Public
Copyright 1999 Cisco Systems, Inc. All Rights Reserved.
Page 1 of 12
W
HITE
P
APER
Combined Wisdom on the Upstream
Recommendations for Shortening the Return Path to Increased Revenue
Executive Overview
The demand for two-way services presents tremendous
opportunity and challenges. Few operators would dispute
the revenue potential, but not everyone agrees on what
deployments cost or how long they take. Everyone has heard
or told stories about two-way rollouts. Often a single
multiple service operator (MSO) will have one city with a
successful, rapid deployment, and another that stumbles,
often missing key milestones and deadlines. Why are there
such differing and seemingly unpredictable results?
Cisco engineers have observed and assisted in thousands
of deployments worldwidein every stage of the deployment
process, in every conceivable plant environment. The Cisco
team has seen extremely lucrative operations and been called in
to help companies recover from near technical-turned-nancial
disasters. Based on this collective experience, consensus
suggests there is a simple formula for success: do it right
the rst time.
Without implying an overly simplistic process or trivial
investment, the Cisco experience indicates that the successful
companies are those that carefully plan deployments, buy
sufcient equipment, and aggressively train technicians
at the beginning of the process (Figure 1). Those that deploy
by trial and error often end up dealing with public relations
or nancial challenges.
In many markets where MSOs have spent years with
single-digit growth rates, the new revenue growth enabled by
two-way digital services has revitalized the entire company.
The payoffs of a rapid, successful deployment are signicant.
Experience has shown that the most successful companies
spend 10- to 15-percent of the upgrade budget to involve
local technicians, contractors, and system engineers early in
the planning stages, equipping them with the necessary
training, tools, and skills. The resultant savings in customer
service expenses, reduced outages, and increased product
quality, coupled with higher subscription rates earlier on,
more than cover those early investments.
Based upon this experience, Cisco offers a number of
practical guidelines that can help ensure success. Included
are suggestions on equipment features, training programs,
planning checklists, and sources for additional information.
HFC upgrades are often multi-million dollar endeavors
involving everything from trenching to the latest in fusion
splicing or 870 MHz ampliers. The latest technology, most
sophisticated equipment, and signicant planning create the
MSO architecture for the future. The corporate ofce has
substantial familiarity with the products, features, and
proceduresbut the local sites, and especially the technicians
and contractors often lack this early exposure.
Public
Copyright 1999 Cisco Systems, Inc. All Rights Reserved.
Page 2 of 12
Figure 1 Cisco Systems Has Invested in Two Headends and Extensive Training Labs to Facilitate Rollouts
The Opportunity
Some operators simply do not believe it is possible to quickly
or cost-effectively roll out a two-way network, but the list of
successful implementations is very long, and growing quickly.
Below are a few relevant, encouraging facts that might
convince the skeptics of both the market opportunities and
the potential for rapid, reliable deployments:
The cable industry estimates that there are nearly two
million cable modem subscribers, and the numbers are
growing at 27 percent per year. (Source: Cahners)
Only two years after standardization, data-over-cable service
interface specication (DOCSIS)-certied cable modems are
now available from 13 major equipment vendors, with more
expected soon. Sales of these standards-based devices have
rapidly overtaken sale of proprietary products. Earlier this
year the DellOro group reported that DOCSIS-based cable
modem termination system (CMTS) equipment had a larger
installed base of upstream ports than all proprietary
products combined.
Analysts report that new services offered by cable operators
over two-way networks will result in a 400-percent revenue
growth in just ve years. Applications such as data, voice,
digital TV, Web, e-commerce transactions, and new
forms of advertising and messaging are revolutionizing
entertainment. Customers pay $30 to $40 per month for
each major service category, and advertising revenue for
narrow cast or target market placements continues
to grow (Figure 2).
Figure 2 New World Opportunities
Often digital subscriber line (DSL) will appear in a market
within six months of two-way cable service launch. All
providers recognize the importance of early market
shareonce a consumer selects either DSL or cable for
broadband services there is no incentive to switch.
Reported churn rates are under two percent, and mostly
represent people moving out of the area. Once at the new
location, that customer actively seeks out whatever
broadband provider they can nd, and re-subscribes.
