the ring topology? 3.Does the versatility of drop and insert rings extend to drop and insert chains? 4.What is the main problem with implementing the ring made from a chain of drop and insert multiplexers? 5.In what way is it sometimes possible to produce a physical ring? 6.How can cross connects providing complete traffic routing flexibility be viewed? 7.What happens to a network of meshed cross connects when a transmission link fails? 8.Are there several potential solutions to the problem of database integrity? What are they? 9.How many strategies are there for the deployment of SDH equipment? What are they? 10.Why are many PTOs currently using the ad hoc approach and deploying the latest STM-16 transmission systems?
Exercise 6 a) Translate into Russian in writing part 42.6.1 paragraphs 4,5(from “The main problem…” up to …”Finally… “. b) Translate into Russian in writing part 42.6.2.
Exercise 7 Make a short report on deployment strategies of SDH equipment.
Part 7 (42.7-42.9) 2. Impact of broadband standards 3. Frame relay 4. Switched Multimegabit Data Service (SMDS) 5. Fibre Distributed Data Interface (FDDI) 6. Future technologies 7. Integrated circuits 8. Optical interfaces 9. Optical amplifiers 10. Optical switching 11. Memory and processing power 12. Conclusion
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Последнее изменение этой страницы: 2016-12-12; просмотров: 171; Нарушение авторского права страницы; Мы поможем в написании вашей работы! infopedia.su Все материалы представленные на сайте исключительно с целью ознакомления читателями и не преследуют коммерческих целей или нарушение авторских прав. Обратная связь - 18.220.194.29 (0.013 с.) |