Chapter 12 - Review

Network+ Guide to Networks, Chapter 12 Review

Voice and Video over IP

In Chapter 1, you learned that convergence is the use of one network to simultaneously carry voice, video, and data communications. For most of the twentieth century, voice and data signals traveled over separate networks. The PSTN (Public Switched Telephone Network), based on Alexander Graham Bell’s circuit-switched model, carried telephone calls and fax transmissions. Packet-switched networks, such as the Internet, took care of e-mail, Web pages, file transfers, and access to other data resources. In the latter part of the twentieth century, the two types of networks began intersecting. However, this intersection is not necessarily seamless or efficient. In some cases, it requires modems to convert digital data into analog signals and vice versa. Networks achieve more unified integration, however, by packetizing voice—that is, digitizing the voice signal and issuing it as a stream of packets over the network. In the last 15 years, telecommunications carriers, network service providers, data equipment manufacturers, and standards organizations have focused on ways to deliver voice, video, and data over the same networks. These converged networks, as they are called, may be cheaper and more convenient, but they also require new technology. This chapter describes a variety of voice and video-over-IP applications, plus the protocols and infrastructure necessary to deliver them.

Terminology

In discussions of convergence, the use of multiple terms to refer to the same or similar technologies is common. This is partly a result of a market that developed rapidly while many different vendors marketed their own solutions and applied their preferred terminology. The terms used throughout this chapter are those most frequently cited by standards organizations such as the ITU and IETF. Before you learn how voice and video-over-IP services work, it’s useful to understand the meaning of these terms. One important term is IP telephony, the use of any network (either public or private) to carry voice signals using the TCP/IP protocol. IP telephony is more commonly known as VoIP (Voice over IP). VoIP can run over any packet-switched network. Virtually any type of data connection can carry VoIP signals, including T-carriers, ISDN, DSL, broadband cable, satellite connections, Wi-Fi, WiMAX, HSPA+, LTE, and cellular telephone networks. When VoIP relies on the Internet, it is often called Internet telephony. But not all VoIP calls are carried over the Internet. In fact, VoIP over private lines is an effective and economical method of completing calls between two locations within an organization. And because the line is private, its network congestion can be easily controlled, which often translates into better sound quality than an Internet telephone call can provide. But given the Internet’s breadth and low cost, it is appealing to consider the Internet for carrying conversations that we currently exchange over the PSTN. Voice is not the only nondata application that can be carried on a converged network. Other applications include IPTV (IP television), in which television signals from broadcast or cable networks travel over packet-switched networks. Videoconferencing, which allows multiple participants to communicate and collaborate at once through audiovisual means, is another example of using networks to carry video information. Streaming video refers to video signals that are compressed and delivered in a continuous stream. For example, when you choose to watch a television show episode on the Web, you are requesting a streaming video service. You don’t have to download the entire episode before you begin to see and hear it. When streaming videos are supplied via the Web, they may be called Webcasts. One way to distribute video signals over IP is multicasting. As described in Chapter 4, in multicasting, one node transmits the same content to every client in a defined group of nodes, such as a subnet.

IPTV, videoconferencing, streaming video, and IP multicasting belong to the range of services known as video over IP. Over time, voice and video services over packet-switched networks have matured, and as a result more users rely on them. These users, in turn, have demanded better integration with traditional data services, such as e-mail and Web browsing. In Chapter 1, you learned that unified communications (sometimes called unified messaging) is a service that makes several forms of communication available from a single user interface. In unified communications, a user can, for example, access the Web, send and receive faxes, e-mail messages, voice mail messages, instant messages, or telephone calls, and participate in videoconference calls—all from one console. This overview of the terms used when discussing converged services gives you a sense of how many applications fall into this category. Now you are ready to learn how they work.

VoIP Applications and Interfaces

VoIP (pronounced “voyp”) has existed in various forms for over a decade. Although organizations were slow to adopt it at first, as networks became faster, more reliable, and more accessible, use of VoIP increased dramatically. Significant reasons for implementing VoIP include the following:

Lower costs for voice calls—In the case of long-distance calling, using VoIP over a WAN allows an organization to avoid paying long-distance telephone charges, a benefit known as toll bypass. For example, an organization that already leases T3s between its offices within a region can use the T3s to carry voice traffic between colleagues.

Supply new or enhanced features and applications—VoIP runs over TCP/IP, an open protocol suite, whereas the PSTN runs over proprietary protocols. This means developers with enough skill and interest can develop their own VoIP applications, making the possibilities for new VoIP features and services endless. It also means that off-the-shelf VoIP applications can be modified to suit a particular organization’s needs.

Centralize voice and data network management—When voice and data transmissions use the same infrastructure, a network manager needs only to design, maintain, and troubleshoot a single network. Furthermore, on that network, VoIP devices can provide detailed information about voice transmissions, such as the date, time, and duration of calls, in addition to their originating number and caller names.

Voice and data can be combined on a network in several different configurations. VoIP callers can use either a traditional telephone, which sends and receives analog signals, a telephone specially designed for TCP/IP transmission, or a computer equipped with a microphone, speaker, and VoIP client software. And on any VoIP network, a mix of these three types of clients is possible. The following sections explain how analog and digital voice networks are integrated and describe equipment necessary to accomplish such integration.

Analog Telephones

If a VoIP caller uses a traditional telephone, signals issued by the telephone must be converted to digital form before being transmitted on a TCP/IP-based network. In fact, even if the entire VoIP connection is digital, voice signals still need to be converted from their natural, analog form into bits. This conversion involves first compressing and encoding analog signals, functions that occur at the Presentation layer of the OSI model. Any method for accomplishing this conversion is known as a codec (a word that derives from its function as a coder/decoder). Detailing the wide variety of voice and video codecs is beyond the scope of this book. However, to successfully implement converged networks, you should understand what types of equipment are necessary to accomplish analog-to-digital conversion. One possibility is to connect an analog telephone to a VoIP adapter, sometimes called an ATA (analog telephone adapter). The ATA might be a card within a computer workstation or an externally attached device that allows for one or more telephone connections. The traditional telephone line connects to an RJ-11 port on the adapter. The ATA, along with its device drivers and software on the computer, converts analog voice signals to IP packets and vice versa. Figure 12-1 shows an ATA that supports two telephone connections. A second way to achieve this conversion is by connecting an analog telephone line to a switch, router, or gateway capable of accepting analog voice signals, converting them into packets, then issuing the packets to a data network—and vice versa. Like the switches, routers, and gateways you learned about earlier in this book, VoIP-enabled devices come with a variety of features, including support for NAT, VPN protocols, encryption, and more. Figure 12-2 shows a VoIP router that accepts up to four telephone lines. Next to the bank of eight RJ-11 ports for incoming analog lines are two RJ-45 ports to connect the router to an Ethernet network. A third example of an analog-to-digital voice conversion device is a digital PBX or, more commonly, an IP-PBX.(PBX stands for private branch exchange, which is the term used to describe a telephone switch that connects and manages calls within a private organization.) In general, an IP-PBX is a private switch that accepts and interprets both analog and digital voice signals. Thus, it can connect with both traditional PSTN lines and data networks. An IP-PBX transmits and receives IP-based voice signals to and from other network connectivity devices, such as routers or gateways. Most IP-PBX systems are packaged with sophisticated software that allows network managers to configure and maintain an organization’s phone system. For example, the system can be set up to ring a user’s desk phone and cell phone simultaneously. And because an IP-PBX stores call information electronically, source and destination numbers, call times, durations, and voice-mail messages can be accessed via a Web interface. A special type of IP-PBX is one that exists on the Internet. Instead of installing an IP-PBX on its WAN, an organization might contract with a service provider for call management services in a hosted PBX arrangement. (Hosted PBXs may also be called virtual PBXs, although this term is a trademark of the VirtualPBX Company.) Organizations that choose a hosted PBX don’t need to install or maintain hardware or software for call completion and management. In a fourth scenario, the traditional telephone connects to an analog PBX, which then connects to a voice-data gateway. In this case, the gateway connects the traditional telephone circuits with a TCP/IP network (such as the Internet or a private WAN). The gateway digitizes incoming analog voice signals, compresses the data, assembles the data into packets, and then issues the packets to the packet-switched network. When transferring calls from a packet-switched network to a circuit-switched network (for example, if you call your home telephone number from your office’s IP telephone), the gateway performs the same functions in the reverse order.

