CHAPTER 1
PREAMBLE
1.1 Introduction
The ability to communicate properly over long distances has become an integral part of society today. Businesses are expanding to different regions in the world, but need to keep the same deadlines. This means it is necessary for employees in two different regions to communicate with each other over long distances, cheaply and trouble free. The public switched telephone network (PSTN) has developed itself to accommodate these requirements.
The Internet has become a very popular means of communication in a very short period of time. It was set up as a network where people could share files and access other people’s work. It has since established itself as a massive communications infrastructure that provides many services such as electronic mail. In the recent years it has further developed itself into providing Internet Telephony or Voice over Internet Protocol (VoIP). This allows users to make voice or video calls over the Internet. All the user needs is a computer with a network connection, a soundcard, and a microphone. VoIP enables a lot of big companies to combine their communications and their networking infrastructures. This is the biggest advantage that VoIP has over the regular telephone system. It means that voice and multimedia services are joined together. This means that a number of calls can be made on the one line, as well as having a multimedia broadcast.
The fact that you are putting elements that would use one line each, down a single line, means that costs are significantly cut in the management and leasing of lines.
There is even no need to change the communication infrastructure that already exists in the company. The companies PABX (private automatic branch exchange) only has to connect to a VoIP gateway, so IP calls can be made. Although VoIP seems to be taking off more with the corporate market the emergence and interest of the general public with Broadband should mean that IP telephony service could soon be implemented to its full extent in the home environment.
Quality of service, and loss issues are the major factors that are holding the growth of VoIP back. The issue is how to guarantee that packet traffic for a voice or other media connection will not be delayed or dropped due to interference from other lower priority traffic. Things to consider are Latency, Packet loss, Burstiness of Loss and Jitter. For efficiently supporting the QoS, a number of key areas are to be addressed. One key issue is Call admission Controller.
Call Admission Control is a concept that applies to voice traffic only-not data traffic. If an influx of data traffic oversubscribes a particular link in the network, queuing, buffering, and packet drop decisions resolve the congestion. The extra traffic is simply delayed until the interface becomes available to send the traffic, or, if traffic is dropped, the protocol or the end user initiates a timeout and requests a retransmission of the information.
CAC is therefore a deterministic and informed decision that is made before a voice call is established and is based on whether the required network resources are available to provide suitable QoS for the new call.
1.2 Statement of the Problem
In existing Call Admission Controller, the new call (admission) is accepted, only if codec specified in the call request is available, otherwise the call admission is rejected. As we know each codec consumes bandwidth based on the type of codec used. This CAC works on the principle of single-shot rejection. This principle of CAC leads to maximum rejection of calls and minimum utilization of bandwidth. In order to overcome these drawbacks we have designed a new CAC, which is called as Adaptive Call Admission Controller.
In proposed Call Admission Controller, the following process will takes place, the new call is established or admitted based on the availability of the network resources. When a new call request is placed, the CAC processes this request and admits the call if a codec specified in the call request is available, otherwise the CAC will offer the different codec based on the availability of total bandwidth. This principle of CAC shows that it utilizes the channel efficiently and avoids the rejection of calls. It utilizes the network resources like bandwidth efficiently, which guarantees the QoS of network.
1.3 Dissertation Objectives
This project is based on simulation studies of Adaptive Call Admission Controller in VoIP, which maximizes resource utilization and minimizes the call rejection by using adaptation method. Here we have used C ++ language for programming on Linux platform.
1.4 Scope of the Dissertation
The project is helpful for designing the Adaptive Call Admission controller in VoIP Networks from which we can give accurate service to the subscribers. Based on the simulation results we can design CAC, which utilizes the Network resources efficiently.
1.5 Methodology
In this dissertation, the following different aspects are considered and implemented.
1. Study of VoIP Systems- Uses, Working and Issues
2. Study of QoS- QoS Issues and QoS requirements in VoIP.
3. Study of CAC- Principles, Existing CAC’s
4. Design and Implementation of ACAC.
CHAPTER 2
VOICE OVER INTERNET PROTOCOL
2.1 Introduction
The term VoIP refers to the transfer of Voice over the Internet Protocol (IP) of the TCP/IP protocol suite. Using "VoIP" technology we can make traditional telephone calls from either computer or phone to other computer or phone using both public switched telephone network (PSTN) and internet (which is packet switched network). All you need is an Internet connection for VoIP. This technology really changes everything because it allows people to receive phone calls from anywhere that an internet connection exists, just in the same way you can receive your emails anywhere that you can connect to the internet.
