Search

A photonics-aided dual-functional system that can simultaneously implement high-speed communication and high-resolution radar detection is proposed. In the proposed system, high-speed orthogonal frequency division multiplexing signals and broadband linear frequency modulated signals are simultaneously realized based on photonic heterodyning. Based on this system, a net rate of 46.55 Gbit/s and radar with a 15 GHz bandwidth in the W-band were achieved.

The ICT infrastructures of medium and large organisations that offer ICT services (infrastructure, platforms, software, applications, etc.) are becoming increasingly complex. Nowadays, these environments combine all sorts of hardware (e.g., CPUs, GPUs, storage elements, network equipment) and software (e.g., virtual machines, servers, microservices, services, products, AI models). Tracking, understanding and acting upon all the data produced in the context of such environments is hence challenging. Configuration management databases have been so far widely used to store and provide access to relevant information and views on these components and on their relationships. However, different databases are organised according to different schemas. Despite existing efforts in standardising the main entities relevant for configuration management, there is not yet a core set of ontologies that describes these environments homogeneously, and which can be easily extended when new types of items appear. This paper presents an ontology network created with the purpose of serving as an initial step towards an homogeneous representation of this domain, and which has been already used to produce a knowledge graph for a large ICT company.

Verifying the intelligent substation Chinese operation order is an important step to ensure the accuracy of substation operations. The current manual verification method based on experience has strong subjectivity, low efficiency and low reliability. In view of this, this paper proposes an automatic verification method of operation order based on the Chinese semantic deep learning model: Global vectors for word representation-convolution Recurrent Neural Network (Glove-RCNN). By combining the text vectorization model Glove and text mining model RCNN, this method is feasible in realizing the semantic discrimination and automatic verification of correctness and error of the operation order texts. The experimental result shows that the MicroF score and MacroF score of the operation order verification model is 93.98% and 94.19%, which can accurately judge the correctness of the operation order, so as to effectively improve the verification efficiency.

The move towards more complex and dynamic telecommunication networks renders the need for more automation in the management system stringent. This is especially true in the context of 5G networks, where understanding context for providing the right management recommendations is crucial. Our work focuses on gathering context from the current status of the network and correlating it with useful information from existing documents in the network provider and operator's domain. We propose an architecture that, based on collected and interlinked information, can provide automatic recommendations for addressing existing problems in the network. More specifically, our approach is based on creating a knowledge-graph to direct workflow recommendations for network incidents. We implemented this system using the actual OSS product documentation and real network events and we show that our approach adds new value for automated network management, including improved efficiency and customer experience.

Scientific evaluation of the main transformer equipment suppliers is the premise to ensure the cost control and safe operation of the power grid. Based on the analysis of various factors affecting the main transformer equipment supplier, the evaluation index of the main transformer equipment supplier is constructed. Considering the fuzziness of the evaluation index, the index weight is determined by AHP, and the evaluation model of main transformer equipment supplier based on fuzzy comprehensive evaluation is established. The example calculation shows that the fuzzy comprehensive evaluation method can fully reveal the evaluation index information and reflect the actual situation of main transformer equipment suppliers scientifically and reasonably.

As an important state-owned energy backbone enterprise related to national energy security and the lifeline of national economy, power grid enterprises are responsible for ensuring the quality and safety of equipment and promoting the development of equipment quality. Therefore, continuously improving the quality evaluation of main transformer equipment is very important to improve the quality of power grid main transformer equipment. This paper first constructs the main transformer equipment quality evaluation index system from equipment procurement, equipment installation, equipment use and equipment decommissioning and scrapping; on this basis, the scoring method of each evaluation index is given, and the final main transformer equipment quality evaluation score is calculated by combining subjective and objective weight determination method; Finally, according to the evaluation of 500kV main transformer equipment of provincial power grid company price score, to evaluate the performance of equipment quality provided by suppliers.

Clustering algorithms are used in a large number of big data analytic applications spread across various application domains, including network management. We propose a novel Split-merge-evolve algorithm for clustering data into k number of clusters. The algorithm randomly divides data into k clusters initially, then repeatedly splits bad clusters and merges closest clusters to evolve the final clustering result. A key metric during the clustering process of the Split-merge-evolve algorithm is a user chosen or defined clustering quality metric or internal evaluation. The algorithm evolves the clustering result towards the user expected high quality result, although there is no ground truth or labelled data involved during the clustering process. The algorithm design is flexible in its implementation, with various common techniques, such as centroid and connectivity based measures that can be used in its implementation. The algorithm is easy to implement and effective. With 4 datasets, including 2 real life datasets in our experiments, the Split-merge-evolve algorithm performs better than both most commonly used K-means and Agglomerative hierarchical algorithms.

