Friend or Foe The VoIP Opportunity for Telcos.pdf

Document Sample
Friend or Foe The VoIP Opportunity for Telcos.pdf Powered By Docstoc
					Friend or Foe?
The VoIP Opportunity for Telcos
Executive Summary
While there has been much debate on convergence and the question of who might be best
positioned to deliver integrated voice, data, and video solutions to the end user, there is little
debate that Voice over Internet Protocol (VoIP) presents a market opportunity for service
providers. Deemed both as a great innovation or the ultimate destroyer of the local telephone
business, VoIP is in reality neither and yet a little bit of both at the same time.

With the explosive growth of the Internet and the continued introduction of new and
innovative IP-based services that take advantage of the Internet’s ubiquitous platform, the
telecommunications industry has undergone a significant transformation. As local exchange
carriers look to maximize ROI on their networks, grow service revenues, and minimize customer
churn, VoIP presents both an opportunity and a challenge.

Clearly, service providers that are first to market with service bundles that include high-value
voice, data, and video offerings will gain competitive advantages in addition to cost
efficiencies. Fortunately, with the right strategy and technical architecture for
transformation, telcos can seize the VoIP opportunity today. For telcos considering a VoIP
platform, an Ethernet and IP-based access network approach provides the best solution, laying
the foundation for the next generation of IP-centric applications.

Meeting the VoIP challenge, Occam Networks provides the Broadband Loop Carrier (BLC)
solution. With technology that makes IP and Ethernet ready for the public carrier network, the
Occam BLC solution uniquely provides local exchange carriers with the delivery platform they
require for the services of today and the future.


VoIP is here, and it is no longer a myth as a mainstream technology. A variety of key economic
and technology drivers have dramatically shaped the evolution and availability of VoIP.

Broadband Accessibility

 As the Internet has become a ubiquitous platform for communications, end users have
benefited from the availability of low and no-cost options for Internet connectivity. With time
spent online rising significantly since the onset of dial-up access, so too has the need for
bandwidth, fueled by the mass-market penetration levels of broadband access. With increasing
demand for bandwidth intensive gaming, multimedia, P2P, and enterprise applications,
consumers and businesses alike have widely adopted broadband and have demonstrated a
willingness to pay a premium for performance.

To meet the early demand for broadband using Digital Subscriber Line (DSL) services, many
service providers have augmented their networks with a mix of switched and cell technologies.
But to meet the service delivery and bandwidth needs of the future, the access network must
change. Access networks based on TDM/SONET/ATM do not scale efficiently to support the
future IP-based services world. As broadband penetration continues and individual subscriber
bandwidth needs increase, the need for a highly scalable, integrated, IP-based access
infrastructure becomes clear.

Demand for Cost Efficiencies

The development of VoIP technologies has presented a fundamental shift. Combined with
unregulated environments, new service providers quickly emerged to offer IP telephony options
that enable their customers to partially or completely bypass the toll charges associated with
the traditional Public Switched Telephone Network (PSTN).

Competitive threats from wireless, cable, and emerging VoIP service providers have introduced
enormous downward pressure in the cost of voice services, across local, long-distance, and
international telecommunications markets alike. Hundreds of VoIP service providers have
entered the market, without the service requirements or the cost structure of the traditional
access network. But for incumbent telcos, the access network is the key asset, and migration
to a single integrated IP platform that’s capable of providing voice, Internet, and advanced
services provides the infrastructure with a low cost structure required to compete long-term.

Revenue Opportunities through Higher-Value Services

But advances in technology have brought more than just a cheaper way to provide basic
services. In addition to cost savings, an IP-enabled network allows for value-added services to
be introduced as VoIP applications continue to evolve. Virtual private networking,
conferencing, IP Centrex, voicemail, messaging, video, and integrated video telephony services
are all potential applications that can be added to a VoIP service provider’s portfolio.

