Comparative Study for Performance Analysis of VOIP Codecs Over WLAN in Nonmobility Scenarios

Voice over IP (VoIP) applications such as Skype, Google Talk, and FaceTime are promising technologies for providing cheaper voice calls to end users over extant networks. Wireless networks such as WiMAX and Wi-Fi focus on providing perfection of service for VoIP. However, there are numerous aspects that affect quality of voice connections over wireless networks. The adoption of Voice over Wireless Local Area Network is on tremendous increase due its relief, non-invasive, economic expansion, low maintenance cost, universal coverage and basic roaming capabilities. However, expansion Voice over Internet Protocol (VoIP) over Wireless Local Area Network (WLAN) is a challenging task for many network specialist and engineers. Voice codec is one of the most critical components of a VoIP system. In this project, we evaluate the performance analysis of various codecs such as G.711, G.723 and G.729 over Wi-Fi networks. NS2 WiFi simulation models are designed. Performance metrics such as Mean Opinion Score (MOS), average end-to-end latency, and disconcert are evaluated and discussed.<br><br>1. In this paper, our area of interest is to compare and study the performance analysis of VoIP codecs in Non-mobility scenarios by changing some parameters and plotting the graphs throughput, End to end Delay, MOS, Packet delivery Ratio, and Jitter by using Network Simulator version.<br><br>2. In this paper we analyze the different performance parameters, Recent research has focused on simulation studies with non- mobility scenarios to analyze different VoIP codecs with nodes up to 5. We have simulated the different VoIP codecs in non-mobility scenario with nodes up to 300.

1. In this paper, our area of interest is to compare and study the performance analysis of VoIP codecs in Non-mobility scenarios by changing some parameters and plotting the graphs throughput, End to end Delay, MOS, Packet delivery Ratio, and Jitter by using Network Simulator version. 2. In this paper we analyze the different performance parameters, Recent research has focused on simulation studies with non-mobility scenarios to analyze different VoIP codecs with nodes up to 5. We have simulated the different VoIP codecs in non-mobility scenario with nodes up to 300.

KEYWORDS:
Voice over Internet Protocol, Wireless Local Area Network (WLAN), Codecs, throughput, delay.

INTRODUCTION:
The recent Voice over IP (VOIP) applications such as Skype, Google Talk, and Face Time have changed the way people communicate to each other. Due to the low cost, people find VOIP as an alternative to the expensive traditional Public Switched Telephone Network (PSTN). VOIP has set of parameters that defined its Quality of Service (QoS) such as end to delay, jitter, packets loss, Mean Opinion Score (MOS, and throughput [13]. The existing wireless networks such as Wi-Fi offer flexibility to support such applications. At the time the IEEE 802.11 (Wi-Fi) technology showed great success as cheap wireless internet access. The Motive of this survey paper is to analyse of Qos in VOIP [13].

VOIP
Voice over IP (VoIP) is the real-time application that is probably the most widely-spread on today's networks. I'll provide here some basic facts related to VoIP. Figure below shows the endto-end path as needed for VoIP communication (a similar path exists in the opposite sense for a bi-directional connection). An audio input device, such as a microphone, is required at the sending end. The audio signal is appeared into digital form by an analog-to-digital converter. Due to the packet-switched nature of computer networks, voice data has to be packetized and encoded prior to being transmitted. Encoding (as well as decoding) is done by codecs that transform sampled voice data into a specific network-level representation and back. Most of the codecs are defined by standards of the International Telecommunication Union, the Telecommunication division (ITU-T) [14].
Each of them has different properties regarding the amount of bandwidth it requires but also the comprehend quality of the encoded speech signal. After binary information is encoded and packetized at the sender end, packets enclose voice data can be pass on the network. Voice packets interact in the network with other application packets and are routed through shared relation to their destination. At the receiver end they are decapsulated and decode. Decode may include other steps as well, the most typical being dejittering. Other examples are defect improvement and packet loss niche. The flow of digital data is then converted to analogue form again and played at an output device, usually a speaker [14].

WI-FI
Wi-Fi is commonly used in residential, business, and public areas. It is notable that the perceived throughput in Wi-Fi does not match the real throughput. Also, all users share the access to the channel which is very critical for all real time traffic in general and especially VoIP. The Wificonnection with low capacity has serious impact on Qos. Beside the high traffic generated by users, both protocols, VoIP and Wi-Fi, create large headers which result in high drawback on VoIP performance [14]. Fig.1 End-to-end data path for VoIP communication [13]

QOS ISSUES OF VOIP APPLICATION
Quality of Service (QoS) is what determines if a technology can successfully deliver high value services such as voice and video. QoS is referred as the ability to control the mixture of bandwidth, delay, jitter, and packet loss in a network in order to deliver a network service Ensuring high voice call quality over VoIP traffic. QoS for VoIP is defined using different parameters. The BE class which is used for data stream with no support for latency and PDR. In this paper, the quality of services for VoIP is measured in terms of end to end latency, jitter and data loss. Latency is defined as the time required for a frame or a packet to travel from the source to its final destination [13].
The main source of delay is categorized into: Devolution Delay and receiver processing delay, efficiency calculation inoperable or inadequate, technological impact, sequencing the packets which outcomes queuing delay etc. An absolute value of delay difference between shortlistedpackets to arrive at receiver is called as jitter. It is not guaranteed that all the packets will follow similar route and confrontation the similar routes to reach the target over the network, and added with the backlogin the network normally resulting in data frames outcomeof order and with shift latency. No disconcert means a network with constant latency and no modification. The amount of data that can actually be pass over the movement channel is called flow capacity. It is used to Estimate the competence of network. The ratio between the quantum of information and the sum of use data, control data and retransmitted data if error is concluded as throughput of a network [10].
Nowadays, people get advantage of the existing data networks by enjoying various ways of communication e.g. text messages, voice calls, and video calls. The traditional phone networks cannot compete with these type of services due to low equipment's and operating cost, and the ability of integrating voice and data in applications. The QoS for VoIP can be measured by evaluating three performance metrics: Mean Opinion Score (MOS), Jitter, and end-to-end delay [18].