The evolution of telecom networks towards 5G comes with challenges that permeate the entire network structure and its management principles. In this paper, we outline how challenges such as the need to support different types of users and new network business models, the movement towards virtualization, as well as the need for more automation drive the requirement for a holistic and flexible way of managing the networks. We detail the architectural principles that underline such a cognitive management framework and we exemplify its use through a scenario based on software defined networking, where we combine machine learning, control and automation in the context of flexible resource provisioning in the Radio Access Network. Our experiments were conducted in collaboration with a major telecom operator and clearly show the advantages of introducing intelligence and automation into the network.

With emerging Network Functions Virtualization (NFV), Software Defined Networking (SDN) paradigms in Network Management (NM), new network devices, features can immediately become available. Available network resources, services can be altered, optimized in real time to gain the maximum benefit. However, this requires real time analytics information sent to SDN controllers rather than traditional manual offline or batch analytics which deliver outputs in hourly or monthly in NM Systems. As a result, real time analytic is becoming a critical element for NM. In this paper, we describe a configuration free (i.e., non-parametric) streaming data anomaly detection analytic engine for automatic NM system development. We describe the design principles, innovative algorithm design, architecture, implementation of the engine in relation to streaming data, mobile NM. Finally, we present use cases, evaluation results.

Much work has been done in the area of telecommunications transmission parameters and, in particular, Channel Quality Indicators, to analyse the best ways to improve the performance and quality of communications. However, there is a gap in published papers when it comes to real-world experiences with CQI reports and network-side analysis of transmission parameters that can provide information about user equipment and network elements behaviour. In this paper, we present our experiences with a real-world analysis of dense LTE networks, focusing on two main aspects — what information can CQI reports give us about the behaviour of user equipment in the network, and what information do they provide in relation to the individual network elements. We show what insights can be derived from a network wide statistical analysis of CQI reports and how these insights can be further used by operators to understand potential problems in their network.

In telecommunications network troubleshooting, analytical applications are widely used. Such applications typically use CEP (Complex Event Processing) and SQL queries for data processing and network analysis. Performance engineers need in-depth knowledge of both the telecommunications domain and telecommunications data structures in order to create the required queries. Moreover valuable information contained in free form text data fields such as “additional_info”, “user_text” or “problem_text” can also be ignored. This work proposes CADMANT: Context Anomaly Detection for MAintenance and Network Troubleshooting. Traditional approaches focus on a specific record type and create specific cause and effect rules. With the CADMANT approach all free form text fields of alarms, logs, etc. are treated as text documents similar to Twitter feeds. CADMANT uses distance based outlier detection within sliding windows to detect abnormal terms at configurable time intervals. The CADMANT approach provides automated analysis without the requirement for SQL/CEP queries and provides distinct network insights in comparison to traditional approaches.

Service-based business processes are often developed and deployed by single organizations. In distributed, shared resource environments like the cloud on the other hand, consumers share resources owned by cloud providers. %Higher levels of resource sharing gives more economy of scale for providers in the software-as-a-service (SaaS) or business process-as-a-service (BPaaS) space. This requires multi-tenancy capability for service processes that provides customized behaviour for on shared process implementations to meet the varying needs of different process consumers as tenants of the process resource. In this paper, we define a distributed multi-tenant architecture for BPEL processes provided as a service. A single-version BPEL process is deployed by a provider and offered for all process consumers, combined with a customization and management functionality to create a unique experience for different consumers (process tenants). We provide two core components: a policy model for consumers to express customization/business requirements of service processes and a coordination framework for policy enforcement between consumers and providers to achieve on-the-fly customization of service processes.

Browsing and searching network information for observation, analysis and troubleshooting is an inherent part of using the features and functions of any Network Management System. Enterprise search has capabilities for handling various data types and sources and big data scalability, and is becoming an emerging technology for such network management functions development. In this paper, we give an overview on work done in our research and prototype team regarding an advanced search project. We provide a brief report on search fundamental knowledge and study of Solr search platform stack. It answers common questions from management and development teams regarding adopting search technology for production development, and gives a Solr reference card for developers. We also introduce two advanced search features, user experience based search recommendation and anomaly detection enhanced search ranking from our research work. Two features are developed to make network searches more efficient as it can help user quickly locate the most valuable search results, but the concept can be adopted for search applications in other domains.

