E-model MOS estimate precision improvement and modelling of jitter effects

INFORMATION AND COMMUNICATION TECHNOLOGIES AND SERVICES VOLUME: 10 | NUMBER: 4 | 2012 | SPECIAL ISSUE 276 © 2012 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING E-MODEL MOS ESTIMATE PRECISION IMPROVEMENT AND MODELLING OF JITTER EFFECTS Adrian KOVAC1, Michal HALAS1 1Department of Telecommunications, Faculty of Electrical Engineering and Information Technology, Slovak University of Technology, Ilkovicova 3, 812 19 Bratislava, Slovak Republic [email protected]ba.sk, [email protected] Abstract. This paper deals with the ITU-T E-model, which is used for non-intrusive MOS VoIP call quality estimation on IP networks. The pros of E-model are computational simplicity and usability on real-time traffic. The cons, as shown in our previous work, are the inability of E-model to reflect effects of network jitter present on real traffic flows and jitter-buffer behavior on end user devices. These effects are visible mostly on traffic over WAN, internet and radio networks and cause the E-model MOS call quality estimate to be noticeably too optimistic. In this paper, we propose a modification to E-model using previously proposed Pplef (effective packet loss) using jitter and jitter-buffer model based on Pareto/D/1/K system. We subsequently perform optimization of newly added parameters reflecting jitter effects into E-model by using PESQ intrusive measurement method as a reference for selected audio codecs. Function fitting and parameter optimization is performed under varying delay, packet loss, jitter and different jitter-buffer sizes for both, correlated and uncorrelated long-tailed network traffic. Keywords Call quality, E-model, jitter, jitter buffer, MOS, network traffic, packet loss. 1. Introduction The Internet, VoIP and in general IP traffic are known to possess the property of being self-similar, long-range dependent (LRD) or in other words “bursty”. The behavior of a “bursty” traffic differs from ideal stochastic model of independent packets when trying to evaluate traffic interarrival times via wellknown distributions. This property translates into the failure of general queuing models, such as M/M/1/k, which counts on Exponential and Poisson characteristics of input stream and service time, to describe the situation of incoming VoIP stream at buffer on the receiver’s side. In our article, we analyze and improve original EModel designed to give real-time estimate of VoIP call quality in MOS scale based solely on network performance parameters and codec type. Our work is applicable to the E-model of version 04/2009 and newer, which still after numerous updates, does not incorporate the effects of jitter. While the performance of the EModel estimate is satisfactory under good network conditions, the E-Model MOS estimate becomes too optimistic under slightly and moderately impaired network conditions as shown in our previous work [1], [2] and [3]. Our measurements and simulation showed that the performance and estimate accuracy of E-Model deteriorates unacceptably beyond network jitter (calculated by RFC 1889) over 20 ms for all tested codecs including G.711 with and without PLC, G.723.1 ACELP and MP-MLQ, G.726 and G.729. Figure 1 shows an example of E-Model MOS inaccuracy of VoIP network connection in the following manner:  “MOS E-Model” – represents MOS as estimated via software on receiving side by reading network performance from RTCP protocol not accounting for the effects of local jitter buffer.  “MOS measured” – represents MOS estimated by measuring software – IX-Chariot – based of the net voice input packet stream entering the decoder behind buffer.  “MOS modified E-Model” – shows estimate performed via software using E-Model [4] incorporating the effects of jitter and buffer size based on actual codec configuration and data about network performance from RTCP without physically observing or interfering with packet stream behind jitter buffer. As we can observe, the actual discrepancy of EModel estimate, being around 1.00 MOS scale under 40 ms jitter is unacceptable for all purposes. These network conditions are not unreal and are common on WiFi and mobile connections. INFORMATION AND COMMUNICATION TECHNOLOGIES AND SERVICES VOLUME: 10 | NUMBER: 4 | 2012 | SPECIAL ISSUE © 2012 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 277 Fig. 1: Comparison of MOS estimates for G.729 codec at 40 ms RFC jitter and 40 ms buffer size, 0 % packet loss under varying network delay. 