Home

Welcome to the intercom wiki!

Sender algorithm (audio):

Input:

1. P = number of bit-planes in the DWT domain.
2. D = number of DWT levels.
3. N = number of samples processed per iteration.

Steps:

0. X <- P-1.
1. Capture N samples.
2. Compute the D-levels DWT of the array of samples. The DWT should be orthogonal.
3. Compress the samples (now represented by an array of DWT coefficients) using a bit-plane encoder. Supose P planes. Generate a packet of code-stream for every plane of bits.
4. Send X+1 packets.
5. Receive the number of packets X transmitted on time (before the maximum latency).
6. Go to step 1.

Receiver algorithm (audio):

Input:

1. P = number of bit-planes in the DWT domain.
2. D = number of DWT levels.
3. L = latency of the encoding system in seconds.
4. N = number of samples processed per iteration.

Steps:

1. X <- 0.
2. while iteration_time < L: 2.1. Receive a packet. 2.2. X <- X+1. 2.3. Progressively reconstruct the array of N DWT coefficients.
3. Send X.
4. Compute the D-levels inverse DWT.
5. Play the array of samples.
6. Go to step 1.

Some useful links:

1. Audio Compression using Wavelet Techniques.
2. High-Quality Audio Compression Using an Adaptive Wavelet Packet Decomposition and Psychoacoustic Modeling.
3. PyWavelets - Discrete Wavelet Transform in Python.
4. Novel Image Compression Using Bit Plane Slicing.
5. Text coding.
6. Audio Coding Fundamentals.
7. Compresión Lossy de Imágenes en el Dominio Wavelet.
8. Wavelet Filter Banks in Perceptual Audio Coding.
9. A Scalable and Progressive Audio Codec.
10. Progressive Audio Coding.