Amr

Introduction to AMR

Place your image on top of this gray box. If no graphic is applicable, delete gray box and notch-out behind gray box, from the Title Master 

Jonathan Hopkinson 20/8/03 Motorola Confidential Proprietary MOTOROLA and the Stylized M Logo are registered in the US Patent & Trademark Office. All other  product or service names are the property of their respective owners. © Motorola, Inc. 2002.

Existing types of Speech Coding • Currently there are two speech coding schemes used in GSM.  – FR - Full Rate (The original coding scheme)  – EFR - Enhanced Full Rate ( An enhancement of FR)

• EFR offers better quality speech. • Most modern mobiles are EFR capable. • The majority of Network Operators have EFR capable Networks. • Motorola’s BSS has supported EFR since GSR3 - with a GDP board fitted at the Transcoder function.

Why change from what we’ve got ? •The traditional “fixed” coding schemes (FR or EFR) use a fixed amount of bandwidth - some dedicated to encode the speech and the rest used to offer data protection - Forward Error Coding, Checksums etc. •When poor RF conditions exist, there may be “errors” in the speech data which cannot be corrected or recovered. The “errors” will be heard as an interruption in the speech. •With EFR or FR, we can only achieve a “best effort” with the coding scheme and protection available - AMR offers some improvements.

AMR Concept •  AMR allows multiple different coding schemes to be used. The higher speech bandwidth schemes offer  good speech quality - with lower error protection. The lower speech bandwidth schemes offer slightly lower  speech quality - with much higher error protection.

4.75 kbps

5.15 kbps

5.9 kbps

6.7 kbps

7.4 kbps

7.95 kbps

10.2 kbps

12.2 kbps

13 kbps

EFR  FR  Better Protection Better Speech Quality

Mos Scores Vs C/I (Static)

AMR Codec Rates • Up to 4 codec modes are allowed in AMR at any one time - This is known as the Active Codec set. • The Active codec set may contain something like :12.2, 7.95, 5.9 and 4.75 kbps. •  AMR switches between the 4 rates - independently on the Uplink and the Downlink - based on the prevailing RF conditions on each link. • Under good RF conditions - a Rate of 12.2 kbps may be used which is similar to EFR. Under poor  conditions 4.75 kbps may be used - offering lower  quality speech - but higher protection against errors.

AMR Codec Rate Switching •  AMR will switch modes within the active codec set, based on C/I and RxQual calculations. C/I 12.2 kbps Hysterisis 7.95 kbps Hysterisis 5.9 kbps Hysterisis 4.75 kbps

Improvements offered by AMR • Improved overall speech quality due to the adaptive algorithm switching to more robust coding schemes when RF conditions are poor. • Possibility of “Cell extension” - Because AMR offers a more robust link - calls with adequate speech quality may be possible over a longer range • Possibility to use AMR as an input to HDPC - for  example to reduce MS/BTS power based on Speech Quality - as well as traditional RxLev and RxQual.

Half Rate AMR •  AMR allows for Half Rate speech channels. • Half Rate can mean twice as many subscribers using the same amount of infrastructure (RF) equipment. This is a very attractive proposition to Network operators. • The Half Rate codec modes are only suitable when the C/I is ~> 15dB. • Half Rate AMR will be possible on Motorola infrastructure - new hardware is required at the BSC and Transcoding functions. • Half Rate offers worse speech quality - but more traffic capacity. - Call tariffs will probably sell Half Rate.