Voice Over Internet Protocol (VoIP) - Internet Telephony

Implementation on Linux environment and practical issue resolution.
Voice Over Internet Protocol (VoIP) - Internet Telephony
Source: https://pixabay.com/images/id-568056/

**NOTE: I wrote this article in 2006 during my master's studies. Therefore, references and links are old and may not represent the current status. It got based on an old 3G technology. Current 5G would not have the issues mentioned in this literature review.**

Abstract:

The telephone has become a major cost to businesses.  To control cost more businesses and individuals are adapting to Internet and/or other hybrid networks.  Voice over Internet Protocol (VoIP) uses the Internet Protocol (IP) data network and/or hybrid networks that can be used to operate as a telephone. However, VoIP is very demanding and challenging web-based technology. VoIP offers several key advantages including cost-saving, increasing organization efficiency, promoting interoperability within the different platforms, and is easy to manage as a single network for the transmission of voice and data. However, VoIP has problems such as reliability, predictability, standards and security, privacy, emergency calling, and backup power for VoIP-handsets to hinder the prevalent deployment of VoIP. Therefore, deploying VoIP requires proper planning. Literature was reviewed for proper planning and implementation of VoIP.

Table of Contents:

Abstract

  1. Introduction 

    1. Origin of Project

  2. Community views VoIP 

  3. Data networking views of VoIP

  4. The Advantages and Disadvantages of Using VoIP over Linux 

    1. Faster Development Time and Costs

    2. Reliability and Call Quality 

    3. Linux Implementation Stability

  5. Criteria of Selection

    1. Implementation in Linux

    2. Hardware and Software Requirement

    3. The layout of network Diagram and Configuration

    4. Implementation Process in Linux

    5.  Pilot Study

    6. Discussion

  6. Future of VoIP

  7. Conclusion

  8. Research Proposal 

  9. List of Mind-Maps 

  10. List of Figures 

  11. List of Tables 

  12. Appendix I 

  13. Glossary of Terms 

  14. Table Concept Matrix

  15. References 

1.0 Introduction:

There has been a general recognition that VoIP will become a key technology in the future. Many studies support the use of VoIP implemented in windows environments. However, literature was reviewed to identify the consumer’s requirements and the technology to implement VoIP in a Linux environment.

There are some doubts about the effectiveness and validity of using the Linux operating system for VoIP. However, these doubts can be corrected or modified to suit our course. Regardless of other operating platforms available for VoIP deployment, the objective is to deploy VoIP in a Linux environment for small businesses and for individuals, as VoIP over Linux is highly reliable and cost-effective.

Data-networking engineers have developed protocols for sending and receiving data packets. Although many protocols for data networking have been developed, the Internet Protocol (IP) has become the most important one, as it has proved to be remarkably scalable and adaptable. IP networking has become significantly important for data transfer and for communication, because of its convergence ability, where it takes different types of data such as voice, video, and application data and transfers them over the same IP network (Walker & Hicks 2004).

The TCP/IP family of protocols forms the basis of the Internet and most of the corporate networks. Computer programs send and receive data over an IP network by making program calls to the TCP/IP software known as protocol stack in their local computer. The TCP/IP stack in the local computer exchanges information with the TCP/IP stacks in the target computer to accomplish the transfer of data (Walker & Hicks 2004).

Therefore, IP for most IT managers seems the logical and practical choice for the inclusion of a voice network while considering integration. IP is a connectionless protocol that delivers data on a best effort basis. The originally designed protocol is not suited for voice transmission. The VoIP developers should consider ‘Delay’ as the lacking factor of IP when deploying VoIP. However, many personal consumers and small businesses do not mind deterioration in quality caused by the delay because of cost-effectiveness.

“The internet traffic has been increasing with more than 100% annual growth rate since 1995”(Asatani 2003).  VoIP integration with an existing data network makes ease of maintenance as a single network and it has the ability to provide multi-services to its users. E-commerce and/or Internet call-center are also one of the major platforms for VoIP.

Graph from Cisco shows the global VoIP minutes of usages in billions per year indicates that Internet telephony is growing, as has been described by many studies. (Asatani 2003; McElligott 2004)

Figure 01: VoIP International use of minutes in billions

Source: www.cisco.com/.../ ns68/covia_bc/covia_b3.jpg

Although, quality is part of the system, which an end-user has to decide, when an industry provides a technology, it has to fulfil certain quality standards. Therefore, quality of service (QoS) becoming a key issue in the VoIP services within the IT industry.

…QoS refers to choices about how different ‘users’ or ‘different application traffic’ gets treated on the network. To make changes in your network so that VoIP traffic gets priority over file-transfer traffic, … industry needs QoS techniques (Walker & Hicks 2004, p. 142)

QoS in VoIP is still under development or at its early stage of development in order to achieve quality grade services in comparison with PSTN (Borthick 2002). A significant amount of research has been conducted on QoS management in order to provide a quality grade, voice services over the IP network.

When constructing a business case for VoIP deployment, hardware and software platforms are one of the major factors to be considered for planning, analysis, and assessment for a successful deployment. There are benefits and obstacles of using VoIP over a Linux network instead of a windows platform when implementing a voice network over an existing IP network.

1.1 Origin of Project:

I started finding cost-effective solutions to save telephone costs for my previous employer. Skype (VoIP softphone) was used for calling other computers and to local landlines. Computer to computer calling is crystal-clear over VoIP using Skype as a softphone and free of charge but computer to landline or computer to mobile, have some marginal charges. There are several VoIP phone cards in the market today. Phone cards provide landline-to-landline calling using VoIP at reasonable prices. I would like to set up my own VoIP network between Australia and India. This project would start with a search on the quality of hardware and software for the deployment of VoIP.

