D
D
a
a
t
t
a
a
o
o
v
v
e
e
r
r
C
C
a
a
b
b
l
l
e
e
S
S
e
e
r
r
v
v
i
i
c
c
e
e
s
s
Contents
Chapter 1:
Introduction
2Chapter 2:
Technical Description of Data over Cable Services
8Chapter 3:
Broadband Cable Networks
11Chapter 4:
C
C
h
h
a
a
p
p
t
t
e
e
r
r
1
1
I
I
n
n
t
t
r
r
o
o
d
d
u
u
c
c
t
t
i
i
o
o
n
n
t
t
o
o
B
B
r
r
o
o
a
a
d
d
b
b
a
a
n
n
d
d
In the latter part of the twentieth century, the almost simultaneous arrival of two major innovations—mobile phones and the Internet—not only changed the face of communications, but also gave fresh impetus for economic growth.
The convergence of mobile and Internet technologies still seems likely to come to such fruition, though the indications are that it will take longer than expected. But in the meantime, a new technology is emerging that promises to provide a unifying platform for three converging industrial sectors: computing, communications and broadcasting. That technology is “broadband”.
Because of the nature of broadband (you have to use it to understand the benefits it offers), market take-off requires a certain critical mass of users. Currently, around one in every ten Internet subscribers worldwide has a dedicated broadband connection (Figure 1, top chart), though many more share the benefits of high-speed Internet access, for instance, through a local area network (LAN), at work or at school. The world leader for broadband is the Republic of Korea (Figure 1, lower chart), which is around three years ahead of the global average in terms of converting Internet users to broadband. There, a critical mass was attained as early as 2000, when prices fell below US$ 25 per month; from which point onwards take-off was rapid (see Figure 1, bottom chart). Over 93 per cent of Internet subscribers in Korea use broadband.
Around the world, there were around 63 million “broadband” subscribers at the start of 2003 compared with 1.13 billion fixed-line users and 1.16 billion mobile phone users. Broadband users enjoy a range of service speeds from 256 kbit/s up to 100 Mbit/s. The number of subscribers is growing rapidly, with a 72 per cent increase during 2002. Digital subscriber line (DSL) is currently the most commonly deployed platform, followed by cable modems, Ethernet local area networks (LAN), fixed-wireless access, wireless LANs (WLAN), satellite and other technologies.
“The term “broadband” is like a moving target. Internet access speeds are increasing all the time.”
Broadband is commonly used to describe recent Internet connections that are significantly faster than today’s dial-up technologies, but it is not a specific speed or service.
Recommendation I.113 of the ITU Standardization Sector defines broadband as a transmission capacity that is faster than primary rate ISDN, at 1.5 or 2.0 Mbit/s. Elsewhere, broadband is considered to correspond to transmission speeds equal to or greater than 256 kbit/s, and some operators even label basic rate ISDN (at 144 kbit/s) as a “type of broadband”. In this report, while not defining broadband specifically, 256 kbit/s is generally taken as the minimum speed.
Wired connections account for the vast majority (over 98 per cent) of current connections—although wireless technologies are starting to grow quickly. Of the fixedline connections, digital subscriber line (DSL) and cable modem technologies are the most popular (Figure 2, top chart). Until 2000, the majority of broadband users were using cable modems, and this is still the most popular form of access in North America. But worldwide, ADSL now accounts for more than half the connections, being particularly popular in Asia and Western Europe.
Where fixed-line connections are not so readily available or convenient to use, a number of wireless technologies such as WiFi have been gaining in popularity too.
“Broadband is increasingly seen as a catalyst for economic success. Supplying broadband is therefore an issue for both the private and public sectors.”
Egypt emphasizes that access to broadband has to be both available and affordable to all users, assuring that broadband access has to be driven by both the private and public sector within the regulatory framework—particularly where effective competition is present in the market—and supported by NTRA interventions only when necessary to correct market failure. MCIT & NTRA have developed national policies and strategies for broadband promotion, and for bringing broadband to regions, or to communities, that would not be among the first to be served through the operation of market forces.
C
C
h
h
a
a
p
p
t
t
e
e
r
r
2
2
T
T
e
e
c
c
h
h
n
n
i
i
c
c
a
a
l
l
D
D
e
e
s
s
c
c
r
r
i
i
p
p
t
t
i
i
o
o
n
n
o
o
f
f
D
D
O
O
C
C
S
S
I
I
S
S
Residential Internet and online usage has managed to grow tremendously despite maddeningly-slow speeds available through existing dial-up telephone modem connections, limited to 53 Kbps or less. Touted as an interactive extravaganza, surfing the World Wide Web more typically offers narrowband users a click-and-wait experience. The growing frustration of existing online users is driving demand for higher-speed connections.
Cable TV Primer
Cable systems were originally designed to deliver broadcast television signals efficiently to subscribers' homes. To ensure that consumers could obtain cable service with the same TV sets they use to receive over-the-air broadcast TV signals, cable operators recreate a portion of the over-the-air radio frequency (RF) spectrum within a sealed coaxial cable line.
Traditional coaxial cable systems typically operate with 330 MHz or 450 MHz of capacity; whereas modern hybrid fiber/coax (HFC) systems are expanded to 750 MHz or more. Logically, downstream video programming signals begin around 50 MHz, the equivalent of channel 2 for over-the-air television signals. The 5 MHz - 42 MHz portion of the spectrum is usually reserved for upstream communications from subscribers' homes.
Each standard television channel occupies 6 MHz (US) / 8 MHz (Europe) of RF spectrum. Thus a traditional cable system with 400 MHz of downstream bandwidth can carry the equivalent of 60 analog TV channels and a modern HFC system with 700 MHz of downstream bandwidth has the capacity for some 110 channels.
There are three types of systems that are categorized by their highest operating frequency.
Bandwidth Operating Frequencies(RF range) Number of channels
170 MHz 50 MHz-220 MHz 12-22 (single coax)
Small
220 MHz 50 MHz-270 MHz 30 (single coax) 280 MHz 50 MHz-330 MHz 40 (single coax)
Medium
350 MHz 50 MHz-400 MHz 52 (single coax)/104 (dual coax) 400 MHz 50 MHz-450 MHz 60 (single coax)/120 (dual coax) 500 MHz 50 MHz-550 MHz 80 (single coax)
DOWNSTREAM SIGNALS: RANGES OF OPERATING FREQUENCIES AND CHANNELS
Cable Modem Access Networks
To deliver data services over a cable network, one television channel (in the 50 - 750 MHz range) is typically allocated for downstream traffic to homes and another channel (in the 5 - 42 MHz band) is used to carry upstream signals.