IP Telephones

Most new VoIP installations use IP telephones (or IP phones), which, unlike traditional phones, transmit and receive only digital signals. When a caller uses an IP telephone, her voice is immediately digitized and issued from the telephone to the network in packet form. To communicate on the network, each IP telephone must have a unique IP address, just as any client connected to the network has a unique IP address. The IP telephone looks like a traditional touch-tone phone, but connects to an RJ-45 wall jack, like a computer workstation. Its connection may then pass through a connectivity device, such as a switch or router, before reaching the IP-PBX. An IP-PBX may contain its own voice-data gateway, or it may connect to a separate voice-data gateway, which is then connected to the network backbone. Figure 12-5 illustrates different ways IP telephones can connect with a data network. IP telephones act much like traditional telephones. For example, they feature speed-dialing, call hold, transfer, and forwarding buttons, conference calling, voice-mail access, speakers and microphones, and an LCD screen that displays caller ID and call hold information. IP telephones come in both mobile and wired styles. More sophisticated IP telephones offer features not available with traditional telephones. Because IP telephones are essentially network clients, like workstations, the number and types of customized features that can be programmed for use with these phones is limitless. For example, IP telephone screens can act as Web browsers that allow users to complete a call by clicking on a telephone number. Another benefit of IP telephones is their mobility. Because IP telephones are addressable over a network, they can be moved from one office to another office, connected to a wall jack, and be ready to accept or make calls. Compare this with the traditional method of moving telephone extensions, which requires reprogramming the extension’s location in a PBX database. A user would have to wait for the network administrator to perform this change before her telephone extension would work in a new location. With IP telephones, however, the user is free to move to any point on the network without missing a call. One issue that faces IP telephones is the need for electric current. A conventional analog telephone obtains current from the local loop. This is necessary for signaling—for example, to make your phone ring and to provide a dial tone. However, IP telephones are not directly connected to the local loop. Instead, most obtain electric current from a separate power supply. This makes IP telephones susceptible to power outages in a way that analog telephones are not. It also points to the need for assured backup power sources in organizations that rely on IP telephones. In some VoIP installations, IP telephones obtain current via their Ethernet connection using PoE (Power over Ethernet). A typical IP phone is shown in Figure 12-6. Using IP telephones is not the only way to benefit from a fully digital voice connection. Instead, an off-the-shelf workstation can be programmed to act like an IP telephone, as described in the next section.

Softphones

Rather than using traditional telephones or IP telephones, a third option is to use a computer programmed to act like an IP telephone, otherwise known as a softphone. Softphones and IP telephones provide the same calling functions; they simply connect to the network and deliver services differently. Before it can be used as a softphone, a computer must meet minimum hardware requirements (which any new workstation purchased at an electronics store would likely meet), be installed with an IP telephony client, and communicate with a digital telephone switch. In addition, softphone computers must have a sound card capable of full-duplex transmission, so that both the caller and the called party can speak at the same time. Finally, a softphone also requires a microphone and speakers or a headset. Skype, the popular Internet telephony software, is one type of softphone. After a user starts the softphone client software, he is typically presented with a graphical representation of a telephone dial pad, as shown in Figure 12-7.

The interface might also present a list of telephone numbers in the caller’s address book, so that the caller can click on the number he wants to call. And like IP telephones, the program features buttons for call forwarding, speed dialing, conferencing, and so on—except that on a softphone, these buttons are clickable icons. Unlike many traditional phones, softphones allow the user to customize the graphical interface. For example, an administrative assistant who spends most of his time calling clients and vendors on behalf of his supervisor can position a list of clickable, frequently called numbers in the foreground of his default interface. One difference between IP telephones and softphones is that a softphone’s versatile connectivity makes it an optimal VoIP solution for traveling employees and telecommuters. For example, suppose you are a district sales manager with a home office and you supervise 32 sales representatives throughout the Pacific Northwest. Your company uses VoIP, with an IP-PBX connected to the company headquarters’ LAN. At your home office, you have a desktop workstation equipped with a sound card, headset, and softphone software. You also lease a DSL connection to your local carrier, which allows you to log on to your company’s LAN from home. After logging on to the LAN, you initiate the softphone client and then log on to the company’s IP-PBX. By logging on to the IP-PBX, you access your personal call profile and indicate to the IP-PBX that your calls should be routed to your home computer. However, because you are a district sales manager, you only spend half of the time working from home. The other half of the time you travel to visit your sales representatives across the region. During that time, you use a laptop that, like your home workstation, is equipped with a sound card, headset, and the softphone client software. While on the road, you use remote connectivity software to access your company’s LAN, and then initiate your softphone client. Now your calls are directed to your laptop computer, rather than your home workstation. No matter where you are, you can establish a remote telephone extension, if the computer has the appropriate software and hardware installed. Figure 12-8 depicts the use of softphones on a converged network. Besides their extreme mobility, another advantage to softphones is the capability for convenient, localized call management. Like IP phones, softphone clients can easily track the date, time, and duration of calls, in addition to their originating number and caller names. A softphone user can also, for example, export call information to a billing or accounting program on the same workstation. This feature simplifies record keeping and billing for professionals—such as lawyers or consulting engineers—who bill their customers by the hour. Now that you understand the variety of ways VoIP services may be implemented, you are ready to learn about the different types of video services that packet-switched networks may carry.

Video-over-IP Applications and Interfaces

Cisco Systems, the largest supplier of networking hardware in the world, estimates that by 2015, over two-thirds of the traffic carried by the Internet will be video traffic, with online video viewers totaling one billion people. This extraordinary growth is thanks to the large quantity of video content from providers such as Netflix and Hulu, the increasing volume of devices accessing the Internet, including smartphones and tablet PCs, plus the decreasing cost of bandwidth and equipment. Given these networking trends, it’s important to understand the types of video services that TCP/IP networks can carry and the hardware and software they rely on. The following sections divide video-over-IP services into three categories: streaming video, IPTV, and videoconferencing. However, divisions between these services are not always clear, as you’ll learn.

Also bear in mind that no matter what the application or distribution method, every video-over-IP transmission begins with digitizing the audio and visual signals using one of several popular video codecs, such as MPEG-4. Details of video codecs are beyond the scope of this book.

