Simulation results of
Abstract
The paper deals with admission control methods used in Internet Protocol (IP) Multimedia Subsystem. The purpose of implementing AC methods in IP Multimedia Subsystem (IMS) is to control the access of incoming connections to network resources. At the Institute of Telecommunications, we have built an experimental laboratory, which is used to test these methods. In this paper, we focus on Internet Protocol Television (IPTV) services; therefore, we have created a variable bit-rate IPTV traffic generator that is used as the input to the network, so we can test the behavior of selected AC methods. They are implemented in a simulated IPTV service provider access network, so we can examine the effects of variable bit-rate IPTV streams on the decisions made by those methods. To calculate the required bandwidth of an input stream, two simulation scenarios with different number of input packets were performed. One of these AC methods was modified where the peak input rate parameter of an IPTV stream was replaced by the average bit-rate of this stream. At the end of this paper, we discuss the achieved results.
Keywords
- admission control
- IP Multimedia Subsystem
- IPTV
1. Introduction
The usage of admission control (AC) methods in connection-oriented network is well known (e.g. CAC methods in ATM networks). But there is also a need to study AC methods in connectionless networks. The Resource and Admission Control Sub-system (RACS) block in an IMS network [1, 2] is responsible for admission and resource control. The functional architecture of RACS block is described in the standard document ETSI ES 282 003 v1.1.1. It is one of the most important blocks of the IMS architecture, and it decides whether a service or connection will be accepted or rejected. The document itself or the available scientific literature does not mention which admission control method or algorithm should be implemented in the RACS block.
The main task of AC methods is to provide sufficient bandwidth for each user service so that required Quality of Service (QoS) [3, 4] will be ensured. AC methods are defined within an IMS network node through the users who are accessing services.
A decision rule is the important part of an AC method. Whether a new request to the network will be accepted depends on the ability of the AC method to retain the QoS for both, existing services [5–7] and the new service that requests additional network resources.
Various methods for Quality of Service rating are used, e.g. subjective or objective. Subjective methods are based on feelings of users during the service provision. For admission control methods, we need to define objective parameters. The main goal of telecommunication operators is to ensure QoS parameters [8, 9] on required levels together with effective bandwidth utilization. The admission control takes a key role in service provisioning (VoIP, IPTV). A wrong AC decision can inflict degradation of QoS parameters for existing [10–12] and for newly accepted data flows.
The key feature of every AC method is the ability to precisely estimate the required bandwidth of an incoming data flow. This estimation is usually based on a theoretical analysis of the network traffic and its accuracy is limited by various simplifications that are used. For example, such a simplification is the use of constant packet lengths or constant times between consecutive packets within a stream [13]. There are many admission control methods and they can be classified into two groups:
Parameter-based admission control (PBAC) methods and
Measurement-based admission control (MBAC) methods.
In these papers [14, 15], various methods and algorithms for admission control have been proposed.
2. Simulation model
The purpose of our simulations is to verify the proposed IPTV traffic generator and to identify suitable AC method for IPTV services. Simulations were realized in the MATLAB environment. Input data flows were generated using the IPTV generator defined in [16]. The principle of simulations is depicted in Figure 1.
The users generate requests for IPTV streams that are received in the network node (router) which uses a defined AC method. Only one request from one user can originate at a time. At the beginning of the simulation, there were no users connected into the test network. If any connection request is rejected by admission control method, then every new connection request is also rejected.
2.1. Simulation principles
For simulation purpose, we need to convert packet departure times into transmission rate of IPTV flows. For this conversion, we need to know the packet size and number of packets sent per defined time interval. The ratio of these two values gives transmission rate. For conversion, it is important to know how frequently the router calculates the parameters of data flows. In our simulations, we used two versions of conversion—conversion for every 1000 packets (
Four simulations of AC methods were performed—
3. Simulations
The following simulation parameters were observed and evaluated:
Number of accepted connections
Average link utilization (%)
Loss (%).
Each method was evaluated for both versions of transmission rate conversion of input data flows—transmission rate conversion for every 1000 packets (
If any connection request is rejected by admission control method, then every new connection request is also rejected. From that moment (in the graphs depicted by vertical black line), the observed parameters are evaluated.
3.1. Simulation of Measured Sum method
Acceptance of a new requesting connection into the network is based on Eq. (1):
where
Version | Accepted connections | Link utilization (%) | Loss (%) |
---|---|---|---|
A | 155 | 95.42 | 0.25634 |
B | 141 | 91.83 | 0.704579 |
The resulting data flows for version A are depicted in Figure 2.