MEAN OPINION SCORE (MOS):
MOS is a scale from 1 to 5 which measures the quality of the voice.

JITTER:
The variation in arrival time of consecutive packets is called jitter. In Predecoding, packets should get deliver and some are not deliver in particular size buffered. Jitter should determine by latency of packets over a interval of time.

PACKET END-TO-END DELAY:
The amount of packet sent from source to destination which has measured in interval of time.It includes network latency, code and decode delay and data shrinkage.
Electronic copy available at: https://ssrn.com/abstract=3389769 Table 1 shows the guidelines for voice quality measurement for both jitter and end-to-end delay as it is provided by ITU Telecommunication Standardization Sector (ITU-T).

THROUGHPUT:
The throughput corresponds to the amount of data in bits that is transmitted over the channel per unit time. Generally the throughput measured in bits per second (bps).

SIMULATION SET UP:
Exponential traffic voice (created packet while talk period for 1.00ms and no packet is created during 1.3ms of silent period) is a scenario in the implementation using SIP to fulfill sending voice packet from end to end nodes.

TOPOLOGY
Considering that, the network topology consists of 1 main base station, along with node for sending voice n_voice and node for receiving call is n_null for each server, which means n_voice1, n_voice2, n_null1 and n_null2. In addition, one node for testing, cooperating with each server is attached with sending or receiving node (n_test1 and n_test2).

RESULT AND DISCUSSION:
In this Section, we compare the capabilities of the three VOIP codec's studied in this paper. To evaluate more reliable performance of G.711, G.723 and G.729 VoIP codec's in same simulation environment (50 to 300 mobile nodes). Performance metrics are calculated from trace file, with the help of AWK program. The simulation results shown in the following section in the form of line graph with description. The result shows the comparison between the three codecs on different QoS parameters in a VoIP network with Non-mobility scenarios.

GRAPHS OF NON-MOBILITY SCENARIO IN WI-FI:
8.1 NON-ADAPTATION:

DELAY
From the below figure and table Delay was gained at destination node against various dimension of networks and varied the simulation time uniformly for each codec. This data may be delivered over a physical or logical link, or pass through a certain network node. It is clear that G.729 gives less delay when the nodes are less. G.711 and G.723 codecs gives less delay. G.729 had a high delay. From these graphs it is clear that delay increase with increase in non-mobility nodes. Electronic copy available at: https://ssrn.com/abstract=3389769  Electronic copy available at: https://ssrn.com/abstract=3389769     Electronic copy available at: https://ssrn.com/abstract=3389769  Average voice traffic sent and received is presented in Figure above. Any network to be more efficiency these two traffic must be equal. The traffic received by the network with codec G.729 is less deviated from the traffic sent comparatively with codec G.723 and G.711. This analysis indicates that the noise added in the G.729 network is less when compared to the other networks, so this codec is more efficient. Electronic copy available at: https://ssrn.com/abstract=3389769 8.2 ADAPTATION:

DELAY
From the below figure and table Delay was gained at destination node against various dimension of networks and varied the simulation time uniformly for each codec. This data may be delivered over a physical or logical link, or pass through a certain network node. it is clear that G.729 gives more delay when the nodes are more. G.711 and G.723 codecs gives less delay. G.729 had a high delay. From these graphs it is clear that delay decrease with increase in non-mobility nodes.   Electronic copy available at: https://ssrn.com/abstract=3389769 approximately varying throughout the simulation. The average voice throughput variation in case of codec G.723 is lower than the other two codecs at the earlier time of simulation. The PDR variation in case of G.729 lies between two other audio codecs. So audio codec G.723 gives better results than audio codecs G.711 and G.729 respectively.   So audio codec G.729 gives better results than audio codecs G.711 and G.723 respectively.

JITTER
From the below figure and table, the variation of the codec is minimum and approximately constant throughout the simulation. The average voice jitter variation in case of codec G.711 is higher than the other two codecs at the earlier time of simulation. But after some time it falls down. The jitter variation in case of G.729 lies between two other audio codecs. The voice jitter threshold for smooth communication in VOIP network is about 1ms so audio codec G.729 gives better results than audio codecs G.711 and G.723 respectively.

CONCLUSION:
In this project, we evaluated the performance of three different VoIP codecs over Wi-Fi networks. The VoIP performance is measured in three scenarios using NS2 MOS, jitter, and average end-toend delay are analysed as performance parameters which define QoS in VoIP. G. 711 codec showed the better codec for VoIP over Wi-Fi.

FUTURE WORK:
This project considered voice calls from fixed nodes. The impact of mobility on VoIP performance is suggested as future research. Also, for future study, various codecs must be investigated on Wi-Fi networks to observe the performance of VoIP and the QoS parameters should be improved to get the maximum throughput and PDR, minimum delay and jitter.