Traffic Channel for Adaptive Multi-Rate Full Rate Speech

Description

The TCH/AFS (Traffic Channel employing AMR Full rate Speech) is a logical channel in GSM that utilizes the Adaptive Multi-Rate (AMR) speech codec in its full-rate mode. Unlike the fixed-rate TCH-FS, the TCH/AFS incorporates a fundamental advancement: adaptability. The AMR codec defines multiple speech codec modes (e.g., 12.2, 10.2, 7.95, 7.40, 6.70, 5.90, 5.15, 4.75 kbps) and multiple channel coding schemes. The TCH/AFS channel operates by dynamically selecting the most appropriate combination of speech bit rate and channel coding protection based on real-time measurements of the radio link quality. This selection, known as Link Adaptation or Codec Mode Adaptation, is performed by the Base Station Controller (BSC).

Technically, the channel occupies one full physical timeslot per TDMA frame, similar to TCH-FS. However, the payload within that timeslot is variable. When radio conditions are good, the system selects a high speech bitrate mode (like 12.2 kbps) with less robust channel coding, maximizing voice quality. When conditions degrade (e.g., low signal, high interference), the system switches to a lower speech bitrate mode, freeing up bits within the same gross bit budget to be used for more powerful Forward Error Correction (FEC), thereby increasing error robustness and maintaining call continuity at a slightly lower audio quality. This adaptation can occur periodically during a call. The channel structure still follows a multiframe, carrying speech data and associated SACCH signaling.

Architecturally, the TCH/AFS requires support in the mobile station, the BTS, the BSC, and the TRAU. The BSC acts as the adaptation controller, using measurement reports from the mobile and BTS to command codec mode changes. The TRAU must be capable of transcoding to/from the various AMR rates. The role of TCH/AFS is to provide a superior user experience and more efficient network operation compared to fixed-rate channels. It optimizes the fundamental trade-off in wireless communications: between source coding (speech quality) and channel coding (error protection). By adapting to conditions, it provides the best possible voice quality the radio link can support at any given moment, reduces the likelihood of dropped calls in poor coverage, and can even improve overall network capacity by allowing calls to persist in conditions where a fixed-rate channel would fail.

Purpose & Motivation

The TCH/AFS was created to overcome the limitations of fixed-rate speech channels (TCH-FS and TCH-HS) which could not adapt to varying radio conditions. The primary problem it solved was the all-or-nothing nature of earlier codecs: in poor conditions, a fixed full-rate call would suffer severe quality degradation or drop, while a fixed half-rate call might offer poor quality even when the radio link was excellent. The AMR technology, and by extension the TCH/AFS channel, introduced intelligence to this process.

The historical motivation was the pursuit of enhanced voice quality and network robustness as GSM networks matured and users expected more consistent service. 3GPP standardized AMR to provide a future-proof, adaptable voice codec. The TCH/AFS channel implementation allowed this codec to be deployed within the existing GSM channel structure. It addressed the key limitation of static resource allocation by making the voice service resilient and quality-aware. This was particularly important for improving service in cell edge areas and indoors, directly impacting customer satisfaction and network performance metrics. It represented a shift from network-centric, fixed resource planning to a more user-centric, adaptive service delivery model for voice.

Key Features

• Employs the Adaptive Multi-Rate (AMR) speech codec in full-rate configuration
• Dynamic Link Adaptation between multiple speech and channel coding rates
• Optimizes voice quality in good conditions and call robustness in poor conditions
• Occupies one physical timeslot, with adaptive payload content
• Codec Mode Adaptation is controlled by the BSC based on radio measurements
• Provides backward compatibility and interworking with legacy fixed-rate channels

Evolution Across Releases

Defining Specifications