A headend Cable Modem Termination System (CMTS) communicates through these channels with cable modems (CM) located in subscriber homes to create a virtual local area network (LAN) connection. Actually the term "Cable Modem" is a bit misleading, as a Cable Modem works more like a LAN interface than as a modem. The external Cable Modem is the small external box that is connected to the computer normally through an ordinary Ethernet connection. The downside is that you need to add a (cheap) Ethernet card to your computer before you can connect the Cable Modem. A plus is that you can connect more computers to the Ethernet. Although internal PCI Cable modem cards are also available.
The cable modem access network operates at Layer 1 (physical) and Layer 2 (media access control/logical link control) of the OSI Reference Model. Thus, Layer 3 (network) protocols, such as IP traffic, can be seamlessly delivered over the cable modem platform to end users.
A single downstream 6 MHz television channel may support up to 27 Mbps of downstream data throughput from the cable headend using 64 QAM
transmission technology. Speeds can be boosted to 36 Mbps using 256 QAM. Upstream channels may deliver 500 Kbps to 10 Mbps from homes using
16QAM or QPSK modulation techniques, depending on the amount of spectrum allocated for service.
This upstream and downstream bandwidth is shared by the active data subscribers connected to a given cable network segment, typically 500 to 2,000 homes on a modern HFC network.
An individual cable modem subscriber may experience access speeds from 500 Kbps to 1.5 Mbps or more − depending on the network architecture and traffic load − blazing performance compared to dial-up alternatives.
In addition to speed, cable modems offer another key benefit: constant connectivity. Because cable modems use connectionless technology, much like in an office LAN, a subscriber's PC is always online with the network. That means there's no need to dial-in to begin a session, so users do not have to worry about receiving busy signals. Additionally, going online does not tie up their telephone line.
Shared Network Platform Performance
Most cable modem systems rely on a shared access platform, much like an office LAN. Because cable modem subscribers share available bandwidth during their sessions, there are concerns that cable modem users will see poor performance as the number of subscribers increases on the network.
If congestion does begin to occur due to high usage, cable operators have the flexibility to add more bandwidth for data services. A cable operator can simply allocate an additional 6 MHz video channel for high-speed data, doubling the downstream bandwidth available to users. Another option for adding bandwidth is to subdivide the physical cable network by running fiber-optic lines deeper into neighborhoods. This reduces the number of homes served by each network segment, and thus, increases the amount of bandwidth available to end users. Cable operators are focusing on providing high-speed intranet access instead of straight Internet access for a simple reason: a network connection is only as fast as its slowest link. Clearly, the benefit of a 1-Mbps cable link is lost if a subscriber tries to access content stored on a Web server that is connected to the Internet though a 56-Kbps line. The solution to this dilemma is to push content closer to the subscriber, ideally right down to cable headend. This is done by "caching" or storing copies of popular Internet content on local servers, so when a cable modem subscriber goes to access a Web page, he or she will be routed to the server in the headend at top-speed, rather than being required to voyage out onto the congested Internet.
C
C
h
h
a
a
p
p
t
t
e
e
r
r
3
3
B
B
r
r
o
o
a
a
d
d
b
b
a
a
n
n
d
d
C
C
a
a
b
b
l
l
e
e
N
N
e
e
t
t
w
w
o
o
r
r
k
k
s
s
There has been much controversy in recent years over the provision of Internet access by cable providers, and whether they should be required to offer subscribers “open access” to Internet Service Providers (ISPs) that compete with their own ISP. The debate has ranged from political to economic to technical issues.
The Controversy Broadband cable services were introduced over cable in the mid-1990s, a few years before the commercial introduction of broadband over telephone and wireless technology.
Three Types of Cable Systems
The cable television industry includes three primary types of cable systems:
• “Branch and Tree” architecture offering one-way transmission only;
• “Hybrid Fiber/Coaxial” (HFC) two-way capable systems integrating fiber optic and coaxial cable; and
• “Fiber-to-the-Curb” (FTTC) enhanced two-way systems with increased reliability, capacity, and scalability.
All three categories include a central facility known as the “headend,” which serves as the central location for all technical operations. The headend receives and processes the various programming signals and then sends these transmissions to the subscribers over the cable plant. The headend building contains video modulators, network administration equipment, and the equipment used for signal receiving, processing, and transmitting, such as satellite and off-air antennas. In some systems, some of the functionality of the headend is distributed to “hubs” that deploy the headend equipment closer to the subscriber.
Generally, the remainder of the cable system can be referred to as “cable plant,” which includes all coaxial and/or fiber optic lines over which signals are sent, amplifiers and
nodes to boost and distribute the signal, and power supplies to run and maintain the system.
Branch and Tree System
In some leading countries, in cable networks, “Branch and tree” coaxial cable topology refers to the architecture of cable systems that have typically not been upgraded since 1995. These systems are also known as “legacy” systems because their architecture dates from the earliest days of cable in the 1950s and 1960s.
Branch and tree systems utilize dated technology that reflects the origin of cable television as a one-way entertainment medium with no status monitoring systems or architectural redundancy. Early cable television systems started as centralized antennas on hills that received over-the-air television signals and transmitted them by cable to homes that could not receive over-the-air signals. In later years, cable systems added additional signals to their offerings by receiving programming over satellite dishes. In this way, cable became a transmission medium for superstations, national news, sports, and movies channels as well as for the original local broadcast stations. Cable was able to offer more programming alternatives and better quality than over-the-air television. The dated architecture of branch and tree systems precludes two-way and other advanced services. All-coaxial systems cannot offer two-way services other than rudimentary pay-per-view and telemetry. Two-way operation is precluded by the large amount of system noise in the upstream direction and by the lack of fiber optics and, therefore, of significant capacity. A branch and tree system is based on one trunk. This is in contrast to more recent architectures described below, in which the system is segmented (essentially, multiple trunks are created by construction of neighborhood fiber optic nodes that translate and boost the signal) to enable each node to reuse channels and thereby multiply capacity for cable modem users.
Branch and Tree Architecture
The headend is at the center of a branch and tree cable system. It serves as the control center and reception point for all of the programming materials carried on the system. The trunk cables transport television signals from the headend to the most distant points in the franchise service area.
