Adaptive multiple rate (AMR) speech codec design incorporates multiple codecs for use in full-rate or half-rate mode that are determined by channel quality. AMR operates in eight sub modes that incorporate bit exact versions of both 12.2 kbps US1/GSM EFR and 7.4 kbps IS-641 full-rate speech coders. The AMR can increase voice capacity by 150% compared to the GSM.
The AMR codec allows dynamic management of voice quality and error control to provide good voice quality even under adverse radio conditions. AMR is not only used in GSM, but also in EDGE and UMTS networks It is designed to work with both GSM full-rate (one user per each of eight time slots in each radio channel) and GSM half-rate (two users per time slot).
AMR defines multiple voice encoding rates, each with a different level of error control . The AMR codec dynamically responds to radio conditions by using the most effective mode of operation at each moment of time. Compared to the GSM EFR codec, AMR can operate under much worse radio conditions.
AMR offers the following benefits:
1) Greater spectral efficiency, hence higher capacity from higher Frequency Reuse with frequency hopping.
2) Better voice quality throughout the cell, particularly at cell edges and deep inside buildings, and increased overall coverage.
3) The potential of operating with toll-quality voice in half- rate mode, which reduces network costs as with a heavily loaded network.