2G GSM Audio Codec and Vocoder: Types, Working, AMR, AMR-WB and Handover Explained

Vocoder / Codec Basics

Vocoder (voice coder) or speech codecs are widely used in voice communication systems. In GSM, audio codecs are used to compress speech signals so they can be transmitted efficiently over limited bandwidth channels. If speech were digitized directly without compression, it would require a very high data rate and wide bandwidth.

Since communication channels have limited capacity, speech data must be compressed before transmission and then reconstructed at the receiver. Modern audio codecs often use Linear Prediction techniques, which model the human vocal tract mathematically. These methods estimate the signal and provide high compression while maintaining acceptable audio quality.

Codec Techniques Used in GSM

CELP (Code Excited Linear Prediction)

CELP is one of the most important speech coding techniques. It was introduced in 1985 and significantly improved voice quality compared to earlier codecs. It uses a method called Analysis-by-Synthesis, where the encoder selects the best possible signal representation by comparing multiple generated signals and choosing the most accurate one. CELP forms the basis for several other codecs such as ACELP and VSELP.

ACELP Codec

ACELP (Algebraic Code Excited Linear Prediction) is an advanced version of CELP. It uses algebraic codebooks to improve efficiency and voice quality.

VSELP Codec

VSELP (Vector Sum Excitation Linear Prediction) was used in early GSM systems. However, it performs poorly in noisy environments and is now largely replaced by modern codecs.

GSM Audio Codecs Overview

Codec Name Bit Rate (kbps) Technology
Full Rate 13 RPE-LPC
EFR 12.2 ACELP
Half Rate 5.6 VSELP
AMR 12.2 – 4.75 ACELP
AMR-WB 23.85 – 6.60 ACELP

GSM Full Rate (RPE-LPC) Codec

The RPE-LPC (Regular Pulse Excited Linear Predictive Coding) was the first codec used in GSM systems. It balanced complexity and performance using LPC with pulse excitation. However, over time, its voice quality was considered limited, leading to the development of improved codecs.

GSM Enhanced Full Rate (EFR) Codec

The Enhanced Full Rate codec was introduced to improve voice quality. It uses ACELP technology and provides better audio clarity compared to the original full-rate codec.

GSM Half Rate Codec

The half-rate codec allows a single channel to carry two voice calls, effectively doubling network capacity. It operates at 5.6 kbps using the VSELP algorithm. Although efficient, it offers lower voice quality and is mainly used during high network congestion.

AMR (Adaptive Multi-Rate) Codec

AMR is the most widely used GSM codec. It supports multiple bit rates and adapts dynamically based on network conditions. Key features of AMR include:

Mode Bit Rate (kbps) Type
AMR 12.2 12.2 Full Rate
AMR 7.95 7.95 FR/HR
AMR 4.75 4.75 FR/HR

AMR-WB (Wideband) Codec

AMR-WB (Adaptive Multi-Rate Wideband) provides improved speech quality by supporting a wider frequency range (50 Hz to 7000 Hz). It operates at a sampling rate of 16 kHz and uses ACELP-based algorithms. This results in clearer and more natural sound compared to standard codecs.

Bit Rate (kbps) Description
6.60 Used during poor network conditions
12.65 Main standard rate
23.85 High-quality audio

GSM Handover

Handover (or handoff) is the process of transferring an ongoing call from one cell or base station to another without interruption.

Types of GSM Handover

Handover Process

In GSM systems, mobiles continuously monitor signal strength and report it to the network. Based on this data, the network decides when to perform a handover. This process is called Mobile Assisted Handover (MAHO). The network assigns a new channel and time slot to the mobile, ensuring seamless communication during movement.

Inter-System Handover

Inter-system handover occurs when switching between different technologies such as GSM and UMTS (3G). Types include:

These handovers are more complex as they involve different network technologies and protocols.

Conclusion

GSM audio codecs and vocoders play a vital role in efficient voice communication by compressing speech data while maintaining quality. Over time, technologies have evolved from basic codecs like RPE-LPC to advanced systems like AMR and AMR-WB, significantly improving voice clarity and network performance.