Current Status and Future Directions of Software Architectures for Telecommunications - Case Study Example

?Contents 2 2 Introduction 2 3 Overview of Techniques in Artificial Intelligence (AI) 7 4 Research Methodology 9 5 Results 12 After the through introduction the description of operation management model of telecommunications operators. A simple overview of AI techniques is provided in the upcoming topics and the research method is also introduced in the later parts of this studies. A detailed analysis of the application of AI in telecommunications is also provided in the study. Finally, a discussion and conclusions are given in the end. 2 Introduction The telecommunication engineering is considered as the most specific and high technology service industry. The constricted integration and Information technology applications related to the business procedures are its main characteristics. Moreover, in order to promote competition in this industry, the telecommunication industry is constantly deploying Information technology applications. The telecommunication environment is determined Due to its unique characteristics such as, distribution, continuous enlargement in network size and specific fault tolerance techniques. The application procedures are defined by the above characteristics. However, in telecommunication, the software systems need to overcome with a new variety of telecommunication protocols and a number of hardware platforms and network architectures. In fact, these systems consist of other characteristics for example, high cost, concurrency, high reliability needs, diversity and complexity (Patel 2002). The telecommunication operations are facing different challenges nowadays, for instance globalization and creative technology procedures. Thus, the industry would become more competitive in order to endure global market along with other competitors. Thus, in order to enhance the quality of services we need to increase cost. The developing economics and state-owned operators are basically privatized. This will help to provide improved and better services to the customers. On the contrary, several larger operators fortunately grab the best opportunities that are offered by the global telecommunication providers. For example, BT and Vodafone, this is due to the economies of scale factor present in the telecommunication networks. However, the technologies that are related with the Internet has been developing immensely thus contributing towards growing future of Internet and IP services and applications. In modern days, the clients are allowed to choose their own vendor and service providers according to their needs and can assemble their personal solutions. The new technologies such as 3G and mobile Internet speed up the production of new services. Therefore, the unchanging market has been converted into an increasingly user-driven market place. This change in telecommunication depends upon the operator’s capability to generate services and applications that are demanded by the customers. There is a vast range of software platforms, innovative variety of services and an increased customer based operations are available in order to manage an effective telecommunication infrastructure basics. The telecom operators must have the capability in order to develop themselves promptly, to implement and to maintain services for the customers demand anytime and anywhere. In addition, the artificial intelligence (AI) techniques are initially implemented to the telecommunication industry ages ago. During 1988, the artificial intelligence technique (IA) was first implemented in telecommunication industry. Moreover, it was initially explained by the Macleish who describes the relationship between the two rapidly evolving technologies i.e. artificial intelligence (AI) and telecommunications. During that time, the major utilization related to the artificial intelligence (AI) in telecommunications was considered as the initial system that is designed for diagnosing complex equipment in all off-line modes. Currently, the DSS are integrated with artificial neural networks. The ANN is defined as “a computing system made up of a number of simple, highly interconnected processing elements, which process information by their dynamic state response to external inputs.”(Neural Network. 2007). Due to the integration of ANN in the clinical based DSS, the results were significantly better as the system was able to do multi-tasking i.e. can process multiple instructions related to demographic information, patient registration information, patient medical history, invent prescriptions and diagnose patients. Moreover, the state of the art systems are able to calculate the likeliness of a potential disease that most likely may affect the patient. Likewise, this system also facilitates best clinical practices for the nursing staff. The features of the system allow the patient and the physician to select the relevant options and access to appropriate information. For instance, a decision is taken to give a dose of three months to the patient. The system will schedule and organize the vital actions that need to be performed along with clinical care factors. There are certain challenges while operating Clinical DSS, as it is accessed by physicians, hospital staff, system administrators and system experts. However, system experts need contribution of the physicians while improving the system functionality. Physicians are involved in the design stage, implementation stage and system improvement stage. Moreover, the decisions recommended by the DSS systems affect the traditional medical practice. As systems are advancing, increasing negative factors for physicians of their natural behavior towards patients. Furthermore, there is a significant negative impact if one of the systems does not comply with the requirements. In order to overcome the above-mentioned issues, physicians give first priority to their experience and decisions. As the DSS only provides relevant information that may be used to heal or cure a disease for a patient, there are certain cases where physicians only assist themselves by the DSS and do not take final decisions based on information provided by these systems. Moreover, physicians may discard the suggestions and instruction of these systems on a