Streaming Video

You have already learned that streaming video is a service in which audiovisual signals are compressed and delivered over the Internet in a continuous stream. If you have watched a YouTube video on the Internet, you have used streaming video. Because most networks are

TCP/IP-based, most streaming video belongs to the category of video over IP. Among all video-over-IP applications, streaming video is perhaps the simplest. A user needs only to have a computer with sufficient processing and caching resources, plus the appropriate audiovisual hardware and software to view encoded video. On the transmission end, video can be delivered by any computer with sufficient capabilities to store and send the video. This might be a streaming server dedicated to the task or any computer that performs video streaming among other tasks. Streaming video can traverse any type of TCP/IP network, though it often relies on the Internet. One popular way of providing video streams is to make them available as stored files (saved in any one of a number of popular video formats) on a server. The viewer then chooses to watch the video at his convenience, typically from a Web browser. Upon receiving a request, the server delivers the video to the viewer. This type of service, in which the video file remains on the server until it is specifically requested by the user, is known as video-on-demand. When you choose to watch a news report from your local TV channel’s Web page, for example, you are making use of video-on-demand. In another form of streaming video, the video is issued live—that is, directly from the source to the user as the camera captures it. For example, suppose you wanted to watch a political debate that’s being broadcast by a TV network. You could access the network’s Web site and watch the debate as it happens using live streaming video. One drawback to live streaming is that content cannot be edited before it’s distributed. Another potential drawback is that viewers must connect with the stream as it occurs, whereas they can use video-on-demand at their convenience. In addition, video-on-demand allows viewers to control their viewing experience, for example, by pausing, rewinding, or fast-forwarding. Figure 12-9 illustrates video-on-demand and live streaming video services. The distinction between on-demand video and live streaming video is just the beginning. You also need to consider the number of clients receiving each service. For example, an IT manager in one office might use his laptop and its built-in camera to capture and issue a video of himself explaining a technical topic to one of his employees in another office. This is an example of point-to-point video over IP. Or he might issue the video stream to a whole group of employees in a point-to-multipoint manner. You might recall the terms unicast and multicast from Chapter 4 and assume that point-to-multipoint streaming video means multicast transmission. That’s not necessarily the case. In IP multicasting, a source issues data to a defined group of IP addresses. In fact, many streaming video services—and nearly all of those issued over a public network, such as the Internet—are examples of unicast transmissions. In a unicast transmission, a single node issues a stream of data to one other node. In a VoIP call, for example, one IP phone addresses another in unicast fashion. If many Internet users watch CSPAN’s streaming video on the Web simultaneously, the CSPAN source would issue encoded audiovisual signals to each viewer via separate unicast transmissions. In the example of an IT manager sharing a video discussion with his employees in another office, the transmission might be unicast or multicast, depending on how he configured it. Finally, streaming video services may also be classified according to the type of network they use, private or public. Watching YouTube videos or TV episodes on Hulu are obviously cases of streaming video issued over a public network, the Internet. Examples of streaming video on private networks include educational videos delivered over the private networks of schools, businesses, or other organizations. For example, a guest speaker’s presentation at a college’s main campus auditorium could be filmed and transmitted via live streaming to classrooms in the colleges’ satellite campuses, all without ever leaving the college’s private network. Most, though not all, examples of streaming video take place over public networks. The following section describes a video-over-IP service that typically makes use of private networks.

IPTV (IP Television)

In Chapter 7, you learned about the networks that telecommunications carriers and cable companies have established to deliver high-bandwidth Internet connections to their customers. These networks are now being used to deliver digital television signals using IPTV. In fact, because of digital video’s value as an added service, your local telephone company is likely investing significant sums into the hardware and software that make IPTV possible. Because telecommunications carriers are leading the way with IPTV installation, this section concentrates on their network architecture and components but bear in mind that cable companies are investing in this technology, too. Several elements come together to deliver digital video to consumers, as shown in Figure 12-10. Each element and its role are described next. To begin, a telco accepts video content at a head-end. The content may include signals captured from satellite video feeds, national or regional broadcasts, or local content, such as live feeds of city council meetings. Typically, this content arrives in analog format, and at the head-end, an encoder converts it to digital format. At the telco’s CO (central office), one or more servers manage customer subscription information, encrypt video to comply with digital rights regulations, publish channel listing information, and associate each video input with its own channel, among other things. Also at the CO, each video channel is assigned to a multicast group. Multicasting makes sense for delivering IPTV. First, it’s a simple way of managing content delivery. For example, 2000 of the telco’s customers might choose to watch a Monday night football game. Rather than supply the video via 2000 separate unicast transmissions, the carrier can issue one multicast transmission to the entire group of 2000 subscribers. The second advantage to using multicasting has to do with local loop capacity. As you know, fiber to the home is still rare in the United States, and most local loops rely on copper cabling. Therefore, throughput is limited. In an environment where customers may choose from hundreds of channels, supplying all the choices to every customer at all times would overwhelm the local loop’s capacity. Even supplying more than a few channels would be too much. Instead, the telco transmits only the content a subscriber has chosen. When an IPTV user changes the channel, therefore, she is merely opting out of one IP multicast group and opting into another. Recall from Chapter 4 that multicasting is managed by IGMP (Internet Group Management Protocol). Therefore, IGMP underlies all IPTV implementations at the Network layer of the OSI model. However, IGMP can only identify group members. To ensure efficient content delivery to a multicast group, routers communicate using a multicast routing protocol. Several multicast routing protocols exist. Which protocol the network uses is less important than the fact that all Layer 3 devices communicate using the same multicast routing protocol. A compressed, digital video signal travels over the telco’s network just as a data signal would. For example, if a subscriber obtains DSL service from the carrier, the signal would go from the telco’s router to a DSLAM (DSL access multiplexer), either at the CO or at a remote switching facility, and then to the subscriber’s DSL modem. If the subscriber obtains service wirelessly—for example, via WiMAX—the signal would be issued from the carrier’s router to an antenna on a tower, and then to an antenna and WiMAX connectivity device at the subscriber’s home. After passing through the DSL modem or home WiMAX device, the video signal is decoded and issued to a television by a set-top box.  Besides decoding the video signal, set-top boxes communicate with content servers to manage video delivery. For example, the set-top box delivers TV program schedules from the content server to subscribers and sends a subscriber’s channel request to the content server. In cases where IPTV providers allow pay-per-view or video-on-demand programming, set-top boxes manage requesting and delivering those services. Set-top boxes may also allow a user to browse the Internet from his TV. Figure 12-11 shows one type of set-top box. A significant advantage of delivering video services over a telecommunications carrier’s or cable company’s network is that those firms control the connection end to end. This means they can better monitor and adjust its QoS (quality of service).

Later in this chapter, you’ll learn some techniques for controlling the QoS of voice and video transmissions. The next section describes a third popular video-over-IP service, videoconferencing.

Videoconferencing

So far in this chapter, you have learned about unidirectional video-over-IP services—that is, video delivered to a user who only watches the content, but does not respond with her own. In most examples of videoconferencing, connections are full-duplex, and participants may send and receive audiovisual signals. This allows two or more people in different locations to see and hear each other in real time. As you can imagine, the cost savings and convenience of such a service make it especially attractive to organizations with offices, clients, or consultants scattered across the nation or the globe. Besides replacing face-to-face business meetings and allowing collaboration, uses for videoconferencing include the following:

Telemedicine, or the provision of medical services from a distance—For example, a physician can view and listen to a patient in another location. Often, the patient is accompanied by a nurse or physician’s assistant, who might administer tests and supply information about the patient’s condition. For patients who live far from major medical facilities, this saves the cost, time, and potential health risks of having to travel long distances. NASA is developing telemedicine capabilities for diagnosing patients in space.

Tele-education, or the exchange of information between one or more participants for the purposes of training and education—A significant benefit of tele-education is the capability for one or a few experts to share their knowledge with many students.

Judicial proceedings, in which judges, lawyers, and defendants can conduct arraignments, hearings, or even trials while in different locations—This not only saves costs, but may minimize potential security risks of transporting prisoners.

Surveillance, or remotely monitoring events happening at one or more distant locations—Unlike previously mentioned videoconferencing applications, surveillance is typically unidirectional. In other words, security personnel watch (and perhaps also listen) to live video feeds from multiple locations around a building or campus, but do not send audiovisual signals to those locations.

Hardware and software requirements for videoconferences include, at minimum, a means

for each participant to generate, send, and receive audiovisual signals. This may be accomplished by workstations that have sufficient processing resources, plus cameras, microphones, and videoconferencing software to capture, encode, and transmit audiovisual signals.

Instead of a workstation, viewers may use a video terminal or a video phone, a type of phone that includes a screen, such as the one shown in Figure 12-12. These devices can decode compressed video and interpret transport and signaling protocols necessary for conducting videoconference sessions. When more than two people participate in a videoconference, for example, in a point-to-multipoint or multipoint-to-multipoint scenario, a video bridge is required. A video bridge manages multiple audiovisual sessions so that participants can see and hear each other. Video bridges may exist as a piece of hardware or as software, in the form of a conference server. For an organization that only occasionally uses videoconferencing, Internet-accessible video bridging services can be leased for a predetermined period. Organizations such as universities that frequently rely on videoconferencing might maintain their own conference servers or supply each auditorium, for example, with its own video bridge.

