Mobile Communication: Source Coding, Speech Processing, and Channel Coding, Slides of Mobile Communication Systems

An in-depth exploration of source coding and speech processing in mobile communication, including source coding techniques like rpe-ltp, speech compression, and channel coding methods such as block coding and convolution coding. The lecture covers error protection mechanisms like interleaving and the importance of speech traffic channel coding.

Typology: Slides

2011/2012

Uploaded on 07/07/2012

luucky
luucky 🇮🇳

4.5

(2)

86 documents

1 / 12

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
Mobile Communication
Physical Layer-04
Lecture by
docsity.com
pf3
pf4
pf5
pf8
pf9
pfa

Partial preview of the text

Download Mobile Communication: Source Coding, Speech Processing, and Channel Coding and more Slides Mobile Communication Systems in PDF only on Docsity!

Mobile Communication

Physical Layer-

Lecture by

TODAY’S SESSION

Source coding and speech processing Source coding Channel coding Convolution coding interleaving

SOURCE CODING AND SPEECH PROCESSING

Speech compression Process

  • Regular Pulse Excitation Long-Term Prediction (RPE-LTP)
  • Partial speech signal is encoded according to wave encoding
  • Rest of the speech signal is transmitted as parameters
  • Receiver reconstructs the signal through speech synthesis
  • Wave encoding example : PCM

SIMPLIFIED RPE-LTP

LPC-Filter Parameters LTP – Filer Parameter RPE Signal

  • 160 samples per 20 ms are received at the coder,
  • Speech signal decomposed into three parts
  • 36 parameters are generated by the filters,
  • RPE signal is compressed by removing redundancy
  • 160 samples are compressed to 260 bits
  • Data rate: 260 bits per 20 ms, or 13 kbps

CHANNEL CODING

Block coding

  • Input data blocks are added with parity and extra zeros to achieve a suitable size for the next step i.e. convolution coding

Convolution coding Convolution codes add redundancy to the original data, A puncturing code can also be used , that deletes some of the extra bits added by convolution code Convolution codes are used for error detection

SPEECH TRAFFIC CHANNEL CODING

Input 260 bits of speech data (full rate) 120 bits of speech data (half rate) The speech input data is divided into two classes Class-I 182 bits , these bits are of high impact on speech need better protection and are encoded differently 95 bits for half rate Class-II 78 bits are of less importance and are not protected 17 bits for half rate

BLOCK CODING FOR LOGICAL CHANNEL

INTERLEAVING