These VoIP applications significantly enhance the service set, opening new markets to the
service provider and allowing for new incremental revenues. Attractive pricing bundles, multi-
year term contract commitments, introductory offers, and promotion activities can all be
employed to win new business, increase average customer spend, and positively affect the
bottom line. Of equal importance is churn reduction. As more services are subscribed to,
there is less incentive to switch service providers, especially if alternative offerings lack similar
quality, selection, and flexibility.

Maximizing Existing Infrastructure and Maintaining Interoperability

While all telcos are looking for ways to reduce costs, critical considerations when weighing
proposed solutions are network integration, support, and maintenance. A key benefit of VoIP is
that it can be integrated as a technology into today’s existing networks. VoIP is extendable
into existing PSTN switches, wireless networks, and local access networks. But with numerous
VoIP solutions available, telcos are faced with the difficult decision of how best to migrate to
VoIP while maximizing their return on previous investment in their existing network

Selecting non-proprietary, open standards-based solutions provides carriers with the flexibility
required for long-term integration and interoperability while preserving network value. As IP
has evolved to be the predominant platform, carriers must consider the need to maintain
flexibility and simplify engineering and maintenance costs by adopting solutions at the network
and application layers that provide for interoperability.

What is Voice over IP?
With VoIP, analog voice is digitized into packets and transported over a network using the
standard Internet protocol (IP). VoIP can be supported on any IP-enabled network, including
the Internet, Intranets, LANs, WANs, and physical media including POTS, xDSL, and cable lines.

 What is VOIP?
 Voice over Internet Protocol
 • Used for Transport of Voice, Fax, Voice Messaging
 • Transported over an IP network or the Internet rather than the PSTN
 • Converts Analog Voice to Digital Format
 • Compression / Translation of the Signal into IP Packets
 • Conversion back to Analog at the receiving end

Table 1. What is VoIP?

In the early to mid 1990’s as VoIP technology was being developed, early supporters expected
that the efficiency of VoIP would quickly lead to the emergence of converged voice and data
services that would replace the traditional telephone, private branch exchange (PBX), and
central office equipment of existing circuit-switched networks. With some of the first
applications employing VoIP technology between LAN-based PC clients, these applications,
while not carrier ready, demonstrated the ability to bypass traditional PBXs and long distance
networks. Using PC-based microphones and headsets, early VoIP callers were generally
constrained by poor voice quality and few people with similar technology to whom they could

With clear cost-efficiencies, continuous advancement in both VoIP technology and broadband
availability fostered VoIP services that gained traction within the corporate private network
and consumer long distance markets. With large and medium-sized businesses making LAN and
WAN investments and upgrades by the late 90’s, VoIP solutions became widely adopted in the
corporate market. Advancements in service innovation, quality of service, standards,
reliability, and interoperability have enabled corporations to carry their traffic more cost-
effectively using VoIP on their private networks. Businesses have shifted from employing
separate telephone and LAN networks within the office to a single IP-based network. And by
utilizing WANs, corporations have realized tremendous cost saving advantages by connecting
their central and remote offices with VoIP technology.

Combined with the penetration of mobile, VoIP technology has completely changed the
landscape of the consumer and business long distance markets as well, with traditional market
leaders exiting segments and facing absorption into ILECs at a fraction of their market value
just a decade ago. With regulation lagging technology advancement, new VoIP service provider
entrants are offering domestic and international long distance services at a fraction of the
rates once charged by the “big three” long distance carriers.

Facing off competitive threats from new Internet Telephony Service Providers (ITSPs),
Competitive Local Exchange Carriers (CLECs), and cable providers that have emerged to grab
market share from traditional carriers by offering VoIP services in local markets, many ILECs
have developed or are planning to deploy their own VoIP and higher-value service offerings. As
local access networks migrate to all-packet network architectures with improved network
intelligence, ILECs will have the technical capability to offer a multitude of services, such as
video on demand and digital video broadcast services.

Figure 1. Original VoIP Applications

Evolution of VOIP Transport and Control Protocols
The rapid evolution of VoIP technology has produced numerous VoIP solutions and protocols,
raising technical considerations when considering how best to deploy a VoIP solution. A decade
ago, as some of the first VOIP solutions began to appear commercially, it became clear that
there was a need to coordinate architecture standards to provide for platform interoperability
and interconnection amongst networks.