2.0 Community views of VoIP:

It is essential to know the foundation of Plain Old Telephone System (POTS) with a clear understanding of how signal information is passed between traditional telephone instruments in order to configure different types of voice over data network products.

…Signaling in a voice network is twofold. First, signaling is necessary to convey information about a potential connection. Second, signaling enables a connection to be established through a voice network…(Held 2000)

This signaling has its set of rules and international standardization in order to provide reliable voice services. The quality and reliability of the PSTN is an outcome of decades of knowledge, experience, and innovation, which results that people being less likely to tolerate reduced quality or performance or dropped calls. PSTN reliability is sometimes referred to as ‘five nines’.

Walker and Hicks, (2004) describe ‘five nines’ that the entire network must be available and functional for 99.999 percent of the time. According to this principle over the period of one year:

365 days * 24 hours/days * 60 minutes/hours* 0.00001

= 5.256 minutes

Therefore, the network can be down for a grand total of fewer than six minutes over one year.

This quality and reliability cannot be achieved without major rules and standardizations; therefore, the International Telecommunication Union (ITU) does this for PSTN. Besides, standardizing the technology of the PSTN it also initially provides standards through its specific division known as the Telecommunication standardization sector or ITU-T. The standards are crucial to the success of technologies like VoIP (Walker & Hicks 2004).

…Defining voice over the Internet as an application rather than a service, Federal Communication Commission (FCC) Chairman Michael Powell argued against regulating VoIP as traditional telephony… “Internet voice will unleash a torrent of innovative products and services”, Powell said in an address at the National Press Club in January 2004…the burden should be placed squarely on government to explain why regulations are needed… (Carlson 2004, p. 20)

There are several studies that have skeptical views, indicating that within the IT industry, it is less likely to find a data network with ‘five nines’, which implicates, that VoIP network, will never reach the standards of PSTN. Only the bravest and the smartest will deploy VoIP (Hall 2005) The evidence provided by Hall, (2005) appears to be somewhat one-sided, given that there is important evidence to the contrary, for example, approaches taken by few authors describing how quality and reliability could be achieved are (Borthick 2002; Isenberg 1999; Keepence 1999).

It has been argued that VoIP needs serious research in order to provide quality grade services which could be in terms of internationally standardizing the protocols for VoIP and deployment of QoS management services.

3.0 Data networking views of VoIP:

Internet telephony has been very popular among IT managers as most of the network managers find it very attractive and cost-effective in terms of integrating voice and data networks into one. This makes it easier for network managers to maintain a single data network. There are various definitions and claims about VoIP within the IT industry. Held (2000) describes Internet telephony as a term used primarily to refer to the use of software in conjunction with a sound card and microphone to provide individual PC users with the ability to initiate and receive calls over the Internet. Walker and Hicks, (2004) describe convergence as taking a different type of data-voice, video, and other application data and transferring them over the same IP network.

Providing data networks has its own set of rules and standardizations referred to as protocols. Several studies indicate that there are many protocols for data networking. Among those, Internet Protocol (IP) has become the most important protocol and has proved to be the most adaptable and remarkably scalable. In the similar fashion of ITU, the Internet Engineering Task Force (IETF) has led the standardization efforts in the data network community(Walker & Hicks 2004). This has resulted in IP networks gaining respect within the IT community, which has changed the way people think about data communication.

IP networks are best-effort networks that were designed for non-real-time applications, whereas VoIP requires timely packet delivery with low latency, jitter, packet loss, and sufficient bandwidth in order to achieve quality grade service over VoIP (Chen et al. 2003; Salah 2005; Uhl 2003).

4.0 The Advantages and Disadvantages of Using VoIP over Linux:

There are several promises given by the IT community in terms of benefits from using VoIP over the traditional telephony PSTN. Apart from all the promises made for VoIP, the IT industry is more or less excited about potential cost savings over a period and the reduced infrastructure of the converged network in a VoIP deployment. According to Walker and Hicks, (2004), there are two main types of benefits to VoIP which he describes are hard benefits and soft benefits.

Hard benefits come with a clearly-defined cost savings… on the other hand, soft benefits do not necessarily save money, or, if they do, they do not always save an easily calculated amount of money… both type of benefits are critical to final ROI, most organizations focus more on hard cost savings, because they are easier to quantify … (Walker & Hicks 2004)

When it comes down to business decision-making, the choice of the operating system on which to build the services is very critical to every business owner. The selection of a correctly operating system would be ideal for getting successful deployment. Linux is a choice because of wider vendor support, services designed on the Linux platform works on a wide range of hardware, which makes the case that businesses are less likely to buy proprietary equipment to make services work. Information about the product drivers and support are regularly updated globally and also available for evaluation and reviewing; Linux is a tool that makes a great deal of sense for a company with a limited budget that wants to implement VoIP services over an existing IP network.

Wider choices in Linux are one of the main factors for its global adaptation, which makes developers customize services according to their specific needs and requirements. Linux can be customized to support individual customer needs. There are several cost benefits to using the Linux operating system such as wide vendor support, limited licensing cost, and wider choices. Low licensing cost makes Linux an attraction for businesses with limited IT budget and needs to deploy VoIP services.  A case study conducted at bell Labs for open source tools and practice in a commercial setting justifies that the recent acceptance of the Linux and Apache project is an excellent example of the open-source phenomenon (Gurbani, Garbert & Herbsleb 2005).

4.1 Faster Development Time and Costs:

Linux is desirable among the developers for open source seeking faster convergence and deployment. There are thousands of programmers and developers contributing to the source code and putting their valuable efforts to makes Linux better and better adding new features to support drivers for a wide range of hardware, whereas proprietary software used by windows does not allow this to happen (Bradner 2004; Gurbani, Garbert & Herbsleb 2005; Herlein 2000 ).