In a branch and tree system, the cable headend receives signals over two general types of antennas: off-air television antennas for local channels and satellite antennas (dishes) for long distance signals. For optimal signal reception, the antennas and headend are often located on a hilltop or other raised land area. Off-air antennas, which receive 55 to 890 MHz signals, are located on towers and aimed at television broadcast stations. Satellite dishes, which receive signals in the C band (3.7 to 4.2 GHz) and Ku band (11.7 to 12.2 GHz)45, are aligned with their transmitting satellites in geosynchronous orbit.
Local television stations sometimes deliver their programs directly to the headend over fiber optic cable to bypass the reception and processing issues associated with radio frequency (RF) transmissions.
Branch and tree systems use coaxial cable to deliver these signals to subscribers. A signal traveling through coaxial cable must be regenerated every one-half to 0.8 Km by an amplifier. The amplifier serves to boost the signal, but also introduces noise and distortion into the signal and is a potential point of failure in the system.
The size of the area served by a single coaxial cable system is limited by the maximum number of trunk amplifiers that can be connected in series, or “cascaded,” and still be capable of providing a satisfactory signal to the most distant subscriber. As illustrated in Figure 5, the trunking network functions as the backbone for the cable system. Typical systems have trunk cable runs comprised of between 15 and 40 amplifiers in series from the headend to the most distant subscriber, which means that the most distant subscriber is about 8 to 30 Km from the headend.
In order to service larger areas, cable operators must construct multiple headends or hubs or must devise special interconnection networks for connecting the systems. Multi-channel microwave links, “super-trunk” cable, and point-to-point fiber optic links are generally the most common technologies used for interconnection.
Branch and Tree Bandwidth and Frequencies
Branch and tree systems have only sufficient channel capacity to support one-way, analog television signals. They typically range from 330 to 550 MHz, or 40 to 75 television channels. Figure 6 illustrates how frequencies are allocated for cable television systems and how branch and tree system capacity supports only analog television channels in contrast to the categories of systems discussed below, which can also support digital TV and interactive applications.
Technical Limitations of Branch and Tree Architecture
There are significant technical limitations with this architecture. The large physical size of the network results in a large number of potential points of failures. All subscribers beyond a failure point experience system outages if a failure occurs in a trunk amplifier located between the headend and the end of the network. In a large cable system, an individual trunk cable might be part of a link that serves tens of thousands of subscribers. A failure at or near the headend can result in a substantial number of subscribers experiencing an outage.
Maintaining the system is an expensive and extensive task, because every trunk amplifier must be checked and adjusted relative to the other amplifiers, a challenge comparable to tuning a group of musical instruments.
Branch and Tree Architecture Precludes Two-Way Service and Open Access.
Hybrid Fiber/Coaxial System
Since the mid-1990s, most American cable networks have incorporated fiber optic technology. These systems use fiber optic cable to link the headend to neighborhood coaxial cable in an architecture called Hybrid Fiber/Coaxial (HFC). In the neighborhoods, the traditional coaxial cable distribution remains, but with upgrades to enable two-way operation. Figure 7 illustrates HFC architecture.
Generally, the evolution of cable networks from the branch and tree configuration to modern HFC networks has entailed construction of fiber optics from the headend to intermediate “hubs” and then eventually to “nodes” in each neighborhood. The nodes contain active devices that convert the fiber optic signals to RF signals for delivery over existing coaxial cable. This architecture has enabled the provision of two-way services and has greatly increased the reliability and quality of the signals offered over the cable system.
Advantages of HFC Architecture
The use of fiber optic cable in HFC systems provides a significant number of advantages over all-coaxial branch and tree systems. These improvements include:
• Fiber backbone with greater capacity than coaxial trunk cables;
• Ability to segment neighborhoods based on nodes, increasing available capacity for each subscriber;
• Reduction in active components, decreasing noise;
• Higher reliability and more cost effective maintenance; and
• Fiber replacing much of the coaxial cables plant, reducing susceptibility to unwanted electromagnetic interference.
All of these improvements make it possible for HFC systems to offer high-speed Internet service with several times the speed of conventional phone line services. In practice, properly operating cable modem networks operate about three times as fast as telephone services in the upstream direction and up to twenty-six times as fast in the downstream direction. HFC systems offer significant reliability, as well as the capability to monitor problems and outages, so that customer complaints are not the sole form of status monitoring, as they are in branch and tree systems. As the Internet becomes a more critical part of economic and emergency infrastructure, that reliability becomes crucial. Customers rely on the telephone infrastructure for critical services and will increasingly demand the same reliability from cable modem infrastructure for Internet and telephone services.
Technical Description of HFC Architecture
Hub and Node Segmentation
In an HFC system, signals leave the headend through laser transmitters that convert signals from RF format into light. Narrowcast lasers send signals bound for specific nodes, and higher power broadcast lasers transmit video signals that are shared by all nodes. Broadcast laser transmissions typically transmit the video programming to all
transmitted by way of hubs and are then received by nodes that convert the signal into RF for coaxial distribution to subscribers. Nodes also contain laser transmitters that send upstream data originating from subscribers back to laser receivers in the headend.
Distribution facilities, known as “hubs,” interconnect fibers to the neighborhood node areas and are intermediate between headend and node in a metropolitan area system. The hubs vary in size depending on the design philosophy or complexity of the network; however, they are usually stand-alone facilities with continuous backup battery power. The hub facilities receive their signals from the headend, usually by two discrete transmission paths to ensure that loss of an interconnection cable at one location will not create a single point of failure.
Hubs connect over fiber optic cable to neighborhood nodes, where the fiber interfaces with the coaxial distribution cable. The area served by a neighborhood node is referred to as the node area. Systems are typically designed with node areas that support between 100 and 2,500 residential dwelling units. Smaller node size allows for higher two-way capacity, along with greater system reliability.
The number of amplifiers between the headend and subscriber is reduced to less than eight in an HFC system. The shorter cascade lowers the signal degradation and reduces the number of potential failure points. An HFC system might typically have a capacity of 750 to 860 MHz, used to support a variety of analog and digital video services, two-way interactive data, and telephony.