The term "VoIP technology" covers a range of technologies, including voice-over-IP (VoIP) and fax-over-IP services, which are carried over both the Internet and private IP-based networks. VoIP is part of packet voice, which includes voice-over-asynchronous-transmission-mode (ATM) and frame-relay networks, which run faster than IP but are less common. VoIP connects across combinations of PCs, Web-based telephones, and phones connected via public telephone lines to remote voice gateways. Because information travels in discrete packets, it doesn't need to rely on a continuously available switched circuit.
Using VoIP we can enhance the traditional PBX by combining voice and data services onto a single network. The end user devices (also called client device) are normally referred to as VoIP phone are used in VoIP. Development of the 'VoIPphone' will require the development of a ' system on a chip' which combines digital signal processing (DSP) functions, micro-controller (MCU) functions, analog interface, telephone user interface and associated glue logic.
2.2 Uses of VoIP
VoIP service is deployed in enterprise and service provider network for various reasons. Most of these can be categorized into following.
Better bandwidth utilization by:
Using compression
Exploiting silence periods during conversations
Sharing of equipment for voice and data traffic (unified processing)
Introduction of new services:
Conferences, distance learning, etc.
2.3 Working of VoIP
The basic steps involved in originating an VoIP call are conversion of the analog voice signal to digital format and compression/translation of the signal into Internet protocol (IP) packets for transmission over the Internet; the process is then reversed at the receiving end -- switching the digital format back to analog so the telephone call goes through as normal.
VoIP calls originate on any broadband line: coaxial cable, DSL (Digital Subscriber Line), wireless or even satellite. The call is routed to the VoIP Company, where a computer converts the sound into data packets – similar to the packets used to transfer internet data such as email. Sending data by packets is far more efficient as it enables the same line to handle more information simultaneously. These data packets are sent through any of the Internet's multiple networks to a recipient of the call. The caller can receive the call via a wireless provider, a broadband provider, or a local phone carrier.
In order to understand VoIP it is essential to have a complete understanding of what the difference between circuit switching and packet switching. A normal telephone uses circuit switching for phone calls, which involves routing of your call through the switch at your local carrier to the person you is calling. The connection of two points in both directions is known as circuit. Packet switching on the other hand is more efficient in transmitting data since small amount of data, which is called a packet, is sent from one system to another.
In a VoIP system, once the called party answers, voice must be transmitted by converting the voice into digitized form, then segmenting the voice signal into a stream of packets. The first step in this process is converting analog voice signals to digital, using an analog-digital converter. Since digitized voice requires a large number of bits, a compression algorithm can be used to reduce the volume of data to be transmitted. Next, voice samples are inserted into data packets to be carried on the Internet. The protocol for the voice packets is typically the Real-time Transport Protocol (RTP). RTP packets have special header fields that hold data needed to correctly re-assemble the packets into a voice signal on the other end. However, voice packets will be carried as payload by UDP (User datagram protocol) protocols that are also used for ordinary data transmission.
2.4 ADVANTAGES
The following are some of the advantages of V0IP:
VoIP is cost effective - using VoIP products long distance phone calls and international calls can be made within the price of a local call. The caller simply connects to the Internet (with the price of a call to his local Internet provider) and using the appropriate software, calls other computers running similar VoIP applications, or even other telephones anywhere in the world. (Performing a PC-2-Phone conversation requires a VoIP Gateway to be present at the remote location).
A growing amount of communication operators throughout the world utilize VoIP as the modern communication method for long distance calls, enabling a full Phone-2-Phone conversation, which is carried over an IP network. (Fully transparent to the caller) Using the IP networks for voice transportation allows for a greater deal of phone calls to be made simultaneously, thus reducing the operators' costs. Furthermore, large companies can use their intranet as their internal enterprise phone network (iPBX). This enables lower maintenance fees, and cheap communication to remote sites and branches of the organization.
Convergence - By using VoIP, a company can have one comprehensive solution handling both data and telephone communication, all on the same platform and supported by a single vendor. This allows companies to use a single system for all their communication needs and prevents the overhead caused when dealing with several software packages and platforms.
Maintenance (Upgrades to existing services or introducing new services) can be easily done, as most applications are actually software based and do not require any hardware replacements and configuration.
Smart Net - Being software based, VoIP products and services enable various smart solutions. Achieving the same management capabilities on standard PSTN requires substantial hardware changes whilst most of the VoIP solutions can easily managed by a click of a mouse. An example could be the routing of a phone call to a subscriber in a predefined way: a schedule is set, and all phone calls are diverted to different locations (e.g. Home, Cellular, Business, and Voicemail) according to this schedule. Performing this capability in VoIP networks is trivial (Software solution), while achieving the same functionality on standard PSTN requires a great deal of effort.