To establish and manage videoconferencing sessions, video bridges depend on signaling protocols, which are described in the following section.

Signaling Protocols

In VoIP and video-over-IP transmission, signaling is the exchange of information between the components of a network or system for the purposes of establishing, monitoring, or releasing connections as well as controlling system operations. Simply put, signaling protocols set up and manage sessions between clients. Some functions performed by signaling protocols include the following:

·         Requesting a call or videoconference setup

·         Locating clients on the network and determining the best routes for calls or video transmissions to follow

·         Acknowledging a request for a call or videoconference setup and setting up the connection

·         Managing ringing, dial tone, call waiting, and in some cases, caller ID and other telephony features

·         Detecting and reestablishing dropped calls or video transmissions

·         Properly terminating a call or videoconference

On the circuit-switched portions of the PSTN, a set of standards established by the ITU known as SS7 (Signaling System 7) typically handles call signaling. You should be familiar with this term, as it might appear in discussions of interconnecting the PSTN with networks running VoIP.

In the early days of VoIP, vendors developed their own, proprietary signaling protocols, which meant that if you wanted to use the Internet to call your neighbor, you and your neighbor had to use hardware or software from the same manufacturer. Now, however, most VoIP and video-over-IP clients and gateways use standardized signaling protocols. The following sections describe the most common of these.

H.323

On the circuit-switched portions of the PSTN, a set of standards established by the ITU known as SS7 (Signaling System 7) typically handles call signaling. You should be familiar with this term, as it might appear in discussions of interconnecting the PSTN with networks running VoIP. H.323 is an ITU standard that describes an architecture and a group of protocols for establishing and managing multimedia sessions on a packet-switched network. H.323 protocols may support voice or video-over-IP services.

Before learning about H.323 protocols, it’s helpful to understand the set of terms unique to H.323 that ITU has designated. Elements of VoIP and video-over-IP networks have special names in H.323 parlance. Following are five key elements identified by H.323:

H.323 terminal—Any node that provides audio, visual, or data information to another node. An IP phone, video phone, or a server issuing streaming video could be considered an H.323 terminal.

H.323 gateway—A device that provides translation between network devices running H.323 signaling protocols and devices running other types of signaling protocols (for example, SS7 on the PSTN).

H.323 gatekeeper—The nerve center for networks that adhere to H.323. Gatekeepers authorize and authenticate terminals and gateways, manage bandwidth, and oversee call routing, accounting, and billing. Gatekeepers are optional on H.323 networks.

MCU (multipoint control unit)—A computer that provides support for multiple H.323 terminals (for example, several workstations participating in a videoconference) and manages communication between them. In videoconferencing, a video bridge serves as an MCU.

H.323 zone—A collection of H.323 terminals, H.323 gateways, and MCUs that are managed by a single H.323 gatekeeper. Figure 12-13 illustrates an H.323 zone comprising four terminals, one gateway, and one MCU.

Now that you understand the elements that belong to an H.323 network, you are ready to learn about the H.225 and H.245 signaling protocols, which are specified in the H.323 standard. Both protocols operate at the Session layer of the OSI model. However, each performs a different function. H.225 is the H.323 protocol that handles call or videoconference signaling. For instance, when an IP telephone user wants to make a call, the IP telephone requests call setup (from the H.323 gateway) via H.225. The same IP telephone would use H.225 protocol to announce its presence on the network, to request the allocation of additional bandwidth, and to indicate when it wants to terminate a call. Another H.323 Session layer protocol, H.245, ensures that the type of information—whethvoice or video—issued to an H.323 terminal is formatted in a way that the H.323 terminal can interpret. To perform this task, H.245 first sets up logical channels between the sending and receiving nodes. On a VoIP or video-over-IP network, these logical channels are identified as port numbers at each IP address. One logical channel is assigned to each transmission direction. Thus, for a call between two IP telephones, H.245 would use two separate control channels. Note that these channels are distinct from the channels used for H.225 call signing. They are also different from channels used to exchange the actual voice or video sign(for example, the words you speak during a conversation or the pictures transmitted in videoconference). In addition to the H.225 and H.245 signaling protocols, the H.323 standard also special interoperability with certain protocols at the Presentation layer, such as those responsible coding and decoding signals, and at the Transport layer. Later in this chapter, you’ll learn about the Transport layer protocols used with voice and video services. ITU codified H.323 as an open protocol for multiservice signaling in 1996. Early versions the H.323 protocol suffered from slow call setup, due to the volume of messages exchanged between nodes. Since that time, ITU has revised and improved H.323 standards several times and H.323 remains a popular signaling protocol on large voice and video networks. AftH.323 was released, however, another protocol for VoIP call signaling, SIP, emerged and attracted the attention of network administrators.

SIP (Session Initiation Protocol)

SIP (Session Initiation Protocol) is a protocol that performs functions similar to those performed by H.323. SIP is an Application layer signaling and control protocol for multi-service-packet-based networks. The protocol’s developers modeled it on HTTP. For example, text-based messages that clients exchange to initiate a VoIP call are formatted like an HTTP request and rely on URL-style addresses. Developers also aimed to reuse as many existing TCP/IP protocols as possible for managing sessions and providing enhanced services. Furthermore, they wanted SIP to be modular and specific. SIP’s capabilities are limited to the following:

·         Determining the location of an endpoint, which in SIP terminology refers to any client, server, or gateway communicating on a network; this means SIP translates the endpoint’s name into its current network address.

·         Determining the availability of an endpoint; if SIP discovers that a client is not available, it returns a message indicating whether the client was already connected a call or simply didn’t respond.

·         Establishing a session between two endpoints and managing calls by adding (inviting), dropping, or transferring participants.

·         Negotiating features of a call or videoconference when it’s established; for example, agreeing on the type of encoding both endpoints will employ.

·         Changing features of a call or videoconference while it’s connected

SIP’s functions are more limited than those performed by the protocols in the H.323 group. For example, SIP does not supply some enhanced features, such as caller ID, that H.323 does. Instead, it depends on other protocols and services to supply them. As with H.323, a SIP network uses terms and follows a specific architecture mapped out in the standard. Components of a SIP network include the following:

User agent—This is any node that initiates or responds to SIP requests.

User agent client—These are end-user devices, which may include workstations, tablet computers, smartphones, or IP telephones. A user agent client initiates a SIP connection.

User agent server—This type of server responds to user agent clients’ requests for session initiation and termination. Practically speaking, a device such as an IP telephone can act as a user agent client and server, thus allowing it to directly contact and establish sessions with other clients in a peer-to-peer fashion. As you have learned, however, peer-to-peer arrangements are undesirable because they become difficult to manage when more than a few users participate. User agent clients and user agent servers are considered user agents.

Registrar server—This type of server maintains a database containing information about the locations (network addresses) of each user agent in its domain. When a user agent joins a SIP network, it transmits its location information to the SIP registrar server.

Proxy server—This type of server accepts requests for location information from user agents, then queries the nearest registrar server on behalf of those user agents. If the recipient user agent is in the SIP proxy server’s domain, then that server will also act as a go-between for calls established and terminated between the requesting user agent and the recipient user agent. If the recipient user agent is not in the SIP proxy server’s domain, the proxy server will pass on session information to a SIP redirect server. Proxy servers are optional on a SIP network.

Redirect server—This type of server accepts and responds to requests from user agents and SIP proxy servers for location information on recipients that belong to external domains. A redirect server does not get involved in establishing or maintaining sessions. Redirect servers are optional on SIP networks.