2. Brief E-model Description Mean opinion score (MOS) is a measure based on subjective user satisfaction with overall listening and conversational quality on five grade scale from 5 (best) to 1 (worst). MOS can be estimated by subjective methods based on physical listening tests or by objective methods relying on and working solely with real-time measured network performance parameters (delay, packet loss) which unfortunately does not include jitter and jitter buffer size. E-model defined by ITU-T G.107 [4] is widely accepted objective method used for estimation of VoIP call quality. E-model uses a set of selected input parameters to calculate intermediate variable – R factor, which is finally converted to MOS value. Input parameters contribute to the final estimate of quality in an additive manner as expressed in (1): AIIIRR effeds  0 , (1) where R0 represents the basic SNR, circuit and room noise, Is represents all impairments related to voice recording, Id covers degradations caused by delay of audio signal, Ie-eff impairment factor presents all degradations caused by packet network transmission path, including end-to-end delay, packet loss and codec PLC masking capabilities, A is an advantage factor of particular technology. We focus at Ie-eff parameter, which is calculated as (2):   plpl pl eeeffe BP P III   95 , (2) where Ie represents impairment factor given by codec compression and voice reproduction capabilities, Bpl is codec robustness characterizing codec’s immunity to random losses. The values are given for 8 kHz sample rate codecs in ITU-T G.133 appendix [6]. Ppl parameter represents measured network packet loss in %. In this paper, we propose a substitution of Ppl parameter for Pplef further described in section IV of the paper. 3. Jitter Buffer Effects on MOS 3.1. Model Implementation Presumptions Timescale of our interest is in order of seconds under practical real-time conditions what is supported by the following facts: Jitter J is calculated from 16 consequent interarrival times. Jitter buffer size is in order of tens to hundreds of milliseconds for practical VoIP call purposes. e.g., with standard packetization of 20 ms we get 320 ms buffer size when considering buffering 16 packets. Regarding the traffic, following holds true: the interarrival time is “exactly second-order self-similar” with Hurst parameter H = 1− β/2 and Eq. (3) holds true:        22 2 1kkr . (3) The variance of input packet stream can be considered constant for the short time-scale we operate on as induced from the results from [9] and [15]. The Hurst parameter value from short-term point of view in order of seconds is constant and can be put equal to H = 1. 3.2. Network Delay Description and Statistics Voice packets are generated at sending device – IP phone – as a homogenous flow with constant transmit intervals depending mostly on packetization interval set in the codec. VoIP packets that traversed transport network have their regular spacing disrupted irregularly. Internet traffic arrival times and delay can be successfully statistically modeled by long-tailed Generalized Pareto distribution (GPD), [8], [9], [10], [12] ,[14]. We use GPD to further describe VoIP input packet stream. Delay distribution of received packets is in Fig. 2. Fig. 2: Distribution of Pareto-related packet arrival times. Real-time change of network parameters causes variations in network delay. Differences between packet arrivals are not constant and arrival times oscillate between minimal delay Ta-min and infinite delay, which is effectively a lost packet. Mean value of the process exists and is interpreted as an End-to-End delay Ta (one of the INFORMATION AND COMMUNICATION TECHNOLOGIES AND SERVICES VOLUME: 10 | NUMBER: 4 | 2012 | SPECIAL ISSUE 278 © 2012 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING input parameters for E-model). Real packet path usually consists of a mixture of different networks with different devices