TETRA数字集群系统语音编解码算法研究

研究了陆地集群无线通信（TETRA）数字集群移动通信系统的语音压缩编解码算法的基本原理。TETRA语音压缩编解码器使用一种低复杂度的ACELP（代数码激励线性预测）算法，通过线性预测和磁量量化等方法降低比特率，并通过自适应码本和代数码本相结合的方法降低了CELP声码器的复杂度，兼顾了较好的语音质量和较低的比特率。在微机上通过C评议和Matlab对TETRA的语音编解码算法进行了实时仿真，语音信号由微机声卡录入和播放。仿真结果表明，该算法具有速率低，重建语音可懂度高，计算复杂度低等优点，适于在移动通信中应用。

Algorithm for TETRA speech encoding and decoding

This paper describes the basic theory of speech compression encoding and decoding algorithm in TETRA (terrestrial trunked radio) system. TETRA speech encoding and decoding uses ACELP algorithm with low complexity, which linear prediction and vector quantization are used to lower the bit rate and two codebooks (adaptive codebook and algebraic codebook) are used to reduce the complexity of CELP vocoder. By means of that a good speech quality and low bit rate are gotten. ANSI C Language and Matlab tools are used to simulate the algorithm on a PC in real time. The speech is recorded and played by the PC sound card. The simulation result shows that the TETRA speech encoding and decoding algorithm has generally good speech quality, low complexity and low bit rate. The reconstructed speech is well understood. The algorithm is suitable for mobile communications.