Figure 12-14 shows how the elements of a SIP system may be arranged on a network. In this example, user agents connect to proxy servers, which accept and forward addressing requests and also make use of redirect servers to learn about user agents on other domains. For purposes of illustration, the registrar server, proxy server, and redirect server are shown as separate computers in Figure 12-14. However, on a SIP network all might be installed on a single computer. Some VoIP vendors prefer SIP because of its simplicity, which makes SIP easier to maintain than H.323. And because it requires fewer instructions to control a call, SIP consumes fewer processing resources than H.323. In some cases, SIP is more flexible than H.323. For example, it is designed to work with many types of Transport layer protocols, not just one. One popular system based on SIP is Asterisk, an open source IP-PBX software package. Companies that provide telephone equipment, such as 3Com, Avaya, Cisco, and Nortel, also supply SIP software with their hardware. SIP and H.323 regulate call signaling and control for VoIP or video-over-IP clients and servers. However, they do not account for communication between media gateways. This type of communication is governed by one of two protocols, MGCP or MEGACO, which are discussed in the following sections.

MGCP (Media Gateway Control Protocol) and MEGACO (H.248)

You have learned about gateways in the context of WANs and VPNs. Gateways are also integral to converged networks. A media gateway accepts PSTN lines, converts the analog signals into VoIP format, and translates between SS7, the PSTN signaling protocol suite, and VoIP signaling protocols, such as H.323 or SIP. You have also learned that information (or “payload,” such as the speech carried by a VoIP network) uses different channels from and may take different logical or physical paths than control signals. In fact, to expedite information handling, the use of separate physical paths is often preferable. The reason for this is that if media gateways are freed from having to process control signals, they can dedicate their resources (for example, ports and processors) to encoding, decoding, and translating data. As a result, they process information faster. And as you have learned, faster data processing on a converged network is particularly important, given quality and reliability concerns. However, gateways still need to exchange and translate signaling and control information with each other so that voice and video packets are properly routed through the network. To do so, gateways rely on an intermediate device known as an MGC (media gateway controller). As its name implies, an MGC is a computer that manages multiple media gateways. This means that it facilitates the exchange of call signaling information between these gateways. It also manages and disseminates information about the paths that voice or video signals take between gateways. Because it is software that performs call switching functions, an MGC is sometimes called a Softswitch. For example, suppose a network has multiple media gateways, all of which accept thousands of connections from both the PSTN and from private TCP/IP WAN and LAN links. When a media gateway receives a call, rather than attempting to determine how to handle the call, the gateway simply contacts the media gateway controller with a message that essentially says, “I received a signal. You figure out what to do with it next.” The media gateway controller then determines which of the network’s media gateways should translate the information carried by the signal. It also figures out which physical media the call should be routed over, according to what signaling protocols the call must be managed, and to what devices the call should be directed.

After the media gateway controller has processed this information, it instructs the appropriate media gateways how to handle the call. The media gateways simply follow orders from the media gateway controller. MGCs are especially advantageous on large VoIP networks—for example, at a telecommunications carrier’s CO. In such an environment, they make a group of media gateways appear to the outside world as one large gateway. This centralizes call control functions, which can simplify network management. Figure 12-15 illustrates this model. (Note that in this figure, as on most large networks, the media gateways supply access services.) MGCs communicate with media gateways according to one of several protocols. The older protocol is MGCP (Media Gateway Control Protocol). MGCP is commonly used on multiservice networks that support a number of media gateways. It can operate in conjunction with H.323 or SIP call signaling and control protocols. A newer gateway control protocol is MEGACO. MEGACO performs the same functions as MGCP, but using different commands and processes. Like MGCP, MEGACO can operate with H.323 or SIP. Many network engineers consider MEGACO superior to MGCP because it supports a broader range of network technologies, including ATM. MEGACO was developed by cooperative efforts of the ITU and IETF, and the ITU has codified the MEGACO protocol in its H.248 standard.

Bear in mind that this chapter describes only some of the signaling protocols used on converged networks. In fact, some softphones, VoIP servers, and videoconferencing software packages (for example, Skype) use proprietary protocols, which means that these devices or applications will only work with other devices or applications that use the same proprietary protocols.

Now that you are familiar with the most popular session control protocols used on converged networks, you are ready to learn about the transport protocols that work in tandem with those session control protocols.

Transport Protocols

The protocols you just learned about only communicate information about a voice or video session. At the Transport layer, a different set of protocols is used to actually deliver the voice or video payload—for example, the bits of encoded voice that together make up words spoken into an IP telephone. Recall that on a TCP/IP network, the UDP and TCP protocols operate at the Transport layer of the OSI model. TCP is connection oriented and, therefore, provides some measure of delivery guarantees. UDP, on the other hand, is connectionless, and does not pay attention to the order in which packets arrive or how quickly they arrive. Despite this lack of accountability, UDP is preferred over TCP for real-time applications such as telephone conversations and videoconferences because it requires less overhead and, as a result, can transport packets more quickly. In transporting voice and video signals, TCP’s slower delivery of packets is intolerable. However UDP’s occasional loss of packets is tolerable—that is, as long as additional protocols are used in conjunction with UDP to make up for its faults.

RTP (Real-time Transport Protocol)

One protocol that helps voice and video networks overcome UDP’s shortcomings is the RTP (Real-time Transport Protocol). RTP operates at the Application layer of the OSI model (despite its name) and relies on UDP at the Transport layer. It applies sequence numbers to indicate the order in which packets should be assembled at their destination. Sequence numbers also help to indicate whether packets were lost during transmission. In addition, RTP assigns each packet a time stamp that corresponds to when the data in the packet were sampled from the voice or video stream.

This time stamp helps the receiving node to compensate for network delay and to synchronize the signals it receives. RTP alone does not, however, provide any mechanisms to detect whether or not it’s successful. For that, it relies on a companion protocol, RTCP.

RTCP (Real-time Transport Control Protocol)

RTCP (Real-time Transport Control Protocol or RTP Control Protocol) provides feedback on the quality of a call or videoconference to its participants. RTCP packets are transmitted periodically to all session endpoints. RTCP allows for several types of messages. For example, each sender issues information about its transmissions’ NTP (Network Time Protocol) time stamps, RTP time stamps, number of packets, and number of bytes. Recipients of RTP data use RTCP to issue information about the number and percentage of packets lost and delay suffered between the sender and receiver. RTCP also maintains identifying information for RTP sources. The value of RTCP lies in what clients and their applications do with the information that RTCP supplies. For example, if a call participant’s software uses RTCP to report that an excessive number of packets are being delayed during transmission, the sender’s software can adjust the rate at which it issues RTP packets. RTCP is not mandatory on networks that use RTP. In fact, on large networks running high bandwidth services, such as IPTV, RTCP might not be able to supply useful feedback in a timely manner. Some network administrators prefer not to use it. It’s important to realize that although RTP and RTCP can provide information about packet order, loss, and delay, they cannot do anything to correct transmission flaws. Attempts to correct these flaws, and thus improve the quality of a voice or video signal, are handled by QoS protocols, which are discussed next.

QoS (Quality of Service) Assurance

Despite all the advantages to using VoIP and video over IP, it is more difficult to transmit these types of signals over a packet-switched network than it is to transmit data signals.

First, more so than data transmissions, voice and video can easily be distorted by a connection’s inconsistent QoS. When you talk with your friend, you need to hear his syllables in the order in which he uttered them, and preferably, without delay. When you watch a movie over the Web, you want to see the scenes sequentially and without interruption. In general, to prevent delays, disorder, and distortion, a voice or video connection requires more dedicated bandwidth than a data connection. In addition, it requires the use of techniques that ensure high QoS. QoS is a measure of how well a network service matches its expected performance. From the point of view of a person using VoIP or video over IP, high QoS translates into an uninterrupted, accurate, and faithful reproduction of audio or visual input. Low, or poor, QoS is often cited as a key disadvantage to using VoIP or video over IP.  But although early attempts at converged services sounded and looked dreadful, thanks to technology improvements, these services now achieve quality comparable to the PSTN (in the case of VoIP) and cable television (in the case of video over IP). Network engineers have developed several techniques to overcome the QoS challenges inherent in delivering voice and video over IP. The following sections describe three of these techniques, all of which are standardized by IETF.