New age multimedia - Because we treat voice as data and due to the fact that we can use voice services and telephony services from our PC, We can use voice applications as another application on our computer. In that way we can use the same hardware to browse the net, talk over the phone and work on other applications at the same time and without having to switch between devices. This idea is also part of the convergence advantage that was brought up here in this section.
Evolution towards better communication services - For all the reasons mentioned above, VoIP is an evolution towards better communication services. We can combine voice with streaming video for conference calls, allow better multimedia by using all sorts of web applications and offer customers with better communication services (such as the smart net) in order to get a communication package that will be adjustable and will be configured to supply every customer needs.
2.5 Main Issues of VoIP
For VoIP to become popular, some key issues need to be resolved. Some of these issues stem from the fact that IP was designed for transporting data while some issues have arisen because the vendors are not conforming to the standards.
The key issues are discussed below.
2.5.1 Interoperability
In a public network environment, products from different vendors need to operate with each other if Voice over IP is to become common among users. To achieve interoperability, standards are being devised and the most common standard for VoIP is the H.323 standard.
2.5.2 Security
This problem exists because in the Internet, anyone can capture the packets meant for someone else. Using encryption and tunneling can provide some security. The common tunneling protocol used is Layer 2 Tunneling protocol and the common encryption mechanism used is Secure Sockets Layer (SSL).
Integration with Public Switched Telephone Network (PSTN) While Internet telephony is being introduced; it will need to work in conjunction with PSTN for a few years. We need to make the PSTN and IP telephony network appear as a single network to the users of this service.
2.5.3 Scalability
As researchers are working to provide the same quality over IP as normal telephone calls but at a much lower cost, so there is a great potential for high growth rates in VoIP systems. VoIP systems need to be flexible enough to grow to large user market and allow a mix of private and public services.
VoIP technology can yield big cost savings to both corporations and consumers. It is more efficient than the plain old telephone service (POTS) and is poised to undergo huge growth. Before that growth can occur, however, designers have to address the issues listed above.
Along with the issues listed above, providing better voice quality to the customer is another major challenge. VoIP introduces a number of potential impairments that can impact voice quality adversely, such as the use of lossy low-bit-rate codec’s, the effects of tandem encoding/transcoding, longer delays, and packet loss. Most of these impairments are either not present or are negligible in circuit switched networks. Thus new techniques for delivering and maintaining voice quality are needed for VoIP networks. The impairments that a voice call experiences can be classified as either architectural or load dependent.
Architectural components include IP phone codec’s and their configuration parameter settings as well as fixed components of delay such as processing delays at each network element along the path and the end-to-end propagation delay. These architectural components define an upper bound on the best voice quality that could be achieved in a given network. If the upper bound is unacceptable, then changes in equipment and configurations will be required. In general, if the architecture is satisfactory, then low packet loss and delay are sufficient to ensure good voice quality.
Load dependent impairments include packet loss, queuing delay, and jitter. As load increases, these parameters deteriorate and begin to degrade voice quality. The voice quality a user experiences depends on the behavior of the entire end-to-end connection. This connection may cross multiple network domains each with its own set of controls and management methods. Since impairments across the connection are cumulative, it is possible that each network domain delivers acceptable voice quality while the end-to-end connection does not.
The networks service offering to the end applications can be measured quantitatively and qualitatively by means defining network Quality of Service. Managing voice Quality of Service across multiple domains requires SLAs (Service Level Agreements) between service providers and use of signaling protocols to indicate the desired QoS.
In the next chapter we have discussed about QoS, and requirements of QoS for voice. The minimum QoS requirements needed for better voice quality in VoIP networks.
CHAPTER 3
QUALITY OF SERVICE
3.1 What is QoS?
QoS is a set of tools and mechanisms available to network administrators to provide predictable service levels to IP packets as they transverse an IP network. Many protocols and applications are not critically sensitive to network congestion. File Transfer Protocol (FTP), for example, has a rather large tolerance for network delay or bandwidth limitation. To the user, FTP simply takes longer to download a file to the target system. Although annoying to the user, this slowness does not normally impede the operation of the application.
On the other hand, applications such as voice and video are particularly sensitive to network delay. If voice packets take too long to reach their destinations, the resulting speech sounds choppy or distorted. QoS can be used to provide predictable service levels to these applications. Critical business applications can also use QoS. Companies whose main business focus relies on Systems Network Architecture (SNA)-based network traffic can feel the pressures of network congestion. SNA’s handshake protocol is very sensitive to delay and normally terminates a session when it does not receive an acknowledgment in time. Unlike TCP/IP, which recovers well from a missed acknowledgment, SNA does not operate well in a congested environment. In these cases, providing low latency treatment for SNA traffic could be a proper approach to QoS.
