Browse Prior Art Database

Adaptive Transmission of Redundant Data on a Wireless Network to Reduce Overall Bandwidth Utilization

IP.com Disclosure Number: IPCOM000130430D
Publication Date: 2005-Oct-28
Document File: 1 page(s) / 8K

Publishing Venue

The IP.com Prior Art Database

Abstract

Wireless networks are inherently lossy. Protocols typically use ACK's, NAK's and retries to recover the lost data. Thus, the lost data results in additional network bandwidth usage for the packets to be resent, and to find out when these resends are necessary. This also adds latency to the transmission.

This text was extracted from a Microsoft Word document.
This is the abbreviated version, containing approximately 84% of the total text.

WIRELESS REDUNDANCIES TO REDUE RETRIES

Adaptive Transmission of Redundant Data on a Wireless Network to Reduce Overall Bandwidth Utilization

Disclosed Anonymously

Wireless networks are inherently lossy.  Protocols typically use ACK's, NAK's and retries to recover the lost data.  Thus, the lost data results in additional network bandwidth usage for the packets to be resent, and to find out when these resends are necessary.  This also adds latency to the transmission.

High availability storage devices use a technology called RAID (Redundant Array of Independent Disks) which stores extra information so that if there is a problem reading some data, the same information can be recovered from the extra information.  The simplest scheme is to have two copies of everything.  More sophisticated schemes such as RAID-5 add one extra unit of information for every 4, reducing the overhead from 100% to 25%.

A wireless data protocol could use the same technique when a datagram is broken into smaller packets. One example would be to send an extra packet for every 4, and use the RAID-5 algorithm to recover the data if one of the packets is lost.

Although this results in a guarantee of using more than the minimum required bandwidth, it can actually save bandwidth and reduce latency if the network is losing a significant number of packets (because retries are no longer necessary.)

As a refinement, the protocol stack can monitor the rate of traffic loss, and adjust the amount of redundant...