3.2. Simulation of Hoeffding Bound method
Acceptance of the new requesting connection into the network is based on Eq. (2):
For parameter
are current bandwidths of already accepted connections in the given time moment. Parameter
Version | Accepted connections | Link utilization (%) | Loss (%) |
---|---|---|---|
A | 137 | 83.83 | 0 |
B | 125 | 80.78 | 0.0238593 |
3.3. Simulation of Acceptance Region method—variant 1
Acceptance of the new requesting connection into the network is based on Eq. (4):
Parameter
Version | Accepted connections | Link utilization (%) | Loss (%) |
---|---|---|---|
A | 138 | 85.02 | 0 |
B | 126 | 82.81 | 0.03564 |
3.4. Simulation of Acceptance Region method—variant 2
Acceptance of a new requesting connection into the network is based on Eq. (5):
For this method, two simulations were performed. For the first simulation, the theoretically described parameters were used. For the second simulation, the calculation of parameter
Version | Accepted connections | Link utilization (%) | Loss (%) |
---|---|---|---|
A | 137 | 83.83 | 0 |
A | 138 | 85.02 | 0 |
B | 124 | 80.78 | 0.023832 |
B | 135 | 87.91 | 0.247054 |
At first glance, it is a small change, but the simulation results are different. The version
4. Evaluation and comparison of simulation results
Simulation results are stated in Table 5. It is proven that the number of accepted connections into the network together with evaluation of link parameters depends on the conversion interval of parameters of input data flows. Obtained results of parameters for
AC method | Version | Accepted connections | Link utilization (%) | Loss (%) |
---|---|---|---|---|
A | 155 | 95.42 | 0.25634 | |
B | 141 | 91.83 | 0.704579 | |
A | 137 | 83.83 | 0 | |
B | 125 | 80.78 | 0.0238593 | |
A | 138 | 85.02 | 0 | |
B | 126 | 82.81 | 0.03564 | |
A | 137 | 83.83 | 0 | |
A | 138 | 85.02 | 0 | |
B | 124 | 80.78 | 0.023832 | |
B | 135 | 87.91 | 0.247054 |
Based on the simulation results, we can suggest
5. Conclusion
The paper deals with admission control methods in IMS networks. We have simulated four admission control methods—
In the future work on our experimental IMS laboratory, we intend to implement selected admission control methods into the access part of the IMS network architecture. Then, we will evaluate implemented method in real time for IPTV services.
Acknowledgments
This article was created with the support of the Ministry of Education, Science, Research, and Sport of the Slovak Republic within the KEGA agency project - 007STU-4/2016 “Progressive educational methods in the field of telecommunications multiservice networks” and VEGA agency project - 1/0462/17 “Modeling of qualitative parameters in IMS networks.”
References
- 1.
Al-Begain K, Ch B, Galindo LA, Fernandez DM. IMS: A Development and Deployment Perspective. United Kingdom: John Wiley & Sons; 2006. p. 316 ISBN: 978-0-470-74034-7 - 2.
Chiang W-K, Kuo P-C. IMS-based centralized service continuity. Wireless Personal Communications. 2013; 68 (3):1177-1195. DOI: 10.1007/s11277-012-0503-z - 3.
Frnda J, Voznak M, Sevcik L. Impact of packet loss and delay variation on the quality of real-time video streaming. Telecommunication Systems. 2016; 62 (2):265-275. DOI: 10.1007/s11235-015-0037-2 - 4.
Mikulec M, Fajkus M, Voznak M, Johansson O. Multiple transcoding impact on speech quality in ideal network conditions. Advances in Electrical and. Electronic Engineering. 2015; 13 (5):552-557. ISSN: 1336-1376. DOI: 10.15598/aeee.v13i5.151 - 5.
De Rango F, Tropea M, Fazio P, Marano S. Call admission control with statistical multiplexing for aggregate MPEG traffic in a DVB-RCS satellite network. In: Proceedings of GLOBECOM '05. IEEE Global Telecommunications Conference; 2005; St. Louis, MO; 2005. p. 3231-3236 - 6.
Fazio P, Tropea M, Veltri F, Marano S. A novel rate adaptation scheme for dynamic bandwidth management in wireless networks. In: IEEE 75th Vehicular Technology Conference (VTC Spring), Yokohama; 2012. p. 1-5 - 7.
Tropea M, Veltri F, De Rango F, Santamaria AF, Belcastro L. Two step-based QoS scheduler for DVB-S2 satellite system. In: IEEE International Conference on Communications (ICC), Kyoto; 2011. p. 1-5 - 8.
Kovac A, Halas M, Orgon M. E-model MOS estimate improvement through jitter buffer packet loss modelling. Advances in Electrical and Electronic Engineering. 2011; 9 (5):233-242 - 9.
Roka R, Certik F. Simulation and analysis of the signal transmission in the optical transmission medium. In: 5th International Conference on Simulation and Modeling Methodologies, Technologies and Applications, SIMULTECH 2015; Colmar; 2015. p. 219-226 - 10.
Hegr T, Bohac L, Kocur Z, Voznak M, Chlumsky P. Methodology of the direct measurement of the switching latency. Przeglad Elektrotechniczny. 2013; 89 (7):59-63 - 11.
Voznak M, Kovac A, Halas M. Effective packet loss estimation on VoIP jitter buffer. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). 2012; 7291 :157-162. DOI: 10.1007/978-3-642-30039-4_21 - 12.
Voznak M. Speech bandwith requirements in IPsec and TLS environment. In: Proceedings of the 13th WSEAS International Conference on Computers - Held as part of the 13th WSEAS CSCC Multiconference; 2009. p. 217-220 - 13.
Davy A, Botvich D, Jennings B. Revenue optimized IPTV admission control using empirical effective bandwidth estimation. Broadcasting, IEEE Transactions. 2008; 54 (3, Part: 2):599-611 - 14.
Frank K, Zachary S, Ziedins I. Notes on effective bandwidths. In: Stochastic Networks: Theory and Applications, Royal Statistical Society Lecture Notes Series. New York, NY, USA: Oxford University Press; 1996. p. 141-168 - 15.
Yi-ran G, Suo-ping W, Hai-ya W. A structural comparison of measurement-based admission control algorithms. The Journal of China Universities of Posts and Telecommunications. 2006; 13 (3):81-86 - 16.
Klucik S, Lackovic M. Modelling of H.264 MPEG2 TS traffic source. Advances in Electrical and. Electronic Engineering. 2013; 11 (5):404-409. E-ISSN: 1804-3119, ISSN: 1336-1376. DOI: 10.15598/aeee.v11i5.870