An Error Correction Technique For The Wireless Channel

An Error Correction Technique For The Wireless Channel Using Speech Codec 2008. 02. 18 Youkyung Oh

System block diagram Input Speech 8 k. Hz Sampling and 16 bit Quantize CELP Encoder (Source Coding) 8 kbps 8. 8 kbps CRC Encoder Convolutional Code Interleaver Channel Output Speech Filtering CELP Decoder CRC Check and Error Frame Correction Convolutional Code Decoding Deinterleaver

Speech codec r Speech codec – Waveform coders • PCM ( Pulse Code Modulation ) – Vocoders • LPC ( Linear Predictive Coding) – Hybrid coders • • CELP ( Code Excited Linear Prediction ) QCELP ( Qualcomm Code Excited Linear Prediction ) EVRC ( Enhanced Variable Rate Coding ) CS-ACELP ( Conjugate-Structure Algebraic CELP ) r CS-ACELP – Encoder : encoder. exe Ex) C: encoder enc_input. bin enc_output. bin – Decoder : decoder. exe Ex) C: decoder dec_input. bin dec_output. bin

Cyclic Redundancy Check r CRC : Frame check sequence

Cyclic Redundancy Check r CRC encoder – General CRC circuit when divisor is – 80 bits(10 byte) + 8 bits – Notice that we should only encode important 40 bits r Error Correction with CRC – Do Nothing, Insert Silence, Parameter Repeat, Parameter Interpolation

Convolutional Code r Parameters – code rate = k/n, constraint Length = K – k input, n output with K registers K=3, code rate = 1/2 Input bit Previous state Output bits Next state 0 00 00 00 11 10 0 10 10 01 11 00 1 01 00 10 0 11 01 01 1 11 10 11

Convolutional Code r State Diagram & Trellis Diagram input 0 input 1 00 Output branch word a=00 11 11 00 b=10 c=01 Encoder State 10 01 d=11 10 01 Input bit 0 Input bit 1

Convolutional Code r Decoder Trellis diagram (1/2, K=3)

Convolutional Code r Viterbi Decoding – Select a survivor at time 4

Interleaver r 4 X 6 Block Interleaver

Channel r Binary Symmetry Model 0 – P(0|1)=P(1|0)=p – P(0|0)=P(1|1)=1 -p 1 r Jake’s model makes consecutive errors p-1 p p p-1 0 1

Prelab/Reference r Prelab – – – Explain the advantage and disadvantage of three speech codecs What is the difference of CS-ACELP codec from CELP codec? Explain “fading” Investigate another decoding schemes of convolutional codes Why do we use a interleaver? What is the adequate length of the interleaver in general fading channels – Investigate another interleaving schemes r Reference – Fred Halsall, Data Communications, Computer Networks and Open Systems – Sklar, Digital Communications, Prentice-Hall – J. G. Proakis, Digital Communications, Prentice Hall, 2001, 4 th ed – Shu Lin, Daniel J. Costello, Error Control Coding, Prentice-Hall