6.1 Physical Layer Access

6.1.1 Synchronous Lines

Most WAN lines run over physical time-division multiplexed (TDM) networks. These are synchronous circuits, which means that they must have the same clock so that the receiving side knows exactly when each frame bit is received.

There are many synchronous line speeds available. These line speeds are based on the basic digital signal level 0 (DS-0) rate of 64 kbps.

The T1 carrier is the most common in North America. It can carry 24 DS-0s at for a capacity of 1.544 Mbps (24x64 kbps). Each of the 24 DS-0s carries data traffic. However, less than a full range of individual channels might be used. Fractional T1 service uses one or more 64 kbps and the DS-0 channel to provide the desired bandwidth.

The E1 carrier is the most common in Europe and other countries, and can carry 30 DS-0s for a capacity of 2.048 Mbps (30x64 kbps). Like the T1, less than a full range of individual channels might be used. This fractional E1 service also uses one or more 64 kbps and the DS-0 channel to provide the desired bandwidth.

The T3 carrier is a dedicated phone connection. It consists of 672 individual DS-0 channels and supports data rates of approximately 45 Mbps. The T3 is also commonly called DS-3 and carries 28 T1 lines.

6.1.2 Encoding Schemes of T1s and E1s

Bipolar 8-zero substitution (B8ZS) is a T1 carrier line code in which bipolar violations are deliberately inserted if the data contains a string of 8 or more consecutive 0s. B8ZS ensures a sufficient number of transitions to maintain system synchronization when the user data stream contains an insufficient number of 1s. The European hierarchy uses B8ZS at the T1 rate. Alternate mark inversion (AMI) is a T1 carrier line code in which 1s are represented by positive or negative voltage, alternatively. A 0 is represented by 0 volts.

6.1.3 SONET and SDH

Synchronous Optical Network (SONET) is an ANSI standard that defines physical interface rates, which allows data streams at different rates to be multiplexed. SONET defines Optical Carrier (OC) levels and is widely used in North America. These rates are listed below in Table 7.1.

SONET uses the Synchronous Transport Signal (STS) as its frame format and STS level 1 (STS-1) as the basic signal rate of 51.84 Mbps. Each SONET frame is constructed of 9 rows by 90 columns of octets for a total of 810 octets. These 810 octets are transmitted in 125 ^secs, or 8000 frames per second.

Synchronous Digital Hierarchy (SDH) is the international standard defined by the International Telecommunication Union (ITU-T) for transmission over fiber optic cables. It is widely used outside of North America. It defines the hierarchy of rates starting at 155.52 Mbps. The rates for SDH are listed below.

Optical Carrier (OC) and SDH Levels

STM Level

SDH Level

Rate

OC-1

-

51.85 Mbps

OC-3

STM-1

155.52 Mbps

OC-12

STM-4

622.08 Mbps

OC-48

STM-16

2.488 Gbps

OC-192

STM-64

9.952 Gbps

Asynchronous Transfer Mode (ATM) can be used to transmit network layer packets over a SONET/SDH network. ATM is use to establish connections and provide traffic management over the SONET network. Alternatively, Packet over SONET (PoS) can be used.

PoS adds less overhead from frame headers than ATM and maps IP directly onto SONET/SDH. PoS has three main components: a link-layer protocol, octet framing to map onto the SONET payload, and data scrambling for data security and reliability. For the link layer, the Point-to-Point protocol (PPP) over SONET/SDH is defined in RFC 2615. SONET/SDH links are provisioned as point-to-point circuits, making PPP a suitable choice for the link-layer protocol. PPP treats SONET/SDH as octet-oriented synchronous links. Octet-oriented framing (PPP with HDLC-like framing) is defined in RFC 1662. Data scrambling is defined in RFC 2615 and prevents packets with bit patterns that might cause synchronization problems, emulates the SDH set-reset scrambler pattern, and replicates the STS-n frame alignment word.

PoS specifies STS-3c/STM-1 (155 Mbps) as the basic data rate, with a usable data bandwidth of 149.760 Mbps.

6.1.4 Dynamic Packet Transport (DPT) and Spatial Reuse Protocol (SRP)

Dynamic Packet Transport (DPT) protocol was developed by Cisco to be a resilient optical packet ring technology that is optimized for data transmission. DPT uses dual, counter-rotating rings that are referred to as inner and outer, which can be used for data and control packet transmission concurrently. DPT operates by sending the data packets in one direction on one fiber ring and the corresponding control packets in the opposite direction on the other fiber ring.

In this way the full capacity of the fiber rings can be utilized for data and control traffic as it not necessary to reserve half of the capacity for redundancy.

OC-48c
The SONET/SDH level designations can include a "c" suffix as in OC-48c. The "c" indicates a "concatenated" channel. This means that the entire payload rate is available for use by a single flow of cells or packets.

DPT uses the Spatial Reuse Protocol (SRP) MAC layer protocol was designed to be scalable and provide optimized IP packet aggregation and transport in LANs, MANs, and WANs. It can support up to 128 nodes running at high speeds (OC-48c/STM-16c and OC-192c/STM-64c). The fairness algorithm and packet priority ensures IP packets with bounded end-to-end delay requirements are delivered successfully.

SRP uses destination stripping to increase bandwidth capacity. With SRP, the packet travels on the ring from the source to the destination, and the destination strips the packet, which frees up bandwidth on other segments of the ring for use by other stations.