These facilities provided by Linux to developers reduce the individual cost for an organization of having an onboard team of developers working on a project to improve the features of an application or developing a driver for specific hardware in order to work with existing applications and to support services.

The quality and reliability of VoIP services is an issue as users are accustomed to the ‘Five Nines’ quality of PSTN and they expect to have the same outcome from VoIP services. There is also a concern about the downtime of a converged network that may directly affect the organizational performance and image.

The project must start with a clear understanding and focus for the targeted goal and therefore, the implementation plan has to be well scheduled and budgeted and be well planned with a clear view of ongoing management and training costs in order to get successful deployment. However, I did not find any specific article on the Linux implementation plan, notwithstanding that I am working on this aspect. According to the quality of literature related to a similar topic indicated, there are a few questions any organization should consider answering before deployment.

  • Does an organization have complete information about its current network environment?
  • What does an organization want to acquire out of VoIP deployment?

The first point can be further divided into a few sub-sections such as current telephone usage of an organization, what reliability and call quality is expected out of VoIP deployment, and willingness to deploy VoIP over an existing IP data network. I will try to find methodological solutions for these questions in my Project B including with implementation plan in order to achieve a successful deployment. However, I discuss a few key points such as:

4.2 Reliability and Call Quality:

Several studies (Dickerson 2005; Follett 2005; Golden 2005) have shown that Linux is a valid and reliable operating platform. Linux has qualities such as rapid development environment; culture for openness and faster integration gives an extra advantage for its selection over other operating. When organizations deploy VoIP, users expect a high level of reliability from their phone system compared with PSTN.

4.3 Linux Implementation Case Study :

Rosen (2005) describes the open-source software as…

… there is no easy conceptual basis for integrating the language of freedom into the legal language of software licensing… where the word Free is currently used in software licensing contexts, it usually means zero, as in free of charge or free of defects… neither of these meanings in intended by open source licenses…(Rosen 2005)

Rosen (2005) further describes four kinds of software freedom:

  • The freedom to run the software for any purpose
  • The freedom to study how the software works and to adapt to your needs
  • The freedom to redistribute copies of the software
  • The freedom to improve the software and distribute your improvements to the public

While Rosen (2005) used RedHat Linux 9.0, there are several versions of Linux that could be used to create the heterogeneous environment to allow Linux and Windows to interoperate. Many people use cost as a criterion for the selection of technology. However, cost should not be the only factor. The technology should be efficient and not be Platform-Dependent in order to work efficiently. There are several other factors such as service quality and support that need to be considered for the successful deployment of VoIP.

Open-source community-based software (Linux) incurs costs indirectly, as open-source-community programmers are not obliged to help. This means that organizations that use Linux would have to have their own in-house programmers to address problems. It is good for businesses that the end-users cannot deploy the technology by themselves.

Chou, et al. (2001, pp. 73-88) found

…that device drivers have errors rates up to three to seven times higher than the rest of the kernel. …that the largest quartile of functions have error rates two to six times higher than the smallest quartile. …that newest quartile of files have error rates up to twice that of the older quartile, which provide evidence that code “hardens” over time. Finally Chou et al. (2001, pp. 73-88) found that bugs remain in the Linux kernel an average of 1.8 years before being fixed…(Chou et al. 2001)

Chou, et al. (2001, pp. 73-88) addresses questions like Do drivers account for most errors? Do bugs cluster? How do different operating system kernels compare in terms of code quality?

They found that drivers account for over 90% of the block, free, and Inter bugs, and over 70% of the Lock, Null, and Var bugs. Since drivers account for the majority of the code (over 70% in this release), they should also have the most bugs. This effect is more pronounced when code size is corrected.

Device-Drivers are the most important factors for an application to function effectively and efficiently. This shows users have to be very careful when deploying an open-source product.

Figure 05: The size of the Linux tree that (Chou et al. 2001) checked over time. Version 1.1.13

Figure 06: The absolute number of projected errors in (Chou et al. 2001) study

Figure 07: This graph shows drivers have an errors rate up to 7 times higher than the rest of the kernel in (Chou et al. 2001) study

There are many vendors making hardware based on open-source software. But there is no guarantee of efficient support and hardware functionality. For example, HP releases its Open Source Network Node Manager software based on Open-Source has functionality issues such as a bug in software associated with it. HP’s response to functionality issues is that it will be corrected in the next version or in the next upgrade (Hamblen 2004).

Golden (2005) found some weaknesses of open-source software.

  • Because of the need for consensus and the large community size, the pace of work and decision-making can seem very slow. There really is not any way around this.
  • Despite the strong informal rules of participation, immature behavior sometimes (rarely) occurs. This can take the form of name-calling, but the more common form is one-upmanship based on technical opinion
  • Occasionally, the community will engage in protracted discussions about off-topic subjects.

However, Rosen (2005) showed individual users often do not have to be concerned with the intricate conditions of these open-source licensing issues, warranties, technical hurdles, patent defenses, or other esoteric legal issues. There is very little litigation over open-source software as there is very little chance of success in the absence of licensing.

5.0 Criteria of selection:

Several open-source software is currently used for VoIP applications. However, the software had to fulfill a list of criteria to be selected for this project. The following criteria had to be fulfilled: should contain source code so that it could be modified, should have limited licensing cost, should have wide vendor support, and should have wider choices. Open-source software such as Linux includes source code where users can modify according to their requirement and need whereas Windows and other proprietary software do not allow modification.

Linux works on a wide range of hardware so implementers and/or application designers have the choice to buy specific types of equipment from a number of vendors in order to make the new applications work effectively. Updated information about all products and services compatible with Linux is easily available on the Internet. For new hardware in the market that is not supported by Linux, the manufacturers of hardware are likely to have an idea of when it will be supported on Linux.

