Description
TDMA-US1 is a speech codec specified for TDMA-based cellular systems, operating at a rate of 12.2 kbit/s. It is defined in 3GPP specifications as a codec with technical characteristics similar to the GSM Enhanced Full Rate (GSM-EFR) codec. The 'US1' designation typically refers to its application in the United States market for systems like IS-136. Like TDMA-EFR and GSM-EFR, it is based on the ACELP (Algebraic Code Excited Linear Prediction) coding technique.
The codec's operation involves processing speech in segments (frames). It analyzes each frame to determine parameters for a linear predictive synthesis filter and an excitation signal. The excitation is constructed from entries in an algebraic codebook (representing innovation) and an adaptive codebook (representing pitch periodicity). These parameters are quantized and encoded into a 244-bit block for every 20 ms frame, resulting in the 12.2 kbit/s rate. This relatively high bit rate, compared to other cellular codecs, allows for less aggressive compression, preserving more of the original speech signal's characteristics and yielding very high subjective quality.
In the network architecture, TDMA-US1 functions within the speech processing chain. It is implemented in the terminal's audio codec and in the network's Transcoder and Rate Adaptation Unit (TRAU) or Media Gateway. The encoded bitstream is packetized and transported over the allocated traffic channel on the TDMA air interface. Specification 3GPP TS 26.093 provides the detailed algorithmic description and test sequences for this codec, ensuring that different network elements and handsets can interoperate while delivering consistent, high-fidelity voice service to the end user.
Purpose & Motivation
TDMA-US1 was developed to provide a high-bit-rate, premium quality speech codec option for TDMA (IS-136) networks, particularly in North America. The context was a competitive landscape where GSM networks globally were deploying the high-quality GSM-EFR codec. TDMA operators needed a comparable offering to ensure their voice service quality was not perceived as inferior.
The problem it addressed was the limitation of earlier TDMA codecs which either had lower quality (full-rate) or required more complex variable-rate operation. TDMA-US1 offered a straightforward, high-quality fixed-rate alternative. Its creation was motivated by the desire for standardization and quality harmonization. By aligning technically with the widely adopted GSM-EFR, it allowed for potential economies of scale in chipset development and testing, and ensured that TDMA subscribers could experience voice clarity on par with users of other leading digital technologies. It served as a quality benchmark for TDMA voice services before the widespread adoption of more advanced, adaptive multi-rate codecs.
Key Features
- Fixed-rate speech codec operating at 12.2 kbit/s
- Based on the ACELP coding algorithm, similar to GSM-EFR
- Processes speech in 20 ms frames, producing 244 bits per frame
- Designed to deliver high-fidelity, near-toll-quality voice
- Standardized for interoperability in TDMA (IS-136) systems
- Includes comprehensive test sequences for implementation validation
Evolution Across Releases
Initially standardized within 3GPP as a reference speech codec. The specification defined the complete algorithmic model, bit-exact fixed-point implementation, and conformance test data for the TDMA-US1 codec, establishing it as a benchmark for high-quality voice in TDMA systems.
Defining Specifications
| Specification | Title |
|---|---|
| TS 26.093 | 3GPP TS 26.093 |