THE internet ENCYCLOPEDIA 1 volume 3 phần 8 pptx

P1: IML Morris WL040/Bidgoli-Vol III-Ch-53 September 15, 2003 12:23 Char Count= 0 VOICE OVER INTERNET PROTOCOL (IP)652 This occurs because the electronics of a typical telephone network cause the far end of almost every call connection to echo back a weakened version of the transmitted signal to the originating end. Although this echo is always present, it is not noticeable until the round trip delay be- comes long enough. All of these deficiencies of a transmission channel have the effect of reducing the channel capacity, or, throughput, of a transmission link. Channel capacity refers to the theoretical upper limit on how much information, in terms of bits per second, can be transmitted through a channel (Wozencraft & Jacobs, 1965). Circuit-Switched Connections of a Call in a Conventional Telephony Network VOIP networks must interoperate with conventional circuit-switched networks, in particular the ubiquitous public switched telephone network or PSTN. A VOIP network that cannot interoperate with, and therefore exchange calls with, the PSTN would be of little value to most telephone users. A network’s value increases as it can be used to reach more users. Like other networks, the PSTN uses a hub and spoke architecture of transmission links and switches (Bell Lab- oratories, 1977, 1983). Each user’s telephone is typically connected by an individual circuit (usually referred to as local loop) to a central office switching hub. In a small town, there may be only one central office, whereas in a large major metropolitan area, there could be several dozen central offices. In each central office hub, there is usually at least one switch. The switches used in traditional telephony are called circuit switches. A circuit switch will route a call over a dedicated path from one transmission link to another for at least the duration of the call. See Figure 4. As will be described more fully below, this contrasts with the packet switching used in the Internet, which does not maintain a dedicated path connecting network links, but instead passes information along shared paths on a packet-by-packet demand basis. Circuit Switching Node Link Link Dedicated Path Created for Call Duration Figure 4: Circuit switched connection between two links. Originating the Call As Figure 5 shows, when a caller picks up a telephone to make a call, the local switch in the caller’s local central office (Node A) provides a dial tone that the caller hears through the telephone. That dial tone indicates to the caller that the local telephone switch has seized his or her local telephone line and is ready to receive the digits of the telephone number that he or she wishes to dial. The processes of seizing the telephone line and returning dial tone are types of supervisory signaling. The caller then enters, or dials, the telephone number of the called telephone. If the calling telephone is enabled with touch-tone signaling, the calling telephone signals to the local switch the digits dialed by using different combination of two touch-tones for each digit dialed. Such transmitting of routing information is called address signaling. The caller’s local switch Node A detects these tone combinations and, in turn, determines the telephone number the caller is trying to call. Circuit Switching to the Call’s Destination Each circuit switch has a routing look-up table for determining how to handle a call based on some or all of the calling and called numbers. If the local switch does not have adequate information in its routing tables to make a routing decision, most modern switches will hold the call Link Link Local Loop Local Loop Interoffice Transmission Link Interoffice Transmission Link Interoffice Transmission Link Calling Telephone Called Telephone Office" "Central Node A Node B "Toll Office" SS7 Packet Switch SMS Database Node C "Central Office" Figure 5: Conventional circuit switched telephony network. P1: IML Morris WL040/Bidgoli-Vol III-Ch-53 September 15, 2003 12:23 Char Count= 0 VOICE OVER INTERNET PROTOCOL 653 and launch a packet data message to an external database (called a service management system or SMS). That message is routed from the switch to the SMS over a par- allel data network called a Signaling System 7 (or SS7) packet network. The SMS returns routing instructions, via the SS7 packet network, to the switch that originated the routing query. The SS7 signaling standard came out of the conventional telephony industry. If the local switch determines that the call needs to be connected to a local loop attached directly to that local switch, the switch will ring the local loop of the called telephone. At the same time, the switch returns a ringing sound signal to the calling party. On the other hand, if the local switch determines that the called party’s telephone is busy, the local switch returns a slow busy signal to the calling party. As soon as the called party picks up its phone, the local switch creates a circuit-switched dedicated audio communications path between the calling party’s local loop and the called party’s local loop, thus completing a dedicated audio transmission path between the calling telephone and the called telephone. For the call’s duration, that dedicated audio path is available for the transmission of voice signals, in both directions, between the calling and called party’s telephones. If one or more of the links used to co- nstruct that path are digital, a CODEC in the connecting switch or other connecting device will make the neces- sary analog-to-digital conversion so that the voice signal may seamlessly traverse the boundaries between the analog and digital links. If the called telephone’s local loop is not directly connected to the calling telephone’s local switch, the local switch must establish a connection to the called telephone through another switch or switches. For most modern PSTN switches, this requires that the originating switch launch an SS7 message, over the SS7 network, to reserve a path through other switches in the network for completing the desired communications path to the call’s destination. If the far end switch where the called party is located determines