The cost of software includes the purchase cost and licensing cost. The purchase cost is very high when it is designed for a particular purpose and the licensing cost is very high when proprietary structures are put in place. These costs make it very difficult for small businesses and individuals to afford this facility on their existing data networks.

Linux is an open-source operating system. Users are free to choose vendors for their specific needs of an application in order to get vendor support and custom programming solutions. Users have the option of doing their own custom programming or they can hire programmers. The Linux programming can be done in-house or by outsourcing. Other proprietary operating systems do not allow this to happen, only software developing companies or OS companies can make any repair, upgrade, or modification.

The choice of the operating system on which to build VoIP services is a business decision, which has to be a rational decision. Linux operating system was selected for this project, as the target was small businesses and individuals with limited budgets. These people cannot afford the licensing fee of proprietary operating systems, upgrades, purchasing costs, and maintenance costs. Linux was selected for this project as it fulfilled all the selection criteria.

5.1 Implementation in Linux:

There are several types of communication hardware and software providers enabling advanced telephony services that can be rapidly and effectively be deployed on converged and/or hybrid networks. One of these advanced services is VoIP. VoIP replaces the traditional hardware switches and other legacy communication equipment with software switches and voice communication protocols, specifically designed for transmitting voice over IP protocol.

In VoIP enabled environment, the user initiates a call by picking up a handset, which may be an old-style analog phone, a new digital phone, or a VoIP handset plugged into an Ethernet port. When the user dials the number by pressing keys on the handset or by using Internet telephony application software such as soft-phone provided by Skype on the net on the computer. Then a routing server finds the phone number in its database and connects it. However, if the call is from outside the network the routing server routes the call to a gateway device that acts as a portal to the Public Switched Telephone Network (PSTN) for call completion (Walker & Hicks 2004).

In order to operate VoIP services efficiently in a Linux network, it requires a wide range of hardware and software. There are several applications designed on the Linux platform for VoIP services. To attract and maintain customers due to competition Internet providers provide additional services of VoIP applications. Internet service providers are offering innovative features and services such as integrated Internet access with VoIP to their customers.

5.2 Hardware and Software Requirement:

Good quality telephone call requires minimum hardware to run VoIP application over a Linux network. It requires a processor with a minimum power of 500 MHz, a network card with 100 Mbps to eliminate bottlenecks, a full-duplex sound card for two-way communication, and several gigabytes of hard drive space to store log files and call record database. There are several hardware and software vendors and suppliers in the marketplace. This setup is for an individual and/or small business this is why not using Asterisk is very complex to set up. Using a simple Linux setup will provide to make calls from PC to PC, PC to Phone, Phone to PC (to receive calls made from phone) and if not answered will be diverted to message bank.

There are three common protocols in use those are SIP, H.323, and Skype (Proprietary). Session Initiation Protocol (SIP) is open source and widely used. SIP supports Linux phones such as Kphone and linphone. SIP is a simple and more flexible protocol. Therefore, has been chosen for this project. H.323 is older and it is complex and has not been widely supported. Skype is the proprietary protocol that does not come with source code and one must use Skype’s servers this is why does not fulfill the selection criteria of this project.

SIP is peer 2 peer. However, a server is required to be used as a Registration server. One SIP phone may directly call another SIP phone such as Linux-based Kphone clients can call another Kphone client over the network. This needed UDP connection to port 5060 makes both machines communicate with each other. Moreover, for better communication Session Initiation bit is required when voice data flows between two VoIP phones. The call-initiation process may use a different server. In the case of dynamic IP addresses used by one or both machines, the SIP URL in the phonebook does not get up to date. Therefore, a Registration Server is required for the Client VoIP handset to sign into and use the server as a central point of contact. Registration server enables other VoIP handsets to be identified on the network, organize calling to multiple locations, and in the case of no-response call diversion to voicemail. Registration Server permits two VoIP handsets to communicate with each other if either handset is behind the firewall. Registration Server also provides Outbound Proxy which enables

two VoIP handsets to communicate to each other in the use of a private network or if the handset is behind the NAT (Network Address Translation) gateway. A list of hardware and software for VoIP deployment in Linux is shown in Table 01.

Table 01: Hardware and Software for VoIP deployment in Linux

Equipment for Linux Based

Application Providers

Open IP PBXs or FreeBX

www.vovida.com

IP Telephony Gateway (2):

Business Communication Manager based on Nortel Corporate Linux

Nortel

IP Phones:

Skype SoftPhone or Zultys Phone

www.zultys.com

Fast Ethernet/Gigabit Ethernet Switches:

Linux Box can be configured as Switches

Red Hat-9.0

Linux Network or Linux-VPN

Red Hat-9.0

PCs and Servers:

Linux Box can be configured as a server

Red Hat-9.0

Voice Quality Tester:

Linux based BCM 50 from Nortel

Nortel

Data Traffic Simulator:

Vovida Linux based VoIP System

www.vovida.com

Application Traffic Simulator:

Nortel Corporate Linux

Nortel

WAN Simulator

QoS Appliances:

SIPFoundry Linux based VoIP solution

www.sipfoundry.org

www.nwfusion.com

VPN/Firewall Appliances:

Linux Box can be configured as Firewall

Red Hat-9.0

Fast Ethernet/Gigabit Ethernet Analyzers:

Linux Box can be configured as Network Analyzer

Red Hat-9.0

Billing Server:

Vovida Linux based VoIP System

www.vovida.com

Training:

Individual Knowledge of Linux

Red Hat-9.0 documentation on

www.redhat.com

Table 03: List of Systems Requirement for Hardware Packages in Windows, Mac, Linux

Hardware Package

Operating System

Systems Requirements

Engine Voice Box

Windows and

Macintosh

  1. Broadband connection, ADSL or Cable
  2. 512 kbps or faster speed recommended
  3. Spare Ethernet port on modem used or router to connect Voice Box
  4. Ports enabled on the firewall: UDP 5060(SIP), UDP range 16382-20382 (RTP), both incoming and outgoing.
  5. Sufficient download quota: 10 minutes talk time uses around 2.2MB

Netcomm LSV40 USB handset

Windows

  1. Broadband service with SIP compatible router
  2. OR Laptop computer running Windows 98SE or Above
  3. 64MB RAM
  4. 5MB free disk space
  5. CD ROM drive
  6. Free USB 1.1 port
  7. Optionally, a sound card (for special ring tone audio).