3.2 Applications for Quality of Service
When would a network engineer consider designing QoS into a network?
Here are a few reasons to deploy QoS in a network topology:
To give priority to certain mission-critical applications in the network
To maximize the use of the current network investment in infrastructure
To provide better performance for delay-sensitive applications such as voice and video
To respond to changes in network traffic flows
Often we find that the simplest method for achieving better performance on a network is to throw more bandwidth at the problem. In this day and age of Gigabit Ethernet and optical networking, higher capacities are readily available. More bandwidth however, does not always guarantee a certain level of performance. It may well be that the very protocols that cause the congestion in the first place will simply eat up the additional bandwidth, leading to the same congestion issues experienced before the bandwidth upgrade.
A more judicious approach is to analyze the traffic flowing through the bottleneck, determine the importance of each protocol and application, and determine a strategy to prioritize the access to the bandwidth. QoS allows the network administrator to have control over bandwidth, latency, and jitter within the network.
Deploying certain types of QoS techniques can control four parameters (bandwidth, latency, jitter, and packet loss). Within many corporate networks today, QoS is not widely deployed. With the push for applications such as multicast, streaming multimedia, and VoIP, the need for certain quality levels is more inherent, especially because these types of applications are susceptible to jitter and delay. The end user immediately notices poor performance. End users experiencing poor performance typically generate trouble tickets, and the network administrator is left troubleshooting the performance problem. A network administrator can proactively manage new, sensitive applications by applying QoS techniques to the network.
It is important to realize that QoS is not the magic solution to every congestion problem. It may very well be that upgrading the bandwidth of a congested link is the proper solution to the problem. However, by knowing the options available, you will be in a better position to make the proper decision to solve congestion and quality issues.
3.3 QoS in Internet
QoS is defined as providing service differentiation and performance assurance for Internet applications. Service differentiation provides different services to different applications according to their requirements. Performance assurance addresses bandwidth, loss, delay and delay variation (jitter). Bandwidth is the fundamental network resource, as its allocation determines the application’s maximum throughput and, in some cases, the bounds on end-to-end delay. Jitter is a secondary quality-of-service metric, since a playout buffer at the receiver can transform it into additional constant delay.
In Internet QoS is ahead using two approaches. They are
A. Differentiated Service Approach
B. Integrated Service Approach
A). Differentiated Service Approach: A Differentiated service approach works on the principle of providing service differentiation on per packet basis. Each packet has information about QoS requirements in the form of encoded value in the packet header. This encoding is called as DSCP (differentiated service code points). Each node in the Internet honors this QoS requirement in per packet basis.
B). Integrated Service Approach: Integrated service approach works on the principle of providing service differentiation on per connection basis. An application that requires QoS guarantee initiates a connection establishment process, during which the QoS parameter negotiated. Every node in the Internet honors this QoS requirement for the packet belonging to the particular session.
3.4 QoS Requirements of VoIP
The barer (data traffic) and the signaling traffic of VoIP require different QoS. VoIP deployments require provisioning explicit priority servicing for VoIP bearer stream traffic and a guaranteed bandwidth service for Call-Signaling traffic. These related classes have been examined separately in [23].
Voice (Bearer Traffic)
A summary of the key QoS requirements and recommendations for Voice (bearer traffic) are:
Voice traffic should be marked to DSCP EF (Expedite Forwarding) per the QoS Baseline and RFC 3246.
Loss should be no more than 1 %.
One-way Latency (mouth-to-ear) should be no more than 150 ms.
Average one-way Jitter should be targeted under 30 ms. 21–320 kbps of guaranteed priority bandwidth is required per call (depending on the sampling rate, VoIP codec and Layer 2 media overhead).
Voice quality is directly affected by all three QoS quality factors: loss, latency and jitter.
Loss causes voice clipping and skips. The packetization interval determines the size of samples contained within a single packet, assuming a 20 ms (default) packetization interval, the loss of two or more consecutive packets results in a noticeable degradation of voice quality. VoIP networks are typically designed for very close to zero percent VoIP packet loss, with the only actual packet loss being due to L2 bit errors or network failures.