RSVP (Resource Reservation Protocol)

RSVP (Resource Reservation Protocol), specified in RFC 2205, is a Transport layer protocol that attempts to reserve a specific amount of network resources for a transmission before the transmission occurs. In other words, assuming it is successful; RSVP will create a path between the sender and receiver that provides sufficient bandwidth for the signal to arrive without suffering delay. You can think of RSVP as a technique that addresses the QoS problem by emulating a circuit-switched connection. 

To establish the path, the sending node issues a PATH statement via RSVP to the receiving node. This PATH message indicates the amount of bandwidth the sending node requires for its transmission, as well as the level of service it expects. RSVP allows for two service types: guaranteed service and controlled-load service. Guaranteed service assures that the transmission will not suffer packet losses and that it will experience minimal delay. Controlled-load service provides the type of QoS a transmission would experience if the network carried little traffic. Each router that the PATH message traverses marks the transmission’s path by noting which router the PATH message came from. This process continues until the PATH message reaches its destination. But the reservation is not yet complete. After the destination node receives the PATH message, it responds with a Reservation Request (RESV) message. The RESV message follows the same path taken by the PATH message, but in reverse. It reiterates information about bandwidth requirements that the sending node transmitted in its PATH message. It also includes information about the type of service the sending node requested. Upon receiving the RESV message, each router between the destination node and the sender allocates the requested bandwidth to the message’s path. This assumes that each router is capable of interpreting RSVP messages and also has sufficient bandwidth to allocate to the transmission. If routers do not have sufficient bandwidth to allocate, they reject the reservation request. After each router in the established path has agreed to allocate the specified amount of bandwidth to the transmission, the sending node transmits its data. It’s important to note that RSVP messaging is separate from the data transmission. In other words, RSVP does not modify the packets that carry voice or video signals. Another characteristic about RSVP is that it can only specify and manage unidirectional transmission. Therefore, for two users to participate in a VoIP call or a videoconference, the resource reservation process must take place in both directions. Because it emulates a circuit-switched path, RSVP provides excellent QoS. However, one drawback to RSVP is its high overhead. It requires a series of message exchanges before data transmission can occur. Thus, RSVP consumes more network resources than some other QoS techniques. Although RSVP might be acceptable on small networks, it is less popular on large, heavily trafficked networks. Instead, these networks use more streamlined QoS techniques, such as DiffServ.

DiffServ (Differentiated Service)

DiffServ (Differentiated Service) is a simple technique that addresses QoS issues by prioritizing traffic. It differs significantly from RSVP in that it modifies the actual IP datagrams that contain payload data. Also, it takes into account all types of network traffic, not just the time-sensitive services such as voice and video. That way, it can assign voice streams a high priority and at the same time assign unessential data streams (for example, an employee surfing the Internet on his lunch hour) a low priority. This technique offers more protection for the time sensitive voice and video services. To prioritize traffic, DiffServ places information in the DiffServ field in an IPv4 datagram. (For a review of the fields in an IP datagram, refer to Chapter 4.) In IPv6 datagrams, DiffServ uses a similar field known as the Traffic Class field. This information indicates to the network routers how the data stream should be forwarded. DiffServ defines two types of forwarding: EF (Expedited Forwarding) or AF (Assured Forwarding). In EF, a data stream is assigned a minimum departure rate from a given node. This technique circumvents delays that slow normal data from reaching its destination on time and in sequence. In AF, different levels of router resources can be assigned to data streams. AF prioritizes data handling, but provides no guarantee that on a busy network, packets will arrive on time and in sequence. This description of DiffServ’s prioritization mechanisms is oversimplified, but a deeper discussion is beyond the scope of this book. Because of its simplicity and relatively low overhead, DiffServ is better suited to large, heavily trafficked networks than RSVP.

MPLS (Multiprotocol Label Switching)

Another QoS technique that modifies data streams at the Network layer is MPLS (multiprotocol label switching). As described in Chapter 5, to indicate where data should be forwarded, MPLS replaces the IP datagram header with a label at the first router a data stream encounters. The MPLS label contains information about where the router should forward the packet next. Each router in the data stream’s path revises the label to indicate the data’s next hop. In this manner, routers on a network can take into consideration network congestion, QoS indicators assigned to the packets, plus other criteria. MPLS forwarding is also fast. This is because, in MPLS, a router knows precisely where to forward a packet. On a typical packet-switched network, routers compare the destination IP address to their routing tables, and forward data to the node with the closest matching address. With MPLS, data streams are more likely to arrive without delay. On a network supplying clients with voice and video services, fast transmission is desirable. A network’s connectivity devices and clients must support the same set of protocols to achieve their QoS benefits. However, networks can—and often do—combine multiple QoS techniques.

Chapter Summary

■ The use of a network (either public or private) to carry voice signals using the TCP/IP protocol is commonly known as VoIP (Voice over IP). VoIP services can operate over any type of transmission medium and access method that support TCP/IP.

■ When VoIP relies on the Internet, it is often called Internet telephony. But not all

VoIP calls are carried over the Internet. In fact, VoIP over private lines is an effective and economical method of completing calls between two locations within an organization.

■ An organization might use VoIP to save money on telephone calls, centralize management of voice and data services, or take advantage of customizable call features.

■ Many types of clients and network designs are available with VoIP networks. Clients can be traditional analog telephones, IP telephones, or softphones (a computer running telephony software and connected to a microphone and headphones). In each case, analog voice signals are first converted to digital signals by a voice codec (coder/ decoder).

■ Analog VoIP clients may connect to IP networks in one of four ways: using an internal or external ATA (analog telephone adapter); connecting directly to a router or to a voice-data gateway that digitizes call information; connecting directly to an IP-PBX capable of handling both analog and digital voice connections; or connecting to an analog PBX, which then connects to a voice-data gateway.

■ Digital VoIP clients typically connect to a digital PBX or other connectivity device with VoIP capabilities.

■ A special type of digital PBX is one that is hosted by a service provider. Organizations that choose a hosted PBX don’t need to install or maintain hardware or software for call completion and management. Hosted PBXs are also known as virtual PBXs.

■ Rather than analog telephones, many VoIP installations use IP telephones, which transmit and receive only digital signals. To communicate on the network, each IP telephone must have a unique IP address. The IP telephone connects to an RJ-45 wall jack, like a computer workstation.

■ One significant benefit to using IP telephones is their mobility. Because IP telephones are addressable over a network, they can be moved from one office to another office, be connected to a wall jack, and be ready to accept or make calls.

■ Rather than using traditional telephones or IP telephones, a third option is to use a computer programmed to act like an IP telephone, otherwise known as a softphone. Softphones and IP telephones provide the same calling functions; they simply connect to the network and deliver services differently. A softphone’s versatile connectivity makes it an optimal VoIP solution for traveling employees and telecommuters.

■ Streaming video refers to video signals that are compressed and delivered in a continuous stream. It may be made available as files stored on a streaming server for video-on-demand or as real-time feeds in live streaming. Streaming video may be delivered in a point-to-point or point-to-multipoint fashion, though both types typically use unicast transmission.

■ In IPTV (IP television), television signals from broadcast or cable networks travel over packet-switched connections. Telecommunications carriers and cable companies supply IPTV over their existing networks. They accept video content from satellite feeds, national or regional broadcasters, and local sources at a head-end. IPTV channels are delivered in multicast fashion to consumers, where they are decoded and issued to televisions by set-top boxes.