Since that time, four predominant standards for VoIP control protocol have evolved, namely
H.323, MGCP, H.248/Megaco, and SIP. With a variety of overlap and applicability for different
networks and applications, each has its own strengths and weaknesses. In addition to these
protocol stacks, other protocols are used such as RTP and RTCP to assist with transport and
quality of service.

                     IP       Internet Protocol
                     UDP      User Datagram Protocol
                     RTP      Real-time Transport Protocol
                     RTCP     Real-time Control Protocol
                     H.323 ITU early Multimedia standard
                     MGCP     Media Gateway Control Protocol
                     H.248    ITU / IETF cooperative effort (Megaco)
                     SIP      Session Initiation Protocol

                             20           8            12        XXX
                             IP          UDP           RTP     Payload

Table 2. Transport and Signaling Protocols


Introducing H.323 in June 1996, the International Telecommunications Union (ITU) drafted a
standard set of specifications for packet-based multi-media communications. Originally
defining a protocol for video conferencing over LANs, the first iteration of this standard
supported point-to-point and point-to-multipoint communications. As the opportunity to
implement VoIP using PC clients emerged, a variety of software vendors, including Microsoft,
adopted the standard and began designing H.323 compliant gateways that were interoperable
with the Internet and the PSTN.

H.323 employs a distributed architecture and specifies how network endpoints can connect to
each other or to the PSTN via H.323 gateways. With carriers testing H.323 for potential
deployment, it was discovered that an H.323 gateway could only handle a few thousand
connections and that it lacked the scalability for carrier-class deployment. The H.323 standard
has since been updated with several iterations to address concerns including scalability, but as
connections are made directly by end users and not controlled centrally, it does not adapt
itself well to the carrier model.

Figure 2. Media Gateways

MGCP divides the traditional voice switching function across media gateways and media
gateway controllers, also known as call agents. With MGCP, the intelligence for the call is
located outside the gateways, administered by a central call agent. With point-to-point and
point-to-multipoint communications possible, the call agent acts as a central media controller,
responsible for administering call connections.

Due to the centralized control and logging, this master-slave architecture lends itself well to
service providers looking to implement large-scale residential VoIP, VoIP gateways for business,
or trunking connections between networks. A key advantage of MGCP’s centralized
architecture is that as advanced services are deployed, upgrades are concentrated at core
network resources which serve the application, such as the media server, without necessarily
requiring equipment upgrades at the network endpoints. With an emphasis on simplicity and
reliability, MGCP has become a widely used standard and enables service providers to develop
reliable and cost-efficient local access systems. It is compatible with Network Control System
(NCS) deployed by cable companies for the first generation of VoIP over broadband systems.of
subscribers makes it easier for a telco to recover capital investments more quickly.


In June 2000, the ITU and ITEF agreed upon a common standard which defined support for
media gateways. The ITU refers to it as H.248, while the ITEF refers to the same standard as
Megaco, derived from the phrase “media gateway control.” Largely based on MGCP, H.248
extends MGCP to support a broader range of networks such as ATM and also focuses on
establishing standards for IP telephony equipment.

Similar to MGCP, H.248 utilizes a master-slave architecture. A centralized controller
administers all communication flows. The controller specifies the media types shared amongst
the media gateways and network endpoints. Additionally, network performance and traffic
statistics are logged by the controller, providing central auditing and billing functionality.

Designed with consideration for future technology advancements, H.248 provides an open,
extendable framework that enables service providers to support new media types as they are
introduced. Other than basic POTS functionality, H.248 does not require intelligence at the
network endpoints, making it well-suited to meet the needs of Telcos deploying next
generation services to residential and small business customers.

