Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

Light-Emitting Diodes. (Monographs in Electrical and Electronic Engineering.) By A. A. Bergh and P. J. Dean. Pp. viii+591. (Clarendon: Oxford; Oxford University: London, 1976.) £22.

/pdf/light-emitting-diodes-ubzde6g9qc.pdf

Light-emitting diodes

Machine Learning (ML) has been enjoying an unprecedented surge in applications that solve problems and enable automation in diverse domains. Primarily, this is due to the explosion in the availability of data, significant improvements in ML techniques, and advancement in computing capabilities. Undoubtedly, ML has been applied to various mundane and complex problems arising in network operation and management. There are various surveys on ML for specific areas in networking or for specific network technologies. This survey is original, since it jointly presents the application of diverse ML techniques in various key areas of networking across different network technologies. In this way, readers will benefit from a comprehensive discussion on the different learning paradigms and ML techniques applied to fundamental problems in networking, including traffic prediction, routing and classification, congestion control, resource and fault management, QoS and QoE management, and network security. Furthermore, this survey delineates the limitations, give insights, research challenges and future opportunities to advance ML in networking. Therefore, this is a timely contribution of the implications of ML for networking, that is pushing the barriers of autonomic network operation and management.

/pdf/a-comprehensive-survey-on-machine-learning-for-networking-3t5h3kbpe8.pdf

A comprehensive survey on machine learning for networking: evolution, applications and research opportunities

Radio access technologies for cellular mobile communications are typically characterized by multiple access schemes, e.g., frequency division multiple access (FDMA), time division multiple access (TDMA), code division multiple access (CDMA), and OFDMA. In the 4th generation (4G) mobile communication systems such as Long-Term Evolution (LTE) (Au et al., Uplink contention based SCMA for 5G radio access. Globecom Workshops (GC Wkshps), 2014. doi:10.1109/GLOCOMW.2014.7063547) and LTE-Advanced (Baracca et al., IEEE Trans. Commun., 2011. doi:10.1109/TCOMM.2011.121410.090252; Barry et al., Digital Communication, Kluwer, Dordrecht, 2004), standardized by the 3rd Generation Partnership Project (3GPP), orthogonal multiple access based on OFDMA or single carrier (SC)-FDMA is adopted. Orthogonal multiple access was a reasonable choice for achieving good system-level throughput performance with simple single-user detection. However, considering the trend in 5G, achieving significant gains in capacity and system throughput performance is a high priority requirement in view of the recent exponential increase in the volume of mobile traffic. In addition the proposed system should be able to support enhanced delay-sensitive high-volume services such as video streaming and cloud computing. Another high-level target of 5G is reduced cost, higher energy efficiency and robustness against emergencies.

Non-Orthogonal Multiple Access (NOMA) for cellular future radio access

Today’s telecommunication networks have become sources of enormous amounts of widely heterogeneous data. This information can be retrieved from network traffic traces, network alarms, signal quality indicators, users’ behavioral data, etc. Advanced mathematical tools are required to extract meaningful information from these data and take decisions pertaining to the proper functioning of the networks from the network-generated data. Among these mathematical tools, machine learning (ML) is regarded as one of the most promising methodological approaches to perform network-data analysis and enable automated network self-configuration and fault management. The adoption of ML techniques in the field of optical communication networks is motivated by the unprecedented growth of network complexity faced by optical networks in the last few years. Such complexity increase is due to the introduction of a huge number of adjustable and interdependent system parameters (e.g., routing configurations, modulation format, symbol rate, coding schemes, etc.) that are enabled by the usage of coherent transmission/reception technologies, advanced digital signal processing, and compensation of nonlinear effects in optical fiber propagation. In this paper we provide an overview of the application of ML to optical communications and networking. We classify and survey relevant literature dealing with the topic, and we also provide an introductory tutorial on ML for researchers and practitioners interested in this field. Although a good number of research papers have recently appeared, the application of ML to optical networks is still in its infancy: to stimulate further work in this area, we conclude this paper proposing new possible research directions.

/pdf/an-overview-on-application-of-machine-learning-techniques-in-3eel2cbu3o.pdf

An Overview on Application of Machine Learning Techniques in Optical Networks

A service-aware dynamic network slicing scheme for wavelength resource supporting delay-sensitive services in 5G fronthaul is proposed. The experiment results demonstrate that this scheme is able to effectively reduce the end-to-end delay of services.

Service-aware Network Slicing Supporting Delay-Sensitive Services for 5G Fronthaul

Optical packet-switched networks (OPNs), which consist of edge nodes and core nodes, are regarded as one of the most promising solutions to meet the increasing service requirements of the global Internet and to utilize efficiently the huge bandwidth of dense wavelength division multiplexing (DWDM) technology. In particular, the edge nodes perform adaptation function which interfaces between the legacy packet networks (i.e. IP networks) and OPNs. Namely, the edge nodes assembly IP packets from electronic layer into a larger optical packet. The function of packet assembly has become an important issue to be addressed in OPNs, since it not only increases the bandwidth utilization, but also reduces the self-similarity of the traffic. 
Studied in this paper are three kinds of typical packet assembly schemes in unslotted OPNs, i.e. the timer-based one, the length-threshold one and the hybrid algorithm. Based on a generic architecture of the edge nodes, these assembly mechanisms are examined by simulations under various self-similar traffic loads. The impacts of these algorithms on the traffic, including statistics of generated optical packets and self-similarity of optical packet traffic, are compared. The simulation results indicate that optical packet assembly can fulfill traffic shaping by improving key traffic characteristics of interest in network performance evaluation, such as the long-term correlation and so on. Moreover, the hybrid algorithm proves to be a better solution when both the assembly time and the packetisation efficiency have been also considered at the same time.

Comparison study of length-threshold, timer-based and hybrid algorithm for optical packet assembly

We propose a NOMA scheme combined with OFDMA for visible light communications, which offers a high throughput, flexible bandwidth allocation and a higher system capacity for a larger number of users.

Non-Orthogonal Multiple Access for Visible Light Communications

This paper presents a comparative study of polar codes constructed by Gaussian Approximation(GA) considering the characteristic of Visible Light Communication(VLC). The impact of key parameters in coding and decoding scheme is investigated through numerical simulations.

Experimental Research on Polar Code based on Gaussian Approximation for VLC system

The rapid growth of the network traffic and the diversity of the network requirements bring network service providers new demands and challenges. In order to achieve better efficiency and resource utilization, we need to solve the problems caused by the heterogeneous network environment, such as the low resource utilization and the control and management collaboration problem of heterogeneous networks. The promotion of the software-defined network concept brings the network operators a new way of thinking to solve the problem caused by heterogeneous network environment, especially the convergence of packet and optical network. We propose a control architecture for packet and optical network convergence, including how to be compatible with the existing optical network control architecture. A series of network performance parameters have been demonstrated under various network environments on our testbed.

Min Zhang

Papers

Service-aware Network Slicing Supporting Delay-Sensitive Services for 5G Fronthaul

Comparison study of length-threshold, timer-based and hybrid algorithm for optical packet assembly

Non-Orthogonal Multiple Access for Visible Light Communications

Experimental Research on Polar Code based on Gaussian Approximation for VLC system

Experimental demonstration of software-defined optical network for heterogeneous packet and optical networks