and technologies. Each device adds a degree of uncertainty in packet delivery time. Overall delay statistics is a sum of all partial statistics at each device. Pareto distribution is well suited to describe delay, which has lower bound, no upper bound and finite mean value. Probability density function of Pareto (PDF) is given by Eq. (4) and cumulative distribution function (CDF) by Eq. (5):                      1 1 ,, 1 1      x xf , (4)     1 ),,( )( 11)(         x xf , (5) where  = std. deviation,  = shape parameter,  = location parameter (minimal value of random variable with Pareto distribution),  is an offset of Pareto distribution from zero on the time axis and represents minimal delay Ta min (Fig. 2). The shape parameter must meet the condition  < 0 and to get valid results from Eq. (4) and (5)  ≤ x ≤  -  /  . 4. Proposed E-Model Modification to Impairment Factor Based on simulation results and measurements, the optimal shape parameter  giving the smallest overall MSE error of differences between measured and estimated Ploss by equations (6) and (7), is published in our previous work [3]. Ploss denotes the probability of a packet arriving with greater delay than being the jitter buffer size. The delayed arrival does not immediately mean that the packet is lost. The buffer can start rebuffering and start a playback with a delay correction during the silent period of conversation, when the sequence of delayed packet is longer. The final effect is then just a short-term increased average two-way network delay.                    dx x xP Tbuff loss 1 1 1 1 ),,,(        Tbuff PDFdx 0 1 , (6)      1 1         x Ploss , (7) where  = scale,  = shape and  = location parameter (min. value of random variable with Pareto distribution),  is an offset of Pareto curve from zero on the time axis and represents minimal network delay Ta-min (Fig. 1.) and x = Tbuff is an actual size of jitter buffer in milliseconds. Actual buffer loss of a packet occurs, when the two consequent packets are delayed and only a single such delay occurs in a short-term period. Then the probability of a packet lost on a buffer, Ploss_buffer is in relation of correlation of delays of the consecutive packets as shown in Tab. 1. Optimal value of sought shape parameter  was pro to be between values -0,1 and -0,2 depending on actual network traffic characteristics giving good results across a wide range of LAN IP networks. Our experiments and consequent analysis show, that the value of -0,1 is acceptable. Experimentally, we have verified, that there is a possibility to find and describe actual packet loss on jitter buffer, regardless on the burstiness (could be measured by Hurst parameter) of the input packet stream, by upper and lower bound for loss Ploss_buffer. These bound can be described by equations (8) and (9) as follows. Equation (8) represents lower bound of packet loss PLOWER_BOUND when the autocorrelation of subsequently delivered packets’ delay is highest (thus the function squared). This function after substitution,  = -0,1 and  = 0 according to our previous work [1], [2] and [3], where x = buffer size in [ms], becomes a compound function. To obtain correct results, a following condition must be obeyed: If x  10  , then Eq. (8) is valid; else PUPPER_BOUND = 0. Equation (9) represents upper bound of packet loss PLOWER_BOUND when the autocorrelation of subsequently delivered packets’ delay is lowest (thus the function is not squared). This function after substitution,  = -0,1 and  = 0 according to our previous work [1], [2] and [3], where x = buffer size in [ms], becomes a compound function. To obtain correct results, a following condition must be obeyed: If x  10  , then Eq. (9) is valid; else PLOWER_BOUND = 0. After substitution of actual values of parameters into Eq. (2), and with reordering capability where x = packet size in [ms]: ),,,( _  xP BOUNDLOWER   2 1 1 2 1                       x , (8) ),,,( _  xP BOUNDUPPER   2 1 1 1                       x , (9) 2 11,0 1 20 __          x PLOWERbufferloss , (10) 2 11,0 1 10 __          x PUPPERbufferloss . (11) INFORMATION AND COMMUNICATION TECHNOLOGIES AND SERVICES VOLUME: 10 | NUMBER: 4 | 2012 | SPECIAL ISSUE © 2012 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 