NVOD/VOD
Transaction
Advertising
Communications
Telephony
Messaging
Information
Web
News
ETV/Synch
Digital
Analog
$10
$10
$10
$25
$25
$10
$30 $30
Packet
Packet
Packet
Packet
Packet
Packet
Packet
RF
Per Sub (1997) Per Sub (2002) Format
Basic and
Premium
Broadband
Packet
Revolution
400%
Public
Copyright 1999 Cisco Systems, Inc. All Rights Reserved.
Page 3 of 12
Localized content, enabled by data-over-cable networks,
has proven to be a major competitive advantage.
With service averaging $40/month, one operator installs
1200 to 1300 modems per week using 61 full time data
installation technicians. They cover a service area of 511,000
subscribers. Installation schedules are 4-5 weeks out, with
minimal advertising. Another 30 technicians perform voice
installations. Digital set top roll outs are growing faster than
the supplier can ship the boxes.
Common Challenges
All cable systems around the world are the same. They rely
on the same fundamental architecture, the same fundamental
issues of craftsmanship, and the same broadband radio
frequency (RF) plant (hybrid ber coaxial [HFC]). There may
be varying levels or patterns of ingress noise or interference,
and different channel plans (National Television Standards
Committee [NTSC], Phase Alteration Line [PAL] or
Systemme Electronic pour Couleur Avec 3 Memoir
[SECAM]). But whether there is a powerful transmitter
within a few meters of an open drop, or an apprentice boat
builder using an arc welder in a garage, every cable operator
faces the same challenge: maintaining a stable return-path
channel (Figure 3). The fundamental processes of installation
and troubleshooting are equally applicable across every
environment, and the challenges can be met and handled with
the same basic set of plans, procedures, and equipment.
Figure 3 Spectrum Analyzer Plot of 5 to 42 MHz Upstream, with Multiple Services Operating at the Wrong Levels
Max Hold
Single Sweep
Upstream Channel
20.00 MHz
Low Band
Noise
Upstream Carriers
Public
Copyright 1999 Cisco Systems, Inc. All Rights Reserved.
Page 4 of 12
Technicians sometimes make the mistake of thinking that the
upstream and downstream are totally unrelated. Figure 4
illustrates an example of common path distortion (CPD).
CPD results from faults in the HFC network creating mixing
products of the forward signals and returning them in the
upstream. Temporarily removing the forward signals while
monitoring the return path is a quick way to verify CPD. While
many have focused upon faulty connections creating diode
effects, some have realized that improper forward alignment
can be just as damaging. Loose module screws, faulty power
packs, even failed power doubling hybrids can cause CPD.
Figure 4 Upstream Plot from Wavetek Wandel & Gotterman Pathtrack Return Path Monitoring System, Showing CPD
Many of these upstream impairments are intermittent, as shown in a three-day plot of a single frequency in the return band (Figure 5).
Figure 5 Sample CPD Performance History Report
Time Window Summary Report for 24 Hours from a Single Return Path
Suspected Common
Path Distortion
Peak distortion
repeated at
6 MHz intervals
Average a little
lower (may have
stopped or been
intermittent through
the day)
Sample CPD Performance History Report
5.0 7.6 10.2 12.8 15.4 18.0 20.6 23.2 25.8 28.4 31.0 33.6 36.2 38.8 41.4 44.0
20
10
0
-10
-20
-30
-40
-50
-60
Amplitude (dBmV)
Frequency (MHz)
Single Frequency Time Window for 24 Hours from Same Node
Graph shows one
of the CPD beat
frequencies
(31.25 MHz)
over time
CPD is high until
around 12:00 pm
when it was fixed
or just stopped.
Sample CPD Performance History Report
12:00 4:45 9:30 2:15 7:00 11:45 4:30 9:15 2:00 6:45 11:30 4:15 9:00 1:45 6:30 11:15
PM PM PM AM AM AM PM PM AM AM AM PM PM AM AM AM
10
0
-10
-20
-30
-40
-50
-60
-70
Amplitude (dBmV)
Time
Public
Copyright 1999 Cisco Systems, Inc. All Rights Reserved.
Page 5 of 12
In the old days with only analog cable television (CATV),
many signicant and intermittent problems could disturb
customer service, but could rarely end it. In Digital, a half dB
decrease in anything can interrupt the customers program. In
some cases, a one-half-second hit might disrupt service for
several minutes. Three years ago, a 30 minute citywide power
failure resulted in a 4 day cable modem outage for 17,000
customers. Today a Cisco based solution has demonstrated
full restoration 7-10 minutes.