Excessive latency can cause voice quality degradation. The goal commonly used in designing networks to support VoIP is the target specified by ITU standard G.114, which states that 150 ms of one-way, end-to-end (mouth-to-ear) delay ensures user satisfaction for telephony applications. A design should apportion this budget to the various components of network delay (propagation delay through the backbone, scheduling delay due to congestion, and the access link serialization delay) and service delay (due to VoIP gateway codec and de-jitter buffer). If the end-to-end voice delay becomes too long, the conversation begins to sound like two parties talking over a satellite link or even a CB radio. While the ITU G.114 states that a 150 ms one-way (mouth-to-ear) delay budget is acceptable for high voice quality, lab testing has shown that there is a negligible difference in voice quality Mean Opinion Scores (MOS) using networks built with 200 ms delay budgets. Cisco recommends designing to the ITU standard of 150 ms, but if constraints exist where this delay target cannot be met, then the delay boundary can be extended to 200 ms without significant impact on voice quality.
Call-Signaling Traffic:
The following are key QoS requirements and recommendations for Call-Signaling traffic:
Call-Signaling traffic should be marked as DSCP CS3 per the QoS Baseline (during migration, it may also be marked the legacy value of DSCP AF31).
150 bps (plus Layer 2 overhead) per phone of guaranteed bandwidth is required for voice control traffic; more may be required, depending on the call signaling protocol(s) in use.
3.5 VoIP in Internet QoS perspective
The advantages of reduced cost and bandwidth savings of carrying voice-over-packet networks are associated with some quality-of-service issues unique to packet networks.
Quality of service is fundamental to a VoIP network’s operation. A VoIP application is much more sensitive to delays than its traditional data counterparts. When downloading a file, a few seconds’ slowdown is negligible. In contrast, a mere 150-millisecond delay can turn a crisp VoIP call into a garbled, unintelligible message.
In the VoIP vernacular, this is the latency problem. Latency turns traditional security measures into double-edged swords for VoIP. Tools such as encryption and firewall protection can help secure the network, but they also introduce significant delay. Latency isn’t just a QoS issue, but also a security issue because it increases the system’s susceptibility to denial-of-service (DoS) attacks. To succeed in a VoIP network, a DoS attack need not completely shut down the system, but only delay voice packets for a fraction of a second. The necessary impediment is even less when latency-producing security devices are slowing down traffic.
Another QoS issue jitter, refers to nonuniform delays that can cause packets to arrive and be processed out of sequence. The Real-Time Transport Protocol, which is used to transport voice media, is based on UDP, so packets received out of order can’t be reassembled at the transport level, but must be reordered at the application level, introducing significant overhead. Even when packets arrive in order, high jitter causes them to arrive at their destination in spurts. To control jitter, network designer’s can use buffers and implement QoS-supporting network elements (especially routers) that let VoIP packets “play through” when larger data packets are scheduled ahead of them. The buffers can use one of several strategies to determine when to release voice data, including several schemes that adapt the playout time during a conversation.
QoS also encompasses packet loss. In addition to the traditional packet loss issues associated with data networks; even VoIP packets that reach their destinations can be rendered useless by latency and jitter. Compounding the packet loss problem is VoIP’s reliance on RTP, which doesn’t guarantee packet delivery.
The good news is that VoIP packets are small, containing a payload of only 10–50 bytes, or approximately 12.5–62.5 ms, with most implementations at the shorter end of the range. The loss of such a minuscule amount of speech is indiscernible, or at least unworthy of complaint, by a human VoIP user.
The bad news is that these packets are rarely lost in isolation. Most causes of packet loss affect all packets being delivered around the same time. Therefore, although losing one packet is fairly inconsequential, probabilistically it means the loss of several packets, which severely degrades a VoIP network’s QoS. Packet losses as low as 1 percent can make a call unintelligible, depending on the compression scheme used. A 5-percent loss is catastrophic, no matter how good the codec. This sensitivity increases a DoS attack’s effectiveness. To be successful, a DoS attack needs to flood or disrupt the network only enough to stop 5 percent of packets from being delivered on time.
Thus, an enterprise’s hardware should support QoS to deliver VoIP traffic at high speed and with preference over less urgent data traffic. An enterprise can use routers that forward packets based on type-of-service bits, for example, or provide a separate queue for VoIP traffic. Anton Kos and colleagues significantly reduced jitter and latency using priority-based network elements. Call
Admission Control can help minimize packet loss by detecting network saturation and preventing VoIP packets from embarking on journeys they can’t complete. Victoria Fineberg covers CAC and other network-specific QoS issues in more depth.
3.6 Overcoming the issues
Quality of Service mechanisms are one way to address these types of issues by correctly prioritizing and handling traffic for applications sensitive to certain network conditions. QoS gives carriers and service providers a measure of reassurance that their networks can deliver voice, video and data packets to end customers in a timely fashion, although it does not eliminate the occurrence of network impairments that impact the service. In order to provide high levels of voice QoS, it is important to understand what happens to VoIP voice quality in a congested and overloaded network.
