Networking Unit 6: X.25, Frame Relay and ATM

Packet switching groups transmitted data into packets, independent of content or structure.
Common packet switching protocols include Frame Relay, IP, and X.25.
Packets contain address information for routers to route them to destinations.

A standard suite of protocols for packet switching covering Layers 1 to 3 of the OSI model.
X.25 packets can carry up to 128 bytes of data.
Features include packet assembly, delivery, error checking, and retransmission, making it suitable for noisy networks.
Multiconversation support through multiplexing and virtual communication channels.
Original use was for voice over analog telephone lines, now seen in ATMs and credit card networks.
As corporate networks adopt IP, many X.25 functions are being shifted to cheaper technologies like Ethernet.

Layer 1 (Physical Layer): Manages electrical signaling; includes standards like V.35, RS232.
Layer 2 (Data Link Layer): Implements HDLC standard (LAPB) ensuring an error-free link; results in relatively high latency due to error correction.
Layer 3 (Network Layer): Handles end-to-end communication, connection setup, and flow control; supports two virtual connection types:
- Switched Virtual Circuits (SVCs): Temporary connections similar to phone calls.
- Permanent Virtual Circuits (PVCs): Always-on connections resembling leased lines.

Default maximum packet size is typically 128 bytes; ranges from 64 to 4096 bytes.
Optimized for low-speed lines (under 100kbps); inefficiency arises at higher speeds due to latency.
Basis for modern packet-switched protocols like TCP/IP and ATM.

A standardized WAN technology using variable-size packets (frames) for cost-efficient data transmission.
Primarily designed for intermittent traffic between LANs and WAN endpoints.
Relies on a permanent virtual circuit (PVC) for continuous connections without full leased line costs.
Faster than X.25 as it operates over low-error rate lines and offloads error correction to endpoints.
Frame relay protocol uses a control plane for connection management and a user plane for data transfer.

Control Plane: Establishes and terminates logical connections, ensuring reliable data link control using LAPD at the data link layer.
User Plane: Transfers user data and includes functionalities like frame delimiting, error detection, and congestion control.

Flag Field: Denotes start and end of the frame using the 01111110 pattern with bit stuffing.
Address Field: Identifies the virtual connection with Data Link Connection Identifiers (DLCI).
Information Field: Maximum payload often negotiated, supporting up to 1600 octets.
Frame Check Sequence (FCS): Utilizes cyclic redundancy check (CRC) for error detection.

A cell relay protocol supporting various traffic types (voice, data, video) through fixed-sized packets (cells).
Employs a connection-oriented model requiring the establishment of virtual circuits prior to data transfer.
Scalable architecture; upgrades can be made easily by adding switches or links.
Underpins the notion of LAN Emulation (LANE) to integrate legacy networking technologies.

Physical Layer: Manages transmission mediums and data stream conversion.
ATM Layer: Responsible for cell multiplexing, transport, and virtual circuit management.
ATM Adaptation Layer (AAL): Prepares user data for cell conversion, segments data, and supports various traffic types through defined protocols.

AAL Type 0: Raw cells without specific fields.
AAL Type 1: Supports constant bit rate, synchronous traffic.
AAL Type 2: Handles time-dependent variable bit rate traffic, suitable for voice.
AAL Type 3/4: For connection-oriented asynchronous and connectionless data; supports additional headers.
AAL Type 5: Similar to AAL 3/4 with a streamlined header format.### AAL 5 Overview
AAL 5 is an ATM adaptation layer designed for connection-oriented higher-layer protocols.
Intended to streamline transport by reducing protocol processing and transmission overhead.
Accommodates variable bit rate and supports both connection-oriented and connectionless data.
Does not require segment tracking or error correction, which simplifies processing compared to AAL 3/4.
Uses the Payload Type Indicator (PTI) bit to signal the last cell in a transmission.
Services employing AAL 5 include classic IP over ATM, Ethernet over ATM, SMDS, and LAN Emulation (LANE).

An ATM cell consists of a 5-byte header and a 48-byte payload.
Two cell formats defined: UNI (User-Network Interface) and NNI (Network-Network Interface), with UNI being most common.
Header fields include:
- GFC (Generic Flow Control): 4 bits, default is four zero bits.
- VPI (Virtual Path Identifier): 8 bits for UNI, 12 bits for NNI.
- VCI (Virtual Channel Identifier): 16 bits.
- PT (Payload Type): 3 bits, used for cell type designation.
- CLP (Cell Loss Priority): 1 bit, indicates cell loss importance.
- HEC (Header Error Control): 8-bit CRC used for single and multi-bit error detection.

HEC employs a polynomial-based CRC algorithm, facilitating error correction and detection.
Single-bit header errors can be corrected, while multi-bit errors prompt dropping of errored cells until a valid one is found.

In UNI cell format, GFC enables local flow control/system submultiplexing, allowing multiple users on one network connection.
On NNI, the 4-bit GFC is reallocated to extend VPI to 12 bits, allowing for a large addressing capability (near 2^12 VPs and almost 2^16 VCs).