Session Initiation Protocol (SIP)

Playing an increasingly important role in the development of telecommunications networks is
the Session Initiation Protocol (SIP), an application layer protocol introduced by the IETF.
Unlike the more network centric model of MGCP, the SIP specification is a client-server
architecture with peer-to-peer signaling and data control protocols. With a decentralized
architecture, SIP assumes that there is some level of intelligence at the network endpoint,
which is responsible for initiating any communications session. The SIP protocol supports a
wide range of session types including telephone, conferencing, and multimedia

Designed for interoperability and similar in architecture to the Hypertext Transfer Protocol
(HTTP), the original SIP specification (RFC 2543) continues to be advanced with additional RFC
update specifications. The SIP specification introduces three server elements that assist with
network client addressing and messaging:

Figure 3. Distributed SIP Architecture

      Registration Server – Handling all registration update requests, the SIP registration server
      or registrar maintains the current locations of SIP network end-points.

      Proxy Server – The SIP proxy server is capable of both processing requests and relaying
      requests to other proxy servers and network end-points.

      Redirect Server – The SIP redirect server receives SIP requests and redirects it to a
      destination server that can process the request.

With a distributed architecture and the need for intelligence at the endpoints, SIP is commonly
employed between two soft-switches or between a soft-switch and an IP PBX. It has been
widely utilized for IP Centrex applications within corporate VoIP networks where IP phones or
integrated access devices (IAD) are deployed. As intelligent endpoints, an IAD or IP phone is
capable of converting voice and telephone control signals into IP packets.

Figure 4. Simplified SIP Call setup and teardown

With its decentralized architecture, as advanced applications such as multimedia are deployed
within a SIP-enabled network, upgrades to the endpoint devices may be required to take
advantage of new applications. IAD hardware may be provided by the service provider, but
these devices are now also widely available at consumer retail outlets. As proxy server control
can be a factor affecting session logging, billing, and security, SIP may be more suitable for
corporate VoIP than ILEC VoIP network deployments.

VOIP Quality of Service Considerations
Implementing VoIP and transmitting voice in packet form over an IP network introduces several
performance considerations which may affect service quality, including latency, jitter, packet
loss, and echo.

Real Time Protocol (RTP)

RTP, defined by the Internet Engineering Task Force (IETF) in RFC 1889, establishes the
underlying protocol for exchanging real-time audio and video information. For VoIP
applications, it provides the necessary media encoding, sequencing, time stamping, and
monitoring features that are required for real-time communications. RTP is incorporated by
the other umbrella VoIP transport protocols.

Real Time Control Protocol (RTCP)

Enabling traffic quality monitoring, Real-time Transport Control Protocol is used in conjunction
with the RTP protocol. Useful for gathering performance-related information on RTP traffic, it
is helpful for identifying typical VoIP quality issues such as delay, jitter, and packet loss.


Latency represents the time required for a packet to travel from origination to endpoint
throughout the network. For a standard toll call, a latency measurement of 100-150ms is
generally acceptable for voice quality service. As latency extends beyond 150ms, there is a
noticeable lag in synchronization which can become disruptive to standard voice conversations.
And beyond 250ms, speakers can be negatively affected by talker overlap. Several factors can
contribute to overall latency:

         Accumulation Delay. Variances in coder-decoders (CODECs) used to digitize voice traffic
         can lead to accumulation delay, sometimes referred to as algorithmic delay. Generally,
         the larger the packet size, the larger the potential accumulation delay as the packet is
         held and filled with data. The use of efficient CODECs and smaller packet sizes help
         minimize accumulation delay.

         Processing Delay. This represents the time it takes to encode analog data samples into
         packets and the reverse process at both ends of the packet network. These encoding
         delays are a function of both the processor execution time and the types of CODEC

         Network Delay. Network delay is a function of the time it takes to serialize the data and
         carry it over the physical mediums employed in the network. Network delay is inversely
         proportional to the speed of the link and is also affected by the processing time required
         as the packet traverses the network.

         Queuing Delays. With packet networks capable of supporting multiple service classes,
         queuing delays can be a factor if proper bandwidth is not available and traffic is not
         prioritized efficiently by service class at any given time. With careful network design
         and class of service management, these delays can be minimized or non-factors.