In a circuit-switched voice network, bearer traffic of each active call is allocated with dedicated time-slots at call set up time throughout the life of the call. However, in IP networks, the voice packets for different calls are multiplexed and share the packet forwarding capacity at each router. When new calls are admitted into an already congested network, both new and existing calls will experience increased packet loss and delay. Voice quality will degrade and eventually conversation will become impossible. Therefore it is critical to introduce mechanisms to mitigate the possibility of congestion.
An option often quoted for small networks and initial deployments with a limited number of subscribers are to overprovision the network – i.e., to provide a network capacity much beyond the traffic loads expected. While such an approach may be sufficient initially, it is not scalable or viable in the long run when users demand quality voice services.
A much more effective method in delivering guaranteed voice QoS in VoIP networks is to apply Call Admission Controller to limit the loading as well as to design the network capacity to meet the forecasted traffic and quality of service requirements including call blocking probability.
CHAPTER 4
CALL ADMISSION CONTROLLER
CAC is a deterministic and informed decision that is made before a voice call is established and is based on whether the required network resources are available to provide suitable QoS for the new call. The purpose of a Call admission controller is to decide, at the time of call arrival, whether or not a new call should be admitted into the network. A new call is admitted if and only if its Quality of Service constraints can be satisfied without jeopardizing the QOS constraints of existing calls in the network.
4.1 NEED OF CALL ADMISSION CONTROLLER
The call admission control function is an essential component of any VoIP system. In order to better understand what call admission controls does and why it is needed, consider the example shown in Figure 1.
Figure 1 Functioning of CAC in Circuit-switched and Packet-switched Network
As shown on the left side of Figure 1, traditional TDM-based (Time division multiplexing-based) PBXs operate within circuit-switched networks, where a circuit is established each time a call is set up. As a consequence, when a legacy PBX is connected to the PSTN or to another PBX, a certain number of physical trunks must be provisioned. When calls have to be set up to the PSTN or to another PBX, the PBX selects a trunk from those that are available. If no trunks are available, the PBX rejects the call and the caller hears a network-busy signal. Now consider the VoIP system shown on the right side of Figure 1. Because it is based on a packet-switched network (the IP network), no circuits are established to set up a VoIP call. Instead, the IP packets containing the voice samples are simply routed across the IP network together with other types of data packets. Quality of Service is used to differentiate the voice packets from the data packets, but bandwidth resources, especially on IP WAN links, are not infinite. Therefore, network administrators dedicate a certain amount of "priority" bandwidth to voice traffic on each IPWAN link. However, once the provisioned bandwidth has been fully utilized, the VoIP system must reject subsequent calls to avoid oversubscription of the priority queue on the IP WAN link, which would cause quality degradation for all voice calls. This function is known as call admission control, and it is essential to guarantee good voice quality in a multisite deployment involving an IP WAN. To preserve a satisfactory end-user experience, the call admission control function should always be performed during the call setup phase so that, if there are no network resources available, a message can be presented to the end-user or the call can be rerouted across a different network (such as the PSTN).
4.1.1 Call admission principles
As mentioned previously, call admission control is a function of the call-processing agent in an IP-based telephony system, so in theory there could be as many call admission control mechanisms, as there are IP-based telephony systems. However, most of the existing call admission control mechanisms fall into one of the following two main categories [4]:
• Topology-unaware call admission control — Based on a static configuration within the call-processing agent.
• Topology-aware call admission control — Based on communication between the call processing agent and the network about the available resources.
4.2 CAC in VoIP networks
A variety of QoS mechanisms other than CAC exist for the purpose of designing and configuring packet networks to provide the necessary low latency and guaranteed delivery required for voice traffic. These QoS mechanisms include tools such as queuing, policing, traffic shaping, packet marking, and fragmentation and interleaving. These mechanisms differ from CAC in the following important ways:
They are designed to protect voice traffic from data traffic contending for the same network resources.
They are designed to deal with traffic already present on the network. CAC mechanisms extend the capabilities of the QoS tool suite to protect voice traffic from being negatively affected by other voice traffic, and to keep excess voice traffic off the network. Figure 2 shows why CAC is needed. If the WAN access link between the two PBXs has the bandwidth to carry only two VoIP calls, admitting the third call will impair the voice quality of all three calls.
Figure 2 VoIP Network without CAC
The reason for this impairment is that the queuing mechanisms provide policing, not CAC, which means that if packets exceeding the configured or allowable rate are received, these packets are simply tail-dropped from the queue. There is no capability in the queuing mechanisms to distinguish which IP packet belongs to which voice call, so any packet exceeding the given arrival rate within a certain period of time will be dropped. Thus, all three calls will experience packet loss, which is perceived as clips by the end users.
