March 1999

AAL2: Voice Over ATM Builds On New Standard

BY RICH STANKEVICH

Attention all network planners - from developers to systems integrators, from VARs to interconnects: The advent of a new scheme to map voice and other bursty real-time traffic (i.e., video) onto ATM cells means the flexibility and bandwidth savings promised by voice over ATM now can be realized. This ability to handle voice was a prime motivation for creating many wide-area ATM nets, and all network planners should make themselves familiar with the new ATM Adaptation Layer 2 (AAL2) standard.

AAL2's Variable Bit Rate (VBR) service handles voice over ATM far more efficiently than the Constant Bit Rate (CBR) service of AAL1, with its inefficiently utilized, permanently allocated bandwidth. AAL2 delivers the standards-based capability to aggregate voice traffic over ATM without prior restrictions of PBX point-to-point. Before AAL2, users wanting to implement voice over ATM had to live with AAL1's limitations or adopt a proprietary solution, which meant increasing network efficiency but negating interoperability. The new standard means ATM switches can extend the benefits of ATM's statistical gain to real-time voice.

Access connections using this new scheme can transport voice circuits over the same facilities as data circuits, minimizing the use of precious bandwidth. Furthermore, adopting voice circuits into ATM at a customer premise allows for a more efficient transport to voice switches. This lowers the investment in expensive resources, allowing carriers to distribute their voice switches more widely in the network. First, however, we must be clear about when the networking muscle of AAL2 should be flexed.

WHEN TO USE AAL2
Because AAL2 offers significant bandwidth-efficiency benefits over AAL1, it should be used when all ATM nodes are AAL2-capable. If a node cannot support AAL2, configuring AAL1 connections for voice allows transport of voice through the non-AAL2 portion of the network. In addition, some applications are adequately supported with AAL1 and, in fact, cannot benefit from using AAL2.

Take, for example, a network where user access lines are traditional voice lines connecting with a voice end-office, but where the network's primary purpose is to provide Internet access. In this case, the ATM network is not part of the voice network, but rather provides interconnection from the Class 5 voice switch to the ISP. The ATM network traffic is all modem data, which is adequately served with AAL1, and which cannot take advantage of the statistical gain associated with AAL2 voice traffic.

WHAT AAL2 OFFERS
A new adaptation layer was required to provide the flexibility for network operators to control delay on voice services and to overcome the excessive bandwidth needed by using structured circuit emulation. AAL1 simply cannot be extended to meet these new ATM networking requirements. The new ATM Adaptation Layer, AAL2, as specified in ITU-T (International Telecommunication Union - www.itu.int) Recommendations I.363.2, I.366.1, and I.366.2, carries the specific mandate to provide efficient voice-over-ATM services.

AAL2 is designed to make use of the more statistically multiplexible VBR ATM traffic classes, providing bandwidth-efficient transmission of low-rate, short, and variable packets for delay-sensitive applications. This means it is not limited to ATM connections using the CBR traffic class. It also supports voice applications using higher layer requirements, such as voice compression, silence detection/suppression, and idle channel removal.

AAL2's structure lets network administrators take traffic variations into account in the design of an ATM network optimized to match traffic conditions. AAL2 also enables multiple user channels on a single ATM virtual circuit and varying traffic conditions for each individual user or channel. Its structure also provides for both the packing of short length packets into one (or more) ATM cell(s) and the mechanisms to recover from transmission errors.

Compared to AAL1 and its fixed payload, AAL2 handles a variable payload within cells and across cells. This provides a dramatic improvement in bandwidth efficiency over either structured or unstructured circuit emulation using AAL1.

HOW IT WORKS
AAL2 structure, as defined in ITU-T Recommendation I.363.2, is divided into two sub-layers: the Common Part Sub-layer (CPS) and the Service Specific Convergence Sub-layer (SSCS). Following is an overview of both layers:

CPS: Fully defined in the ITU I.363.2, the CPS provides the basic structure for identifying the AAL users, assembling/disassembling the variable payload associated with individual users, error correction, and the relationship with the SSCS. Each AAL2 user can select a given AAL-SAP (service access point) associated with the Quality of Service (QoS) required to transport that individual higher layer application. AAL2 makes use of the service provided by the underlying ATM layer. Multiple AAL connections can be associated with a single ATM layer connection, allowing multiplexing at the AAL layer. The AAL2 user selects the QoS provided by AAL2 through the choice of the AAL-SAP used for data transfer.