Netgear TA612V

Windows

  1. Broadband connection
  2. Voice over IP service
  3. Analog touch phone-tone telephone
  4. CD-ROM drive
  5. Windows 9x running a TCP/IP network

Snom 100 VoIP Phone

Linux

  1. SIP, H.323/H450, and Asterisk as well as data protocol HTTP, TAPI, and LDAP
  2. 16MB RAM
  3. Motorola QUICC processor running Linux 2.4.18 kernel
  4. Broadband connection

Table 04: List of Systems Requirement for Skype in Windows, Mac, Linux

Windows:

  • 400 MHz processor
  • Windows 2000 or Windows XP
  • 128 MB RAM
  • 15MB free disk space on the hard drive
  • Sound card, speakers, and microphone
  • Broadband internet connection or dialup internet connection at 33.6 kbps or higher

Mac:

  • Macintosh computer with G3 or G5 processor
  • Mac OS X v 10.3 (Panther) or newer
  • 128MB RAM
  • 20MB free disk space on the hard drive
  • Microphone
  • Broadband internet connection or dialup internet connection at 33.6 kbps or higher

Linux

  • 400 MHz processor
  • 128MB RAM
  • 10MB free disk space on the hard drive
  • Full duplex sound card, speaker, and microphone
  • Broadband internet connection or dialup internet connection at 33.6 kbps or higher

5.3 Layout of network Diagram and configuration:

Source: White paper from

According to bell labs, VoIP networks may have to interwork with existing time-division multiplexing (TDM) – based voice networks. From the point of view of the internetworking bell, labs give basic three distinct scenarios as shown in figure 1. VoIP service scenarios are such as

  1. Network Interworking, where the packet network is deployed in between TDM networks such as PSTN and the end host/devices are traditional voice devices, such as analog or integrated service digital network (ISDN) phones.
  2. Service Interworking, where an IP endpoint communicates with traditional voice devices over a packet and TDM network.
  3. Native VoIP service, where IP endpoints communicate with each other over a packet network.

Source:

Source: http://www.voicetronix.com/hda.htm

A first step towards the setup is testing the hardware to check that all the VoIP handsets, headphones, and microphones working properly. Make sure that all sound cards are full-duplex and enable the “20dB Mic Boost” using a mixer such as a gnome alsamixer that reduces the echoing.

Second downloading the software such as Linux-based soft-phones linphone or Kphone. Linphone is chosen for this project because it has a better interface and documentation. Linphone may require a diagnostic program named sipomatic. It is recommended that first install libgai0-devel for the gnome-applet libraries and afterward use the standard configuration in order to compile the linphone.

The third step is to register with a SIP server. FreeWorldDialup is a well-established SIP server and provides services free. Therefore, has been chosen for this project.

The fourth step would be configuring a linphone using Configuration Guide or online documentation. Configuration may vary depending on the use of private and public addresses. Local ISPs provide a range of public addresses at affordable prices.

Table 05: Linphone Configuration (may or may not work for most linphone depending on the version)

Setting (Linphone 0.12)

Sip:NNNNNN@fwd.pulver.com

Use sip register

Server address

Sip: fwd.pulver.com or sip: fwdnat.pulver.com:5082

Using registrar as outbound proxy

Number of buffers size in milliseconds

200ms works fine

NAT traversal enable

Firewall IP address

The public IP of Firewall or Gateway

RTP port used for audio

Default (7078)

Source: www.pulver.com

The fifth step would be testing the configurations and upgrade. One may require an incoming phone number for receiving calls from landlines to PC. This can be done by local ISPs that provide VoIP services such as ‘CallUK’ does within the UK and is free.

The final step is outgoing phone calls to other phones including international calls this requires signing with a service provider such as Skype, CallUK.com and IconnectHere.com one may choose accordingly depending on the requirement.

5.4 Implementation Process in Linux:

Successful deployment of VoIP in a Linux environment requires knowledge of Linux or open-source software, proper planning, assessment, and pilots. The first step in the deployment process is the assessment of a VoIP-Readiness. If the existing network is not capable of handling VoIP applications it has to be improved by upgrading existing pieces of equipment and links, implementing QoS, obtaining new network service contracts and service-level agreements (SLA). The second step is acquiring the hardware and software for VoIP deployment. The third step is the configuration of hardware and software. The fourth step is testing and addressing problems.

5.5 Pilot Study:

Walker and Hicks (2004) argued that a pilot deployment should be taken before a full-blown deployment. The first five customers will be part of a pilot study. The customers will be interviewed using a structured questionnaire to identify the quality of service and user-friendliness. The problems identified will be addressed or rectified where possible before full deployment.

5.6 Discussion:

The criteria used for the selection of the Linux platform were cost, reliability, security, manageability, and training. The choice of hardware, software, and training depends on the available fund, as small businesses and individuals are always low on funding. Linux based platform was more attractive to small businesses and individuals because of its low cost.

Cost includes acquiring hardware, software, training, and implementation. Security for small businesses and individuals depends upon information stored on the call record database. Security was not a major issue at this stage. For big organizations the security of call records is essential. For secure access within an enterprise data network, it requires authenticated access such as firewall security and/or Virtual Private Network (VPN). Manageability is important for big organizations to maintain and monitor the call-servers and require a qualified, in-house trained staff. Small businesses and individuals do not require administrative intervention.