In a digital network, small measurement or procedural
errors on the return path are not acceptable. 2 dB errors in
the upstream path can quickly add up to 810 dB tilt
and overloaded return ampliers or optics. Return-path
alignment teams must be very familiar with their equipment,
ampliers, and procedures. Even though 542 or 565 MHz
is not as many MHz as the forward path, it represents the
same alignment complexity and requires the same care.
To safeguard against service interruptions, both service
operators and vendors should invest in thorough testing of
structures before real-world deployment. Cisco, for example,
evaluates its cable modems and CMTS equipment in a large
system testbed with more than 4500 cable modems in one
room. Duplicating the real world as much as possible in a lab
presents the opportunity to characterize performance,
identify problem areas, and experiment with solutions
without negatively impacting subscribers.
In the real world, is squirrel chew signicant in this area,
and have all damaged spans been replaced? Often the CAD
department will design the network, and construction
implements the design. No matter how complete the
planning there are always minor deviations, and top of the
hood redesigns. Aside from returning these designs back to
CAD, construction crews should consider the geometry of
the installed product: Is the coax cylindrical or egg-shaped?
Does this span have repetitive dents every few feet? Are
connectors installed according to all specications?
For forward and return path alignment or sweep, often
the contractor has one method, the technicians another, and
sometimes the equipment vendor a third. Before embarking
on a schedule to wire the entire city, ensure all parties come
to agreement on procedures and completeness standards.
Everyone doing it his own way is a recipe for future
disappointment. At tremendous cost and frustration to the
operator attempting to generate revenue.
Figure 6 Cable Modem Signals Disrupted by Upstream Interference Due to Poor Return-Path Alignment
Modem clobbered
by impulsive noise
Impulsive noise
Zero Span Mode Analyzes Noise/Ingress Under Live TDMA Digital Signals
Public
Copyright 1999 Cisco Systems, Inc. All Rights Reserved.
Page 6 of 12
Laboratory test equipment can easily measure the upstream
time-division multiple access (TDMA) burst signals of cable
modems or set-top boxes. Unless such measurements can be
simplied and applied in the real world they are effectively
academic. Cisco has researched methods using typical
CATV test and measurement equipmentsuch as the
HP8591c and TEK 2715. Detailed procedures are published
in the Cisco CMTS documentation and are available on
the Cisco Web site.
In cooperation with major test equipment vendors and
several large MSOs, Cisco has rened an upstream burst
power measurement suitable for DOCSIS cable modems.
Cisco has also released a more detailed procedure for
measurements with adjacent channels (Figure 7).
Figure 7 HP8591c in Time Domain (Zero Span) Mode Looking at Upstream Bursts from Two Different DOCSIS Cable Modems
Modem on
Modem off
Zero Span Mode Analyzes Noise/Ingress Under Live TDMA Digital Signals
Zero Span Zero Span with Noise Carrier Level
Trigger
Line
25 dB Down
Noise Floor
Impulse
Hit
The only method to view in-channel interference.
Provides ability to measure per modem tx/rx power.
Calibrated Upstream Burst Power Measurement Using CATV Spectrum Analyzer
Freq: 21.08 MHz Span: 0 MHz Res BW: 3 MHz Vid BW: 3 MHz Sweep: 80 usec
Preamble
Data Bits
Public
Copyright 1999 Cisco Systems, Inc. All Rights Reserved.
