2018. 2nd Issue

Table of contents 

Full issue  (14 MB) 


Hum Nath Parajuli and Eszter Udvary
Wired-Wireless Converged Passive Optical Network with 4-PAM and Multi-sub-bands FBMC  
Future 5G based passive optical networks (PON) are expected as capable of the simultaneous provision of wired and wireless services for multi-users. In this paper, for the first time, we propose and demonstrate the simultaneous delivery of wired 4-pulse amplitude modulation (4-PAM) and wireless multi-sub-bands filter bank multicarrier (FBMC) signals in one wavelength using one laser source for the future 5G PON. The 4-PAM can be used in cost-efficient intensity modulation direct detection (IM/DD) systems and it provides the double bandwidth efficiency compared to conventional on-off keying (OOK). FBMC is considered as a potential candidate for future wireless 5G due to its high suppression for out of band emissions, which allows combining multiple sub-bands with very narrow band-gaps. Using multi-sub-bands with a narrow band gap, the overall transmission capacity can be increased. In the designed system, the composite wired 4-PAM and wireless multi-sub-bands FBMC signal is generated and transmitted with intensity modulation in optical line terminal (OLT). In the optical network unit (ONU) the wired and wireless signals from the received composite signal are extracted using an electrical square band-pass filter and separately demodulated using digital signal processing techniques. The designed 4-PAM has baseband bandwidth of 4.8 GHz and multi-subbands FBMC consists of 4 sub-bands of 500 MHz each, having very narrow inter-sub-bands gap of 488.28 kHz and the aggregate bandwidth of 2.0015 GHz. The bit error rate (BER) has been evaluated for the performance analysis of the 4-PAM and multi-sub-bands FBMC for two cases (a) separate transmission and (b) composite transmission.

Gábor Szabó and Eszter Udvary
Advanced Approximation of Channel Quality in a VLC CDM System  
Expanding the functionality of LED indoor lighting with visible light communication (VLC) allows an additional communication channel beside wireless radio in buildings. This service may be based on various channel access methods and modulation types. Code division multiplexing (CDM) is a suitable method to such an application, but it is complicated to measure the signal quality which is essential to compare different codes and settings, and necessary for some applications like position-dependent information services. Computing crest factor is a suitable method to estimate quality, but it may be inaccurate in some cases. This paper presents novel methods to approximate the quality of received CDM signals along with the crest factor, aiding the more accurate investigation of the VLC CDM technique.



Gábor Lencse and Youki Kadobayashi
Methodology for DNS Cache Poisoning Vulnerability Analysis of DNS64 Implementations  
The trustworthy operation of the DNS service is a very important precondition for a secure Internet. As we point it out, DNS cache poisoning could be even more dangerous if it is performed against DNS64 servers. Based on RCF 5452, we give an introduction to the three main components of DNS cache poisoning vulnerability, namely Transaction ID prediction, source port number prediction, and birthday paradox based attack, which is possible if a DNS or DNS64 server sends out multiple equivalent queries (with identical QNAME, QTYPE, and QCLASS fields) concurrently. We design and implement a methodology and a testbed, which can be used for the systematic testing of DNS or DNS64 implementations, whether they are susceptible to these three vulnerabilities. We perform the tests with the following DNS64 implementations: BIND, PowerDNS, Unbound, TOTD (two versions) and mtd64-ng. As for the testbed, we use three virtual Linux machines executed by a Windows 7 host. As for tools, we use VMware Workstation 12 Player for virtualization, Wireshark and tshark for monitoring, dns64perf for Transaction ID and source port predictability tests, and our currently developed "birthday-test" program for concurrently sent multiple equivalent queries testing. Our methodology can be used for DNS cache poisoning vulnerablility analysis of further DNS or DNS64 implementations. A testbed with the same structure may be used for security vulnerablility analysis of DNS or DNS64 servers and also NAT64 gateways concerning further threats.