The literature review identified several limitations such as treating VoIP management as an afterthought; ignoring priority and quality handling for voice traffic (over-provisioning bandwidth service instead of managing service); managing data, voice, and QoS on a single network with separate management tools; Assuming that existing data-network is ready to handle the demand of VoIP traffic; waiting for the user to complain before addressing voice quality issue; ignoring the training for in-house staff; taking shortcuts to solve the issues. (Daley & Snaith 2004)

Skype is free telephony software or a softphone that uses P2P technology to provide free of charge voice communication services over a wide area network. The quality of voice conversation over PC to PC is very close to PSTN quality (five nines) but there are some variations in voice quality over PC to Landline or PC to Mobile phone. The objective of this project was to provide technology that provides the same quality of voice conversation for PC to Landline or PC to Mobile phone.

At present voice quality using Skype is not important for my clients, as their objective is to pass the message across to their relatives or business counterparts. However, my objective is to provide the same quality of voice conversation for PC to landline and PC to mobile phone. To achieve a quality grade service one requires the knowledge and understanding of QoS and QoS management.

6.0 Future of VoIP

There has been a general recognition that quality of service (QoS) is becoming a key issue (Armitage 2003; Walker & Hicks 2004). In order to achieve quality grade services in comparison with PSTN services, QoS in VoIP requires further improvement. A significant amount of research has been conducted on VoIP services in order to improve the quality of voice communication over IP networks. There are several methods and techniques to adjust and tune the network devices to improve the call quality.

There is ongoing research on resource management mechanisms for next-generation networks conducted at UTS, Sydney on Linux-based soft routers to measure Quality of Service (QoS) in a Differentiated Service Domain providing resource management for hard end-to-end QoS guarantee.

According to researchers (Nanda, Simmonds & Lee 2002), the advanced networking features with support for Quality of Service (QoS) in the recent Linux kernel (2.2.9+) provide a flexible platform to experiment with resource management in a Differentiated Service (Diffserv) domain. Their research has shown that with Diffserv enabled in the router involved multi-field classification in order to mark Expedited Forwarding (EF) aggregate traffic and protect them from heavy background traffic. The results are as shown in figure 08.

Nanda, Simmonds & Lee (2002) has described that, with Diffserv enabled, the foreground traffic is forwarded with expedited forwarding PHB and continues without any loss for the same interval. The variance in jitter is also bound to less than 1 ms using CBQ. Their research demonstrated how Diffserv expedited, forwarding per-hop behavior could be used to reduce data loss and jitter even in the presence of heavy background traffic.

7.0 Conclusion:

The literature review identified hardware, software, shortcoming, and QoS requirements for VoIP deployment in a Linux environment. To control the cost of telephones more businesses and individuals are adapting to Internet and/or other hybrid networks. Several advantages have been identified such as cost-saving, increase organization efficiency, interoperability within the different platforms, and are easy to manage as a single network for the transmission of voice and data.

From this review, a few positive points have become clear. Firstly, Linux is an alternative to proprietary operating systems for an installation that cannot be ignored. It is an opportunity that most small businesses are taking up and will continue to take up for financial and other benefits. Secondly, Linux, implementation is the best solution for a cost-effective integration with the potential to grow in time with the availability of faster development time and improvement. At the same time, providing the device driver code bugs may cause some serious damage to the organization's image as if they contribute to the failure of an application resulting in business loss. VoIP should be insensitive to its operating system platform; however, if an organization decides to take Open-Source into consideration then issues related to the device –drivers should be resolved beforehand.

Existing work in the VoIP has shown that QoS management has much to offer in a certain type of deployment to achieve quality grade services. Literature review shows that QoS management is necessary to improve quality and reliability. While there is no ideal QoS management technique available yet, it should not be ignored. Further research is required to determine the suitability of so many different QoS techniques and methods to handle many network device interconnections and interactions.

8.0 Research Proposal:

It is now recognized that most IP networks need some QoS guaranteeing mechanism in order to ensure QoS on VoIP applications. There have been several solutions within the IT industry and university laboratories to address the QoS management mechanism issue. There are several QoS schemes and parameters in use within the IT industry to improve VoIP services (Hackforth 2002, p. 48; Walker & Hicks 2004). The objective of ‘Project B’ would be to define and understand the QoS configuration in the Linux environment. This will be achieved through a literature review and information acquired through a pilot study.

9.0 List of Mind-Maps:

Mind-Map-01: Report Layout

Figure-04: Linux implementation graph according to Technology Adoption Process, from Moore’s BookCrossing the Chasm

10.0 List of Figures:

11.0 List of Tables:

Table 01: Equipment Description for VoIP deployment

Table 02: List of Software available in Windows, Mac, Linux

Table 03: List of Systems Requirement for Hardware Packages in Windows, Mac, Linux

Table 04: List of Systems Requirement for Skype in Windows, Mac, Linux

Table 05: Linphone Configuration

12.0 Appendix I

LITERATURE REVIEW GUIDELINES FOR EVALUATION

Prepared by Student ID No: 02021691

1 Purpose of a literature review:

  1. How critical is it?

  2. Does it report within the general body of scientific knowledge?

  3. Does it base on existing knowledge and resources available?

  4. Does it bring the reader up-to-date with other research that has been done in the area, in particular previous, relevant, research in the topic-related area?