HFC systems enable the reuse of system capacity for different neighborhood nodes. In other words, the segmentation of the system into separate nodes enables narrowcasting to individual node service areas, much as if each area were a different cable system. This segmentation enables the system to have adequate two-way capacity for telephone, Internet service, and video-on-demand. With increased network capabilities comes increased flexibility as well as technical complexity, since different combinations of multiple services are available.
HFC architecture enables a system simultaneously to broadcast cable channels system-wide and to narrowcast services that are specific to a neighborhood node. Transmissions
from data, telephony, and pay-per-view can be sent to individual users based on their service node.
Bandwidth
Figure 6 illustrates the allocation of bandwidth in a typical modern cable system. In the forward direction (from the headend to the subscriber) the available bandwidth could be in excess of 800 MHz. In the return path, information sent from the subscriber to the headend, the bandwidth is limited to a narrower range. As shown in Figure 6, the spectrum from 5 to 40 MHz is available for transmissions back to the headend, for a total effective bandwidth of less than 35 MHz. This asymmetry exists because cable was originally designed as a one-way technology maximizing bandwidth to the consumer. Interactive services include pay-per-view and video-on-demand ordering, cable modem network status monitoring, and telephony. If services are to remain in operation during power outages at the subscriber’s home, additional power redundancy must be built into the HFC network. The redundancy may be in the form of power through the network, as is done over standard telephone networks, or power through a battery pack at the subscriber’s home.
The size of the node area is a critical performance parameter because all of the bandwidth for interactive services must be shared among the users connected to the node. For example, a node serving 500 homes with a cable modem penetration of fifty percent might need to service up to 250 users simultaneously. In contrast, a smaller node serving 150 homes with the same penetration level would only be required to service 75 homes simultaneously, essentially providing three and one-half times as much usable bandwidth for each subscriber.
Headend
HFC system headends have similar receiving antennas and processing equipment to branch and tree systems, but with additional equipment to accommodate such two-way services as high speed Internet and telephony.
Redundancy
HFC system headends include system redundancy that was not a priority in branch and tree systems. Redundancy typically includes backup power and redundant HVAC. Redundancy also includes failsafe communications technologies such as SONET backbone rings and data and telephone equipment with redundant power supplies, chassis, and modules. Headend facilities are equipped with battery uninterruptible power supplies and diesel or natural gas generators that continuously power the headend in the event of a power failure. Status monitoring devices in the system headend monitor the signal and power systems in the cable network. Monitoring equipment can then notify maintenance staff of any problems that need attention before the problems affect subscribers.
Staffing Needs
Introduction of advanced cable technologies necessitates a corresponding upgrade in the skills of system staff. A 24-hour staff presence is needed in the headend or data center to detect and troubleshoot problems. Other parts of the network should be configured to alert staff of problems.
Repair personnel must also have expertise in fiber splicing. Customer service and installation staff must be versed in computer hardware and software. Installing cable modems at subscriber homes involves knowing how to install PC peripherals, dealing with a wide variety of customers and their computers, and being able to recognize user hardware and software which may or may not be compatible with the components to be installed. Procedures must in place to escalate problems to regional or national staff or to vendor support in the event that these issues cannot be resolved by system staff.
Operation of a Cable Modem Network
Cable modem network operation is comparable to Ethernet packet data networks, where many users utilize a shared medium. The modem is connected to the network by either the subscriber or an installer. Once on the network, the modem communicates with a cable modem termination system (CMTS), a device that sets the power level of the
All downstream data is sent out in one shared stream, with each modem reading only authorized information addressed to it. Upstream data is arranged into slots, where each modem “speaks” during its assigned time slots. Business or high-end customers may receive more time slots or higher priority.
Cable modem transmission is illustrated in Figure 9.
Digital video and phone services are offered on separate channels. As telephone technologies become integrated with Internet Protocol (IP), voice and video will be capable of being combined into the same channels as cable modem data. The same headend equipment, probably a CMTS, would serve as the headend interface device for all services.
The CMTS also interfaces RF cable plant with the cable operator’s Ethernet or ATM packet data network. As is illustrated in Figure 8, a router connects the CMTS to the Internet backbone, to an associated ISP, or to servers for mail, the web, news, and chat. Various local servers may also connect to the router at the headend for caching of frequently viewed web sites. Other content sources include video servers for video-on-demand that handle subscriber requests for access to scheduled programs.
Caching
An ISP may cache (store locally) the information that subscribers request from the Internet. Content caching may improve network performance. When a user on the network visits a web site, the web server downloads the site from the Internet and sends it to the user’s cable modem and also saves a copy of the site in a cache. As the cache space fills up, the oldest site files on the disk are cleared as the newest files are saved. If a user requests a site while it is cached at the headend, the server can download the site directly from the cache to the user instead of using the Internet to access the web site again. Multiple caches can be linked together to form cache hierarchies as well. If a site is not currently saved on a particular cache, the web server can try to retrieve the site from a cache at a regional ISP operations center, which is still faster than downloading from the Internet. This results in a faster download for the user and reduced traffic on the network. Figure 10 illustrates one use of caching. In this illustration, User 1 requests a site that is not currently cached, and the site is downloaded to the server cache as well as to the User. When User 2 requests the same site, it is obtained from the cache, eliminating the steps of going through the Internet to find and retrieve the site.
Locating Content Locally
Guaranteeing high quality video-on-demand and interactive services may require more extensive data and processing capability at the headend or regional network operations center (NOC) rather than at the facilities of the Internet content provider. In this scenario, content providers such as Intertainer.com, who supply live and stored video and interactive games, station their content sources and processing power at local headends or regional NOCs.
As with site caching, distributing and moving data closer to the subscriber can increase file access speed and reduce bandwidth consumption on the Internet backbone. A content server at the headend can deliver programming to users faster and more reliably than a
remote content server across the public Internet. Delivering content from headend servers also reduces Internet traffic and network congestion.
A network of smaller servers throughout the Internet also increases redundancy and allows different geographic areas to have customized video availability. Three different content server placement scenarios are illustrated in Figure 11.
Limitations of HFC architecture
The shared HFC architecture also creates limitations for the network. For example, security concerns necessitate that packets on the network be encrypted or scrambled to protect the information of subscribers sharing a segment. The architecture also does not offer a ready-made solution to offer a range of service levels to different customers.
physical architecture from the provider of the Internet connection and Internet services, relative to a physical architecture where each user has a dedicated physical connection from a home or business to the ISP’s routers. All of these challenges have solutions that are being tested and implemented in the cable industry.