Tomáš Huszaník, Ján Turán and Ľuboš Ovseník
Demonstration of Multimode Optical Fiber Communication System using 1300 nm Directly Modulated VCSEL for Gigabit Ethernet  
In the recent years, the optical networks have grown to unexpected dimensions. The growth of active users and growing demand for data services set high requirements to network providers. Driving forces of this growth are multimedia, cloud computing and web services which set high bandwith demand. The majority of currently deployed optical networks utilize passive or active network structure using dominantly singlemode optical fiber (SMF). SMF is believed to be the better choice over multimode optical fiber (MMF) for high speed optical fiber communication systems. And in some applications it definitely is. MMF has found use especially for short distance communication as it easily supports distances required for interconnecting building, data centres or campuses. In this paper we present a simulation model of 1000BASE-LX Ethernet with MMF using different optical modulation techniques. The aim of this article is to demonstrate possibilities of MMF based 1000BASE-LX Ethernet with directly modulated vertical-cavity surface-emitting laser (VCSEL).



IEEE International Conference on Computer Communications
IEEE INFOCOM 2019, Paris, France

IEEE International Symposium on Computational Intelligence and Informatics
IEEE CINTI 2018, Budapest, Hungary

IEEE Wireless Communications and Networking Conference
IEEE WCNC 2019, Marrakech, Morocco

IFIP/IEEE International Symposium on Integrated Network Management
IFIP/IEEE IM 2019, Washington DC, USA



Guidelines for our Authors


2018. 1st Issue

Table of contents 

Full issue   (6,1 MB)

The full papers are available only for subscribers - authentication required!  / marked as 


Ferenc Nandor Janky and Pal Varga
Time Synchronization Solution for FPGA-based Distributed Network Monitoring  
Distributed network monitoring solutions face various challenges with the increase of line speed, the extending variety of protocols, and new services with complex KPIs. This paper addresses one part of the first challenge: faster line speed necessitates time-stamping with higher granularity and higher precision than ever. Proper, system-wide time-stamping is inevitable for network monitoring and traffic analysis point of view. It is hard to find feasible time synchronization solutions for those systems that have nation-wide, physically distributed probes.
Current networking equipment reside in server rooms, and have many legacy nodes. Access to GPS signal is complicated in these places, and Precision Time Protocol (PTP) does not seem to be supported by all network nodes in the near future – so high precision time-stamping is indeed a current problem. This paper suggests a novel, practical solution to overcome the obstacles. The core idea is that in real-life, distributed network monitoring systems operate with a few, finite number of probeclusters, and their site should have a precise clock provided by PTP or GPS somewhere in the building. The distribution of time information within a site is still troublesome, even within a server rack. This paper presents a closed control loop solution implemented in an FPGA-based device in order to minimize the jitter, and compensate the calculated delay.

Devender Kumar, Satish Chand and Bijendra Kumar
Cryptanalysis and Improvement of a Two-Factor User Authentication Scheme  
Recently, Wang-Wang have discussed a two birds with one stone: two-factor authentication with security beyond conventional bound. We find that this scheme is vulnerable to the password exposure attack and also does not offer user anonymity, which is an important feature for some of the applications like e-healthcare services, e-banking, etc. In this paper, we provide the solution to these problems.

Marton Szabo, David Haja and Mark Szalay

Cost-Efficient Resource Allocation Method for Heterogeneous Cloud Environments  
In this paper we present a novel on-line NFV (Network Function Virtualization) orchestration algorithm for edge computing infrastructure providers that operate in a heterogeneous cloud environment. The goal of our algorithm is to minimize the usage of computing resources which are offered by a public cloud provider (e.g., Amazon Web Services), while fulfilling the required networking related constraints (latency, bandwidth) of the services to be deployed.We propose a reference network architecture which acts as a test environment for the evaluation of our algorithm. During the measurements, we compare our results to the optimal solution provided by an ILP-based solver.

Tamas Bisztray and Laszlo Bacsardi
The Evolution of Free-Space Quantum Key Distribution  
In this paper we are looking at the milestones that were achieved in free−space quantum key distribution as well as the current state of this technology. First a brief overview introduces the technical prerequisites that will help to better understand the rest of the paper. After looking into the first successful demonstrations of short range free space QKD both indoor and outdoor, we are examining the longer range terrestrial QKD experiments. In the next step we look at some experiments that were aiming to take free space QKD to the next level by placing the sender or the receiver on moving vehicles. After the terrestrial demonstrations we focus on satellite based experiments. Finally, we explore hyper-dimensional QKD, utilising energy−time, polarization and orbital angular momentum (OAM) degrees of freedom.


IEEE ComSoc Affiliate Membership

Guidelines for our Authors


Technical Co-Sponsors





National Cultural Fund, Hungary

National Civil Fund, Hungary