Network operators are increasingly using analytic applications to improve the performance of their networks. Telecommunications analytical applications typically use SQL and Complex Event Processing (CEP) for data processing, network analysis and troubleshooting. Such approaches are hindered as they require an in-depth knowledge of both the telecommunications domain and telecommunications data structures in order to create the required queries. Valuable information contained in free form text data fields such as “additional_info”, “user_text” or “problem_text” can also be ignored. This work proposes An Anomaly Detection Algorithm for MAintenance and Network Troubleshooting (ADAMANT), a text analytic based network anomaly detection approach. Once telecommunications data records have been indexed, ADAMANT uses distance based outlier detection within sliding windows to detect abnormal terms at configurable time intervals. Traditional approaches focus on a specific type of record and create specific cause and effect rules. With the ADAMANT approach all free form text fields of alarms, logs, etc. are treated as text documents similar to Twitter feeds. All documents within a window represent a snapshot of the network state that is processed by ADAMANT. The ADAMANT approach focuses on text analytics to provide automated analysis without the requirement for SQL/CEP queries. Such an approach provides distinct network insights in comparison to traditional approaches.

When searching for data in a telecommunications network management application, large search result sets are common. In order to refine the results to retrieve useful information existing systems normally require additional user intervention such as appending or removing a search keyword, adding a filter, grouping results, etc. This work proposes a Probabilistic Search Recommendation Algorithm to Improve Search Efficiency for Network Intelligence and Troubleshooting (PRISENIT). PRISENIT is a query-based recommendation algorithm intended to improve search efficiency and usability for telecommunication system management. PRISENIT is an extension of an item-based collaborative filtering algorithm. It uses correlation-based similarity and users' implicit feedback in order to improve search efficiency. It learns from previous experiences in order to optimize decision-making. Currently there exists no known query-based recommender adaptation mechanism for network management. Existing search engines use previous user searches to make a suggestion based on the keyword. PRISENIT not only considers search terms, it also considers the influence of filters and features in order to makes network searches more efficient as it removes the necessity for users to manually choose search features or search filters. Experimental results show that PRISENIT can improve user experience in a telecommunications management environment.

Services are offered in an execution context that is determined by how a provider provisions the service and how the user consumes it. The need for more flexibility requires the provisioning and consumption aspects to be addressed at run-time. We propose an ontology-based context model providing a framework for service provisioning and consumption aspects and techniques for managing context constraints for Web service processes where dynamic context concerns can be monitored and validated at service process run-time. We discuss the contextualisation of dynamically relevant aspects of Web service processes as our main goal, i.e. capture aspects in an extended context model. The technical contributions of this paper are a context model ontology for dynamic service context and an operator calculus for integrated and coherent context manipulation, composition and reasoning. The context model ontology formalises dynamic aspects of Web services and facilitates reasoning. We present the context ontology in terms of four core dimensions--functional, QoS, domain and platform--which are internally interconnected.

Service processes are often enacted across different boundaries such as organisations, countries or even languages. Specifically, looking at the quality and governance of data or content processed by services in this context is important to control different constraints in this cross-boundary processing. In order to provide a context-aware solution that takes into account data and data processing requirements, a rule-based constraints specification and adaptation of processes shall be proposed. A domain ontology shall capture the key data/content data types, activities and constraints, which forms the basis of a rule-based policy monitoring solution. A provenance model is at the core of this ontology solution. The key contribution is a domain-specific model and specification template for constraint policy definition, which can be applied to adapt service processes to domain-specific needs.

In many domains such as Telecom various scenarios necessitate the processing of large amounts of data using statistical and machine learning algorithms. A noticeable effort has been made to move the data management systems into MapReduce parallel processing environments such as Hadoop and Pig. Nevertheless these systems lack the features of advanced machine learning and statistical analysis. Frame-works such as Mahout on top of Hadoop support machine learning but their implementations are at the preliminary stage. For example Mahout does not provide Support Vector Machine (SVM) algorithms and it is difficult to use. On the other hand traditional statistical software tools such as R containing comprehensive statistical algorithms for advanced analysis are widely used. But such software can only run on a single computer and therefore it is not scalable. In this paper we propose an integrated solution RPig which takes the advantages of R (for machine learning and statistical analysis capabilities) and parallel data processing capabilities of Pig. The RPig framework offers a scalable advanced data analysis solution for machine learning and statistical analysis. Analysis jobs can be easily developed with RPig script in high level languages. We describe the design implementation and an eclipse-based RPigEditor for the RPig framework. Using application scenarios from the Telecom domain we show the usage of RPig and how the framework can significantly reduce the development effort. The results demonstrate the scalability of our framework and the simplicity of deployment for analysis jobs.