Another limitation of the HFC architecture is that extensive additional fiber construction and terminal equipment are required to scale HFC systems for significantly greater bandwidth per customer. There exists a hard capacity limit per node area. The limitation is imposed by the need for data services to go through HFC-based router equipment in the cable headend. In all existing and planned cable modem systems, the hardware limits each network segment to 40 or less Mbps downstream capacity. In order to increase the capacity available to a subscriber, the cable operator must segment its system to progressively smaller node areas. Even at maximum segmentation, HFC will have a hard limit of 40 Mbps per user. This is in contrast to fiber optic technologies, that transport hundreds of thousands of Mbps, and that can be easily scaled to higher speed as technology advances by changing the equipment at the ends of the fiber and leaving the cable plant itself unchanged.
HFC-based equipment is also more specialized than equipment for fiber optic communications and is thus manufactured by fewer companies. This affords the cable operator less flexibility than an ISP using telephone or carrier facilities.
Fiber-to-the-Curb Architecture
The third category of systems, known as fiber-to-the-curb (FTTC), continues the trend of deploying fiber deep into the network. As nodes are segmented into smaller areas, the number of users on a node decreases and available bandwidth and system redundancy increase. In a variation of FTTC architecture, “fiber-to-the-home” (FTTH) systems deploy fiber all the way into residences.
Technical Description of FTTC Architecture
FTTC architecture is characterized by headends and hubs interconnected with fiber in multiple rings. In addition, fiber rings extend to neighborhood nodes, with 10 to 150 homes per node. The fiber follows city and neighborhood streets past residences, with more than one transmission path to the headend or hub for each node. Redundant transmission paths ensure that loss of an interconnection cable at one location will not create a single point of failure. Although this discussion is specific to cable networks, FTTC principles are also applicable to a carrier who provides its services over twisted-pair telephone lines.
FTTC architecture is illustrated in Figure 12.
As envisioned here, FTTC systems have sufficient capacity to offer individual subscribers a choice between cable-modem based services for the home and small office, and, on the other hand, premium carrier-grade direct fiber optic services. Additional fiber optics enable a residential or business subscriber to obtain fiber optic connection at relatively low installation charge, providing the option of receiving higher speed symmetrical services on pipeline unmanaged by the cable operator. This is an attractive option for a user who requires high capacity. It may also be desirable for a customer who cannot send information through a shared cable modem system because of specialized applications, security needs, or a need to connect directly to a specific network.
In addition to the equipment included in HFC headends, FTTC systems may include digital file servers for demand and interactive television services for video-on-demand subscribers. As more advanced and lifeline services are introduced on the system, more system monitoring equipment may need to be installed in the headend and in the physical plant.
Users desiring Gigabit Ethernet or other premium high-speed service will connect via fiber directly into the headend or hub router or SONET multiplexer, bypassing the CMTS equipment. This can be accomplished by offering direct fiber users a managed service in which they connect to cable company routers, or by offering users opportunity to connect to other service providers in a co-location area in the headend or hub.
Figure 13 illustrates an FTTC headend or network operations center.
FTTC headends include:
• SONET-based fiber multiplexer equipment for telephony and fiber customers.
• Packet switches and routers between customers and the Internet.
• Status monitoring of signal parameters and operation of field equipment.
• Remote monitoring of equipment, HVAC, and intrusion at hub sites.
• Cache servers.
• Co-location of facilities for multiple service providers.
• Servers for interactive television, video-on-demand, subscription video-on-demand, and web content (potentially multiple competing providers in the co-location area).
• Back-office infrastructure for subscriber and service provider provisioning and billing. Multiple survivable Tier 1 connections to the Internet from multiple providers.
• Staffing for 24 hours per day and seven days per week. The advantages of FTTC include the following:
• Fiber optic cable costs approximately the same per-mile as coaxial cable.
• FTTC systems can provide more advanced high-speed interactive services than do HFC systems. An FTTC system can simultaneously offer interactive television, video-on-demand, and higher capacity data and Internet access. The deployment of fiber optics deep into neighborhoods enables the provider to offer all of the applications possible in HFC systems, and to operate with increased reliability and redundancy.
• Reliability is increased by replacement of active electronic components and coaxial cables by temperature-and RF-resistant fiber optic networks. In addition, the subscribers are able to connect via a range of services, including 10/100/1000 Mbps Ethernet, ATM, and dedicated fiber optics known as “dark fiber.”
• Scalability is high with FTTC because of the high density of fibers and coverage of nodes. The system can be upgraded, in its entirety or by neighborhood, to a fully fiber-optic passive optical network (PON) by: 1) constructing fiber to users’ homes, and 2) installing multiplexers at node locations, as shown in figure 14. Migration of FTTC to PON would also increase system scalability with almost unlimited capacity available to each home.
• Once constructed, FTTC architecture more economically facilitates the construction of fiber directly to those subscribers who request additional bandwidth, such as businesses and residents who run home businesses, telecommute, or are early adopters of new technology. With the ability to connect individual users with dedicated fiber optics, capacity is almost unlimited.
• This model thus addresses many of the limitations of HFC technology, and should be of interest to new cable operators and operators constructing networks in new developments, campuses, and apartment buildings. An FTTC system is likely to be the optimal choice when building a new network.
14
C
C
h
h
a
a
p
p
t
t
e
e
r
r
4
4
D
D
A
A
T
T
A
A
O
O
V
V
E
E
R
R
C
C
A
A
B
B
L
L
E
E
S
S
T
T
A
A
N
N
D
D
A
A
R
R
D
D
S
S
A
A
N
N
D
D
S
S
P
P
E
E
C
C
I
I
F
F
I
I
C
C
A
A
T
T
I
I
O
O
N
N
S
S
Introduction
Two international standards have emerged for cable modem products: DOCSIS (Data Over Cable Service Interface Specification), which is the standard in North America and other International markets, and EuroDOCSIS, which is the dominant standard in Europe.
The DOCSIS Standard
The Institute of Electronic and Electrical Engineering (IEEE) 802.14 Cable TV Media Access Control (MAC) and Physical (PHY) Protocol Working Group was formed in May 1994 by vendor engineers to develop an international cable modem standard. The group set a publication goal of December 1995, but missed that target by more than two years.