Page 7 of 12
Table 1 Typical Upstream Ingress Bands (USA and Canada)
From (KHz) To (KHz) GAP (KHz) 200 (KHz) 400 (KHz) 800 (KHz) 1600 (KHz) 3200 (KHz)
5000
5950 950 4 2 1 0 0
6200
7000 800 4 2 1 0 0
7300
9500 2200 11 5 2 1 0
9900
10,100 200 1 0 0 0 0
10,150
11,650 1500 7 3 1 0 0
12,050
13,600 1550 7 3 1 0 0
13,800
14,000 200 1 0 0 0 0
14,350
15,100 750 3 1 0 0 0
15,600
17,550 1950 9 4 2 1 0
17,900
18,068 168 0 0 0 0 0
18,168
21,000 2832 14 7 3 1 0
21,850
24,980 3040 15 7 3 1 0
24,990
25,670 680 3 1 0 0 0
26,100
26,960 860 4 2 1 0 0
27,410
28,000 590 2 0 0 0 0
29,700
40,000 10,300 51 25 12 6 3
Table 1 includes no-y zones and protected airspace
in the upstream 5-40 MHz frequency band. In the rst row,
5000 to 5950 kHz is considered a protected airspace. The
second row starts at 6200, indicating a no-y zone
between 5950 and 6200 kHz. The 10.3 MHz gap between
29.7 and 40.0 MHz indicates this is expected to be the
cleanest part of the band with respect to ingress (Figure 8).
Often a particular ber node in the United States will
include interference in a so-called protected frequency. In
most cases, this interference occurs somewhere inside the
HFC network itselfsome minor malfunction in the
downstream path is wreaking havoc with the return band.
After resolution the band is again clean, and capable of
generating revenue.
Figure 8 Typical 5 to 42 MHz Return with Data
Public
Copyright 1999 Cisco Systems, Inc. All Rights Reserved.
Page 8 of 12
Proven Guidelines for Success
Although it would be impossible to discuss every
environment, system conguration, and service plan, a few
basic guidelines will favorably stack the oddswhatever the
particular business or technical situation. Following these
suggestions can go a long way to ensuring that the
deployment will go quickly and smoothly.
Train technicians at every level. Start at both ends,
including the headend and in-home installation teams.
Verify that they have quality test equipment, and the
knowledge and incentive to use it properly. If the drops are
of poor quality, replace them rather than postponing the
solution with drop lters (traps). If technicians do not
understand how to ensure consistent and reliable operation
of the upstream channel, give them time to study it. Set
quality standards, and utilize available tools and products
to dene actual system availability.
Early CATV designs provided, at best, 97-percent
availability. Modern HFC networks are capable of telco
grade 99.99 percent or better. Train the deployment team to
settle for nothing short of 100-percent qualityin parts,
in systems, and in their own attention to detail. Analog
systems offer some play in the steering, but digital
systems do not. A simple plant extension can cost
thousands of dollars in service calls if the return-path rough
balance of the new amplier shuts down a nodes two-way
services for three days (Figure 9).
Figure 9 48-hour Plot of Proposed Cable Modem Channel after Connection of New Plant Extension
Single Frequency Time Window for 72 Hours from One Return Path
Sample: Loose Plant Performance History
Average noise floor
at 17 MHz varies
consistently by
time of day
Indication of
return path with
an ingress problem
Level may be ok
but maintenance
now may prevent
future problem
7:00AM
11:00PM 7:00AM
11:00PM
from one return path
10
0
-10
-20
-30
-40
-50
-60
-70
Time
17 MHz noise
floor tracked
over time
Average noise
floor stays
fairly flat and
consistent over
the 3 day period
Single Frequency Time Window for 72 Hours from One Return Path
Sample: Tight Plant Performance HIstory
Inconsistent
Problem
(high peak,
low average)
10
0
-10
-20
-30
-40
-50
-60
-70
Time
Amplitude (dBmV)
Three-Day Plot of Proposed Cable Modem Channel on a Segment of "Tight Plant"
Public
Copyright 1999 Cisco Systems, Inc. All Rights Reserved.
Page 9 of 12
Invest in good test equipment. Give technicians the right
equipment to measure how products interact with the cable
network and to determine behavior of the network itself
over time. A wide array of network vendor and third-party
products exist to help technicians get a good picture of
system performance.
Make sure that installation and service teams are prepared
both in equipment operation and in troubleshooting
techniques. They should intimately understand ampliers,
cable modems, and CMTS equipment. Details count. The
Cisco CMTS, for example, maintains important historical
data about modem reliabilitypower-cycling or removing
the boards can destroy valuable information.
Evaluate available features before choosing among the array
of devicesor shopping on price or delivery schedule alone.
Although all DOCSIS cable modems share a common base
feature set, and most are very similar, some modems and
CMTS equipment can go far beyond the basics to reduce
operations costs. Invest in intelligent cable modems and CMTS
products. Part of the Cisco CMTS design is to reduce customer
care and operations expense by at least 10 percent each year.
