Search

As a consequence of increased complexity, technical requirements of services and expectations of end-users, telecommunication networks and systems must still increase their efficiency. Research currently focuses on introducing automation and intelligence into network control and management. Neural Networks form a class of Machine Learning techniques utilizing the supervised learning paradigm. The class is mature and comprises a significant number of algorithms. As a result, Neural Networks-based solutions are widely applicable, also in the context of communications and networking. The aim of this tutorial is twofold. Firstly, to provide fundamentals regarding the Neural Network (NN) method. Secondly, to comprehensively study the examples of applying NN-based solutions to solve problems in different aspects of communications and networking. Studies are supplemented with additional explanations, figures and critical considerations, including pros and cons of using selected methods for particular purposes. This part uniquely complements the tutorial one and facilitates in-depth understanding of NN. Based on the conducted studies, we draw a comparative analysis, summaries and expected future research topics and challenges of using NN in communications and networking.

This paper presents a novel concept of a DC-AC converter that ensures AC voltage modulation on four levels using a single branch and achieves voltage boosting. The proposed boost inverter is based on a single-stage topology and does not use a DC-DC boosting part. Compared with a cascaded boost inverter topology, the proposed converter is bidirectional and allows operation at a lower DC-link voltage, which reduces voltage stress across devices. The inverter topology contains an additional DC capacitor. An adequate voltage level for this capacitor is maintained by a switched-capacitor-based (SC)-based balancing circuit. The balancing process is of low complexity because it only requires the synchronization of the SC branch control signals with the PWM of the inverter. The balancing circuit operates in resonant mode to avoid inrush currents. The main power circuit of the converter does not use diodes, and can be adequate for MOSFET or GaN-based implementations. It can be configured as a single-phase or multiphase system. The results presented in this paper demonstrate the correctness of the converter concept and its modulation, voltage stresses of switches, and balancing process of the auxiliary capacitor. The experimental results confirm the feasibility of this concept and demonstrate the operation and efficiency of the inverter.

With the global Internet traffic continuously growing, network operators face more and more challenges related to the management of their networks. Efficient utilization of the available network resources becomes crucial to maintain the desired Quality of Service level and control the upsurge in operational expenses. The combination of the Software-Defined Networking concept with the multi-layer network architecture can simplify the process of control and management of the network, its layers, and resources. In this paper, we propose a solution to enhance resource utilization in software-defined multi-layer optical networks. The proposed solution takes the advantages of the Software-Defined Networking, Flow-Aware Multi-Topology Adaptive Routing, and Automatic Hidden Bypasses mechanisms to ensure simultaneous, multi-path data transmissions in both IP and optical layers. The Software-Defined Networking controller manages both mentioned mechanisms, selects the best possible bypass, and allocates lightpaths to ensure that the optical spectral efficiency is optimal. The evaluation shows that the proposed solution for multi-layer software-defined network increases the overall network performance and resource utilization.

The article presents an analytical method for optimizing the geometry of the magnetic core of a three-phase compensation choke. The method describes the process of identification the fringing magnetic fields and the corresponding magnetic reluctances of the magnetic core, the flux density calculation even in the case of the core supersaturation and the total losses estimation. It shows finding the trade-off between the size/weight of the inductor and the magnetic core with respect to the overall losses and demonstrates their minimization. The mathematical model for the flux’s identification is based on a standard iterative calculation using the analogy to electrical circuits but includes a new approach to the calculation of fringing magnetic fields caused by the air gapped magnetic core. The presented method is verified by the finite element method (FEM) using the engineering calculation software ANSYS.

This paper presents a novel concept of a high-voltage-gain DC-DC converter. The converter is made up of switched capacitors and passive resonant branches. A significant reduction in the count of switches and low voltage stress on the switches is achieved in this proposed converter topology, in comparison to that of a classical SC series-parallel converter. It is essential from the cost, volume, and efficiency of the converter standpoint. The reduction in the count of switches is threefold. The highest voltage stress on the switches depends on the output voltage and is decreased in the proposed converter as well, as the output voltage is divided into two series-connected capacitors. The presented results demonstrate the operation of the converter with the use of resonant branches, its switching strategies, voltage stresses of switches, efficiency, voltage gain, and output voltage regulation as well as the zero-voltage switching (ZVS) operation. The paper also presents novel issues related to analytical loss modeling, extended concepts of topology, converter start-up, and operation during transient states. The demonstrated concept of the converter, the analytical discussion and its design, as well as the experimental setup and results clearly demonstrate the optimization achievements.

This paper discusses the constraint data interpolation or range restricted interpolation for surface data arranges on rectangular meshes that lie above or below an arbitrary plane and between two arbitrary planes by using partially blended rational bi-cubic spline function with 12 parameters. Common research in range restricted surface interpolation is to construct the constrained surface lie above linear plane. However, in this paper, we consider the constraint surfaces up to degree three (cubic). To construct the constrained surface with shape preserving properties, i.e., the resulting surface will lie below or above single planes or between two respective planes, the data dependent sufficient conditions are derived on four parameters; meanwhile, the remaining eight parameters are free parameters to change the shape of the interpolating surface locally. The proposed scheme is tested with various types of data test, including some well-known functions. From the numerical results, we found that the proposed scheme is easy to use, locally control via free parameters, and require less computation compared with some existing schemes as well as visually pleasant for visualization. Furthermore, based on root mean square error (RMSE) and coefficient of determination (R2), the proposed scheme is better than existing scheme with the value of R2 achieved is in between 0.9701 (97.01%) and 0.9954 (99.54%). This is quite good for range restricted surface data interpolation since we can explain at least 97.01% of the variance in the interpolation by using the proposed scheme. Furthermore, the proposed scheme requires less CPU time (in seconds) compared with the existing scheme.