■ Videoconferencing allows multiple participants to communicate and collaborate at once through audiovisual means. To view a videoconference, each participant must have at least a video terminal or video phone that can decode compressed video and interpret signaling protocols. To send and receive audiovisual signals, participants must also have a device equipped with a camera, microphone, and video-encoding capabilities.

■ Videoconferences that are point to multipoint or multipoint to multipoint rely on video bridges to manage communication among participants. Video bridges may be hardware devices dedicated to this task or software running on conference servers.

■ In VoIP and video-over-IP transmission, signaling is the exchange of information

between the components of a network or system for the purposes of establishing,

monitoring, or releasing connections as well as controlling system operations.

■ Voice and video-over-IP services depend on signaling protocols to request a call or

videoconference setup, locate clients on the network and determine the best routes

for calls to follow, and acknowledge a request for a call or videoconference setup.

Voice and video-over-IP services also depend on signaling protocols to set up the

connection; manage ringing, dial tone, and other telephony features; detect and

reestablish dropped calls or video transmissions; and properly terminate a call or

videoconference.

■ H.323 is an ITU standard that describes an architecture and a group of protocols for

establishing and managing multimedia sessions on a packet-switched network.

■ H.225 is the protocol specified by the H.323 standard that handles call or

videoconference signaling. For instance, when an IP telephone user wants to make a

call, the IP telephone requests a call setup (from the H.323 gateway) via H.225.

■ Another H.323 Session layer protocol, H.245, ensures that the type of information—

whether voice or video—issued to an H.323 terminal is formatted in a way that the

H.323 terminal can interpret.

■ SIP (Session Initiation Protocol) is an Application layer signaling and control protocol

for multiservice, packet-based networks. SIP’s developers modeled it on the HTTP

protocol and aimed to reuse as many existing TCP/IP protocols as possible for

managing sessions and providing enhanced services.

■ SIP does not attempt to perform and control as many functions as the H.323

protocols. Its capabilities are limited to determining the location of an endpoint;

determining the availability of an endpoint; establishing a session between two

endpoints; managing calls by adding (inviting), dropping, or transferring participants;

negotiating features of a call or videoconference when it’s established; and changing

features of a call or videoconference while it’s connected.

■ Some VoIP vendors prefer SIP because of its simplicity, which makes SIP easier to

maintain than H.323. And because it requires fewer instructions to control a call,

SIP consumes fewer processing resources than H.323.

■ Media gateways rely on an intermediate device known as an MGC (media gateway

controller) to exchange and translate signaling and control information with each

other. An MGC facilitates the exchange of call signaling information between these

gateways and manages and disseminates information about the paths that voice or

video signals take between gateways.

■ MGCs communicate with media gateways according to one of several protocols.

The older protocol is MGCP (Media Gateway Control Protocol). MEGACO performs

the same functions as MGCP, but uses different commands and processes. Many

network engineers consider MEGACO superior to MGCP because it supports a

broader range of network technologies, including ATM. The ITU has codified the

MEGACO protocol in its H.248 standard.

■ RTP (Real-time Transport Protocol) operates at the Application layer of the OSI model

and relies on UDP at the Transport layer. It applies sequence numbers to indicate the

order in which packets should be assembled at their destination and assigns each packet

a time stamp that corresponds to when the data in the packet were sampled from the

voice or video stream. This time stamp helps the receiving node to compensate for

network delay and to synchronize the signals it receives.

■ RTCP (Real-time Transport Control Protocol) provides feedback on the quality of a

call or videoconference, such as the extent of delay or packet loss in a transmission.

■ Network engineers have developed several techniques to overcome the QoS challenges

inherent in delivering voice and video over IP. One, RSVP (Resource Reservation

Protocol), is a Transport layer protocol that attempts to reserve a specific amount of

network resources for a transmission before the transmission occurs.

■ DiffServ (Differentiated Service) is a simple technique that addresses QoS issues by

prioritizing traffic. DiffServ places information in the DiffServ field in an IPv4

datagram. In IPv6 datagrams, DiffServ uses a similar field known as the Traffic Class

field. This information indicates to the network routers how the data stream should

be forwarded.

■ Another QoS technique that modifies data streams at the Network layer is MPLS

(multiprotocol label switching). To indicate where data should be forwarded, MPLS

replaces the IP datagram header with a label at the first router a data stream

encounters. The MPLS label contains information about where the router should

forward the packet next. Each router in the data stream’s path revises the label to

indicate the data’s next hop. In this manner, routers on a network can take into

consideration network congestion, QoS indicators assigned to the packets, plus other

criteria.

Review Questions

1.   You have decided to establish a VoIP system in your home. Which of the following devices is necessary to connect your analog telephone to your VoIP server?

a.   Codec

b.   IP-PBX

c.   Softphone

d.  ATA

2.   Skype, the popular Internet telephony service, provides a user with what type of interface?

a.   IP phone

b.   Analog telephone

c.   Softphone

d.   IP-PBX

3.   A company’s use of VoIP on its WAN to avoid long distance telephone charges is known as:

a.   Toll bypass

b.   WAN redirect

c.   Fee gauging

d.   Circuit redirect

4.   Which of the following is the most popular signaling protocol used on traditional, circuit-switched PSTN connections?

a.   SIP

b.  SS7

c.   H.323

d.   MEGACO

5.   Watching a YouTube video on the Web is an example of which of the following types of video-over-IP services?

a.   Videoconferencing

b.  Streaming video

c.   IP multicasting

d.   IPTV

6.   In an IPTV system, which of the following functions does a set top box perform?

a.   Decodes video signals and issues them to a television

b.   Determines the appropriate amount of bandwidth necessary to deliver a requested video and adjusts the connection accordingly

c.   Interprets multicast routing protocols to determine the most efficient means of distributing video signals

d.   Generates video content based on a subscriber’s channel selection

7.   What type of video-over-IP service relies on full-duplex communication?

a.   Webcasting

b.   Streaming video

c.   Videoconferencing

d.   IPTV

8.   What protocol manages addressing for multicast groups?

a.   IGMP

b.  MGCP

c.   MEGACO

d.  H.245

9.   Which of the following protocols would be used by a video bridge to invite a video phone to join a videoconference?

a.   MGCP

b.  H.225

c.   IGMP

d.  RSVP

10. Suppose your organization’s PSTN and VoIP systems are integrated, and that your VoIP system adheres to architecture specified in H.323. Which of the following performs translation between the PSTN’s signaling protocols and H.323 on your network?

a.   H.323 terminal

b.  H.323 gatekeeper

c.   H.323 gateway

d.  H.323 zone

11. You are using Skype to initiate a video call with a friend in another state. Which of the following protocols is generating segments at the Transport layer of this transmission?

a.   ICMP

b.   TCP

c.   FTP

d.  UDP

12. What function does the H.225 protocol provide, as part of the H.323 VoIP

specification?

a. Handles call setup, call routing, and call termination

b.  Controls communication between media gateways and media gateway controllers

c.   Ensures that signals issued to an H.323 terminal are in a format that the terminal can interpret

d. Indicates priority of each IP datagram

13. In SIP, which of the following network elements maintains a database with network address information for every SIP client?

a.   Redirect server

b.  Registrar server

c.   Domain server

d.  Proxy server

14. Which of the following are reasons for choosing SIP over H.323? (Choose two.)

a.   SIP is an older, more reliable standard.

b.   SIP has limited functionality, which makes it more flexible.

c.   SIP messages use fewer processing resources.

d.   SIP includes QoS mechanisms that make it more dependable.

e.   SIP supports a wider range of voice and video codecs.

