Browse Prior Art Database

Contiguous Receive Buffer for TDM Data Streams

IP.com Disclosure Number: IPCOM000010796D
Original Publication Date: 2003-Jan-22
Included in the Prior Art Database: 2003-Jan-22

Publishing Venue

Motorola

Related People

Michael Bercovici: AUTHOR [+3]

Abstract

In a TDM environment each user is statically allocated a communication time slot. The user may or may not use this time slot. When a time slot is not used, it is wasted. In a processor system receiving a TDM data stream, the usual procedure is to convert this stream into parallel form and store it into a memory device for further processing. In such a case the unused time slots would be stored together with the used time slots. This means wasted memory space and processor real time (for processing the unused time slots).

This text was extracted from a Microsoft Word document.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 46% of the total text.

        � � � � � � � � � � �

Contiguous Receive Buffer for TDM Data Streams

Michael Bercovici

Shemesh Yaron

Kahlon Haim

This page was left intentionally blank

� Table of Contents

1        � � � � � � � � � � � � � Introduction        . 6

2        � � � � � � � � � � � � � The programmable contiguous buffer. 8

3        � � � � � � � � � � � � � Conclusion        . 13

List of Figures

Figure 1 TDM Example        . 6

Figure 2 Block Diagram (TDM to CPU direction)         8

Figure 3 Four Frames Deep Buffer , 32 TS/Frame Example        . 10

Figure 4 The Counter         11

Figure 5 Four Frames Deep Buffer , 32 TS/Frame 16 bits/TS Example Timing Diagram        .. 11

ABSTRACT

� In a TDM environment each user is statically allocated a communication time slot. The user may or may not use this time slot. When a time slot is not used, it is wasted. In a processor system receiving a TDM data stream, the usual procedure is to convert this stream into parallel form and store it into a memory device for further processing. In such a case the unused time slots would be stored together with the used time slots. This means wasted memory space and processor real time (for processing the unused time slots).

1          Introduction

A TDM (Time Division Multiplexing) system combines (multiplexes) data streams by assigning each stream a different time slot (TS) in a set. The number of TS in a set varies from one TDM standard to another. Such a set is also called a Frame. The frames are repeatedly or periodically transmitted over a single transmission channel.


Figure 1 TDM Example

Figure 1 TDM Example above illustrates a TDM system heaving 32 TS per frame and 16 bits per TS.

TDM processing is usually a real time activity. In many environments the TDM data is buffered for a number of frames and is subsequently transferred to the CPU memory space for processing. In a symmetric manner, the CPU prepares a TDM buffer in its memory space and transfers it to the TDM system for transmission over the communication channel. The most efficient way to perform these data transfers is by using DMA (Direct Memory Access) that allows data to be sent directly, in both directions, without CPU involvement.

The simplest TDM scheme is the fixed assignment where each TS is pre-assigned to a data source.� This type of assignment is highly efficient when the traffic is continuous (the TS are always full). Under bursty or sporadic traffic a lot of TS may be wasted.

In such a case, the TDM data buffer mentioned before would include a variable number of unused TS that are merely transferred to and from the CPU wasting DMA bandwidth and CPU resources.

2          The programmable contiguous buffer

The solution below is proposing a P-frames programmable buffer (P = ) that includes only the in-service TS ordered in a contiguous manner.


Figure 2 Block Diagram (TDM to CPU direction)

The circuit consists of 2 memories: Active Time Slots Memory (ATSM), Receive Data Memory (RxDM) and a counter. The Active Time Slots Memory (ATSM) holds the addresses to which, the active TDM TS will be written into...