that the called party’s line is busy, no circuit connections will be made, the path reservation for the connecting links is dropped, and the originating switch will return a slow busy signal to the calling party. However, if the reason the call cannot be completed is network congestion, the originating switch will return a fast busy to the calling party. If the called party’s line is not busy and a complete path to the called party’s line can be established, the called party’s local switch rings the called party’s telephone. While the called party’s phone is ringing, the calling party’s local switch is returning an audible ringing signal to the calling party. When and if the called party answers, the switches instantly circuit-switch together the links along the reserved path in order to complete a dedicated path between the calling and called telephones. Finally, it should be noted that some switches commu- nicate interoffice signaling information using non-SS7 interoffice signaling arrangements, such as in-band signaling. Regardless of the supervisory and address signaling arrangement used, the net result of establishing a communications path between the calling and called telephone will be the same. VOICE OVER INTERNET PROTOCOL VOIP literally refers to the transmission of a digitized voice signal using digital packets, routed using the In- ternet protocol or IP. The driving forces for using VOIP are beliefs in its cost savings, flexibility, and the growing desire to combine voice and data transmission on one network. See, e.g., Morris (1998), Cisco VOIP Primer (2002), Matthew (2002). How VOIP Transmission Works Because a VOIP call is transmitted digitally, it begins with a digitization process similar to that used in conventional telephony. First, the voice signal is sampled at a rate greater than the Nyquist sampling rate, and those samples are digitized. Whereas conventional circuit switched telephony transmits the digitized samples in a constant stream of synchronized (i.e., equally spaced in time) digital samples, VOIP transmits the digitized voice communications samples in asynchronous (i.e., unequally spaced in time), sequentially numbered packets of data. Each packet (which may contain many voice samples) contains its own IP formatted address information, which allows it to be routed over an IP network. Unlike conventional telephony, where each sample follows the path of the sample before it, each of the IP packets containing several samples of voice data may take an independent path (which is shared with other data packets) to its destination. With each packet potentially taking a different route, the packets often can arrive at their destination out of order (or sometimes not at all). At the far end, the IP and other routing information is stripped from the packet, the voice samples are temporarily collected in a buffer and reordered as required, and then, if all goes well, the original voice signal is reconstructed, albeit, slightly delayed. For VOIP technology to have the functionality and flexibility of conventional telecommunications, some form of call control, in the form of supervisory and address signaling, is required. Figure 6 shows the basic VOIP transmission scheme. VOIP Signaling VOIP signaling refers to signaling that can be used over an IP network to establish a VOIP call. It provides the needed functionality of supervisory and address signaling. VOIP call signaling comes in two predominant competing schemes: H.323 and Session Initiation Protocol (SIP). Each must satisfy the generic needs of telephonic calling. Specifically, each must provide for the following: The calling phone to address and signal to the called telephone, The called telephone signaling its availability for receiving calls, Establishing a transmission path between the calling and called telephone, The called or calling telephone signaling to the other phone it has hung up, and The tearing down the transmission path once the call is over. P1: IML Morris WL040/Bidgoli-Vol III-Ch-53 September 15, 2003 12:23 Char Count= 0 VOICE OVER INTERNET PROTOCOL (IP)654 Figure 6: VOIP network. Figure 7 shows a generic signaling progression for establishing a VOIP call that might occur using H.323-type centralized call control arrangement. First, the calling telephone dials the called telephone’s number (1). The calling phone forwards the dialed telephone number (address signaling) to a VOIP call controller—which is a special purpose server. That call controller does a lookup (2) in a database for determining the IP address to reach the called telephone. If the called telephone is on the IP network, a call setup signal is routed to the called telephone to ring the called telephone (4). While the called telephone is ringing, a ringing signal is sent back to the calling telephone (3), which telephone, in turn, generates a ringing sound in the caller’s earpiece. When and if the called party answers the telephone, a se- ries of signals to set up the channel path are returned to the calling and called telephones (5, 6, 7). A UDP/IP communications path is then set up between the calling and called telephone (8). The digitized voice is transported using the user datagram protocol (or UDP) in the transport layer, with two protocols, RTP (real time protocol) and RTCP (real time control protocol), rather than TCP—which is often used by non-voice data. UDP is connectionless (e.g., packets can take different routes) and can transport data packets without acknowledging their receipt. UDP is nonstop with less address information overhead. The tradeoff of using a UDP path is lower reliability than TCP. UDP packets may be dropped or arrive out of order, but if they do arrive they do with less delay. This is a good tradeoff for voice communications, which is highly tolerant to dropped packets, but relatively intolerant to delay. RTP over UDP provides packet sequence numbering, so out-of-order and/or missing packets are detectable at Calling Telephone Called Telephone IP Network IP Router Call Control Gateway 1. Dial 2. Look Up Dialed # 3. Generate Audible Ring 4. Ring 5. Off- Hook 7. Open Channel 6. Open Channel 8. RTP (Voice Stream) IP Packet s Voice Mail Figure 