Browse Prior Art Database

Method and Apparatus for look-up-table coded symbol system to substitute white noise packet payload Disclosure Number: IPCOM000029823D
Original Publication Date: 2004-Jul-14
Included in the Prior Art Database: 2004-Jul-14
Document File: 4 page(s) / 69K

Publishing Venue



A program is disclosed for look-up-table coded symbol system to substitute white noise packet payload in the networking system.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 54% of the total text.

Page 1 of 4

Method and Apparatus for look -up-table coded symbol system to substitute white noise packet payload

The following diagram showed a typical digital quantized voice stream before coding:

Voice Voice Voic

Diagram 1: Typical bi-directional digitally quantized voice stream

The diamond represented the bi-directional quantized active voice or useful source information, and the line represent the silent gap. The background nature noise is normally filtered out by the digital non-linear quantization. To add back to the realistic of feeling of channel connection and nature communication, noise packets are added to the voice codec (Coder Decoder).

In a packet switched wired or wireless communication channel, a typical sequence of packets from a voice codec is laying out like this:

Diagram 2: Typical packet sequence in a serial communication channel

If we closely examine the bit sequence of Diagram 2, there are two unutilized (no voice channel payload or system payload) time slot here:

1. The space in between the packet is the maximum separator time gap that can't be utilized due to separation needed in between packets.
2. The AGWN (Additive Gaussian White Noise) packets.

These AGWN packets have to be inserted to add to the realistic feeling of the channel link establishment. An innovative algorithm is created to solve this problem by closely examining the characteristic of the AGWN packets series.

The AGWN packet is normally constructed, according to the following format:

Sync Voice Packet packe

BWGN Channel m

Voice Packet Packet 1

Channel n

Header Header Header Header Payload random byte and trail byte


[This page contains 3 pictures or other non-text objects]

Page 2 of 4

Diagram 3: Typical AGWN packet structure

The AGWN packet is normally formatted like above sequence, beyond a small header section consisted with header ID, address and possible checker and trailer bytes, the vast majority of the packet is occupied by the AGWN payload.

So the situation can be summarized in the following diagram:

Quantiz Source ation

Channel Coding

Modem Coding

Diagram 4: Solution Stage

This problem is that the AGWN being carry out with more and more coding along the output of source coding and send out all the way through modem, thus significantly degraded effective channel utilization. By recognize the nature of the coding sequence, a certain mathematically proven algorithm can be further explored. The dealing with the realization of AGWN can be altered in the bit stream, so that is the background of the patent.

An algorithm for solving this type of problem is created, based on re-distribution of the AGWN packets realization as illustrated in the following diagram:

The original packet:

Diagram 5: Algorithm of re-align the AGWN packet generation

The key invention of the algorithms is that by transmitting the far small sized code of the much larger sized AGWN payload, the whole down stream functional block can effectively concentrate on the voice payload an...