Figure 3 VoIP Network with CAC
Figure 3 illustrates the point at which a CAC decision is reached by the outgoing gateway that insufficient network resources are available to allow a call to proceed.
After the call is rejected, the originating Gateway must find another means of handling of handling the call. There are several possibilities, most of which are dependent on the configuration of the Gateway. In the absence of any specific configuration, the outgoing Gateway provides a reorder tone to the calling party. The reorder tone is called fast busy.
4.3 Existing CAC algorithms
We have many Call admission controller algorithms, some of them are
1. Site-Utilization-Based CAC (SU-CAC).
2. Link-Utilization-Based (L¬U-CAC).
Site-Utilization-Based CAC (SU-CAC): The basic idea of SU-CAC is to do admission control based on the Bandwidth, which is pre-allocated to the sites. Bandwidth pre-allocation is performed at the configuration time (i.e., at off-line). A New arrival call can be admitted if there is enough bandwidth left for the related site, otherwise the call will be rejected. In this strategy, the site can be a Zone to the Gatekeeper (GK).
A Zone is a collection of H.323 endpoints that register to the same GK. The core of SU-CAC is how to do bandwidth pre-allocation to the sites. Bandwidth pre-allocation determines the certainty of QoS that a VoIP system can provide to voice in IP networks. Unfortunately, so far, the VoIP system does not define a proper way to do that. Currently, Bandwidth pre-allocation is performed in a very ad-hoc manner. As a matter of fact, it is the reason why the end-to-end QoS guarantees cannot be achieved in current VoIP system.
The main advantage of SU-CAC is simple, and the admission control can be performed in a fully distributed fashion. It neither sends the probes to test the availability of resources nor dispatches messages to make reservations. However, since the bandwidth has been pre-allocated to the sites at the configuration time, links cannot be fully shared by dynamic calls, and accordingly the high network resource utilization cannot be achieved.
Link-Utilization-Based (LU-CAC): The LU-CAC aims to address this issue. The main idea of LU-CAC is to do admission control directly based on the availability of the individual link bandwidth. With this mechanism, call multiplexing can be performed at the link level; hence the high network resource utilization can be obtained. The disadvantage of LU-CAC is its complexity.
The current VoIP system has to rely on the resource reservation protocols, such as RSVP to do explicit resource reservation within the whole network. To achieve that, all the routers within the network should support resource reservation, which is not practical. Also in the current high-speed network, there are potentially thousands of flows passing through the core-routers. The overhead of the core-routers within the network to support resource reservation is large. The overhead of resource reservation at the core-routers will compromise their main function, i.e., packet forwarding, which will degrade the whole network performance.
In existing Call admission controller systems, when a new call request is placed, the parameters of the call request is sent to the gatekeeper, the gatekeeper processes the call request and forwards the call request messages to the CAC. The main function of the Gatekeeper is to perform the address translation and endpoint location. After processing, the Gatekeeper offers the available codec’s to the CAC. The CAC after receiving the call request messages, it will decide whether to admit or to reject the call based on the availability of the total bandwidth.
In existing Call Admission Controller systems the call is admitted if a particular codec requested by the H.323 endpoint is available .As we know each codec has a particular bandwidth, for example, G.711-alaw requires bandwidth 64 kbps. If particular codec is not available then the call is rejected. It works on the principle of single-shot rejection. The existing Call Admission Controller systems rejects the maximum calls (call rejection ratio is high) and utilization of the link is not efficient, i.e., minimum channel utilization.
4.4 How to Measure CAC algorithm
An efficient Call Admission Controller algorithm is one that can accept calls such that the link utilization is always 100 %. However, due to variations in the resource request and the call arrival intervals, the admission controller could not be able to satisfy the 100 % link utilization.
Thus a good measure of CAC algorithm is the link utilization, higher the link utilization betters the CAC algorithm.
CHAPTER 5
PROPOSED SYSTEM
We have designed and developed an Adaptive Call Admission Controller for VoIP system. The objective of the ACAC is to minimize the call-rejection and to maximize the link-utilization. The ACAC works on the principle of multiple-shot rejection.
5.1 Multiple-Shot Rejection
The existing Call Admission Controller works on the principle of single-shot rejection. It means, when a client places a call request with some parameters, the CAC in gatekeeper processes this request and admits the call if and only if the parameters specified in the call request are available, otherwise the call is rejected. This process of call admission is called single-shot rejection.