7: VOIP H.323-type signaling system using central call control. P1: IML Morris WL040/Bidgoli-Vol III-Ch-53 September 15, 2003 12:23 Char Count= 0 VOICE OVER INTERNET PROTOCOL 655 Calling Telephone Called Telephone IP Network IP Router Call Control Gateway 1. Dial 2. Look Up Dialed # 3. Generate 4. Ring 5. Off- Hook 7. Open Channel 6. Open Channel 8. RTP (Voice Stream) IP Packets Voice Mail 5. Called Phone Does Not Answer Audible Ring Figure 8: Called telephone unavailable with reroute to voice mail with H.323-type signaling. the far end. RTCP provides a separate signaling channel between the end devices (e.g., telephones) to allow exchange of information about packet loss, packet jitter, and packet delays, as well as additional information, such as the source’s name, e-mail, phone, and identification. (Real-Time Transport Protocol [RTP], 2001, August), and (Streaming Video Over the Internet, 2002). Additional requirements for a commercial VOIP system include being able to produce information for billing and/or accounting for calls. Similarly, today’s users demand that it provide other features, such as Caller ID and voicemail. These capabilities may reside in the call controller, IP telephones, and/or other devices in the IP network. Figure 8 shows how voicemail can be provided with H.232. H.323 vs. SIP In addition to H.323, SIP is the major competing standard for VOIP signaling. They have both evolved to offer very similar feature capabilities. H.323 takes a more telecommunications-oriented approach than SIP. SIP takes an Internet-oriented approach. H.323 is the older of the two (Doron, 2001; Paketizer, 2002). It was developed under the International Telecom- munications Union (ITU), a telecommunications standards group, and has gone through various revisions (ITU-T, Recommendation H.323, 2000). The latest version of H.323 standard (H.323 v3) is very robust in that it cov- ers many possible implementations. However, H.323 is considered more difficult to implement than SIP due to its use of binary encoded signaling commands. H.323 v.3 can be implemented with or without a call control server. Thus, an H.323 v.3 end device (e.g., a telephone) can be designed to either set up a call through a call control server, or set it up directly with another end device without using an intervening call control server. An H.323 v.3 call control server can be set up to relay the communications stream during the call, or the end devices can directly establish the communications RTP streaming channel between themselves. The call control server can be either stateless (i.e., not track a transaction or a session state) or stateful (i.e., track a transaction and/or call session state). The significance of this is that in the stateful configuration, H.323 v.3 is not as scalable. Finally, H.323 v.3 employs signaling protocols that can easily be mapped through a gateway for routing calls between the VOIP network and the public switched network. SIP is a Web-based architecture that was developed under the Internet Engineering Task Force (IETF). Like URLs and Web e-mail, SIP’s messages are in ASCII text format that follow the HTTP programming model, i.e., using a grammar similar to that used to create basic Web pages—resulting in slightly lower efficiency in transmitting signaling information, as compared to the more efficiently encoded binary H.323 signaling messages. Also, SIP is very extensible, leading many vendors to implement variations that may be somewhat incompatible (IETF, SIP, 2003; IETF, SIP RFC2543bis, 2002). An address used for routing a SIP messages is of the form SIPAddress@xyz.com. As shown in Figure 9, when a phone wishes to originate a call, it transmits an ASCII SIP “INVITE” message (1) addressed to the SIP address of the called phone (e.g., sales@xyz.com., where “xyz.com” is the domain name of the SIP proxy server for the called phone). Using conventional domain name server (DNS) lookup, the internet routes this e-mail-type message to the SIP call control server, which may act as either a proxy server or a redirect server for the domain of the dialed called telephone address (here, xyz.com). In order to determine where the telephone is located, the proxy server or redirect server will query a location server (2), which will return the routing directions. What happens next depends upon whether the call controller is acting as a proxy server or a redirect server for the called telephone. P1: IML Morris WL040/Bidgoli-Vol III-Ch-53 September 15, 2003 12:23 Char Count= 0 VOICE OVER INTERNET PROTOCOL (IP)656 xyz.com Calling Telephone Called Telephone IP Network IP Router or Switch Proxy & Location Servers Gateway 2. Look Up abc@xyz.com 5. RING to Caller@SIP.COM 6. OK to Caller@SIP.COM 8. ACK to IP of xyz 7. OK 9. RTP (Voice Stream) IP Packets Voice Mail xyz@195.37.78.173 Caller@SIP.COM 4. RING to Caller@SIP.COM 1. INVITE to abc@xyz.com 3. INVITE to xyz@195.37.78.173 xyz.com Figure 9: VOIP call routed using SIP signaling. For a proxy server, the location server will return the IP address for the called telephone. Using that IP address, the SIP INVITE message will be directed to the called telephone (3) along with the calling telephone’s IP address. The called telephone will then return a “RING” message to the proxy server (4), which then forwards that RING message to the originating telephone (5). When the called phone answers, an OK message is returned (6, 7), which includes the called telephone’s IP address. Finally, the originating telephone (which now knows the IP address of the destination phone from the INVITE/OK message exchange) sends an “ACK” message to the IP address of the called telephone (8). The originating telephone then establishes an RTP communications link with the called telephone (9). For a redirect server, the location server will return the redirected (i.e., forwarding) e-mail-style SIP address of the called telephone (e.g., sales@home.com, or joe@home.com). The redirect server will then forward the INVITE request to the proxy or redirect server associated with that redirected address, and then the steps enumer- ated above will take place. Each location server must track the IP address of each of the telephones in