certain period, as they are more confident to cure the patience by their experience and expertise. During 1993, the requirements for Al in telecommunications are highlighted by Muller and also provide a survey of Al methods. These methods provided a long discussion related to the Al in telecommunication. Since these early applications, more and more AI techniques have been applied to telecommunications. Also, during this period, both telecommunications and AI have evolved signi?cantly; therefore a thorough review of AI applications in the telecommunications industry is now timely. Telecommunication operation management model Future information and communication services can be represented as three layers. The infrastructure layer is at the bottom and is supported through the ubiquitous network. The services layer with pervasive computing and universal services is in the middle layer. Mobile and broadband multimedia services are positioned as the highest layer, named the application layer. Suppose if any business organization is considered as a system than, the operations related to an organization must be considered as a procedure that helps to transforms the internal resources of that organization into products required by the market under specific circumstances. The result for the telecommunication operation for any organization is information services instead of tangible products. However, the input of the operation procedures is service needs for information from the customer and market. In order to meet the requirement, the telecommunications operation invests and constructs the networks. The recent services that are provided in the package are the products which are sold to the customers market. The basic flows in telecommunication operator’s operations procedures are Service Flow, Information Flow and Financial Flow. The customer required information flow starts at the customer market and ends at the initial step of the operation procedures. The service flow starts during the second step and ends at the customer market. In fact, the financial flow goes along with the service flow. It is obvious that the main flow is the service flow while, the performance is very important for improving the major effectiveness of the operator. Moreover, the telecom operator’s operation management, as based on the hierarchy of network services and the operation procedures related to the operator. The CMM i.e. Customer and Market Management deals along with the analysis of the clients and market. The major four factors such as Prediction, Decision, Implementation and Feedbacks are important for this module in order to understand the changes took place in customer and market as, well as marketing sales and customer care. The performance of the services and products that are offered to the customers are handled by the Services and Products Management (SPM). The major role of the SPM is to determine the needs of the customers related to the services and products for designing and developing new services and products according to the customers need. Prompt response in order to be in market the key. However the contribution of partners in the value chain is needed in order to enhance data services. Services, products and partnerships life cycle management is one of the most important components. The Network Management (NM) is the technique that is related to the customers, techniques, methods and tools that are needed to improve the telecommunication network management at every stage in its life cycle (i.e. pre-service, in-service and future service). The Network Management (NM) relies on the improved management as it is considered as the major characteristics in telecom. The management in telecom is needed to handle the entire network resources while achieving the services and products that are offered to the clients. The resource management basically handles the areas that cover the performance and fault management and particularly the dynamic bandwidth management and fault protection mechanism. However, the old fashioned functions that are handled by the Network Management (NM) are Fault management, Performance management, Security management and Accounting management. The Customer and Market Management (CMM) are considered as the basis of Services and Products Management (SPM). The operations can be started while utilizing Services and Products Management (SPM) as mentioned by a recent analysis of Customer and Market Management (CMM). However, the Services and Products Management (SPM) is also a basis of Network Management (NM). Due to the network resource needs as determined by the volume of total services and products, the operators may decide to invest and develop network planning strategy. Hence, the quality of services that are provided by the Network Resource management must be suitable for the customers. 3 Overview of Techniques in Artificial Intelligence (AI) AI has been defined by many professionals. Dreyfus and Dreyfus in the year 1986 defined it as ‘autonomous thinking machines that are free of human control’. Another author Daniels in the year 1986 defined AI as ‘the application of knowledge, thought, and learning to computer systems to aid humans’. Shapiro considered AI as a field of science and engineering concerned with the computational understanding of what was commonly called intelligent behavior, and with the creation of artifacts that exhibit such behavior. However, the most specific definition regarding Artificial Intelligence was given by Feigenbaum and Feldman in 1995. It states as “the science of artificial simulation of human thought processes with a computer”. In this article we will focus on the definition given by Feigenbaum and Feldman. This is to mention here that the Artificial Intelligence also consist of several sub fields. These fields are also defined by many authors. Some of the authors are Pandit in 1994 considered that there are six types of sub fields in Artificial Intelligence such as Robotics, Expert Systems, Neural Networks, Natural Language Processing, Speech Recognition and Speech synthesis. Another author named Muller et al. characterizes Artificial Intelligence into three components. First component is meant to perform high level tasks such as planning and designing e.g. Expert Systems, Model-based reasoning and