Tired of waiting for IEEE 802.14, cable operators combined their purchasing power to jumpstart the standards process. In January 1996, MSOs (Multiple System operator) Comcast, Cox, TCI (now AT&T), and Time Warner formed a limited partnership called
Multimedia Cable Network System Partners Ltd. (MCNS) to research and publish their
own cable modem system specifications. MediaOne Group, Rogers Cablesystems and Cable Television Laboratories Inc. also signed on to the initiative.
MCNS released its draft standard, called the Data Over Cable Service Interface Specification (DOCSIS 1.0), to the manufacturing community in March 1997. Vendors immediately began building prototype products and the first public interoperability demonstration of DOCSIS equipment was held in December 1997.
In early 1998 CableLabs began a formal certification program for DOCSIS equipment to ensure products built by different manufacturers are indeed compatible. In March 1998, the International Telecommunications Union (ITU) accepted DOCSIS as a cable modem standard, called ITU J.112.
CableLabs manages a certification process to ensure DOCSIS cable modems manufactured by different vendors comply with the standard and are interoperable. Those products that pass the tests earn the right to affix a seal marked "CableLabs Certified" to their DOCSIS cable modem packaging, informing buyers that the product is guaranteed to interoperate with other certified products.
CableLabs is not formally certifying DOCSIS headend CMTS products since the equipment is purchased directly by cable operators.
In April 1999 CableLabs issued a second-generation specification called DOCSIS 1.1, which adds key enhancements to the original standard, such as improved QoS and hardware-based packet-fragmentation capabilities, to support IP telephony and other constant-bit-rate services. In short, DOCSIS 1.1 provides the bandwidth and latency guarantees required to offer toll-quality voice, dedicated business-class data services and multimedia applications across a shared cable modem access network. The next-generation standard is designed to be backward compatible, enabling DOCSIS 1.0 and 1.1 modems to operate in the same spectrum on the same network.
In addition to 1.1, CableLabs has developed a third-generation standard called DOCSIS 2.0 that adds an advanced PHY to the core specifications to increase upstream transmission capacity and reliability. DOCSIS 2.0 mandates the use of both frequency-agile time division multiple access (FA-TDMA) and synchronous code division multiple access (S-CDMA) technology. DOCSIS 2.0 provides 30 Mbps of capability. All three phases of the DOCSIS specifications have been formally approved by national, regional, and international standards development organizations such as the Society for Cable Telecommunications Engineers (SCTE), the European Telecommunications Standards Institute (ETSI), and the International Telecommunications Union (ITU).
EuroDOCSIS
Through tComLabs in Belgium, cable operators are certifying modems for compliance with a European version of the DOCSIS standard called EuroDOCSIS.
The Euro-DOCSIS (European Data Over Cable Service Interface Specification Standard) defines interface specifications for cable modems, cable modem termination systems and embedded cable modems within a DVB-C set top box involved in high-speed data communication over Hybrid Fibre Coax (HFC) networks. The certification and qualification status gives the manufacturers the assurance that their product is
compliant with the specification and interoperable with other certified/qualified products.
Currently there are 3 Euro-DOCSIS specifications available:
1-
The first specification, called Euro-DOCSIS 1.0 is the basic specification derived from the DOCSIS 1.0 specification offering best-effort data communication. Certification and Qualification for Euro-DOCSIS 1.0 started between June 26 and August 11, 2001. The Euro-DOCSIS 1.0 specification can be found as annex N in the RFI-specification published by CableLabs (also found as annex B.N in ITU-T Recommendation J.112 Annex B). This specification was recently adopted by the SCTE as national standard.
2-
The second Euro-DOCSIS specification is Euro-DOCSIS 1.1. It adds to the Euro-DOCSIS 1.0 specification:
Service flow mechanism that enables quality of service;
Enhanced security based on certificates.
3-
The third specification is Euro-DOCSIS 2.0, offering enhanced capacity in upstream using an advanced (TDMA and S-CDMA) physical layer.
All specifications are backwards compatible.
Some of the details of cable modem requirements are listed below.
Physical Layer
Downstream Data Channel
At the cable modem physical layer, downstream data channel is based on North American digital video specifications (i.e., ITU–T Recommendation J.83 Annex B) and includes the following features:
• 64 and 256 QAM
• 6 MHz–occupied spectrum that coexists with other signals in cable plant
• Concatenation of Reed-Solomon block code and Trellis code that supports operation in a higher percentage of the North American cable plants
• variable length interleaving supports, both sensitive and latency-insensitive data services.
• contiguous serial bit-stream with no implied framing, provides complete physical (PHY) and MAC layer decoupling.
Upstream Data Channel
The upstream data channel is a shared channel featuring the following: • QPSK and 16 QAM formats.
• multiple symbol rates.
• data rates from 320 kbps to 10 Mbps.
• flexible and programmable cable modem under control of CMTS. • frequency agility.
• time-division multiple access.
• support of both fixed-frame and variable-length protocol data units. • programmable Reed-Solomon block coding.
• programmable preambles.
MAC Layer
The MAC layer provides the general requirements for many cable modem subscribers to share a single upstream data channel for transmission to the network. These requirements include collision detection and retransmission. The large geographic reach of a cable data network poses special problems as a result of the transmission delay between users close to headend versus users at a distance from cable headend. To compensate for cable losses and delay as a result of distance, the MAC layer performs ranging, by which each cable modem can assess time delay in transmitting to the headend. The MAC layer supports timing and synchronization, bandwidth allocation to cable modems at the control of CMTS, error detection, handling and error recovery, and procedures for registering new
Privacy
Privacy of user data is achieved by encrypting link-layer data between cable modems and CMTS. Cable modems and CMTS headend controller encrypt the payload data of link-layer frames transmitted on the cable network. A set of security parameters including keying data is assigned to a cable modem by the Security Association (SA). All of the upstream transmissions from a cable modem travel across a single upstream data channel and are received by the CMTS. In the downstream data channel a CMTS must select appropriate SA based on the destination address of the target cable modem. Baseline privacy employs the data encryption standard (DES) block cipher for encryption of user data. The encryption can be integrated directly within the MAC hardware and software interface.
Network Layer
Cable data networks use IP for communication from the cable modem to the network. The Internet Engineering Task Force (IETF) DHCP forms the basis for all IP address assignment and administration in the cable network. A network address translation (NAT) system may be used to map multiple computers that use a single high-speed access via cable modem.
Transport Layer
Cable data networks support both transmission control protocol (TCP) and user datagram protocol (UDP) at the transport layer.
