CCNP Practical Studies: Using DSL to Access a Central Site
This chapter covers the following topics:
ADSL Overview
Cisco 6160 DSLAM Overview
Cisco 6400 UAC Overview
DSL Access Architectures and Protocols
This chapter focuses on Digital Subscriber Line (DSL) technology. DSL, like cable modem, is one of the most popular broadband access methods and will be a new topic on the CCNP exam.
After completing this chapter, you will understand the basic Asymmetric DSL (ADSL) technology, Cisco 6160 DSL Access Multiplexer (DSLAM) configuration, and Cisco 6400 Universal Access Concentrator (UAC) configuration. You will also understand different access architectures and protocols such as Integrated Routing and Bridging (IRB), Routed Bridge Encapsulation (RBE), Point-to-Point Protocol over ATM (PPPoA), and Point-to-Point Protocol over Ethernet (PPPoE).
Note that there are different flavors of DSL technologies. This chapter focuses on ADSL technology.
ADSL Overview
DSL technology introduces a new family of products that can provide high-speed data and voice service over existing copper pairs. Several flavors of DSL exist, but each type can be categorized as either SDSL or ADSL. Symmetric DSL (SDSL) provides equal bandwidth from the customer premises to the service provider (upstream) and from the service provider to the customer (downstream). ADSL provides higher downstream speeds than upstream.
Traditionally, ADSL has been used to provide high-speed data service by encoding data on the local loop by using frequencies (up to 1 MHz) greater than voice (up to 4 kHz) so that existing telephone service would be preserved and would travel simultaneously with the data. At the central office (CO), the voice would be routed to the public switched telephone network (PSTN) using a low-pass frequency filter called a POTS splitter chassis (PSC).
Figure 8-1 depicts a typical end-to-end ADSL system. Beginning at the customer premises, the user's general-purpose computer is connected to the ADSL Terminating Unit-Remote (ATU-R) over an Ethernet connection.
Figure
8-1
Typical End-to-End ADSL System
The ATU-R is typically connected to an external splitter device. In some cases, however, the external splitter is eliminated in lieu of an internal filter in the ATU-R and microfilters attached to plain old telephone service (POTS) devices in the home. From the splitter, the loop is wired to a Network Interface Device (NID) that serves as the demarcation point into the customer premises. From the NID, the loop is connected to a splitter device in the central office that splits off voice traffic and routes it to the PSTN. Data is connected to the ADSL Terminating Unit-Central Office (ATU-C). The user's data traffic is then typically routed across the ATM network to an aggregation gateway or router.
Modulation Methods
Three modulation methods for encoding data onto the local loop are Carrierless Amplitude and Phase (CAP), Discreet MultiTone 2 - Issue 2 (DMT2), and G.lite. DMT was selected as the preferred standard for ADSL modulation. CAP technology is cost-effective and readily available. G.lite is a simplified DMT encoding scheme that provides limited features to facilitate interoperability and minimize end-user interaction.
Table 8-1 shows the maximum data rates for downstream and upstream, line-coding technologies, and maximum reach. Note that the maximum-reach number is best-case, assuming "clean copper."
Table 8-1 ADSL Data Rates
Maximum Data Rate Downlink/Uplink |
Line Coding Technology |
Maximum Reach |
8 Mbps/1 Mbps |
CAP, DMT |
18,000 feet/5.5 km |
1.5 Mbps/640 kbps |
G.lite |
18,000 feet/5.5 km |
Sources of Interference
Many sources of interference can degrade the quality of DSL. For instance, loading coils are used as a low-frequency (300 to 3300 Hz) filter but cannot be present for ADSL operation. Other sources of interference include the following:
Impedance changes
Bridged taps
Crosstalk
Impulse hits
Techniques to Solve Interference
Several techniques exist for adjusting to interference:
Rate-Adaptive DSL (RADSL)Used to adjust the transmission rate.
Reed-Solomon Forward Error Correction (FEC)The process of correcting errors mathematically at the receiving end of a transmission path rather than calling for a retransmission.
Bit interleavingUsed to avoid having consecutive errors delivered to the FEC algorithm at the receiving end of the circuit.
Trellis codingA modulation error-correction technique to improve error performance during reception.