Several patented and revolutionary features included in the
Cisco Cable Technology Suite (CTS) provide detailed
operational information to the MSO. Examples include:
DOCSIS 1.0 and 1.1 quality of service (QoS)
CTS ap list
CTS modem connectivity
CTS dynamic ranging
CTS modem health and performance
Figure 10 The Cable Modem Price Represents About 10 percent of the Monthly
Operating Costs. The Largest Expense is Customer Care and Operations
Proper tools can dramatically reduce the cost and time
associated with tracking down problems. One MSO used the
Cisco Technology Suite features to reduce open trouble tickets
during a massive cable modem rollout. On November 1, 1998
they had 150 open tickets with 1000 customers. Five weeks
later, it was down to 25 with 4200 customers. Before these
tools, they averaged ve trouble tickets resolved per day.
They were able to prioritize the most troublesome points
rst, target fault resolution, and improve efciency ten-fold.
Today, the main customer care call center dispatches based
upon the automated status reportsoften xing problems as
much as two weeks before customers call.
Lease/Connect
Mktng/Acquisition
Net/Server
Cable Modem
Content/Settlement
Install/Provision
Customer Care
and Operations
Line Costs, Service Mix
Caching
Competition, Brand,
Bundle, Offers
Service Mix, Services,
Bandwidth
Retail, Customer
Purchase, Dual Use
Licensing,
Original Content
Self Provisioning
Subscriber Churn
Streamlined Single
Packet Network,
Out of Box Experience,
Online Help
Cash Flow/Sub
BB Data OpEx (98)
Impact and
Dependencies
$2
$2
$3
$3
$4
$5
$10
Table 2 DOCSIS 1.0 Operator Dened QoS Proles
ID Prio
Max
Upstream
Bandwidth
Guarantee
Max Upstream
Bandwidth
Max
Downstream
Bandwidth
Max Tx
Burst ToS
ToS Mask
Value by Create
B Priv
Rate
Enab
IP Prec
Enab
1
0 0 0 0 0 0x0 0x0 CMTS(r) no no
2
0 64000 0 1000000 0 0x0 0x0 CMTS(r) no no
3
1 34500 0 34500 0 0x0 0x0 CMTS yes no
4
2 83200 3800 83200 0 0x0 0x0 CMTS yes no
5
3 1024000 0 3072000 0 0x5 0x2 Mgmt yes no
6
7 2000000 100000 4000000 0 0x5 0x2 SNMP yes no
Public
Copyright 1999 Cisco Systems, Inc. All Rights Reserved.
Page 10 of 12
Table 3 CTS Flap List Trouble Shooter Modem Fault Location and Diagnosis
The Cisco patent-pending CTS ap list provides automated fault detection up to two weeks before customers call in. Using the newly
available Cisco Cable Flap List Troubleshooter operators can quickly classify faults into multiple categoriesmatching the
service technician qualications.
The CTS ap list is 100-percent DOCSIS 1.0 compliant, consumes zero network resources, centralizes, and prioritizes fault
detection. It monitors every modem in the network at least once every 30 seconds.