Application Layer
All of the Internet-related applications are supported here. These applications include e-mail, ftp, tftp, http, news, chat, and signaling network management protocol (SNMP). The use of SNMP provides for management of the CMTS and cable data networks.
Operations System
The operations support system interface (OSSI) requirements of DOCSIS specify how a cable data network is managed. To date, the requirements specify an RF MIB. This enables system vendors to develop an EMS to support spectrum management, subscriber management, billing, and other operations.
Guidelines for this section
After the last introduction, this section shall introduce the most important standards for the transmission of video, audio and data over cable, that’s plus the related specifications that explains the interfaces in the network.
In this section, the standards of 4 organizations are presented. However, these standards are similar.
We will start by the standards and specifications of CableLabs and SCTE, as they are the first to adopt those standards and specifications. These standards and specifications are approved by International organization as ITU and ETSI to obtain international consensus.
After presenting the different standards, this section shall be ended by the relation between the standards and concluding how to use the standards in the licensing procedure.
¾
CableLabs
DOCSIS 1.0:
SP-BPI Baseline Privacy Interface Specification
SP-CMTRI Cable Modem Telephony Return Interface Specification SP-CMCI Cable Modem to Customer Premises Equipment Interface Specification SP-CMTS-NSI Cable Modem Termination System Network Side Interface Specification SP-OSSI Operations Support System Interface Specification
SP-RFI Radio Frequency Interface Specification
DOCSIS 1.1:
SP-BPI+ Baseline Privacy Plus Interface Specification
SP-BPI Baseline Privacy Interface Specification
SP-CMCI Cable Modem to Customer Premises Equipment Interface Specification SP-OSSIv1.1 Operations Support System Interface Specification
SP-RFIv1.1 Radio Frequency Interface Specification
SP-CMTS-NSI Cable Modem Termination System Network Side Interface Specification
BPI ATP DOCSIS 1.1 and 2.0 BPI Acceptance Test Plan
CMCI ATP DOCSIS 1.1 and 2.0 CMCI Acceptance Test Plan
OSSI ATP DOCSIS 1.1 and 2.0 OSSI Acceptance Test Plan
RFI ATP DOCSIS 1.1 and 2.0 RFI Acceptance Test Plan
DOCSIS 2:
SP-RFIv2.0 Radio Frequency Interface Specification
SP-OSSIv2.0 Operations System Support Interface Specification SP-BPI+ Baseline Privacy Plus Interface Specification
SP-BPI Baseline Privacy Interface Specification
SP-CMCI Cable Modem to Customer Premises Equipment Interface Specification SP-CMTS-NSI Cable Modem Termination System Network Side Interface Specification
eDOCSIS:
eDOCSIS
Specification eDOCSIS™ Specification
Existing DOCSIS specifications were created for stand-alone cable modems that provide high-speed broadband services using the hybrid-fiber-coaxial cable infrastructure. The emergence of a class of devices that embeds additional functionality with a Cable Modem, such as packet-telephony, home networking and video, has necessitated the creation of this specification to define additional requirements such as interfaces, management and provisioning models. This is necessary to insure that the Cable Modem will function properly and interact properly with the embedded Service/Application Functional Entities (eSAFEs).
DOCSIS Set-top Gateway (DSG) Interface:
SP-DSG-I01- 020228 DOCSIS Set-top Gateway (DSG) Interface Specification
¾
SCTE
ANSI/SCTE 22-1 2002
(formerly DSS 02-05) DOCSIS 1.0: Radio Frequency Interface
ANSI/SCTE 22-2 2002
(formerly DSS 02-03) DOCSIS 1.0: Baseline Privacy Interface
ANSI/SCTE 22-3 2002
(formerly DSS 02-04) DOCSIS 1.0 Part 3: Operations Support System Interface
ANSI/SCTE 23-1 2002
(formerly DSS 02-09) DOCSIS 1.1 Part 1: Radio Frequency Interface
ANSI/SCTE 23-2 2002
(formerly DSS 02-10) DOCSIS 1.1 Part 2: Baseline Privacy Interface Plus
ANSI/SCTE 23-3 2003
(formerly DSS 02-06) DOCSIS 1.1 Part 3: Operations Support System Interface
ANSI/SCTE 79-1 2003
(formerly DSS 02-01) DOCS 2.0 Part 1: Radio Frequency Interface
ANSI/SCTE 79-2 2002
¾
Relation
between SCTE standards and Cable Labs Standards:
SCTE Cable Labs Details
ANSI/SCTE 22-1 2002
(formerly DSS 02-05)
SP-RFI Radio Frequency Interface Specification
ANSI/SCTE 22-2 2002
(formerly DSS 02-03)
SP-BPI Baseline Privacy Interface Specification
ANSI/SCTE 22-3 2002
(formerly DSS 02-04)
SP-OSSI Operations Support System Interface Specification
ANSI/SCTE 22-2 2002
(formerly DSS 02-03)
SP-BPI Baseline Privacy Interface Specification
DOCSIS
1.0
ـــــــــــــــــــــــــــ SP-CMTRI Cable Interface Specification Modem Telephony Return ـــــــــــــــــــــــــــ SP-CMCI Cable Modem to Customer Premises Equipment Interface Specification
DOCSIS
1.0/1.1/2.0
ـــــــــــــــــــــــــــ SP-CMTS-NSI Cable Modem Termination System Network Side Interface Specification
ANSI/SCTE 23-1 2002
(formerly DSS 02-09)
SP-RFIv1.1 Radio Frequency Interface Specification
DOCSIS
1.1
ANSI/SCTE 23-3 2003(formerly DSS 02-06)
SP-OSSIv1.1 Operations Support System Interface Specification
DOCSIS
1.1/2.0
ANSI/SCTE 23-2 2002 (formerly DSS 02-10)SP-BPI+ Baseline Privacy Plus Interface Specification
ANSI/SCTE 79-1 2003
(formerly DSS 02-01)
SP-RFIv2.0 Radio Frequency Interface Specification
DOCSIS
2.0
ANSI/SCTE 79-2 2002(formerly DSS 02-07)
SP-OSSIv2.0 Operations System Support Interface Specification
eDOCSIS
ـــــــــــــــــــــــــــeDOCSIS™
Specification This specification defines additional features that must be added to a DOCSIS CableModem for
implementations that embed the Cable Modem with another application.