its SIP domain. Thus, each SIP telephone must register with its domain’s location server via a registration server of its telephone service provider. An individual telephone can be registered with any registration server with which the user has a service arrangement. Registration binds each SIP telephone’s IP address to its SIP address in its service provider’s domain. Note that a SIP call control server’s primary purpose is to handle the routing of initial supervisory and address signaling information. Also note that, after the initial exchange of supervisory and address information, SIP end devices establish and maintain the communications channel without involvement of the SIP call control server. Like H.323 v.3, the SIP packets that carry the signaling messages almost always follow a different path from the path taken by the communications data. Finally, with SIP most of the intelligence resides in the end devices, as compared to being in the network, as is the case with conventional telephone networks and with H.323. Because SIP proxy and redirect servers typically do not track a call’s status after the call is set up, SIP is often viewed as being more scalable than H.323. When a call control server is configured to track call status, its resources must bear the added burden of such monitoring. SIP’s ASCII encoding is considered more extensible and open than the binary encoded signaling of H.323 v.3. SIP uses a very generic syntax for messages, which can be customized to fit the needs of different applications resident on end devices. For analysis of the various (and somewhat controversial) comparisons of SIP and H.323, see Dalgica and Fang (1999). Integrating VOIP Into Conventional Circuit-Switched Telephony Networks As previously noted, the value of a telephony network is a direct function of how many telephones are directly or indirectly connected to it; thus a VOIP network must be able to exchange calls (i.e., internetwork) with the PSTN such that a VOIP user can originate telephone calls to and receive telephone calls from a telephone user who is connected to the PSTN. In addition, for conventional telephone providers to deploy VOIP technology inside their networks, VOIP technology’s presence must be impercep- tible to their existing base of telephone users. Figure 10 illustrates how a VOIP call can make a connection from a phone connected to an IP network to P1: IML Morris WL040/Bidgoli-Vol III-Ch-53 September 15, 2003 12:23 Char Count= 0 VOICE OVER INTERNET PROTOCOL 657 Calling Telephone Called Telephone IP Network IP Router or Switch Call Control Gateway 2. Look Up Dialed # 4. Generate 4. Call 4b. Off- Hook 5. Open 6. Open Channel 8. RTP (Voice Stream) IP Packets Voice Mail PSTN 4a. Ring SS7* 7a. Call Setup * SS7 connection if Gateway Is Carrier Provided 7.Call Setup 1. Dial Audible Ring Channel Figure 10: VOIP signaling internetworking with public switched network (PSTN). a phone on a PSTN network. A device, called a gateway, is used to translate signaling messages across the VOIP/PSTN network boundary and to transform the jit- tery voice packets on the IP side of the gateway into a synchronous stream of voice data information (if there is a digital voice circuit on the other PSTN side of the gateway) or an analog voice signal (if there is an analog voice circuit on the PSTN side of the gateway). On the IP side, the VOIP signaling to the gateway looks much the same as the signaling that would be done to another end device on the IP network. ENUM, the Fully Interoperable Numbering Plan ENUM is a new standard for numbering plans that would allow seamless telephone number addressing between conventional and VOIP telephony. See, e.g., Neustar (2003) and IETF, Telephone Number Mapping (ENUM) (2003). It unifies Internet and conventional telephone addressing schemes by mapping E.164 (i.e., conventional telephone) numbers to a URL Internet (and SIP)-friendly format. With ENUM, a single global digital identifier system can serve equally subscribers attached to the PSTN or the Internet. The same identifier can be used to identify multiple devices, such as plain telephones, fax, voicemail, and data services, regardless of whether they are on the PSTN or public Internet, thus conserving scarce E.164 numbers and domain name resources. As an example, a telephone number 1-305-599-1234 would map onto the URL 4.3.2.1.9.9.5.5.0.3.1.E164.arpa. A DNS query on this domain name would return a number of records, each listing a specific service registered to the owner of the E.164 number. Devices and services attached anywhere on the public Internet or PSTN could be registered to this universal, fully portable, single number identifier of their registered owner, allowing mobility around both the PSTN and the public Internet. The ENUM standards raise new issues of privacy, security, and administration. Also, final agreement between Internet and traditional telecom industry represented by the ITU is still pending. Quality of Service Issues Transmission and routing can introduce effects that de- grade the quality of VOIP. VOIP signals are not immune to the deficiencies of the facilities that transport and route them. These deficiencies can cause packet loss, jitter, and delay. Packet loss refers to the loss of packets containing some of the voice samples during transmission. The loss might be caused by high bit error rates in the UDP transmission channel, misrouted packets, and/or congestion causing intermediate routing devices to drop packets. Because voice transmission is very sensitive to delay, the reassem- bly of the voice signal at the receiving end of the call cannot typically wait for the retransmission of erroneous or misrouted packets. However, voice transmission has a high tolerance for packet loss—mainly because the ultimate receiver of the signal (the human ear and/or intervening CODEC) does a good job of interpolating (i.e., filling in the gap) where a packet has been lost. There is a limit to how much packet loss can be tolerated. That limit P1: IML Morris WL040/Bidgoli-Vol III-Ch-53 September 15, 2003 12:23 Char Count= 0 VOICE OVER INTERNET PROTOCOL (IP)658 depends upon several factors, including: (1) the nature of the sound