However, in Adaptive Call Admission Controller, when a call request from a client comes, the ACAC in gatekeeper (in this Project) admits the call if the parameters are available. If specified parameters are not available, then instead of rejecting the call, ACAC in the gatekeeper offers different parameters based on the availability of the resources. The offered parameters are sent to the call- originator, the originator checks whether the offered parameters are supported by the call. If supported then the call is admitted with offered parameters. This process of admitting the call is called multiple-shot rejection.
Figure 4 Proposed ACAC
The above diagram shows various messages that are exchanged between the ACAC, gatekeeper and Ohphone. They are explained below in steps:
1. First Ohphone sends the Admission request (ARQ) to the Gatekeeper.
2. The Gatekeeper receives the request and forwards that request to the ACAC.
3. The ACAC checks the availability of requested resources.
4. If available then the ACAC sends the ACF (Admission Confirm) message to the gatekeeper. If not available then the ACAC will send the different resources to the gatekeeper.
5. The gatekeeper then forwards that message to the Ohphone 1.
6. A) If Ohphone 1 receives ACF from the GK, then Ohphone 1 establish a connection to Ohphone 2 (Go to step 8 directly).
B) Or if it receives the offered resources from the ACAC, then it checks whether it supports or not. If Ohphone 1 supports the offered resources from ACAC, then Ohphone 1 sends the accepted message to the GK.
7. Then gatekeeper forward the message received from the ohphone 1 to ACAC.
8. Then Ohphone 1 will establish a connection to the Ohphone 2.
5.2 ADAPTIVE CALL ADMISSION CONTROLLER ALGORITHM
The steps followed in the proposed adaptive call admission controller algorithm are as follows.
Steps:
1. ACAC receives the call request from the endpoint through GK.
2. ACAC checks the codec bandwidth specified in the new call request.
3. Admits the call if codec bandwidth is less than the total bandwidth.
4. Otherwise ACAC offers different bandwidth to the endpoint by using adaptive method (By performing codec negotiation).
5. Admits the call if endpoint accepts the offered bandwidth from ACAC, otherwise reject the call.
We have assumed the following parameters..,
TB is Total Bandwidth available.
B is the requested Bandwidth from the endpoint.
A call is admitted if the following condition is verified:
Figure 5 Block diagram of Proposed Call Admission Control Algorithm
The block diagram of proposed call admission controller algorithm is shown in the figure 6. Initially an ohphone which acts a H.323 terminal (Client) sends call request to the gatekeeper (GK which acts as server in this project). The gatekeeper forwards that call request to the ACAC algorithm. The call admission control algorithm processes that call request. It checks the availability of bandwidth. If the total bandwidth specified in the configuration file is greater than the bandwidth requested by the H.323 terminal (the bandwidth of Specified codec) then the call is admitted. The total bandwidth is set in the gatekeeper configuration file.
If the total bandwidth specified in the ACAC is less than the bandwidth requested by the H.323 terminal, then instead of rejecting the call, the proposed ACAC offers the default bandwidth based on the availability of the total bandwidth to the H.323 terminal by negotiation. If the terminal accepts the new bandwidth offered by the ACAC, then call is admitted. If terminal rejects the new bandwidth then call is rejected. After every call admission the total bandwidth is decremented. Where T.B is a total bandwidth and B is the codec’s bandwidth.
5.2.1 ACAC call flow example.
Consider two Endpoints OHPHONE A and OHPHONE B as shown in figure 7. Suppose OHPHONE A wants to establish a connection with OHPHONE B in VoIP network. The gatekeeper is used here as mediator between this two endpoints. The procedure is as follows:
First both endpoints should discover the gatekeeper and get register with the gatekeeper.
Figure 6 ARQ from Ohphone to Gatekeeper
Figure 7 Call Admission process
Then OHPHONE A places a call admission (ARQ) request. The gatekeeper receives and processes this request. The processed request is sent to CAC in gatekeeper. Then the gatekeeper offers the available codec’s e.g.: A, B, C, and D..., to CAC based on availability of total bandwidth. As we know each codec require some bandwidth for e.g.: G.711-Alaw requires 64 kbps.
The CAC will decide which codec to offer for a particular call request. In proposed ACAC scheme, if an OHPHONE A requests call admission with a specified codec G.711, the ACAC processes that request and offers the particular codec if it is available, then call is admitted, otherwise if a particular codec is not available, in existing CAC the call is rejected but in proposed CAC instead of rejecting the call admission, the ACAC will offer the available codec’s to the OHPHONE A based on the bandwidth availability by performing codec negotiation. If OHPHONE A accepts the codec offered by the ACAC, then the call is admitted or otherwise it is rejected.
Figure 7 shows the steps followed to establish a call between two endpoints.