279 Data from measurements of real packet loss on jitter buffer and respective lower and upper bounds are present in tabular form in Tab. 1 for one selected data row of 21,121 ms jitter. Tab.1: Measured packet loss vs. calculated upper and lower bounds for 21,121 ms HW jitter and varying buffer size. Jitter buffer size [ms] F(x) 1-F(x) F(x)/2 Lower bound (1F(x))^2 / 2 Upper bound (1-F(x)) / 2 Hardware measured loss 0 0,000000 1,000000 0,000000 0,500000 0,500000 N/A 10 0,386087 0,613913 0,193043 0,188445 0,306957 N/A 20 0,632427 0,367573 0,316214 0,067555 0,183786 0,140333 30 0,785942 0,214058 0,392971 0,022910 0,107029 N/A 40 0,879136 0,120864 0,439568 0,007304 0,060432 0,022667 50 0,934082 0,065918 0,467041 0,002173 0,032959 N/A 60 0,965428 0,034572 0,482714 0,000598 0,017286 0,003000 70 0,982658 0,017342 0,491329 0,000150 0,008671 N/A 80 0,991735 0,008265 0,495868 0,000034 0,004132 0,000333 90 0,996288 0,003712 0,498144 0,000007 0,001856 N/A 100 0,998445 0,001555 0,499222 0,000001 0,000778 0,000000 110 0,999400 0,000600 0,499700 0,000000 0,000300 N/A 120 0,999791 0,000209 0,499895 0,000000 0,000105 0,000000 130 0,999936 0,000064 0,499968 0,000000 0,000032 N/A 140 0,999983 0,000017 0,499992 0,000000 0,000008 0,000000 150 0,999996 0,000004 0,499998 0,000000 0,000002 N/A 160 0,999999 0,000001 0,500000 0,000000 0,000000 0,000000 Fig. 3: Measured packet loss vs. calculated upper and lower bounds for 21,121 ms HW jitter and varying buffer size in a lin-log graph showing waterfall-like loss curves up to the expected measurement accuracy. As has previously been shown in our previous work [1], [2], [3] and several studies in the field of Internet and IP traffic [8], [9], [10], [12], [14] the distribution of packet arrival and interarrival times is long-tailed with long-range dependency (LRD). When considering suitable function for E-model improvement to simulate PESQ results of MOS, it is proficient to simplify the function (12) and find a descriptive function with parameters between upper and lower bounds as stated previously.                    dx x xP Tpacket woloss 1 1 _1 1 ),,,(        Tbuff PDFdx 0 1 . (12) Based on local time invariance and presumptions in section A, supported by the results in [2], [3], we consider distribution functions of interarrival times of two consecutive packets to be in the ratio of 1:1 hence Eq. (12) can be rewritten to (13),   2 1 1),,,( 1 _                       x xP woloss . (13) Tab.2: Preliminary results of MOS given by E-model compared to PESQ estimates for G.711 codec and 20 ms packetization. Pareto Sigma (Traffic) RFC 3550 jitter (calcula -tion) [ms] RFC 3550 jitter (HW) [ms] One-Way Delay Average [ms] (software ) RFC 1889 Jitter Averag e [ms] MOS - PESQ mix (HW) MOS - RTP Emodel (HW) MOS - Emodel (SW) 1 1,186 0,746 2 0,367 4,459 4,410 4,37 5 5,93 4,530 5 4,433 4,300 4,400 4,37 10 11,86 8,572 9 6,571 3,614 3,940 4,33 15 17,79 11,204 17 8,269 3,065 3,052 4,17 20 23,72 14,065 28 9,917 2,558 2,997 3,63 25 29,65 17,087 28 11,696 2,236 2,921 3,44 30 35,58 20,109 30 12,136 1,843 2,872 2,83 35 41,51 21,121 28 12,714 1,779 2,754 2,67 40 47,44 23,570 28 13,5 1,554 2,306 2,2 45 53,37 25,378 37 15,632 1,311 1,983 2,19 50 59,3 28,153 41 14,684 1,300 1,361 2,08 55 65,23 29,311 45 16,667 1,266 1,276 1,98 60 71,16 30,542 51 16,882 1,169 1,237 1,31 65 77,09 32,374 50 18,059 1,169 1,221 1,41 70 83,02 34,749 60 19,188 n/a 1,200 1,26 75 88,95 36,495 69 18,875 n/a 1,180 1,05 80 94,88 37,739 64 21,2 n/a 1,160 1,07 5. Test of Results Iterative distribution fitting was performed using various distributions to find the best fit parameters. These parameters and distributions were put under Kolmogorov-Smirnov and Chi-Squared tests to find best descriptive statistics of Pareto-distributed stream time differences with applied jitter. Results of finding best descriptive statistics with optimal iteratively found parameter set with error of 10e-5 are sorted in Tab. 3. INFORMATION AND COMMUNICATION TECHNOLOGIES AND SERVICES VOLUME: 10 | NUMBER: 4 | 2012 | SPECIAL ISSUE 280 © 2012 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING Tab.3: Best fit parameters of tested distributions. Distribution Best fit distribution parameters Generalized Pareto (GPD) k = 0,19328 σ = 0,0224 μ = -0,00306 Generalized Extreme k = 0,36239 σ = 0,01384 μ = 0,00909 Weibull α = 0,39981 β = 0,01718 Gen. Gamma k = 0,98444 α = 0,40502 β = 0,05293 Log-Pearson 3 α = 6,081 β = -1,175 γ = 1,641 Laplace λ = 39,109 μ = 0,0247 Weibull (3P) α = 0,4745 β = 0,01408 μ = 1,3003e-5 Gamma α = 0,46676 β = 0,05293 Logistic σ = 0,01994 μ = 0,0247 Lognormal σ = 2,8971 μ = -5,504 Statistical tests showed as a proof of concept, that GPD Pareto distribution is also the most suitable one for describing interarrival times of general long-tailed LAN/WAN packet streams impaired by random jitter with equal distribution. This shows also Pareto distribution to be the best compromise between calculation complexity (compared to fractal modelling methods) and statistical significance for modelling also jitter buffer loss behaviour under variable jitter. To explain best fit parameters of GPD from Tab. 3:   in all equations corresponds to optimised σ in Tab. 3. Proposed relation between  and actual jitter J substituted can be expressed in the ratio J/   <1;2>. For actual imposed 40 ms network jitter the optimized parameter was σ = 0,0224 s = 22,4 ms what would field J/  ratio = 22/14  <1;2>. Actual parameter substitution ratio needs further testing.   = shape parameter corresponds to optimised k. Actual shape parameter for our model was chosen to be  = [-k] rounded to one tenth in order to maintain exponent in all equations of integer value for computational effectiveness.   = location parameter corresponds in Tab. 3 to  = -0,00306. It was chosen as  = 0 with negligible effect. Fig. 4: Jitter buffer packet loss Pjitter graph for different jitter. 6. Conclusion Proposed change in equipment impairment factor calculation leads to improved MOS estimate of E-model when network jitter is present. Proposed method is useful for MOS prediction under real network conditions with jitter. Discovered dependence of buffer packet loss at different jitter strengths for different buffer sizes is results in different MOS estimates for E-model and PESQ methods. Proposed equations and modifications to Emodel improve voice quality MOS estimate accuracy when network jitter is present. We use a simplified estimate to calculate expected packet loss on jitter buffer of the receiving device, which is superimposed in a multiplicative way to network packet loss. Resulting packet loss is commonly greater when the jitter buffer is not large enough. The model is able to give results, which are more in concordance with expected results of PESQ intrusive method of quality testing. Acknowledgements This work is a part of research activities conducted at the Slovak University of Technology Bratislava, Faculty of Electrical Engineering and Information Technology, Institute of Telecommunications, within the scope of the project “Prediction, modelling and assurance of quality of VoIP voice services” (PAMKVIS), supported by the STU University grant programme “Grant programme to support young researchers”. The support was also provided by the Center of Excellence for SMART Technologies, Systems and Services II., ITMS 26240120029, co-funded by the ERDF. References [1] KOVAC, A. and M. HALAS. Analysis of Influence of Network Performance Parameters on VoIP Call Quality. In: The conference proceedings of the Knowledge in Telecommunication Technologies and Optics: KTTO 2010. Ostrava: VSB-Technical University of Ostrava, 2010, pp. 2630. ISBN 978-80-248-2330-0. [2] KOVAC, A., M. HALAS, M. ORGON and M. VOZNAK. Emodel MOS Estimate Improvement through Jitter Buffer Packet Loss Modelling. Advances in Electrical and Electronic Engineering. 2011, vol. 9, no. 5, pp. 233-242. ISSN 13361376. [3] KOVAC, A., M. HALAS and M. ORGON. Determining Buffer Behaviour under Different Traffic Conditions as MMPP/D/1/K System. In: RTT 2011. Research in Telecommun Technology: 13th Int. Conference. Techov: Brno University of Technology, 2011, pp. 143-147. ISBN 978-80-214-4283-2. [4] ITU-G.107. The E-model: a computational model for use in transmission planning. Geneva: International Telecommunication Union ITU, 2009. [5] CHROMY, E. J. DIEZKA, M. KAVACKY and M. VOZNAK. Markov Models and Their Use for Calculations of Important Traffic Parameters of Contact Center. WSEAS TRANSACTIONS INFORMATION AND COMMUNICATION TECHNOLOGIES AND SERVICES VOLUME: 10 | NUMBER: 4 | 2012 | SPECIAL ISSUE © 2012 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 281 on COMMUNICATIONS. 