  5. Does it identify the particular topic-related areas of agreement and disagreement?

  6. Does it assist in:

    1. Researching an important question and justifying by putting time, energy, and resources into examining that questions:

    2. Selecting an appropriate population, societies, industries, and sample;

    3. Developing a suitable research design and justifying it;

    4. Identifying useful approaches to analyzing and reporting data gathered;

    5. Discussing the results;

    6. Drawing conclusions;

    7. Discussing the limitations of the methodology used;

    8. Recommending changes in industrial practice and/or area of further research;

    9. Bibliographic function and/or citation of referenced material;

2 Methods:

2.1 Type of information available

2.1.1 Theory;

2.1.2 Research results;

2.1.3 Industrial experience;

Theories are built through the creative integration of research results, industrial experience and hunches.

2.2 Source of information;

  1. Text/Books;
  2. Journal Articles;
  3. Published Reports;
  4. Others;

2.3 What to look for?

  1. Unexpected findings;
  2. Populations that have not been studied;
  3. Characteristics which have been studied;
  4. Suggestions for further research;
  5. Factors impacting on the problems;
  6. A theory that explains the problem of this assignment;
  7. Conflicting theories;

2.4 Methodology

  1. Field and or industry studied and reason for selecting same;
  2. Sample selection and problems experienced in gaining a random or representative sample of the industry of interest;
  3. Suggestions for overcoming the problems;
  4. Research design and reason for using that approach;
  5. Which design worked well and which did not (productive research only);
  6. What recommendation does the author make;
  7. How did they measure the data;
  8. How did they describe their findings; (i.e. Tables, figures, and charts used)
  9. Were their tables and figures informative, easy to follow, meet the “stand-alone” test;
  10. Which methodology they recommend;

13.0 Glossary of Terms:

POTS

Plain old telephone network, the basic analog phone line most people have in their homes.

Fax

Facsimile, a method of sending a copy of a paper documents over an analog telephone line from point A to point B.

Modem

Modulator/demodulator, a device that connects computer systems by transmitting signals using analog signals.

CO

Central office, the telephone company’s switching center that provides POTS service to the surrounding area.

PCM

Pulse code modulation, a way of storing an audio signal as a bit stream.

CELP

Code-excited linear prediction, a compression codec used in advanced VoIP codecs. CELP reduces the number of bits necessary for successful transmission of the sound; it is somehow achieved by degrading the audio signal.

RJ12

A socket or adapter being used to connect analog phone to POTS

RJ45

A socket or adapter being used to connect network devices to each other

GPL

General Public License

BSD

University of California’s Berkeley Software Distribution

OSL

Open software silence

AFL

Academic free license

14.0 Concept Matrix 01: Concept Centric

15.0 References: In Concept Matrix & Report:

References

Allard, H. 2000, 'Understanding QoS How to obtain quality over bursty networks', Teleconference; ProQuest Telecommunication, vol. 19, no. 1, pp. 10-17.

Armitage, G.J. 2003, 'Revisiting IP QoS: What do we care, What have we learned? ' ACM SIGCOMM 2003 RIPQOS Workshop Report, vol. 33, Melbourne, Australia, pp. 81-88.

Asatani, K. 2003, 'IP Telephony and QoS Issues', Proceeding of ICCT2003, Tokyo, Japan, pp. 48-50.

Borthick, S. 2002, 'QoS: Once Impossible, Now Inevitable?' Business Communications Review; ProQuest Telecommunication, vol. 32, no. 10, pp. 20-24.

Bradner, S. 2004, 'Is Part of the future of VoIP Open?' Network World, vol. 21, no. 40, p. 30.

Carlson, C. 2004, 'FCC's Powell: Don't Stifle VoIP', January 19, 2004, no. January 19, 2004, p. 20.

Chen, X., Wang, C., Xuan, D., Li, Z., Min, Y. & Zhao, W. 2003, 'Survey on QoS Management of VoIP', International Conference on Computer Networks and Mobile Computing, IEEE Computer Society, Beijing, China, Columbus, USA, Texas, USA, pp. 0-7695-2033-7692.

Chou, A., Yang, J., Chelf, B., Hallem, S.; & Engler, D. 2001, 'An Empirical Study of Operating System Errors', ACM pp. 73-88.

Daley, P. & Snaith, P. 2004, 'VoIP's SEVEN Deadly Sins', Communication Convergence; ProQuest Telecommunication, vol. 12, no. 2, pp. 14-16.

Dickerson, C. 2005, Open Source Calling, InfoWorld, vol. 27, Infoworld Publishing Group, pp. 14-14<http://search.epnet.com/login.aspx?direct=true&db=afh&an=17552029 >.

Drake, W.D. 1997, 'Using Linux to Solve the Problem of Mixed Operating System Lan Services', ACM SIGUCSS, vol. XXV, no. 1997, pp. 67-72.

Follett, J.H. 2005, 'Fonality Talks Up Low-Cost Open-Source VoIP System', Inform Global, vol. 1152, no. CRN; Jan 20, 2005, p. 10.

Golden, B. 2005, Succeeding with Open Source, Addison-Wesley, Boston, Sydney.

Gurbani, K.V., Garbert, A. & Herbsleb, D.J. 2005, 'A Case Study of Open Source Tools and Practices in a Commercial Setting', ACM, pp. 1-6.

Hackforth, S. 2002, 'A question of standards', Telecommunication International; ProQuest Telecommunication, vol. 36, no. 8, pp. 48-50.

Hall, M. 2005, 'VoIP Scary', Computer World, vol. 39, no. 3, p. 37.

Hamblen, M. 2004, OpenView Users Ask HP for LessComplexity, More Support, Computerworld, vol. 38, Computerworld, pp. 12-12.

Held, G. 2000, Voice & Data Internetworking, second Edition edn, McGraw-Hill, Sydney.

Herlein, G. 2000 'The OpenPhone Project--Internet Telephony for Everyone!' Linux Journal vol. Volume 2000 no. Issue 69es  (January 2000)

Isenberg, D.S. 1999, 'Is quality of service necessary', America's Network, ProQuest Telecommunication, vol. 103, no. 4, p. 20.