DOCSIS
Set-top ـــــــــــــــــــــــــــ
SP-DSG-I01- 020228
This specification defines the interface requirements for transport of a class of service
¾
tComlabs
This organization was responsible for enhancing the DOCSIS standards to be compatible with the European requirements;
EuroDOCSIS 1.0 This specification adds the European specification additions to DOCSIS 1.0, it can be found also in Appendix N in DOCSIS 1. EuroDOCSIS 1.1 This specification adds the European specification additions to DOCSIS 1.1, it can be found also in Appendix N in DOCSIS 1.1 EuroDOCSIS 2.0 This specification adds the European specification additions to DOCSIS 2.0, it can be found also in Appendix F in DOCSIS 1. Gateway
(DSG)
known as “Out-Of-Band (OOB) messaging” between a set-top network controller (or servers) and the customer premise equipment (CPE).
¾
ITU
The ITU has specified the J-series for Cable networks and transmission of television, sound programs and other multimedia signals.
Hereafter, we had divided the related standards in the J-series to the Data over Cable service into two parts; the first contains the standards that explain the technical description of the service. The second part contains the standards that describe the QoS related issues.
Technical description:
J.83
Digital multi-programme systems for television, sound and data services for cable distributionJ.112
Transmission systems for interactive cable television services
Example of linking options between annexes of Rec. J.112 and annexes of Rec. J.83 may be found in Supplement 1 to J series (1998).
Guidelines for the implementation of annex A of Rec. J.112 may be found in Supplement 2 to J series (1998).
J.112 Annex A
Digital video broadcasting: DVB interaction channel for cable TV distribution systemsJ.Sup2
Guidelines for the implementation of Annex A of
Recommendation J.112, "Transmission systems for interactive cable television services" - Example of Digital Video Broadcasting (DVB) interaction channel for cable television distribution
J.112 Annex B
Data-over-cable service interface specifications: Radio frequency interface specificationJ.Imp112
Annex B
Implementor's Guide (04/03) for ITU-T Recommendation J.112 Annex B (03/01)
J.112 Annex C
Data-over-cable service interface specifications: Radio-frequency interface specification using QAM techniqueJ.Sup1
Example of linking options between annexes of ITU-T Recommendation J.112 and annexes of ITU-T Recommendation J.83QoS:
J.87
Use of hybrid cable television links for the secondary distribution of television into the user's premisesJ.140
Subjective picture quality assessment for digital cable television systemsJ.141
Performance indicators for data services delivered over digital cable television systemsJ.142
Methods for the measurement of parameters in the transmission of digital cable television signalsJ.143
User requirements for objective perceptual video quality measurements in digital cable televisionJ.144
Objective perceptual video quality measurement techniques for digital cable television in the presence of a full referenceJ.145
Measurement and control of the quality of service for sound transmission over contribution and distribution networksJ.146
Loop latency issues in contribution circuits for conversational TV programmesJ.147
Objective picture quality measurement method by use of in-service test signalsJ.148
Requirements for an objective perceptual multimedia quality modelGuidelines for ITU-T standards:
Technical description
• Recommendation J.83 provides worldwide specifications for the delivery of digital television services over a cable television network.
This Recommendation has four Annexes (A, B, C and D), that provide the specifications for four different digital television cable systems to be used in different ITU regions (Annex A depends on work done in Europe, Annex B and D depends on work done in North America, Annex C depends on work done in Japan).
The Recommendation defines the framing structure, channel coding and modulation for digital multi-programme television, sound and data signals distributed to the audience by cable television networks, possibly in frequency-division multiplex with existing analogue television signals.
• Recommendation J.112 extends the scope of Recommendation J.83 to make provision for bidirectional data transmission over coaxial and hybrid fiber-coax cables for interactive services.
Like Recommendation J.83, Recommendation J.112 contains several annexes in recognition of different existing media environments. The annexes in Recommendation J.112 should be read in conjunction with the corresponding annexes in Recommendation J.83.
It should be noted that the annexes to Recommendation J.112 describe different variations of the same protocol layers, for use in different ITU regions. However, telecommunications and computer standards that are well established and widely used in the public domain can support connectivity between these variations.
Recommendation J.112 should be studied in conjunction with Supplements 1 and 2 to the J-series Recommendations.
Supplement 1 gives an example of how the interactivity features described in a particular Annex to Recommendation J.112 for use with the transmission system
also be used in conjunction with the transmission system that is specified in a different Annex to Recommendation J.83. E.g. it describes how Annex A to Recommendation J.112, which is intended for use together with the transmission system specified in Annex A to Recommendation J.83, can also be used together with the transmission system specified in another Annex to Recommendation J.83.
Supplement 2 provides guidelines for the implementation of the interaction channel provided by cable television networks.
QoS
• Recommendation J.87 specifies operational rules, which should be followed in order to facilitate the carriage of both analogue and digital television signals of satisfactory quality on the same coaxial cable delivery system for the secondary distribution of television to the home.
• Recommendation J.140 describes a subjective method for assessment of picture quality for digital cable television systems. It concerns all of the television chain from the signal source to user's receiver. This chain may contain satellite links, terrestrial links and/or cable links. The assessment is made using consumer grade receivers assuming a home viewing environment.
• Recommendation J.141 recommends some performance indicators that can be used (among others) to evaluate the performance of digital modems in a hybrid fibre/coax (HFC) cable television network in the presence of continuous or impulsive noise. The Recommendation is based on some characteristics of the modems that are intended for use in the delivery of data services over digital television cable.
• Recommendation J.142 specifies objective methods for the measurement of parameters in the transmission of digital cable television signals. It concerns the end-to-end performance measurement of digital cable television signals from the signal
source to the user's receiver. This transmission chain contains the cable distribution system, consisting of
full-coaxial or hybrid fibre and full-coaxial (HFC) cables, and may also contain the satellite links, terrestrial links and/or broadband network links that may provide sources for
the cable head-end. The Recommendation is applicable for Digital Cable Television Signals using PSK,
QAM and OFDM modulation. Measurement of mutual interference between analogue and digital television signals is also described in the Appendices.
• Recommendation J.143 describes users' requirements for objective measurements of perceptual video quality in digital video systems used in cable television and in similar applications. Such objective perceptual video quality measurements may be required for several applications, as described below.
− To measure the end-to-end performance of digital cable television systems from the signal source to the user's receiver. In this application the transmission chain includes the cable distribution system and may also include the satellite