Table 4 uBR#sh Cable Modem
Mac Addr CableIF Ins Hit Miss CRC P-Adj Flap Time
00d0.ba77.78a1
C5/0 U0 0 5124 440 1 496 538 Oct 13 15:50:13
0000.f027.4b8f
C5/0 U0 61 489 4888 0 28 284 Oct 13 11:02:45
0000.f027.986f
C5/0 U0 72 4610 2144 0 36 272 Oct 13 14:46:22
0000.f027.9ce1
C5/0 U0 81 469 1473 0 0 194 Oct 13 15:54:51
0000.f027.9851
C5/0 U0 4 5017 1292 0 72 169 Oct 13 09:26:33
0000.f027.9931
C5/0 U0 42 99 536 0 30 112 Oct 13 15:53:28
0000.f027.12a1
C5/0 U0 28 2626 480 0 0 71 Oct 13 14:49:43
0000.f027.1a19
C5/0 U0 21 5100 520 0 8 67 Oct 13 14:47:05
0000.f027.1a43
C5/0 U0 24 5095 335 0 8 58 Oct 13 14:46:22
0000.f027.1b3b
C5/0 U0 0 2460 468 0 16 49 Oct 13 14:46:54
Interface SID Online
Timing
Offset
Receive
Power
QoS
Settings
Max MAC
Address IP Address
Modem Mac
Address
Cable3/0/U0
1 Online 3180 0.75 3 1 10.128.18.4 0000.f027.98d1
Cable3/0/U0
2 Online 3172 1.00 4 2 10.128.18.27 0000.f027.983d
Cable3/0/U0
3 Online 3202 0.25 3 1 10.128.18.5 0000.f027.49e9
Cable3/0/U0
4 Online 3261 1.00 3 1 10.128.18.15 0000.f027.9827
Cable3/0/U0
5 Online 3150 0.50 3 1 10.128.18.11 0000.f027.982d
Cable3/0/U0
6 Online 3165 0.50 3 1 10.128.18.6 0000.f027.9937
Cable3/0/U0
7 Online 3279 0.25 3 1 10.128.18.8 0000.f027.9833
Cable3/0/U0
8 Init(d) 3220 0.75 2 1 10.128.18.2 00d0.5836.c5cb
Cable3/0/U0
9 Reject(c) 3222 1.00 2 0 10.128.18.9 0000.f027.99d9
Public
Copyright 1999 Cisco Systems, Inc. All Rights Reserved.
Page 11 of 12
Based on extensive customer feedback, the show cable
modem command now includes all the DOCSIS states, as
well as other useful troubleshooting information, including
last received upstream RF power level and maximum number
of provisioned customer premise equipment (CPE).
Table 5 DOCSIS Cable Modem Registration States
DOCSIS cable modems are required to pass through
successive states during registration and provisioning, using
this information, operators can instantly pinpoint why a
modem is ofine, or unavailable.
Ofine modem considered ofine
Init(r1) modem sent initial ranging
Init(r2) modem is ranging
Init(rc) ranging complete
Init(d) dhcp request received
Init(i) dhcp reply received; IP address assigned
Init(o) option le transfer started
Init(t) TOD exchange started
Online modem registered, enabled for data
Online(d) modem registered, but network access for the CM is disabled
Online(pk) modem registered, BPI enabled and KEK assigned
Online(pt) modem registered, BPI enabled and TEK assigned
Reject(m) modem did attempt to register; registration was refused due
to bad mic
Reject(c) modem did attempt to register; registration was refused due
to bad COS
Reject(pk) KEK modem key assignment rejected
Reject(pt) TEK modem key assignment rejected
Table 6 CTS Service ID Connectivity
The Cisco service ID (SID) connectivity counter provides customer care personnel historical detail on modem link reliability and
helps prioritize and categorize customer calls. (See Table 6.)
By combining these advanced features with self-provisioning tools, MSO can greatly contribute to a timely rollout, and keep
up with customer demand. One operator uses a retail sales model similar to mobile phones. Subscribers simply sign up for cable
modem service at the time they purchase a PC or cable modem, and can be on line the same day.
Summary
By making the nancial and philosophical commitment to succeed, companies can literally be launching services within 90 days
after completing the two-way upgrade. The revenue realized from the rst rollouts can then nance the next phases of the
deployment, even helping to eliminate capital costs. By doing it right the rst time, MSOs can be guaranteed that the digital network
will function reliably, predictably, and protably.
Online Time Ofine Time
SID
First Time
Online
Times
Online
Percent
Online Min. Avg Max Min. Avg Max
3
Oct 15 1999 1 99.99 00:00 15d08h 15d08h 00:05 00:05 00:05
4
Oct 15 1999 1 99.99 00:00 15d08h 15d08h 00:04 00:04 00:04
5
Oct 15 1999 1 99.99 00:00 15d07h 15d07h 00:15 00:15 00:15
6
Oct 15 1999 1 99.99 00:00 15d07h 15d07h 00:10 00:10 00:10
7
Oct 15 1999 1 99.99 00:00 15d07h 15d07h 00:25 00:25 00:25
8
Oct 15 1999 1 99.99 00:00 15d06h 15d06h 00:03 00:03 00:03
9
Oct 15 1999 1 99.99 00:00 15d06h 15d06h 00:10 00:10 00:10
10
Oct 29 1999 2 82.67 18h04m 14h25m 18h04m 00:42 3h01m 6h02m
Copyright 1999 Cisco Systems, Inc. All rights reserved. Printed in the USA. Cisco, Cisco IOS, Cisco Systems, and the Cisco Systems logo are registered trademarks of Cisco Systems, Inc. or its afliates in the U.S.