15. Which of the following devices enable multiple media gateways to communicate?

a.   VoIP router

b.  IP-PBX

c.   MGC

d.  IP phone

16. At what layer of the OSI model does RTP operate?

a.   Transport

b.  Presentation

c.   Session

d.  Application

17. What can RTCP do that RTP cannot?

a.   Issue timestamps for every transmission

b.  Assign sequence numbers to each packet in a transmission

c.   Report on the degree of packet loss and delay in a connection

d.  Modify each IP datagram to assign a priority level

18. How does RSVP help improve QoS?

a.   It assigns a label to each IP datagram that will be read and modified by every router in the data’s path.

b.  It continually assesses the status of likely routes in the transmission’s path and dynamically modifies IP datagrams as they’re issued with instructions for following the best path.

c.   It modifies the Priority field in each IP datagram so that high-bandwidth applications are given precedence over low-bandwidth applications.

d.  It establishes a path between the sender and receiver that is guaranteed to supply sufficient bandwidth for the transmission.

19. The Traffic Class field in an IPv6 datagram serves the same function as which of the following fields in an IPv4 datagram?

a.   TTL

b.  DiffServ

c.   RSVP

d.   Padding

20. On a VoIP network that uses the DiffServ QoS technique, which of the following makes certain that a router forwards packets within a given time period?

a.   Assured Forwarding

b.   Superior Forwarding

c.   Expedited Forwarding

d.   Best-effort Forwarding

Practice Test

1. VoIP can run over any packet-switched network.

a.       True

b.      False

 

2. NASA is developing telemedicine capabilities for diagnosing patients in space.

a.       True

b.      False

3. A(n) ____ emulate and interpret conventional fax signaling protocols when communicating with a conventional fax machine.

fax gateway

 

4. In the case of long-distance calling, using VoIP over a WAN allows an organization to avoid paying long-distance telephone charges, a benefit known as ____.

toll bypass

 

5.  ____ is the protocol specified by the H.323 standard that handles call or videoconference signaling.

a.       H.225

b.      H.245

c.       H.248

d.      H.252

6. Because of its simplicity and relatively low overhead, RSVP is better suited to large, heavily trafficked networks than DiffServ.

a.       True

b.      False

7.  A(n) ____ is a private switch that accepts and interprets both analog and digital voice signals.

a.       H.323 terminal

b.      H.323 gateway

c.       IP-PBX

d.      registrar server

8. To prioritize traffic, DiffServ places information in the DiffServ field in a(n) ____.

a.       toll bypass

b.      IPv4 datagram

c.       set top box

d.      redirect server

9. In ____ Forwarding, a data stream is assigned a minimum departure rate from a given node.

expedited

 

10. When VoIP relies on the Internet, it is often called ____.

Internet telephony

 

11. FoIP (Fax over IP) uses ____ networks to transmit faxes from one node on the network to another.

a.       bridged

b.      multipoint

c.       circuit switched

d.      packet-switched

12. Because IP telephones are addressable over a network, they can be moved from one office to another office, connected to a wall jack, and be ready to accept or make calls.

a.       True

b.      False

13. SIP does not attempt to perform and control as many functions as the H.323 protocols.

a.       True

b.      False

14. IPTV, videoconferencing, streaming video, and IP multicasting belong to the range of services known as ____.

a.       Fax over IP

b.      voice over DSL

c.       video over IP

d.      Webcasts

15.    ____ is the exchange of information between one or more participants for the purposes of training and education.

Tele-education

16. Mobility is a benefit of IP telephones.

a.       True

b.      False

17.   ____ is a simple technique that addresses QoS issues by prioritizing traffic.

a.       RTCP (Real-time Transport Control Protocol)

b.      RSVP (Resource Reservation Protocol)

c.       MPLS (multiprotocol label switching)

d.      DiffServ (Differentiated Service)

 

18. In videoconferencing, a video bridge serves as a(n) ____.

a.       gateway

b.      terminal

c.       H.323 zone

d.      MCU

19. Videoconferencing is a VoIP application.

a.       True

b.      False

20. The term telephony refers to video signals that are compressed and delivered in a continuous stream.

a.       True

b.      False

21. Gatekeepers are optional on H.323 networks.

a.       True

b.      False

22. A(n) ____ manages multiple audiovisual sessions so that participants can see and hear each other.

a.       proxy server

b.      set top box

c.       video-bridge

d.      softswitch

Chapter Test

1. Using VoIP over a WAN allows an organization to avoid paying long-distance telephone charges, a benefit known as ____.

a.       charge bypass

b.      easypass

c.       toll bypass

d.      distance bypass

2. SIP and H.323 account for communication between media gateways.

a.       True

b.      False

 

3.  ____ signaling functions are more limited than those performed by the protocols in the H.323 group.

a.       MEGACO

b.      MGC

c.       SIP

d.      RTCP

4. ____ is the use of one network to simultaneously carry voice, video, and data communications.

a.       Convergence

b.      Divergence

c.       Multicasting

d.      Unicasting

5. SIP and H.323 regulate ____ for VoIP or video-over-IP clients and servers.

a.       control only

b.      call signaling only

c.       call signaling and control

d.      communication between media gateways

 

6.  When streaming videos are supplied via the Web, they are often called ____________________.

Webcasts

 

7.  ____ describes the use of any network to carry voice signals using the TCP/IP protocol.

a.       Internet telephony

b.      Voice telephony

c.       Telephony

d.      IP telephony

8. When more than two people participate in a videoconference, for example, in a point-to-multipoint or multipoint-to-multipoint scenario, a video ____ is required.

a.       switch

b.      gateway

c.       bridge

d.      router

 

 9. In general, a(n) ____ is a private switch that accepts and interprets both analog and digital voice signals.

a.       IT-PBX

b.      Data PBX

c.       IP-PBX

d.      analog PBX

10. It is more difficult to transmit VoIP and video over IP signals over a packet-switched network than it is to transmit data signals.

a.       True

b.      False

 

 11. The popular Internet telephony software, Skype, is a type of ____.

            a.  IP telephone

            b.  teleapplication

            c.  compu-phone

            d.  softphone

 

12. When a caller uses an IP telephone, his or her voice is immediately digitized and issued from the telephone to the network in ____ form.

            a.    frame

            b.    segment

            c.    packet

            d.   circuit

13. An off-the-shelf workstation can be programmed to act like an IP telephone.

a.       True

b.      False

14. IPTV, videoconferencing, streaming video, and IP multicasting belong to the range of services known as ____.

            a.    voice over IP

            b.    data over IP

            c.    video over IP

            d.    Web over IP

15. Many streaming video services - and nearly all of those issued over a public network, such as the Internet - are examples of ____ transmissions.

            a.     broadcast

            b.     telecast

            c.     unicast

            d.     multicast

16. A computer programmed to act like an IP telephone is known as a(n) ____.

            a.    video phone

            b.    softphone

            c.    compu-phone

            d.    streaming server

17.    ____________________ is a simple technique that addresses QoS issues by prioritizing traffic.

DiffServ

 

 18.   ____________________ performs the same functions as MGCP, but using different commands and processes.

MEGACO

 

 19. Many network engineers consider ____ to be superior to MGCP.

            a.     MGC

            b.     RTCP

            c.     SIP

            d.     MEGACO

20.  When using an analog telephone, a VoIP adapter that performs analog-to-digital conversion is known as a(n) ____.

            a.     DTA (digital telephone adapter)

            b.     ATA (analog telephone adapter)

            c.     DTA (data telephone adapter)

            d.     VTA (voice telephone adapter)

21.  ____ is a measure of how well a network service matches its expected performance.

            a.     RSVP

            b.     DiffServ

            c.     QoS

            d.     MPLS

22.  ____________________ is a QoS technique that replaces the IP datagram header with a label at the first router a data stream encounters.

MPLS

 

 23. One drawback to ____ video is that content may not be edited before it’s distributed.

            a.     streaming server

            b.     live streaming

            c.     on demand

            d.     VoIP

24.  IP telephones are directly connected to the local loop.

a.       True

b.      False

25. When VoIP relies on the Internet, it is often called ____.

            a.     telephony

            b.     IP telephony

            c.     voice telephony

            d.     Internet telephony