being transmitted, (2) the correlation in time between the packets lost (i.e., are they bunched together or widely dispersed in time), and (3) the randomness of the losses. Because they do such an efficient job of squeezing out redundancy, some CODECs that use data compression algorithms are not able to recover from the loss of as few as two packets in a row. Jitter can also contribute to lower quality. With IP traffic often taking multiple routes and/or mixed in with bursty IP packet traffic, the interarrival times of the packets at the receiving end may be irregular and/or the packets arrive out of order. Jitter can be overcome by buffering the packets—i.e., temporarily storing them long enough to reorder them before forwarding them to the de- coder. However, buffering has the negative consequence of adding more delay. Delay in the transmission of voice packets can be mini- mized in several ways. One is overbuilding the IP network, i.e., ensuring that there is always excess capacity in the IP network for all traffic, including during peak traffic peri- ods. Alternatively, priority routing can be afforded to voice traffic, at the expense of lower priority data traffic—which is more tolerant of delays. Separate treatment of voice from data can be done by virtual segregation or physical segregation from lower priority traffic. The former can be done at either the network or data link layer. For example, segregation can be done at the data link layer by using separate ATM channels for voice and data traffic and then assigning ATM-based priority treatment to the ATM channels carrying voice traffic. Where voice and data are mixed on the IP network, identifiers can be used to indicate to the intermediate network components (such as routers) that designated traffic (such as voice traffic) should be given priority. An example of this last method is RSVP, the Resource Reservation Protocol. RSVP allows a VOIP application to request end-to-end QoS guarantee from a network (Cisco, VoIP Call Admission Control Using RSVP, 2003). If the guarantee cannot be provided, the call will not be al- lowed to go through. Where the guarantee cannot be se- cured, the traffic might be redirected to an alternate network or blocked (resulting in VOIP users receiving an “equipment busy” signal). At the current time, priority schemes such RSVP typically do not work over the public Internet (with a large “I”). This is because, among other reasons, the economic incentives are not there for intermediate Internet providers to honor any type of prioritization routing scheme, given that their reimbursement is the same for all traffic—regardless of its priority desig- nation. Therefore the current market structure for public Internet backbone routing prevents the realization of a higher quality of service for VOIP traffic over the public Internet. To differentiate themselves, some Internet backbone providers are introducing prioritization schemes such as MPLS-based networks, which are ATM-like in their at- tributes, but operate at a mixture of layer 2 and 3 protocols in pure IP environments. As competition intensifies, public networks are expected to become friendlier to real- time services such as VOIP. Quality of service is discussed in detail elsewhere. The Costs and Savings of Using VOIP One source of VOIP’s cost savings over conventional telephony is its ability to employ transmission more efficiently due to both the extensive use of compression algorithms and the statistical nature of its information transmission. However, such efficiencies will tend to be most signifi- cant to private networks and/or network providers whose transmission networks are capacity-constrained. A second source of VOIP’s cost savings is lower capital cost per call using lower cost switching devices (i.e., In- ternet routers and switches). Again, networks with sunk investments in conventional technology with excess capacity would derive little benefit from such savings. A third source of VOIP’s savings is lower costs of administration, particularly in enterprise environments. A good deal of administrative cost is incurred in enterprises to accommodate the movement of telephone users within the enterprise. Each time a user moves to another office or enters or leaves the firm, the routing tables and direc- tory of a conventional telephone system must be updated, often manually. VOIP’s self-registration feature eliminates these administrative costs. A fourth source of VOIP’s savings comes from the “economies of scope” that VOIP can achieve by its ability to intermingle with other traffic on data networks, eliminating the need to segregate voice and data traffic, as is often done with conventional telephony. These savings are most easily exploited in the LAN and WAN enterprise environment or by data-centric carriers who wish to combine their voice and data traffic. Administrative savings also come from eliminating the conventional regime of separate administrative staffs for voice and data. Finally, VOIP traffic on a data network looks like any other data on that network. This allows some carriers and enterprise users to avoid some of the economically dis- torting taxes that local, federal, and foreign regulatory regimes place on pure voice traffic, but not on data traffic. It is important to remember that the realization of these savings are application-specific and may not be realized in every situation. See, e.g., Morris (1998) and “Cisco Seeks Bigger Role in Phone Networks” (March 15, 2003). Security Issues for VOIP Security is a concern with VOIP, particularly because of the distributed nature of the call control, much of which is handled between end devices (Cisco, SAFE: IP Telephony Security in Depth, 2002). Some of the things that make VOIP attractive, e.g., self-registration of end devices and software-controlled PCs acting as end devices, are also the sources of VOIP’s vulnerability. Security issues include four categories: (1) eavesdropping, (2) toll fraud, (3) identity spoofing, and (4) IP spoofing. Eavesdropping refers to an unauthorized party “lis- tening” to the packets and, in turn, being able to listen to the voice conversation. This problem exists with both VOIP and