AAL2's CPS has the following characteristics:

• It is defined on an end-to-end basis as a concentration of AAL2 channels.
• Each AAL2 channel is a bi-directional virtual channel, with the same channel identifier value used for both directions.
• AAL2 channels are established over an ATM layer Permanent Virtual Circuit (PVC), Soft Permanent Virtual Circuit (SPVC), or Switched Virtual Circuit (SVC).

The multiplexing function in the CPS merges several streams of CPS packets (individual voice circuits) onto a single ATM connection.

SSCS: In ITU-T Recommendation I.363.2, the SSCS is defined as the link between the AAL2 CPS and the higher layer applications of the individual AAL2 users. Two SSCS Recommendations have been developed in the ITU-T: Recommendations I.366.1 and I.366.2.

• I.366.1 (6/98) Segmentation and Reassembly Service Specific Convergence Sublayer for the AAL Type 2 provides an SSCS for AAL2 to handle segmentation and reassembly for data users. Options in I.366.1 allow a basic function similar to AAL5, error protection, and an assured mode of operation for guaranteed operation when used for signaling.
• I.366.2 (2/99) AAL Type 2 Service Specific Convergence Sublayer for Trunking provides a rich set of features for voice users of AAL2. Features in I.366.2 include compressed voice, silence indication, alarm handling, channel associated signaling, dialed digits, and fax demodulation/remodulation. This SSCS also includes formatting voice as 64K PCM, which is a mandatory feature to insure interoperability between networks that may be using AAL1.

In a development related to the ITU-T Recommendations, the ATM Forum (www.atmforum.com) is nearing completion of a specification titled "ATM Trunking Using AAL2 for Narrowband Services." This ATM Forum specification incorporates the ITU-T AAL2 Recommendations by reference to I.363.2, I.366.1, and I.366.2.

MEASURING EFFICIENCY
Because of the complexity of dealing with both fixed and statistical compression in a voice channel (e.g., 32K ADPCM or 8K CS-ACELP and silence suppression), we need a more application-oriented way to appreciate the efficiency of an AAL2 connection. One way to do this is to identify how many voice channels may be carried over a fixed bandwidth ATM trunk between ATM network elements. Optioning all channels for minimum bandwidth (8K CS-ACELP), with trunks designed for 80 percent "off-hook" at busy hour and 50 percent silence content in active voice calls, yields an effective compression ratio of approximately 12:1.

CHOOSING ATM
ATM networks are arguably emerging as the only viable underlying technology for efficient wide-area multiservice networking. Consider that when these voice channels are not busy (for example, overnight), the bandwidth that would otherwise be used for voice is efficiently made available for other applications (that is, remote server archiving or software downloading).

The most important benefit of AAL2 is the substantial reduction of bandwidth needs for supporting voice on ATM networks, driving down network costs. A further benefit is obtained by the inherent flexibility to add features via the SSCS sub-layer structure - different fixed compression techniques and voice channel switching between ATM circuits are examples. The ability to add dynamic interconnection between ATM network elements (switches/concentrators) and existing telephony devices (like PBXs for setup and teardown of ATM voice connections) provides even greater cost/bandwidth efficiencies and flexibility.

In carriers' networks and within corporate enterprises around the world, the promise of voice over ATM is rapidly being realized because the multiservice strengths of ATM WANs now can showcase both technological elegance and a compelling business case.

Rich Stankevich is the product marketing manager, ATM marketing, for General DataComm in Middlebury, CT. Comments or questions should be forwarded to Rich at [email protected]. GDC is a leader in the design, development, and manufacture of multiservice communications systems for service providers and enterprise businesses. For more information, please contact the company at 203-574-1118, or visit their Web site at www.gdc.com.