Jitter is the variation in time between the expected arrival and the actual arrival of a packet.
To account for jitter and to preserve vocal quality, media gateways may employ jitter buffers
and buffer schemes to hold packets so that slower packets may arrive in order. This allows for
packets to be played in the correct sequence. Buffer schemes may also drop packets if the
packets arrive out of order.

Packet Loss

Packet loss occurs when a packet fails to reach its destination and is dropped, a concern in
highly congested or high error rate networks. Schemes can be introduced to account for
packet loss, including retransmitting and repeating packets to fill gaps. Non-real-time
applications like standard http Internet requests usually are more tolerant of packet loss. But
this can have a detrimental effect on real-time applications such as voice traffic if packets are
dropped and conversations become unintelligible. In addition to capacity planning, quality of
service management can alleviate the impact of packet loss by establishing priority preferences
for voice traffic.


Echo can also present a problem in a network if the round-trip delay exceeds 50 milliseconds,
which often occurs with VoIP. Capable of occurring in both circuit and packet-switched
networks, echo is generated by signal reflections within the network. ITU standards G.165 and
G.IEC, a more stringent standard, have been drafted to define performance requirements for
echo cancellers, which alleviate the problem.

Occam Networks Broadband Loop Carrier Solution
With considerations for current and future service needs, the Occam BLC solution is an ideal
choice as a VoIP-enabling platform. Occam’s BLC technology possesses a variety of key
strengths that address the critical economic and technical considerations for VoIP deployment.

    •   Maximized Value of Existing Infrastructure
    •   Lowest Cost Core Technologies
    •   Seamless Operation and Maximum Interoperability
    •   Carrier Class
    •   Service Quality Management
    •   Ethernet Protection Switching Reliability
    •   Extendability

Maximized Value of Existing Infrastructure

The Occam BLC solution enables telcos to implement an IP-based access network and offer VoIP
services while uniquely leveraging existing outside plant facilities and switching infrastructure.
With cost control being a high priority in implementing a converged network, Occam’s BLC
technology is superior in that line access, aggregation, and gateway functions are all supplied
by one integrated network device. This provides a number of significant advantages in terms
of minimizing deployment costs.

Subscriber                  Access Network                         Central Office
• Voice, Data, Video        • Fiber and Copper feeder to           • Voice (POTS and derived) is
• Existing copper plant,      remotes                                connected to traditional
  or FTTP in future         • All traffic carried as IP packets      Class 5 via TR-08 / GR-303
• Analog Lifeline POTS,     • POTS to VoIP (G.711                    T1 lines
  standard phones             uncompressed), RTP, MGCP                     -- or --
• ADSL2Plus, standard       • ADSL data (IP in ATM cells) to         Voice is connected directly to
  modems                      pure IP over Ethernet. ATM ends        a Next Gen switch – no TDM
• Res. GW / IADs for          at the port.                           Gateway
  derived voice lines       • Protected Ethernet: <50ms            • Data is connected directly to
                              recovery from network or node          Internet router – no ATM
                              failure                              • Video traffic directly from
                            • IP QoS: all traffic tagged,            IPTV Headend router – no
                              prioritized and forwarded              ATM
                              according to class

Figure 5. BLC 6000 IP Service Delivery Model

By removing the need for additional Digital Line Access Multiplexers (DSLAM) devices and
collapsing all necessary functionality for hand-off and switching into a single Occam BLC
device, capital equipment expenditures are reduced. The core access network is implemented
utilizing BLC Remote Terminal (RT) devices at the network node level and a BLC Central Office
Terminal (COT) at the Central Office. The simplified infrastructure reduces the amount of
equipment and vendors used, while minimizing installation, integration, and maintenance
support costs.

Lowest Cost Core Technologies

With aggregation occurring at the edge, a single IP integrated access network architecture
makes the most efficient use of available bandwidth, while simplifying switching and call
routing decisions. There is no need to scale multiple networks, with the BLC dynamically
supporting all IP-based services utilizing the same core Ethernet network. Carriers can
conserve bandwidth and increase capacity only when needed. Further, eliminating the
complexity of ATM from the network reduces overall capital expenditures.