and certain other countries. All other trademarks mentioned in this document are the property of their respective owners. The use of the word partner does not imply a partnership relationship between Cisco and
any of its resellers. (9911R) 12/99 BW5738
Cisco Systems has more than 200 ofces in the following countries. Addresses, phone numbers, and fax numbers are listed on the
C i s c o C o n n e c t i o n O n l i n e We b s i t e a t h t t p : / / www. c i s c o . c o m/ o f f i c e s .
Argentina Australia Austria Belgium Brazil Canada Chile China Colombia Costa Rica Croatia Czech Republic Denmark Dubai, UAE
Finland France Germany Greece Hong Kong Hungary India Indonesia Ireland Israel Italy Japan Korea Luxembourg Malaysia
Mexico The Netherlands New Zealand Norway Peru Philippines Poland Portugal Puerto Rico Romania Russia Saudi Arabia Singapore
Slovakia Slovenia South Africa Spain Sweden Switzerland Taiwan Thailand Turkey Ukraine United Kingdom United States Venezuela
Corporate Headquarters
Cisco Systems, Inc.
170 West Tasman Drive
San Jose, CA 95134-1706
USA
http://www.cisco.com
Tel: 408 526-4000
800 553-NETS (6387)
Fax: 408 526-4100
European Headquarters
Cisco Systems Europe s.a.r.l.
Parc Evolic, Batiment L1/L2
16 Avenue du Quebec
Villebon, BP 706
91961 Courtaboeuf Cedex
France
http://www-europe.cisco.com
Tel: 33 1 69 18 61 00
Fax: 33 1 69 28 83 26
Americas Headquarters
Cisco Systems, Inc.
170 West Tasman Drive
San Jose, CA 95134-1706
USA
http://www.cisco.com
Tel: 408 526-7660
Fax: 408 527-0883
Asia Headquarters
Nihon Cisco Systems K.K.
Fuji Building, 9th Floor
3-2-3 Marunouchi
Chiyoda-ku, Tokyo 100
Japan
http://www.cisco.com
Tel: 81 3 5219 6250
Fax: 81 3 5219 6001
References
Listed below are sources and materials that Cisco engineers
and customers have found to be extremely valuable in
preparing for two-way digital network deployments.
Textbooks
Raskin, D. and D. Stoneback.
Broadband Return Systems
for Hybrid Fiber/Coax Cable TV Systems.
Prentice Hall;
1997.
Thomas, J.L. and J. Thomas.
Cable Television
Proof-of-Performance: A Practical Guide to Cable TV
Compliance Measurement Using a Spectrum Analyzer.
Prentice Hall. 1995.
Thomas J.L. and F. M. Edgington.
Digital Basics for Cable
Television Systems.
Prentice Hall. 1998.
Farmer J., D. Large, and W. Ciciora.
Modern Cable
Television Technology: Video, Voice, and Data
Communications
(Morgan Kaufmann Series in
Networking). Morgan Kaufmann Publishers. 1998.
Web Sites
http://cable.doit.wisc.edu
www.cablelabs.com
www.cablemodem.com
www.cisco.com/cable
White Papers
Cisco Technology Suite: Flap List White Paper, Cisco Systems
Report on Dynamic CPD (common point distortion) by
Barry Patel, Comtel, http://cable.doit.wisc.edu/cpd/
cpd2.doc
Financial Studies
Chaners: http://www.dsldigest.com/
newsjumpdsl.cfm?ContentID=-1038707612
617/630-2139 or visit http://www.instat.com/catalog/
cat-cq-modem.htm#q399
Cisco Systems Inc.
Business Analysis for Cable Companies, Robin Opie, et. al.
[email protected] DellOro Group, www.delloro.com
Glossary of Terms
CMTS
Cable modem termination system
CPD
Common path distortion
CTS
Cisco Technology Suite
CPE
Customer premise equipment
DOCSIS
Data over cable system interface specication
DSL
Digital subscriber line, broadband consumer
access via twisted pair
RF
Radio frequency
TDMA
Time division multiple access