conventional analog/digital voice communications. In both cases, the simplest method of prevent- ing this problem is to encrypt the digital signal at its source, the originating telephone. The problem encryption brings is overhead and computational load, which can introduce delay. Also, encryption can create problems P1: IML Morris WL040/Bidgoli-Vol III-Ch-53 September 15, 2003 12:23 Char Count= 0 REFERENCES 659 for law enforcement agencies where they have a warrant to wiretap a telephone conversation. Toll fraud is the problem of unauthorized persons making telephone calls over the network. This can be caused by unauthorized users originating calls on the IP network either on legitimate phones or over illegitimate phones. As with conventional telephony, this can be combated with security codes and other authentication methods. Identity spoofing refers to a hacker tricking a remote user into believing he or she are talking to the person he or she dialed on the IP network, when, in fact, he or she is talking to the hacker. Again, security codes and other authorization methods are helpful here. IP spoofing refers to theft of IP identity, where one end device is able to convince the IP network that its IP address is the same as a legitimate device’s IP address. This allows the device with the fraudulent IP address to inter- cept calls and/or perform toll fraud using that IP address. Spoofing the IP address of a gateway allows eavesdropping on telephone calls. Some generally accepted recommendations for min- imizing many of these security problems are to disable self-registration of VOIP end devices after initial network installation, to segregate voice from data traffic at level 2 or 3, and to use a stateful firewall at the PSTN gateway. As noted above, segregating data from voice services also provides the added benefit of maintaining different quality of service for data and voice. CONCLUSION VOIP holds great promise where the convergence of data and voice can occur. Internetworking and overcoming QOS issues remain some of the biggest challenges. GLOSSARY Analog signal A continuous signal that, at any point in time, can have an infinite number of possible values and that is typically analogous in some characteristic to another signal or physical phenomenon. Asynchronous transfer mode (ATM) A network transfer method, employed at the data link layer (Level 2), for high-speed switched packet routing, which can establish virtual channels with a specified QoS. Channel capacity The theoretical upper rate limit, in bits per second, at which information can be transmitted over a transmission channel. Circuit switch A switch that makes a temporary or permanent dedicated transmission path between two transmission links that are attached to that switch, based on signaling information received prior to establishing the dedicated path. Digital encoding Encoding a signal in the form of a string of 1s and 0s. Digital transmission Transmission of information encoded as 1s and 0s. Internet A global public network based on the “Inter- net protocol,” connecting millions of hosts worldwide, and for which users often pay a flat fee to access, with little or no charge for transmitting each packet of information. (Outside the U.S., Internet access is often measured and charged on a usage basis, e.g., minutes or units of data.) Internet protocol (small “i”) or IP A packet switching protocol used for routing packets over and between and private networks that is “connectionless” (i.e., each packet making up the same message may take a different route to reach the ultimate destination). Node A point of connection between transmission links, which may contain switches, and/or may contain con- verters to interconnect transmission links with differ- ing modalities (e.g., for connecting wire links to wire- less links, non-digitally encoded links to digitally encoded links, or fiber links to copper wire links). Packet router A type of packet switching device that typically routes based on Level 3 network address information (such as an IP address) and typically has the ability to choose optimal routing based on dynamically changing criteria and routing tables. Packet switch A type of packet switching device that routes packets of data between links based on address information associated with each packet. A Level 3 switch uses network addresses, such as IP addresses, to route packets of data. Level 2 switches uses data link layer addresses (which are typically local and/or hard- coded) for routing. Public switched telephone network (PSTN) A circuit- switched network that is provided by regulated com- mon carriers who offer their voice telephone services to the general public. Quality of Service (QoS) A set of performance param- eters or criteria, such a bandwidth, jitter, packet loss, and delay, prespecified for a service. Transmission The movement of information (whether or not digitally encoded) from one point to another via a signal carried over a physical medium, such as wires, fiber, radio, or light. Transducer A device actuated by signal power from one system and supplying signal power in another form to a second system (e.g., a telephone receiver earpiece actuated by electric power of a received transmission signal and supplying acoustic signal power to the sur- rounding air, which the telephone user can hear, or a telephone microphone that has a quartz crystal that produces electrical signal power for transmission over wires from the mechanical acoustic power originating from the telephone user’s voice). Transmission link A transmission path connecting two nodes. Voice communication The transmission of information contained in a voice signal. CROSS REFERENCES See Circuit, Message, and Packet Switching; Digital Com- munication; Internet Literacy; Public Networks; TCP/IP Suite; Web Quality of Service; Wide Area and Metropolitan Area Networks. REFERENCES ASCII Table. Retrieved March 23, 2003, from http://web. cs.mun.ca/∼michael/c/ascii-table.html P1: IML Morris WL040/Bidgoli-Vol III-Ch-53 September 15, 2003 12:23 Char Count= 0 VOICE OVER INTERNET PROTOCOL (IP)660 Bell Laboratories (1977, 1983). Engineering and opera- tions in the Bell System. Murray Hill, NJ: Bell Labo- ratories. Castelli, M. (2002). Network consultants handbook. Indi- anapolis, IN: Cisco Press. Cisco, VoIP call admission control using RSVP (2003). Retrieved March 25, 2003, from http://www.cisco.com/ univercd/cc/td/doc/product/software/ios121/121newft/ 121t/121t5/dt4trsvp.htm Cisco, VOIP primer (2002). Retrieved March 25, 2003, from http://www.cisco.com/univercd/cc/td/doc/product/ access/acs mod/1700/1751/1751 swg/intro.htm Cisco, Understanding codecs: Complexity, hardware support, MOS, and negotiation (2002). Retrieved March 25, 2003, from http://www.cisco.com/warp/ public/788/voip/codec complexity.pdf Cisco, SAFE: IP telephony security in depth (2002). Re- trieved March 25, 2003, from http://www.cisco.com/ warp/public/cc/so/cuso/epso/sqfr/safip wp.htm Dalgic, I., & Fang, H. (1999). Comparison of H.323 and SIP for IP telephony signaling. Retrieved March 25, 2003, from http://216.239.57.100/cobrand univ?q=cache:PNnMM0MUWcsC:www.cs.columbia. edu/∼hgs/papers/others/Dalg9909 Comparison.pdf+ dalgic&hl=en&ie=UTF-8 Doron, E. (2001). SIP and H.323 for voice/video over IP—Complement, don’t compete! Internet Telephony. Retrieved March 25, 2003, from http://www.tmcnet. com/it/0801/0801radv.htm Internet Engineering Task Force (IETF) SIP (2003). Retrived March 25, 2003, from http://www.ietf.org/ html.charters/sip-charter.html Internet Engineering Task Force (IETF) Telephone number mapping (ENUM) (2003). Retrieved March 25, 2003, from http://www.ietf.org/html.charters/enum- charter.html Internet Engineering Task Force (IETF), SIP RFC2543bis (2002). SIP: Session initiation protocol, SIP WG Internet Draft. Retrieved March 25, 2003, from http://www.jdrosen.net/sip bis.html ITU-T, Recommendation H.323 (2000). Retrieved March 25, 2003, from http://www.itu.int/rec/ recommendation.asp?type=items&lang=E&parent=T- REC-H.323-200011-I Morris, R. L. (1998). Voice over IP telephony: Sizzle or steak? Retrieved March 25, 2003, from http://members. aol.com/ ht a/roym11/LoopCo/VOIP.html Neustar (2003). Retrieved March 25, 2003, from http:// www. enum.org Newton, H. (1998). Newton’s telecom dictionary (14th ex- panded ed.). New York: Flatiron Publishing. Cisco seeks bigger role in phone networks (2003, Ma- rch 3). New York Times. Retrieved March 25, 2003, from http://www.nytimes.com/2003/03/03/technology/ 03CISC.html?tntemail1=&pagewanted=print&position =top Packetizer, T. M. (2002). Comparisons between H.323 and SIP. Retrieved March 25, 2003, from http://www. packetizer.com/iptel/h323 vs sip/complist.html Real-Time Transport Protocol (RTP) (2001, August). Re- trieved March 25, 2003, from http://www.cs.columbia. edu/∼hgs/teaching/ais/slides/rtp.pdf Sanford (1999). Packet voice technology: Cheap talk? Retrieved March 25, 2003, from http://www.applied- research.com/articles/99/ARTicle10Sanford.htm Truxal, J. G. (1990). The age of electronic messages. Cambridge, MA: MIT Press. Streaming video over the Internet (2002). Retrieved March 25, 2003, from http://www.streamdemon.co.uk/ tranproto.html Wozencraft, R., & Jacobs, M. (1965). Principles of communications theory. New York: Wiley. P1: JDW Fahy WL040/Bidgolio-Vol I WL040-Sample.cls June 20, 2003 17:43 Char Count= 0 W W Web-Based Training Web-Based Training Patrick J. Fahy, Athabasca University Web-Based Training (WBT): Background 661 Training Principles and Technological Developments Supporting WBT 661 High-Technology and Training 662 Using the Web for Training 663 Strengths 663 Weaknesses 664 WBT’s Challenges 664 New Roles 664 The WBT Environment 665 Individual Differences 665 Economic Factors 666 The Future of WBT 667 Bandwidth and Security 667 Implementing WBT 668 Media and the Future 669 Conclusion 671 Glossary 671 Cross References 672 References 672 WEB-BASED TRAINING (WBT): BACKGROUND As part of corporate health, even survival, companies and training institutions globally have recognized the need to provide relevant and flexible training. Professional development (PD) in the form of upgrading, re-training, and various educational opportunities is seen as enhancing the skills of valued employees, helping organizations maintain their competitive advantage by developing (and thereby retaining) experienced people. Well-designed Web-based training (WBT) can offer valuable advantages over other types of training delivery in a wide variety of public and private environments: training time and travel can be reduced, even eliminated, lowering costs; materials stored on central servers can be continually revised and updated, assuring currency and enhancing quality; training content is more consistent, supporting higher standards; greater efficiency (chiefly the result of individualization) can increase trainee learning and satisfaction, improving motivation; and produc- tion and delivery of training programs may be more sys- tematic, improving the cost-effectiveness of development. At the same time, using the World Wide Web (WWW) for training presents some challenges: existing training materials must usually be redesigned, sometimes exten- sively; bandwidth limitations (often at the user’s end, in the “last five feet” of the communications chain) may re- strict or even prohibit use of multimedia by some trainees; all participants (trainees and instructors) must learn new skills to use WBT effectively; and an initial investment (sometimes substantial) in equipment and expertise may be needed. Other factors in the structure or culture of a training organization may also need to change to make WBT feasible. Dropout rates (admittedly often a problem in WBT) may indicate the health of WBT programs: high rates may mean a mismatch between trainees’ expecta- tions and the instructional design of the training material, or may reveal a lack of leadership or management support (Frankola, 2001). In this chapter, WBT will be discussed from theoretical and practical perspectives: important training principles are reviewed briefly, including basic concepts now com- mon in WBT; practical problems in WBT are considered, as well as the strengths and weaknesses of this mode of training delivery; and finally the prospects for the future of WBT, and some of the pedagogic, technical, and economic assumptions on which the