constraint satisfaction. The second was considered as the distributed artificial intelligence. This type of AI tends to resolve problems along with a collection of agents despite of single agent. The third and the last type are less procedural and it was more concerned with lower levels of task for instance, Perception and Pattern recognition e.g. Neural Networks and Genetic Algorithms. The Adlassing in year 2002 proposed that the areas of Artificial Intelligence consist of natural language and understanding, problem solving capability and search, information representation and reasoning, learning, vision and robotics. The following discussed techniques are mainly available. The Expert Systems (Ess) is a work station program that encodes human expertise. Likewise, the knowledgebase and the interface engine incorporate two foundation components i.e. Natural Language Understanding (NLU) and speech recognition / speech synthesis. The primary objective of natural language processing is to construct a language interface for a computing system that will process input and output in English and Chinese language. Moreover, speech recognition is associated with human voice processing and interpretation and speech synthesis is the output of computer generated human like voice. Furthermore, Machine Learning is linked with translation of the same work in an efficient manner. The algorithm of Machine Learning technology can be called as kernel and Neural networks and generic algorithms are the two core machine learning technologies. Distributed Artificial Intelligence (DAI) takes initiative to resolve issues in a distributed approach. Likewise, the DAI system incorporates agents that are responsible for a sub part of the problem. Dissimilar communication levels, cooperation levels and controlling agents are essential for achieving a global solution. Another technology is called Robotics that focuses on use of controlled motion of a device. Likewise, the controlled motion of these devices is managed by computing devices. 4 Research Methodology Initially, the research can be conducted while using IEEE Xplore. The IEEE Xplore offers full text access for the IEEE transactions, journals, magazines and conference proceedings. Since 1998, the most of the academic publications for the telecommunication industry consist of IEEE Xplore. In fact, by utilizing Artificial intelligence and telecommunication as two keywords, the research will match around 113 papers out of 1,396,453 documents. Secondly the ACM digital library is used. The ACM offers resources related to the computing profession along with the leading-edge publications, conferences and career resources. . Using AI and telecommunications as keywords, 13 papers were obtained. Third, we used the Elsevier SDOS=SDOL, Kluwer Online and ABI databases. With AI and telecommunications as keywords, we matched 72 papers. We also searched the Proceedings of the Fifth International Symposium on Communication Systems Networks and Digital Signal Processing (CSNDSP2006) and IFISF 2006: An International Forum of Information Systems Frontiers, which was published in July 2006. We kept 63 papers for analysis. They were selected using the following criteria: AI is directly applied to telecommunications; The application of AI in a certain area of telecommunications is the main topic of the paper. The research methodology is well described. The research results are available and complete. All of these 63 papers are involved with the following periodicals and international conference Proceedings: Artificial Intelligence Artificial Intelligence for Engineering Design, Analysis and Manufacturing AI Expert Engineering Applications of Artificial Intelligence IEEE Communications Letters IEEE Transactions on Vehicular Technology IEEE Transactions on Evolutionary Computation IEEE Transactions on Neural Networks IEEE Journal on Selected Areas in Communications IEEE Transaction Magazine Optical Fibre Communications Computer Networks and ISDN Systems Speech Communication Computer Communications Computer Networks Telematics and Informatics Information Economics and Policy Information and Software Technology ACM Computing Survey Communications of the ACM Information Systems Research Production and Operations Management Journal of Systems Architecture Interacting with Computers IEICM Transactions of Information and Systems Omega Proceedings of the 2003 International Conference on Cyberworlds Proceedings of the 8th IEEE International Symposium on Computers and Communication (ISCC ‘03) Proceedings of the 36th Hawaii International Conference on System Sciences (HICSS ‘03) ESANN 2004 Proceedings of the European Symposium on Artificial Neural Networks Proceedings of the 5th International Symposium on Communication Systems Networks and Digital Signal Processing (CSNDSP 2006) IFISF 2006: An International Forum of Information Systems Frontiers 5 Results Applications of Artificial Intelligence in CMM The application of Artificial Intelligence that is being utilized in CMM is basically modern. This is due to the change in telecommunication industry as well as innovations in this field. Moreover, the completion in telecommunication industry is quiet difficult due to the natural presence of monopolistic competition in this field. During previous years, the operators had the opportunity over late and new entrants however; it is now very difficult for the operators to deal with the new entrants. This is due to the fact that new entrants cannot survive because of the restrictions imposed by the existing entrants. In fact, the previous operators do not need to focus on the customers’ requirements due to the presence of monopoly I market. Thus no motivation is generated for the operators to provide better services to the clients. But since 1980, the anti-monopoly, deregulation, globalization and technological development in this field have intensified all the competition related problems for this industry. Now it is important for all the operators to fulfill the needs of the customers as, well as the market needs should also be satisfied. References PATEL, A., 2002. Current status and future directions of software architectures for telecommunications. Computer Communications, 25(2), pp. 121-132. Neural Network. 2007. Network Dictionary, , pp. 340-340. Read More