IMS  

 

 

 

EVS

EVS stands for Enhanced Voice Service and is defined in 3GPP 26.445. As you can guess from the terminology itself, this is a kind of 'Voice Service' (like VoLTE) but 'Enhanced'.

Yes, I know you have a question right now. What do you mean by 'Enhanced' in this context ? It is hard to say in single words or a single short sentense. The single sentense would be "It is Enhanced in various aspects comparing to the existing Voice service.

Now, you would ask 'in what aspects is it enhanced comparing to the exisiting (legacy ?) Voice Service' ?  I think we can roughly list up as follows.

  • Enhanced Sampling Rate
  • Enhanced Code Algorithm (Codec Rate)
  • Improved (shorter) Codec delay

EVS is an adaptive codec, which means it can dynamically adjust its bit rate and other parameters based on network conditions and other factors. It supports a wide range of bit rates, ranging from 5.9 kbps to 128 kbps, and can operate at sampling rates of 8 kHz and 16 kHz. This allows it to adapt to a wide range of network conditions and provide high-quality voice communication even in challenging environments.

EVS uses a variety of advanced techniques to improve voice quality and reduce background noise, including:

  • Linear Prediction (LP): This technique is used to predict future speech samples based on past samples, which helps to reduce the amount of data that needs to be transmitted over the network.
  • Code Excited Linear Prediction (CELP): This technique is used to encode speech samples in a compact and efficient manner, while still preserving high levels of speech quality.
  • Wideband Spectral Enhancement (WSE): This technique is used to improve voice quality in challenging network conditions by enhancing the spectral characteristics of the speech signal.
  • Noise Reduction: EVS includes advanced noise reduction techniques to reduce background noise and improve speech quality, even in noisy environments.

Examples of EVS configuration in SDP : Following is an example showing how EVS is configured in SDP.

    SDP Offer :

      v=0

      o=- 0 0 IN IP4 192.168.1.1

      s=Example Session

      c=IN IP4 192.168.1.1

      t=0 0

      m=audio 5000 RTP/AVP 114 115

      a=rtpmap:114 EVS/16000/1

      a=fmtp:114 mode=adaptive;bandwidth=adaptive

      a=rtpmap:115 AMR-WB/16000/1

      a=fmtp:115 mode-set=0,1,2,3,4,5,6,7; mode-change-period=2

     

    SDP Answer :

      v=0

      o=- 0 0 IN IP4 192.168.1.2

      s=Example Session

      c=IN IP4 192.168.1.2

      t=0 0

      m=audio 6000 RTP/AVP 114 115

      a=rtpmap:114 EVS/16000/1

      a=fmtp:114 mode=adaptive;bandwidth=adaptive

      a=rtpmap:115 AMR-WB/16000/1

      a=fmtp:115 mode-set=0,1,2,3,4,5,6,7; mode-change-period=2

       

The details of each of these aspects is pretty long and complicated story. It is story of over 600 pages of 3GPP documents. I will keep updating this page as I am getting further details.

If you are working on this right now and have to implement SIP signaling part for this, 26.445-A.3 Payload Format Parameters can be a good starting point and extend your reading to understand the meaning of each parameters mentioned in 26.445-A.3.

Reference :

[1] The 3GPP Enhanced Voice Services (EVS) codec - Nokia Whitepaper