Cloud computing needs end-user customisation and personalisation of multi-tenant cloud service offerings. Particularly, QoS and governance policy management and monitoring is needed. We propose a user-customisable policy definition solution that can be enforced in multitenant cloud offerings through automated instrumentation and monitoring. Service processes run by cloud and SaaS providers can be made policy-aware in a transparent way.

Service engineering needs to address integration problems allowing services to collaborate and coordinate. The need to address dynamic automated changes - caused by on-demand environments and changing requirements - can be addressed through service coordination based on ontology-based composition and matching techniques. Our solution to composition and matching utilises a service coordination space that acts as a passive infrastructure for collaboration. We discuss the information models and the coordination principles of such a collaboration environment in terms of an ontology and its underlying description logics. We provide ontology-based solutions for structural composition of descriptions and matching between requested and provided services.

peer-reviewed Service-oriented architecture is an architectural approach that can be applied for building autonomous service systems dynamically to satisfy on-demand business requests. During the execution of service compositions, architectural constraint violations relating to functional and non-fucntional system properties need to be handled intelligently and autonomously, possibly requiring architectural reconfigurations. We propose integrated architectural constraint violation handling to deal with architectural quality problems through dynamic reconfiguration. We concentrate on service replacement selection as a remedial strategy for a possible quality violation requiring architectural remedies.

With the emergence of Business Process Outsourcing and Cloud Computing, enterprises are looking for available business processes outside of their organizations to quickly adopt to new business requirements and also reduce process development and maintenance costs. The process execution needs to be governed as policy enforcement might differ between different clients. Since a process is deployed outside of the organizations and serves multiple process clients, distribution and multi-tenancy have become two requirements for runtime governance of service processes. We address this problem by introducing a policy-oriented aspectual business process framework. The runtime governance constraints from process clients are integrated as aspects through dynamic weaving into rpcess execution.

Service-Oriented Architecture (SOA) offers a technical foundation for Enterprise Application Integration and business collaboration through service-based business components. With increasing process outsourcing and cloud computing, enterprises need process-level integration and collaboration (process-oriented) to quickly launch new business processes for new customers and products. However, business processes that cross organisations’ compliance regulation boundaries are still unaddressed. We introduce a distributed aspect-oriented service composition approach, which enables multiple process clients hot-plugging their business compliance models (business rules, fault handling policy, and execution monitor) to BPEL business processes.

Dynamic service composition is suitable for on-demand business requests. For autonomic computing, service composition needs to deal with runtime environment faults, but also with business constraint violations which result from business requirements. We propose an approach for integrated handling of business constraint violations and runtime environment faults for dynamic service composition. We introduce a loosely coupled implementation architecture to maintain the platform-independent nature.

Autonomic, i.e. dynamic and fault-tolerant Web service composition is a requirement resulting from recent developments such as on-demand services. In the context of planning-based service composition, multi-agent planning and dynamic error handling are still unresolved problems. Recently, business rule and constraint management has been looked at for enterprise SOA to add business flexibility. This paper proposes a constraint integration and violation handling technique for dynamic service composition. Higher degrees of reliability and fault-tolerance, but also performance for autonomously composed WS-BPEL processes are the objectives.

Web service composition is a technology that has received considerable attention in the last number of years. Languages and tools to aid in the process of creating composite Web services have been received specific attention. Web service composition is the process of linking single Web services together in order to accomplish more complex tasks. One area of Web service composition that has not received as much attention is the area of dynamic error handling and re-planning, enabling autonomic composition. Given a repository of service descriptions and a task to complete, it is possible for AI planners to automatically create a plan that will achieve this goal. If however a service in the plan is unavailable or erroneous the plan will fail. Motivated by this problem, this paper suggests autonomous re-planning as a means to overcome dynamic problems. Our solution involves automatically recovering from faults and creating a context-dependent alternate plan. We present an architecture that serves as a basis for the central activities autonomous composition, monitoring and fault handling.

Recent cross-organisational software service offerings, such as cloud computing, create higher integration needs. In particular, services are combined through brokers and mediators, solutions to allow individual services to collaborate and their interaction to be coordinated are required. The need to address dynamic management - caused by cloud and on-demand environments - can be addressed through service coordination based on ontology-based composition and matching techniques. Our solution to composition and matching utilises a service coordination space that acts as a passive infrastructure for collaboration where users submit requests that are then selected and taken on by providers. We discuss the information models and the coordination principles of such a collaboration environment in terms of an ontology and its underlying description logics. We provide ontology-based solutions for structural composition of descriptions and matching between requested and provided services.