文件名称：Understanding UMTS Radio Network Modelling, Planning and Automated Optimisation
文件大小：9.15MB
文件格式：PDF
更新时间：2012-04-19 20:13:34
umts
Yet another book on UMTS? Not quite! Our prime goal is to encourage the readership to understand why certain things happen in the UMTS Radio Access Network and others do not, which parameters are strongly coupled and which are not and what the analytical dependencies are between them. Thus, we try to minimise explaining system performance only on a case-by-case basis, which is the general case for many related books on the market, but rather equip the readership with fairly generic mathematical tools which allow complex system performances and dependencies to be understood, analysed and, above all, optimised. Also – ‘automated’ – a small additional word in the title of this book which makes the big difference: a difference to the scope of this book, a difference to the life of thousands of network optimisation engineers, a difference to everybody making use of wireless voice or data services in one way or another. While the 3rd generation (3G) UMTS standard may seem an ‘old hat’ to the euphoric academic research community, the number of people trying to understand, deploy and hone this very sophisticated wireless communication system increases on a daily basis. They can only begin to grasp that, unlike the 2nd generation (2G) GSM standard, UMTS is indeed very flexible across all communication layers in providing a whole raft of services. They soon come to realise, however, that this flexibility comes at the non-negligible price of increased complexity, a prolonged system learning curve and much higher risks in return for investment. Did you know that in a wrongly dimensioned UMTS network a faulty 3G terminal in London may influence a communication link in Edinburgh? Did you know that a 3dB planning error in pilot transmission power, which determines the size of each cell, may potentially cost an operator millions of pounds? Or, put it in other words, why the salary of a 3G-contract’s sales man in Edinburgh is dependent on the transmission power levels in London? If you did not know, this book will give you a clue as to why all the parameters in UMTS are so highly dynamic and interdependent. If you did know, you will appreciate that optimising such systems is both vital and inescapable. Optimisation has been known to civilisation from its very beginnings – the wheel being a prominent example which, by trial and error, fortunately emerged to be round. In contrast, given the vast number of its interdependent UMTS network parameters, optimisation by means of trial and error is clearly not an option. Only the early UMTS test trials and preliminary network rollouts were conducted manually, mainly using the experience of 2G network optimisation engineers. The currently deployed, operational UMTS networks have been partially optimised by means of software programs which yield satisfactory solutions for given input conditions. And here lies the trick! The input conditions may vary on an hourly basis, an example of which is the temporarily varying terminal density in central London that results from the rush hour. Given the highly dynamic nature of UMTS, the optimal radio design would require many parameters to be reconfigured frequently and continuously, something clearly not viable given the large network size, xiv Preface limited processing power and long convergence times of numerical optimisation routines. To introduce automated optimisation routines embedded into UMTS base stations and the network backbone is the natural direction to take. A successful radio network optimisation, be it automated or manual, can only be accomplished by appropriate prior network planning, which in turn must rely on precise network modelling. The book will discuss these three complementary subjects related to the UMTS radio network, i.e. modelling, planning and optimisation. They are dealt with in great theoretical depth facilitating an understanding of the UMTS network behaviour and, importantly, an abstraction of the presented theory to other beyond-3G networks that rely, in one form or another, on CDMA technology. The theoretical analysis is underpinned by professional field experience from the first commercially successful UMTS network implementations, thereby enriching the understanding of a 3G network design. Modelling is examined theoretically and practically at various levels and covers a wide range of aspects that have significant importance on the overall 3G network planning process: simplified as well as very detailed models of the UMTS radio network and its behaviour, modelling of geographical data as well as propagation with a special attention to the wideband character of the radio channel, all in terms of the actual UMTS radio network elements deployed. The important issue of investment business modelling is included as well. The models serve as a basis for development of network planning methods and sophisticated automatic network design procedures. Planning considers various planning stages, starting with business planning and including the following technical requirements: network dimensioning including coverage/capacity considerations, influence of traffic on the required number of both radio and non-radio network elements, detailed network planning with computer aided design and comprehensive aspects that need to be taken into account, such as infrastructure sharing, cross-border co-ordination etc. Optimisation means achieving the highest profit by an operator with the lowest possible expenses and is characterised by good investment business planning as well as tuning the network parameters and infrastructure for optimal performance. This covers the challenges and goals of an automated optimisation processes, the selection of appropriate cost functions and optimisation algorithms as well as the computational complexity of an implementation. Automated network tuning of RRM parameters, as the highest level of optimisation activities, becomes increasingly important for correct network operation. The subject of planning and optimisation in the book also relates to the UTRAN transmission infrastructure, where significant amounts of money are spent by network operators. This part of the network needs to be planned efficiently but is usually somehow neglected and hence requires careful attention. To this end, Chapter 17 concentrates solely on the issue of UTRAN transmission infrastructure planning and optimisation. The theoretical approach, coupled with practical examples, makes this book a complete and systematic compendium, serving a wide spectrum of readership ranging from college students to professional network engineers. The healthy mix of academics, ex-academics, industrial members of both small and large telecom companies having written this compendium guarantees that the important issue of UMTS radio network tuning is reflected in a fair, comprehensive and knowledgeable manner. Ideally, this book ought to be read from the beginning to the end; however, each chapter can be read stand-alone, which is why some natural overlap between the chapters occurs. The reader is also invited to visit the book’s website, where complete lists of acronyms, abbreviations and variables are available, as well as figures and some optimisation examples (http://www.zrt.pwr.wroc.pl/umts-optimisation). This website will also include a dynamic forum, allowing modelling, planning and optimisation experts around the globe to share thoughts and experiences. We dedicate this book to the student who, we trust, will understand the problems associated with current system design and inject new knowledge into future wireless communication system designs; to the network designer and optimiser who, we hope, will comprehend the parametric interdependencies Preface xv and use this to implement automated solutions; and to managers and CEOs who will come to believe that there is hope of effectively running these networks, acquired, not so long ago, for such substantial sums.