Browse Prior Art Database

NEW TECHNIQUE FOR FM TRANSMISSION OF DIGITAL DATA

IP.com Disclosure Number: IPCOM000009628D
Original Publication Date: 2000-Jan-01
Included in the Prior Art Database: 2002-Sep-05
Document File: 2 page(s) / 132K

Publishing Venue

Motorola

Related People

Denise Eribes: AUTHOR [+3]

Abstract

FSK is a commonly used technique for trans- mission of digital data in low cost applications. Unfortunately, several problems are associated with using direct FM. Noise immunity is compromised when the DC component of the incoming data is not near zero. Moreover, there is no direct method for synchronization or clock recovery. Data encoding techniques, such as Manchester Encoding and Alternate Mark Key Encoding, are often used to resolve these problems.

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 48% of the total text.

Page 1 of 2

0 M MOTOROLA Technical Developments

NEW TECHNIQUE FOR FM TRANSMISSION OF DIGITAL DATA

by Denise Eribes, Mike Miller and Paul Sofianos

BACKGROUND

  FSK is a commonly used technique for trans- mission of digital data in low cost applications. Unfortunately, several problems are associated with using direct FM. Noise immunity is compromised when the DC component of the incoming data is not near zero. Moreover, there is no direct method for synchronization or clock recovery. Data encoding techniques, such as Manchester Encoding and Alternate Mark Key Encoding, are often used to resolve these problems.

  However, these methods can constrain imple- mentation feasibility due to complexity and required bandwidth. A new encoding technique has been developed to address problems associated with FSK modulation without added complexity. The solution can be implemented in hardware or software.

PROBLEM

  Direct-conversion FSK modulation is often used in conjunction with a PLL synthesized carrier. For this type of transmission, the digital modulation

source must meet certain requirements. To maintain the best noise immunity at the receiver, the DC component of the incoming data should be as close to zero as possible. In addition, a minimum fre- quency component must be maintained at all times. Failing to meet this condition will drive in the trans- mitter's PLL and receiver's coilless demodulator to "track out" the modulating signal. It is also impor- tant that the maximum frequency component be known. This helps define the modulation index and total bandwidth required for the transceiver.

SOLUTION

  Encoding the raw data prior to RF transmission is necessary for two reasons. First, incoming data stream generally contains a DC component. Additionally, the data stream does not have a pre- defined minimum frequency component. Figure 1 illustrates the encoding scheme that was developed to address these problems. Four additional bits will surround a data byte: I (Invert) Bit,i(Invert Bar) Bit, D/ c (Data/Control ) Bit and D /C (Data/Control).

D1 c

  In this encoding technique each data byte has an is for a data stream to always contain an equal num- assigned "weighted" value that is based on the mnn- ber of l's and O's so that their respective weighted bcr of l's and O's contained in that stream. Every 1 values cancel to zero. Unfortunately, since the data has a weight of +l and every zero has a weight of - 1. stream cannot be controlled or predicted, it is possi- The ideal condition to achieve a zero DC component ble to have long data streams of l's or 0's. To

0 MotamLa, 1°C. m 10 Janua?y 2ooa

[This page contains 15 pictures or other non-text objects]

Page 2 of 2

MOTOROLA

ensure that such condition does not disrupt the desired DC component, an estimate average of all data transmitted is maintained.

  For every new byte, the MCU computes two weighted values by counting l's and O's in the man- ner discussed above. The first value is a derived directly from the ra...