Tài liệu GSM and UMTS (P3) ppt

Chapter 3: The Detailed Specification Work Leading to the GSM Phase 1 Standard used for the Opening of Service (1987–1991) Thomas Haug 1 Mid-1987 was a milestone in the work of GSM. I think everybody in the group felt, despite what they might have felt earlier, that the agreement on the basic parameters in the air interface marked the turning point in the work. Another very important event took place in 1987 after years of preparations, i.e. the Memorandum of Understanding (MoU), signed by – initially – 14 operators who made a firm commitment to implement the system by 1991. This is described by Stephen Temple (Chapter 2, section 4), and I shall not go into it here. Suffice it to say that without a commitment on the part of the operators along those lines, there would hardly ever have been a GSM system in the real world, only on paper, simply because the manufacturers would never have dared to invest the huge sums in hard- and software work required to implement the system. The first set of specifications (or ‘‘Recommendations’’ as we used to call them at that time, keeping in line with the ITU and CEPT terminology) used for tendering is seen in the document GSM 31/88 in the acompanying CD-ROM. Another example of our intention to make the work better known was the Hagen seminar in October 1988 as seen in the documents GSM 187/87 and GSM 120/88. This activity is covered in Philippe Dupuis’s chapter and I need not go into details here. One change in the work rules of CEPT was very beneficial to GSM. Originally, CEPT had always worked as a very closed organisation, not open to outside bodies, and its results were not made public (although they often leaked out). Therefore, many CEPT standards were known to the manufacturers only if a PTT wanted to use a particular standard in connection with a contract for delivery of equipment, and usually they had very little influence in the preparation of standards. In December 1986, however, the UK representa- tive (Stephen Temple, DTI) proposed to CCH that in the future, each delegation to CEPT working groups could bring along up to two representatives of their national manufacturing industry as technical advisors. The argument presented was that the situation had changed 1 The views expressed in this section are those of the autor and do not necessarily reflect the views of his affiliation entity. GSM and UMTS: The Creation of Global Mobile Communication Edited by Friedhelm Hillebrand Copyright q 2001 John Wiley & Sons Ltd ISBNs: 0-470-84322-5 (Hardback); 0-470-845546 (Electronic) since the early days of CEPT, which was conceived in the days of the monopolies, and CEPT could no longer work in a secluded environment. I think it was natural for DTI to propose this since the UK was the country where the break-up of the telecom monopoly first became a reality. The change was implemented by GSM for the first time in June 1987, as it was felt that the crucial meeting in Funchal in February 1987 should be handled by the CEPT members alone. A number of activities were gradually moved from GSM to the MoU forum, above all issues of an operational and commercial nature, and also the IPR questions, which eventually became a very difficult area to deal with. The problems were not yet solved when the Phase 1 version of the system was opened. The technical work of GSM during this period was characterised by an ever increasing workload as more and more problems concerning details became apparent. It also became clear that the group had to rely on its own competence and if necessary its ability to enlist qualified assistance which could work under the leadership of the group, the subgroups, or the Permanent Nucleus (PN). The original idea of CEPT was that GSM should as far as possible ‘‘ subcontract’’ work to other, existing groups in CEPT, but the way in which those groups worked was not usually suitable for the time schedule of GSM. Therefore, GSM and MoU had to start a number of activities in order to deal with those questions. The setting up of ETSI in 1988 did not have an immediate effect on GSM, which remained a CEPT working group until March 1989. The work then proceeded by and large as before, but with the difference that many more delegates took part in the meetings since ETSI was open to all manufacturers. This was in some respects a great advantage, since many of the new members contributed a great deal to the work. The PN changed its name to Project Team 12, but stayed in Paris and went on with its work as before. The PN had a very important task to fulfil, and did it brilliantly. One of its most important tasks was to keep an eye on the various specifications which without exception went through a series of modifications, and it became a substantial burden for the staff at the PN. An efficient tool for keeping the files updated was the system of Rules for Change Requests, most of which came through the subgroups. In some meetings of GSM (SMG) there were more than 150 changes to the specifications, and all those which were approved, which means most of them, had to be worked into the specifications as soon as possible. The tools used today, when everybody has e-mail and a PC, were mostly unknown at that time, so everything had to be copied on paper and distributed in the meetings or shortly