Service Specific Segmentation and Re-assembly sublayer

Description

The Service Specific Segmentation and Reassembly (SSSAR) sublayer is a protocol entity defined in the 3GPP UMTS standards for the Circuit-Switched (CS) domain. It operates within the Iu interface user plane protocol stack, specifically between the Radio Access Network (RAN) and the Core Network (CN). The Iu interface connects the Radio Network Controller (RNC) to the Mobile Switching Center (MSC) for CS services. The primary role of the SSSAR is to adapt the variable-size or fixed-size Service Data Units (SDUs) received from the higher layer (typically the Radio Link Control or RLC layer) into a format suitable for transport over the underlying Asynchronous Transfer Mode (ATM) network using the ATM Adaptation Layer type 2 (AAL2).

Architecturally, the SSSAR sublayer resides above the AAL2 layer and below the user plane protocols for the specific service (e.g., voice codec data stream). It provides a service to its upper layer, which is the transparent transfer of SDUs. Its key operation involves two functions: segmentation and reassembly. On the transmitting side (e.g., in the RNC), the SSSAR sublayer takes an SDU, which could be a speech frame from an AMR codec, and segments it into smaller, fixed-size packets called SSSAR-PDUs (Protocol Data Units) if the SDU is larger than the maximum payload size allowed by the AAL2 Common Part Sublayer (CPS). Each SSSAR-PDU is then prepended with a SSSAR header. This header contains control information such as sequence numbers for in-order delivery and flags to indicate the beginning and end of the original SDU. These SSSAR-PDUs are then passed to the AAL2 layer for further processing and transmission.

On the receiving side (e.g., in the MSC), the process is reversed. The SSSAR sublayer receives the SSSAR-PDUs from the AAL2 layer. Using the sequence numbers and segmentation flags in the headers, it reassembles the original SDU. It must handle potential out-of-order arrival and loss of PDUs, although the underlying AAL2 and ATM networks typically provide a sequenced, reliable connection for CS traffic. Once reassembled, the complete SDU is delivered to the upper-layer service application. This process is 'service specific' because it can be configured with parameters (like maximum PDU size) tailored to the requirements of the specific CS service being transported, such as voice or circuit-switched data.

Purpose & Motivation

The SSSAR sublayer was created to address the fundamental mismatch between the data units generated by circuit-switched services (like voice frames) and the transport mechanism defined for the Iu interface in early UMTS releases: ATM/AAL2. ATM networks are designed to transport fixed-size cells (53-byte ATM cells). AAL2 was introduced to efficiently pack variable-length, delay-sensitive packets (like voice) into these ATM cells to minimize packetization delay and utilize bandwidth efficiently.

Without a segmentation layer, a complete voice frame (e.g., a 244-bit AMR frame) would need to be transported as a single AAL2 packet. However, AAL2 has limitations on packet size and multiplexing many small packets from different users into a single ATM cell is more efficient. SSSAR allows for the optional segmentation of larger upper-layer SDUs into smaller units that can be optimally packed into AAL2 packets. This increases bandwidth utilization on the expensive backhaul links between the RNC and MSC. It also provides a standardized mechanism for the receiver to reconstruct the original data stream reliably.

Before standardized segmentation in the user plane, inefficient transport or proprietary solutions could have been used. The SSSAR provided a 3GPP-standardized method, ensuring interoperability between equipment from different vendors in the UMTS network. It solved the problem of adapting the bursty, variable-rate nature of coded voice traffic to the connection-oriented, cell-based ATM transport, which was the dominant technology for core network interfaces at the time of UMTS R99 design.

Key Features

• Segments upper-layer SDUs into smaller SSSAR-PDUs for efficient AAL2 transport.
• Reassembles SSSAR-PDUs back into the original SDUs at the receiver.
• Adds a control header to each PDU containing sequence numbers and segmentation flags.
• Provides in-order delivery and identification of SDU boundaries for the service data stream.
• Operates in the user plane of the Iu-CS interface between RNC and MSC.
• Configuration is service-specific, allowing optimization for different traffic types (e.g., voice).

Evolution Across Releases

Defining Specifications