Browse Prior Art Database

SYNCHRONIZATION FOR A SPECTRUM EFFICIENT MODULATED SIGNAL

IP.com Disclosure Number: IPCOM000005829D
Original Publication Date: 1990-Mar-01
Included in the Prior Art Database: 2001-Nov-09
Document File: 3 page(s) / 170K

Publishing Venue

Motorola

Related People

Michael L. Needham: AUTHOR [+2]

Abstract

In aspectrum efficient communication system, it is desirable to utilize modulation techniques that occupy minimum transmitted spectrum. One highly advantageous modulation approach is the frequency division multiplexing of informa- tion onto closely spaced multiple subcarriers. This approach was recently used in a new spectrum efficient communica- tions system developed at Motorola. In this system, digital data and processed speech symbols were encoded together for transmission over a radio channel.

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 38% of the total text.

Page 1 of 3

m MmROLA Technical Developments Volume 10 March 1990

SYNCHRONIZATION FOR A SPECTRUM EFFICIENT MODULATED SIGNAL

by Michael L. Needham and Michael D. Kotzin

   In aspectrum efficient communication system, it is desirable to utilize modulation techniques that occupy minimum transmitted spectrum. One highly advantageous modulation approach is the frequency division multiplexing of informa- tion onto closely spaced multiple subcarriers. This approach was recently used in a new spectrum efficient communica- tions system developed at Motorola. In this system, digital data and processed speech symbols were encoded together for transmission over a radio channel.

   The modulated signal was created at baseband by partitioning the data into six separate streams, then passing it through six filters of a polyphase filter bank and summing the outputs. The data may be formatted into any number of discrete amplitude levels depending on the nature of the information to be communicated, and is input to the filter bank, along with necessary synchronization overhead, at a rate of 500 samples per second per filter, or a total rate of 3000 samples per second. Anti-aliasing polyphase filter bank principles are well known in the art, and are commonly used in subband type speech coders. Our particular filters were 80 tap FIR filters with a sampling rate of 8000 Hz, a bandwidth of 250 Hz each (3 dB down), and spaced 250 Hz apart. The resulting signal will then cover a contiguous frequency block from 250 to 1750 Hz, for a total signal bandwidth of 1500 Hz.

   Advantages of the above approach include minimum resulting signal bandwidth, excellent signaling sensitivity, the ability to correct frequency selective impairments, uniform power spectral density across the bandwidth, and spectrum occupancy of the baseband signal away from DC. This last feature allows the signal to be sent over voiceband channels
(e.g., a phone line, or a SSB channel).

   The format of the resulting output signal is shown in Figure 1. Each frame of data is made up of six individual spec- trally isolated data streams. One should recognize that the real signal exhibits considerable frequency overlap among the filters as well as considerable time overlap (controlled intersymbol interference) among the samples from a single filter. At the receiver, a similar polyphase filter bank is used to "dewde" this transmitted signal. Due to the nature of the anti-aliasing properties of the filter bank, any of the frequency overlap or time overlap of the samples is eliminated and the symbol stream from each filter band is recovered without distortion. However, correct decoding of the informa- tion is critically dependent on proper timing synchronization at the receiver.

   The polyphase filter bank essentially modulates the data onto six separate subcarriers. The subcarriers share a common phase reference and are synchronized to the data clock. The decoding of data by the polyphase filter bank at the receiver is an...