after. One change was that the responsibility for the circulation of the security algorithms was moved to the MoU forum. Although this concerned a subject which was more technical than commercial in nature, it was considered preferable to secure the confidentiality of the algo- rithms, since ETSI was to be a totally open body, and it was then felt that the subject belonged to the operators’ group. The specification of the SIM, however, clearly belonged in the ETSI sphere, since the manufacturers were members of that body. Of course, there was an immense amount of work on details to be done in the years that followed, as always in a large project. In addition, the technology of miniaturisation devel- oped quickly as we went along. Therefore, the emphasis on hand-held terminals, which is so natural today, made several changes to our specifications necessary, since we had originally based much of our work on car installed terminals, with quite different parameters in output power, current consumption, volume and weight. However, the foundation that was laid in the spring of 1987 was very firm, and despite many struggles, the goal of our work was never in GSM and UMTS: The Creation of Global Mobile Communication58 doubt. It would be tedious to go through all the details of the meetings – those who are interested can study the reports of the meetings, contained on the CD – but as we all hoped, more and more bits fell into place through the following years, and the aim of the MoU to open the system by 1 July 1991, was felt to be within reach, provided a liberal interpretation of what was meant by ‘‘ the system’’ was acceptable to the operators. In other words, it was seen that a system with a full set of features could not possibly be opened at that date, but a limited one with only the basic features could. That was what is generally termed ‘‘ Phase 1’’ . In order to do that, a freeze of the specifications had to be implemented, so that after 1990, changes should not be made to the specifications of Phase 1 unless it was absolutely neces- sary. However, since there was improvement and elaboration of the features going on all the time, it was necessary to start work on specifications for Phase 2. One very important objective was then to secure the compatibility between the two phases. To a large extent, this was achieved, but in some cases it proved impossible. One problem which was not wholly within the responsibility of the committee, was the issue of type approval. This question was obviously of very great importance for the implementation of the system, in particular because the Pan-European character of the system necessitated mutual recognition of type approval between all the participating. Consequently, type approval in one country would therefore be a final stamp of approval and it would be very difficult to withdraw the approval once it had been granted. To work out the requirements for type approval was the task of the committee, while the MoU followed the development of the test equipment itself. This split in the responsibility may have been one of the reasons for the delay which the manufacturers and the operators complained about, and for a long time, the terminals had to be sold with only an interim type approval. The work on the specifications is further described by Remi Thomas and David Barnes in Chapter 17. At the same time, there was a new task allocated to GSM. The UK expressed a need to have a Personal Communication Network (PCN) for the 1800 MHz range in place quickly. ETSI found after studying the question that since there already was a system specification available which was almost completed and could be considered stable, namely that of GSM, an update of that specification to the PCS frequency range would be the fastest way to meet the UK demand, and so, the 1800 MHz GSM was born. The change mainly involved changes in the radio parameters, channel numbering and so on, but basically, the new 1800 MHz system was a GSM system. The first Phase 1 network to be opened was an experimental one in Finland, an event which occurred on 1 July 1991, as stated in the MoU. In the last meeting of GSM (GSM#32, 30 September – 10 October 1991) a decision was taken to go into a new study area, i.e. the UMTS system, aiming at the next generation of mobile systems to be implemented after the year 2000, and a new Sub-Technical Commit- tee was created for that purpose. Stein Hansen of Norway was appointed chairman of that group 2 . With the decision of ETSI to allocate the work on the UMTS to GSM, it was also decided that the name of the group should be changed to SMG as of January 1992. After meeting No. 2 of SMG in Oostend in April 1992, the author of these lines went into retirement and Philippe Dupuis of France Telecom took over the chairmanship of SMG. Chapter 3: The Specification Work Leading to the GSM Phase 1 Standard 59 2 GSM#32 meeting report in the attached CD ROM folder A1, input document GSM 359/91 in folder A2. . October 1988 as seen in the documents GSM 187/87 and GSM 120/88. This activity is covered in Philippe Dupuis’s chapter and I need not go into details here. One. in this section are those of the autor and do not necessarily reflect the views of his affiliation entity. GSM and UMTS: The Creation of Global Mobile Communication Edited