Browse Prior Art Database

Queue Analysis of Incoming and Outgoing T1 Signalling Bits

IP.com Disclosure Number: IPCOM000037353D
Original Publication Date: 1989-Dec-01
Included in the Prior Art Database: 2005-Jan-29
Document File: 5 page(s) / 83K

Publishing Venue

IBM

Related People

Cazzolla, JJ: AUTHOR [+4]

Abstract

This article describes techniques to accurately measure pulses on incoming and outgoing T1 signalling bits by means of queue analysis.

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 26% of the total text.

Page 1 of 5

Queue Analysis of Incoming and Outgoing T1 Signalling Bits

This article describes techniques to accurately measure pulses on incoming and outgoing T1 signalling bits by means of queue analysis.

T1 communications provides an increasingly popular means of high speed digital communication. A T1 line consists of 24 channels, each operating at 64 Kbps, plus an 8 Kbps framing signal. Every 125 microseconds, a byte of data is received on each channel. The 24 received bytes, along with a framing bit, constitute a frame. These frames are typically grouped together to form a superframe (12 contiguous frames) or an extended superframe (24 contiguous frames). During every sixth frame, the least significant bit (LSB) of each data byte is used instead for signalling information. For this reason, this method of signalling is referred to as "robbed bit" signalling.

These bits are used to convey signalling information, such as seizures, answer supervision, dial pulses, ringing, and flashes. They are used to set up and tear down calls. Various protocols have been defined for the use of these signalling bits and industry standards define timing requirements for dial pulse qualification and certain other signalling functions. In the case of a T1 card, the computerized branch exchange (CBX) software also presents various timing requirements relating to the signalling. In order to meet the timing requirements for qualifying incoming dial pulses, it is necessary for firmware to sample the incoming signalling bits at least every three milliseconds. In addition, a three- millisecond sampling rate is desirable in order to provide sufficient accuracy in measuring other types of signalling pulses.

A conventional approach to sampling signalling bits is to architect the firmware with an executive loop which periodically polls the signalling bits as it processes each channel. An interrupt routine updates the current set of signalling bits every superframe or extended superframe. Each pass through the loop, the signalling bits for each channel would be polled and action would be taken based on their result. Depending upon the number of channels which are supported and the overall function which must be performed during each executive loop, the firmware may not be able to complete a typical executive loop within three milliseconds. It is even less likely that the firmware would be able to complete a worst-case executive loop within three milliseconds. This would result in inaccurate measurement of incoming signalling bit pulses and can lead to misinterpretation of signalling information and misdialed numbers. It is this inaccurate measurement of incoming signalling bit pulses which the method disclosed herein addresses.

These problems are solved by storing the signalling bits in a queue and using an interrupt routine to add a new set every three milliseconds. The executive loop, as it processes each channel, rather than poll the current set of signalling bits, wi...