optimistic predictions depend, are considered. Training Principles and Technological Developments Supporting WBT Pioneering Ideas in Training In the first half of the 20th century, pioneering researchers such as Thorndike (1971), Dewey (1938), Skinner (1971), and Keller (1968) conducted research that began identi- fying fundamental learning principles. (While these fig- ures wrote and researched in the fields of psychology and education, their theories have evolved so that they are now used in the design of effective teaching and training of all kinds, including WBT.) Thorndike’s three fundamental behavioral laws were among the first dis- coveries: (1) repetition strengthens any new behavior; (2) pleasure or reward associated with a particular behavior increases the likelihood the behavior will be repeated, while pain or lack of reward may diminish the likelihood; and (3) an individual’s personal readiness is crucial to the performance of any new skill or behavior (Saettler, 1990). Dewey added that individual trainee differences were crucial in the success of training. Dewey and Piaget (1952) both recognized the importance of each individual learner’s personal background, and advocated that trainees’ experiences and previous learning be considered 661 [...]... Webcasting around the Globe Conclusion Glossary Cross References References 682 682 682 682 6 83 6 83 6 83 684 684 684 684 685 685 685 686 686 Reasons for Using Webcasting and Significance of Webcasting to the Internet and E-commerce World Webcasting offers many benefits to individuals and organizations that need to disseminate information and content The interactivity of computers allows the personalization... (19 75), Patrick Suppes (19 78) , Robert Glaser (19 78) , Robert Gagne and Leslie Briggs (19 79), Leslie Briggs and Walter Wager (19 81 ) , and Walter Dick and Lou Carey (19 78) There were certainly others, but these individuals led the way The Internet as a Training Platform The earliest forms of the Internet emerged as training was being transformed by a new understanding of learning itself The fact that the. .. U.S Population % age 12 + % Internet users % broadband users Male/female Under age 35 Time spent on Internet/ day Tuned to a webcast in the past month Tuned to a webcast in the past week 80 millions 35 % 48% 26% 50%/50% 52% 1 hr 42 minutes 40 millions 17 % 24% 30 % 56%/44% 56% 2 hr 16 minutes 20 millions 9% 13 % 34 % 62% / 38 % 60% 2 hr 49 minutes Source: Arbitron/Edison Media Research 8 Study, 2002 sets Multicasting... wired.com/news/business/0 , 13 67 ,35 2 08, 00.html Chan-Olmsted, S., & Ha, L (in press) Internet business models of broadcasters Journal of Broadcasting and Electronic Media Cyberatlas (2002) The world’s online population Retrieved November 20, 2002, from http://cyberatlas internet. com/big picture/geographics/article/0 , 13 23, 5 911 15 115 1,00.html Digital TV (20 01, February) Webcasting 45–50 Gunzerath, D (2000) Radio and the Internet. .. Nielsen/Netratings Retrieved April 14 , 20 03, from http://netratings.com/pr/pr 03 011 5.pdf Nitschke, A (19 99) Station Internet activities report Retrieved May 10 , 2002, from http://www.nab.org/ Research/Reports/TvstationInternetActivity.asp Olsen, S (2002, June 19 ) Apple: We told you QuickTime was #1! CNET News.com Retrieved April 14 , 20 03, from http://zdnet.com.com/ 210 0 -11 05- 937 379.html Real Networks (2002)... courses (Blocher, 19 97; Kirkpatrick & Cuban, 19 98) The Internet may be about to change that pattern: since May 2000, trends have shown that women as a group exceeded the number of men online, so that by June 20 01 women comprised 40.9 of Internet users and men 39 .8% Interestingly, in relation to the question of the feasibility of the Internet for the training of older workers and women, the largest increase... connected, staying unplugged (19 99) Innovation Analysis Bulletin, 1, 4 Statistics Canada– Catalogue No 88 -0 03- XIE Retrieved March 2000 from http://www.statcan.ca :80 /english/freepub /88 -0 03- XIE/ free.htm Glaser, R (Ed.) (19 78) Advances in instructional psychology Hillsdale, NJ: Lawrence Erlbaum Grow, G (19 91) Teaching learners to be self-directed Adult Education Quarterly, 41, 12 5 14 9 Harapniuk, D., Montgomerie,... expected to grow in number By one estimate, in 20 01 viruses were found in 1 of every 30 0 e-mail messages; at the current rate of proliferation, by 20 08 the ratio will be 1 in 10 , and by 2 0 13 it will be 1 in 2 (“Outbreak,” 20 01) “Virus” is the generic name for all malicious programs, including worms and Trojan horses; the term malware has been suggested for all these malicious forms of code (Seltzer, 2002)... Z (19 89 , March) Seven principles for good practice in undergraduate education AAHE Bulletin, 3 7 Conference Board of Canada (20 01) Performance and Potential 20 01 02 Retrieved May 2002 from http:// www.conferenceboard.ca/pandp/documents/pandp. 01 pdf Cross, P (19 81 ) Adults as learners San Francisco: JosseyBass Dewey, J (19 38 ) Experience and education New York: Macmillan Dick, W., & Carey, L (19 78) The. .. (Digital TV, 20 01) TYPES OF WEBCASTING There are three types of webcasting based on the technology that webcasters use to deliver the content or information to the Internet audience: (1) push, (2) on-demand, and (3) live streaming (Miles, 19 98) Webcasting can be streamed live or be downloaded and stored on the server for later retrieval by the users Table 1 compares the differences between the three types . to Caller@SIP.COM 8. ACK to IP of xyz 7. OK 9. RTP (Voice Stream) IP Packets Voice Mail xyz @19 5 .37 . 78 .17 3 Caller@SIP.COM 4. RING to Caller@SIP.COM 1. INVITE to abc@xyz.com 3. INVITE to xyz @19 5 .37 . 78 .17 3 xyz.com Figure. (19 75), Patrick Suppes (19 78) , Robert Glaser (19 78) , Robert Gagne and Leslie Briggs (19 79), Leslie Briggs and Walter Wager (19 81 ) , and Walter Dick and Lou Carey (19 78) . There were certainly others,. http://web. cs.mun.ca/∼michael/c/ascii-table.html P1: IML Morris WL040/Bidgoli-Vol III-Ch- 53 September 15 , 20 03 12 : 23 Char Count= 0 VOICE OVER INTERNET PROTOCOL (IP)660 Bell Laboratories (19 77, 19 83 ) . Engineering and opera- tions in the Bell

Định dạng
Số trang	98
Dung lượng	1,71 MB