Seamless Operation and Maximum Interoperability

Long-term operations and maintenance costs are also favorably impacted. The Occam BLC
solution provides seamless operation and maximum interoperability with its IP-based
architecture for the access network. Based on open standards, the BLC provides both the
functionality needed now as well as a clear integration path for the future, simplifying long-
term support needs and operational expense.

With an IP network provided from the customer to the core, Occam provides a platform
designed for vendor interoperability at the subscriber and CO ends of the network as new VoIP
and other advanced IP applications are introduced. And, importantly for the subscriber, the
BLC solution is also seamless, enabling the use of traditional telephones.

Between the subscriber and the BLC RT, Occam offers great flexibility by supporting a variety
of interfaces beyond standard POTS. This flexibility enables the use of IP Integrated Access
Devices (IAD) at the subscriber end. Ideally suited to serve the needs of both business and
residential service classes, Occam provides support for Asymmetric Digital Subscriber Line,
ADSL2 Plus, Optical Ethernet, and T1 interfaces.

                                                            As Next Generation
                                                            switching equipment is
                                                            deployed outside the access
                                                            network, migration planning
                                                            and infrastructure
                                                            investments are optimized
                                                            by BLC’s multi-protocol

Figure 6. BLC Support for Next Generation Switching

Occam relies on standards-based technologies in handling traffic from the customer to the CO
and from the CO to the appropriate upstream network (Internet, PSTN, and Video Head-end).
Occam converts POTS to VoIP at the BLC RT where it is forwarded to the CO. And softswitch-
ready, the Occam BLC solution can directly hand-off voice traffic to a softswitch at the CO.
Alternatively, if the voice traffic is destined for a Class 5 switch, a BLC COT can be used to
convert the voice traffic into standard Class 5 TR-08 or GR-303.


Designed specifically to support long-term evolutionary infrastructure planning, the Occam BLC
solution provides telcos with a clear vision for transformation. Widely deployed since 2000,
Occam provides the solution for the access network that preserves competitive advantage,
enabling telcos to offer new and differentiated IP services.

Scalable to cost-efficiently serve from 24 to thousands of subscribers, the Occam BLC can be
deployed within the Central Offices, remote terminals, vaults, and multi-tenant buildings.
Uniquely architected for redundancy and space management, Occam’s Intelligent Blade
Interconnect Architecture (IBIA) provides direct interconnection of intelligent, distributed
processing blades. With options for low or medium-density stacks as well as highest-density
racked applications, Occam’s flexible product line is optimized for space, power, and HVAC

Service Quality Management

The Occam BLC solution provides strong QoS functionality based on the IEEE standards 802.1q
and 802.1p. By utilizing widely adopted standards, the Occam BLC implementation is a
preferred choice for interoperability and allows carriers to meet the diversified content and
bandwidth needs of their end subscribers. The BLC implementation provides compatibility with
outside networks utilizing IP QoS standards and helps ensure that all traffic types receive
appropriate prioritization as the packets enter and traverse the nodes in the access network.

The Occam Advantage
To remain competitive long-term, carriers must consider the benefits of VoIP and an IP delivery
platform, which provide the opportunities for high-value service bundles, new revenue sources,
and network cost reductions. Occam’s Broadband Loop Carrier solution offers a compelling
value proposition, providing a fully integrated platform for the delivery of VoIP, data, and
video services that uniquely and cost effectively gives carriers speed to market and
competitive differentiation. Eliminating the need for costly ATM technologies, Occam’ BLC
architecture simplifies the network with an all IP-services model. Using Ethernet transport to
provide economical, highly scalable bandwidth, Occam enables carriers to minimize spending
and to efficiently scale the access network in-line with subscriber.

Designed for maximum interoperability with both existing carrier infrastructure and next
generation switching, Occam’s BLC delivers the scalability, reliability, and service quality
management requirements for carrier-class service. And, committed to open standards, Occam
provides a fully extendable platform that is the building block to support the VoIP applications
of today while evolving the access network to support the applications of the future.


Shared By:
yanyan yan yanyan yan