2011, vol. 10, iss. 11, pp. 341 – 350. ISSN 1109-2742. [6] ITU-T G.133. Group-delay distortion: Appendix I: Provisional planning values for the equipment impairment factor Ie and packet-loss robustness factor Bpl. Geneva: International Telecommunication Union ITU, 1999. [7] CHROMY, E., T. MISUTH and M. KAVACKY. Erlang C Formula and Its Use in the Call Centers. Advances in Electrical and Electronic Engineering. 2011, vol. 9, no. 1, pp. 7-13. ISSN 1804-3119. [8] KOH, Y. and K. KISEON. Loss probability behavior of Pareto/M/1/K queue. IEEE Communications Letters. 2003, vol. 7, iss. 1, pp. 39-41. ISSN 1089-7798. DOI: 10.1109/LCOMM.2002.806469. [9] KASAHARA, S. Internet traffic modeling: a Markovian approach to self-similar traffic and prediction of loss probability for finite queues. IEICE Transactions on Communications: Special Issue on Internet Technology. 2001, vol. E84-B, no. 8, pp. 2134-2141. ISSN 1745-1345. [10] MIRTCHEV, S. and R. GOLEVA. Evaluation of Pareto/D/1/K Queue by Simulation. International Book Series: Information Science and Computing. Sofia: Institute of Information Theories and Applications FOI ITHEA, 2008. pp. 45–52. ISSN 1313-0455. [11] CHROMY, E. and M. KAVACKY. Asynchronous Networks and Erlang Formulas. International Journal of Communication Networks and Information Security (IJCNIS). 2010, vol. 2, no. 2, pp. 85-89. ISSN 2076-0930. [12] CROVELLA, M. E. and A. BESTAVROS. Self-Similarity in World Wide Web Traffic: Evidence and Possible Causes. IEEE/ACM Transactions on Networking. 1997, vol. 5, no. 6, pp. 835-846. DOI: 10.1109/90.650143. [13] KLUCIK, S. and A. TISOVSKY. Queuing Systems in Multimedia Networks. Elektrorevue. 2010, vol. 15, no. 99. ISSN 1213-1539. [14] ZHANG, W. and J. HE. Statistical Modeling and Correlation Analysis of End-to-End Delay in Wide Area Networks. In: Eighth ACIS International Conference on Software Engineering, Artificial Intelligence, Networking, and Parallel/Distributed Computing (SNPD 2007). Qingdao: IEEE, 2007, vol. 3. pp. 968-973. ISBN 0-7695-2909-7. DOI: 10.1109/SNPD.2007.490. [15] TORAL, H., D. TORRES, C. HERNANDEZ and L. ESTRADA. Self-Similarity, Packet Loss, Jitter, and Packet Size: Empirical Relationships for VoIP. In: International conference on electronics, communications and computers. Los Alamitos: IEEE Computer Society, 2008, pp. 11-16. ISBN 9780-7695-3120-5. DOI: 10.1109/CONIELECOMP.2008.20. [16] HUERLIMANN, Werner. From the General Affine Transform Family to a Pareto Type IV Model. Journal of Probability and Statistics. 2009, vol. 2009, no. ID 364901, pp. 1-10. ISSN 1687-952x. DOI: 10.1155/2009/364901 [17] CHOTIAPANICH. D. and B. C. ARNOLD. Pareto and Generalized Pareto Distributions. Modelling Income Distributions and Lorenz Curves: Economic Studies in Inequality, Social Exclusion and Well-Being. New York: Springer, 2008, vol. 5, pp. 119-145. ISBN 978-0-387-72756-1. DOI: 10.1007/978-0-387-72796-7. About Authors Adrian KOVAC was born in 1986 in Slovakia, has received master’s degree in Telecommunications in 2010 at the Slovak University of Technology, Faculty of Electrical Engineering and Information Technology in Bratislava, where he currently pursues his Ph.D. studies and provides lectures about communication network security and services. Adrian specialises in network security, performance testing and modelling of communication systems‘ behaviour. His current research topic is modelling and measurement of VoIP call quality and secure communication. Michal HALAS was born in 1978 in Slovakia. He graduated from Slovak University of Technology and received the electronic engineering degree in 2003. Since this year he has started postgraduate study at Department of Telecommunications STU Bratislava. In 2006 he received his Ph.D. from Slovak University of Technology Nowadays he works as a lecturer in Department of Telecommunications of Faculty of Electrical Engineering and Information Technology, Slovak University of Technology and topics of his research interests are Speech quality, IMS and IP telephony.