Jiang, W., Koguchi, K. & Schulzrinne, H. 2003, 'QoS Evaluation of VoIP End-points, IEEE Computer Society, vol. 0-7803-7802-4/03, no. 4/03, pp. 1917-1921.

Johnson, J.T. 2003, 'Implementing IP telephony: What to watch for', Network World; ProQuest Telecommunication, vol. 20, no. 9, p. 27.

Kagklis, D., Sykas, E., Manousos, M., Apostolacos, S., Grammatikakis, I. & Mexis, D. 2005, 'Voice-Quality Monitoring and Control for VoIP', IEEE Computer Society, vol. 1089-7801, no. 05, pp. 35-42.

Kang, J.W. & Nath, B. 2003, 'Adaptive QoS Control by Toggling Voice Traffic between Circuit and Packet Cellular Networks', IEEE Computer Society, vol. 0-7803-7974-8, no. 8/03, pp. 3498-3503.

Kathryn, M. 2002, 'Can Open Source Software Change The Microsoft Story in Public School', Australian Computers in Education Conference, Hobart, July 2002, ed. T.a.D. Department of Education, Adelaide, ACEC 20002 Proceedings: refereed paper section, Hobart, pp. 1-20.

Keepence, B. 1999, 'Quality of Service for Voice over IP', paper presented to The Institute of Electrical Engineers.

Lewis, A. 2002, 'Solving the QoS puzzle', Telecommunication International; ProQuest Telecommunication, vol. 38, no. 8, pp. 22-24.

Li, B., Hamdi, M., Jiang, D. & Cao, X.-R. 2000, 'QoS-Enabled Voice Support in the Next-Generation Internet: Issues, Existing Approaches and Challenges', IEEE Communication Magazine, April 2000, pp. 54-61.

Liang, C. 2000, 'A Course on TCP/IP Networking with Linux', paper presented to the JCSC 15: The Consortium for Computing in Small Colleges, Hartford, CT, May 5, 2000.

Lindstrom, A. & DeVeaux, P. 1999, 'Follow the QoS road', America's Network; ProQuest Telecommunication, vol. 103, no. 9, pp. 53-60.

Mahoney, K. & Rumbel, E. 2003, 'Five tips for successful VoIP', Communication News, ProQuest Telecommunication, vol. 40, no. 5, pp. 20-21.

McElligott, T. 2004, PINGTEL RINGS UP OPEN SOURCE, Telephony, vol. 245, Primedia Business Magazines & Media Inc., pp. 14-14<http://search.epnet.com/login.aspx?direct=true&db=afh&an=12391306 >.

Nakamoto, Y. 2004, 'Towards Mobile Phone Linux', IEEE Computer Society, pp. 117-124.

Nanda, P., Simmonds, J.A. & Lee, S. 2002, 'Measuring Quality of Service in a Differentiated Sevice Domain with Linux', IFIP, Internetworking 2002 symposium, Department of Computer Systems, Faculty of IT, UTS, Sydney, Perth, Australia, pp. 1-6.

Nolle, T. 2004, 'Technical Perspective IP QoS: For Real', Telecommunications Americas; ProQuest Telecommunication, vol. 38, no. 10, p. 14.

Qiao, Z., Sun, L., Heilemann, N. & Ifeachor, E. 2004, 'A new method for VoIP Quality of Service control use combined adaptive sender rate and priority marking', IEEE Computer Society, vol. 0-7803-8533-0/04, no. 0/04, pp. 1473-1477.

Rosen, L. 2005, Open Source Licensing 'Software Freedom and Intellectual Property LAw', Prentice-Hall, New Jersey.

Salah, K. 2005, 'On the deployment of VoIP in Ethernet networks: methodology and case study, ELSEVIER Science Direct, vol. XX, no. June 2005, pp. 1-16.

Saran, C. 2005, 'Supplier group to address VoIP security concerns', Computer Weekly; ProQuest Telecommunication, p. 4.

Stefano, N. 2004, 'VoIP: The Regulatory Issues', OVUM 2004, vol. September 2004, no. September 2004, pp. 1-26.

Tolly, K. 2005, 'The Quest for QoS', Network World; ProQuest Telecommunication, vol. 22, no. 32, p. 20.

Uhl, T. 2003, 'Quality of Service in VoIP Communication', International Journal of Electronics and Communication (AEU), vol. 58(2004), pp. 178-182, 1434-8411.

VanderBrug, G. 1999, 'Making QoS work with voice over IP', Communication News, ProQuest Telecommunication, vol. 36, no. 4, p. 56.

Vittore, V. 2005, 'Does QoS Matter for Voice over IP?' Telephony; ProQuest Telecommunication, vol. 246, no. 7, pp. 28-30.

Walker, Q.J. & Hicks, T.J. 2004, Taking Charge of Your VoIP Project, February 2004 edn, Cisco Press, Indiana, USA.

Wang, S., Mai, Z., Magnussen, W., Xuan, D. & Zhao, W. 2002, 'Implementation of QoS-Provisioning System for VoIP', Real-Time and Embedded Technology and Applications Symposium (RTAS'02), vol. 0-7695-1739-0/02, IEEE Computer Society, Texas, USA, pp. 1-10.

Wilkins, D., Cole, D. & Nelson, M. 2004, 'Windows Interoperability with Linux in the Enterprise: A Solution to the High Cost of Licensing, Downtime, and Security Problems', paper presented to the CCSE: Consortium for Computer Science in Colleges, Myers, FL.

About the author
M SINGH

All the best resources in one place

Point of Sale (POS)

MenuBee

Great! You’ve successfully signed up.

Welcome back! You've successfully signed in.

You've successfully subscribed to MenuBee.

Success! Check your email for magic link to sign-in.

Success! Your billing info has been updated.

Your billing was not updated.