DEMODULATION D'UNE FORME D'ONDE NUMERIQUE COMPATIBLE MODULEE EN AMPLITUDE, EN UTILISANT DEUX PROCESSEURS FFT

AM COMPATIBLE DIGITAL WAVEFORM DEMODULATION USING A DUAL FFT

Abrégé français

L'invention concerne un procédé et un appareil pour démoduler une forme d'onde composite d'audiodiffusion numérique (DAB) modulée en amplitude qui contient des porteuses modulées numériquement et qui fait appel à un mélangeur (180) pour convertir un signal reçu (170) en deux signaux, le premier de ces signaux représentant un composant en phase et le second un composant en quadrature de phase, deux convertisseurs de type analogique/numérique (182, 184) pour convertir les deux signaux en signaux numériques et deux éléments de transformée de Fourier rapide (FFT) (188, 190) pour extraire les données séparément des deux signaux numériques. Des signaux des porteuses numériques complémentaires sont récupérés du composant en quadrature de phase et des porteuses numériques non complémentaires sont obtenues par l'addition des données complémentaires et de la sortie de l'élément FFT recevant le composant en phase. On évite que les fuites des signaux modulés en amplitude dans un filtre passe-haut (186) n'interfèrent avec la démodulation des signaux complémentaires de la porteuse en utilisant des canaux de démodulation séparés.

Abrégé anglais

The invention provides a method and apparatus for demodulating a composite AM
DAB waveform which contains digitally modulated carriers and which employ a
mixer (180) for converting a received signal (170) into two signals, the first
of these signals represents an in-phase component and the second of the
signals represents a quadrature component, two analog-to-digital converters
(182, 184) for converting the two signals into digital signals, and two fast
Fourier transform elements (188, 190) for extracting data separately from the
two digital signals. Complementary digital carrier signals are recovered from
the quadrature, and non-complementary digital carriers are derived from a sum
of the complementary data and the output of the in-phase component FFT
process. Leakage of the AM signal through a highpass filter (186) is prevented
from interfering with the demodulation of the complementary carrier signal's
use of separated demodulation channels.

Revendications

8
WHAT IS CLAIMED IS:
1. An apparatus for demodulating a composite signal, the
composite signal including a first carrier signal amplitude
modulated by an analog message signal and a second signal
comprising a plurality of digitally modulated carriers,
comprising:
means for separating the second signal into a third
signal and a fourth signal, the third signal representing an
in-phase component and the fourth signal representing a
quadrature phase component of the composite signal;
first means coupled to receive the in-phase component
for converting the in-phase component to an in-phase digital
signal;
second means coupled to receive the quadrature
component for converting the quadrature component to a
quadrature digital signal;
a first fast Fourier transform means for extracting
in-phase data from the in-phase digital signal; and
a second fast Fourier transform means for extracting
quadrature data from the quadrature digital signal.
2. The apparatus for demodulating according to
claim 1, wherein the quadrature phase component comprises
complementary carriers and non-complementary carriers.
3. The apparatus for demodulating according to
claim 2, wherein the first carrier signal and the complementary
carriers are located in a same frequency region.
4. The apparatus for demodulating according to
claim 1, wherein the means for separating is a quadrature
mixer.
5. The apparatus for demodulating according to
claim 1, wherein the first means and the second means are
analog-to-digital converters.

9
6. The apparatus for demodulating according to
claim 1, further comprising:
means coupled to receive the in-phase data and the
quadrature data, and for combining the in-phase data and the
quadrature data to produce a fifth digital signal, the fifth
digital signal being representative of non-complementary data.
7. The apparatus for demodulating according to
claim 1, wherein the in-phase digital signal is phase coherent
with the first carrier signal-
8. The apparatus for demodulating according to
claim 1, wherein the composite signal is a radio frequency (RF)
signal, and wherein the second signal comprising a plurality of
digitally modulated carriers is modulated in an orthogonal
frequency division multiplexed format.
9. A method of demodulating a composite signal, the
composite signal including a first carrier signal amplitude
modulated by an analog message signal and a second signal
comprising a plurality of digitally modulated carriers, the
method comprising the steps of:
receiving the composite signal;
separating the composite signal into a third signal
and a fourth signal. the third signal representing an in-phase
component and the fourth signal representing a quadrature phase
component of the composite signal;
converting the in-phase component to an in-phase
digital signal, the in-phase digital signal being phase
coherent with the first carrier signal;
converting the quadrature component to a quadrature
digital signal:
extracting in-phase data from the third signal by
means of a first fast Fourier transform; and
extracting quadrature data from the fourth signal by
means of a second fast Fourier transform.

10. The method for demodulating a composite signal
according to claim 9, wherein the quadrature data modulates
complementary quadrature amplitude modulated carriers.
11. The method for demodulating a composite signal
according to claim 9, further comprising:
combining the in-phase data and the quadrature data
to produce a fifth signal, the fifth signal comprising
non-complementary data.
12. The method for demodulating a composite signal
according to claim 9, wherein the first fast Fourier transform
and the second fast Fourier transform are performed
substantially simultaneously.

Description

CA 02233063 1998-02-16
W O 97/08877 PCT~US96/13526
AM COMPATIBLE DIGITAL WAVEFORM DEMODULATION USING A DUAL FFT
R~UND OF THE lNV~. lON
This invention relates to waveform demodulation, and
more particularly to methods of and apparatus for receiving and
demodulating digitally modulated signals and analog amplitude
modulated signals within the same frequency ~h~nnel assignment.
Broadcast and reception of digitally-encoded audio
signals is expected to provide improved audio fidelity.
Several approaches have been suggested. Out-of--band t~ ni ques
provide broadcast of digital radio signals in a specially
designated frequency band. In-band techniques provide
broadcast within substantially vacant slots between adjacent
channels in the existing broadcast band (interstitial approach)
or in under-utilized portions within the same frequency channel
allocations currently used by ~- ~cial broadcasters (in-band
on-channel or IBOC approach). The in-band on-channel approach
may be implemented without the need for additional frequency
coordination and with relatively minor changes to existing
transmitting equipment. It is also a requirement that any
digital audio broadcasting (DAB) technique must not degrade
analog signal reception by conventional analog receivers.
In-band approaches to digital audio broadcasting have
thus far only been proposed in the FM band (88 MHz to 108 MHz),
since the bandwidth of AM channels is relatively narrow as
compared to the FM band allocation. However, high fidelity
digital audio broadcasting in the AM band (530 kHz to 1700 kHz)
would provide AM broadcasting stations with a means to compete
with high-quality, portable audio sources such as cassette
tapes and compact disc players. It would therefore be
desirable to promote an in-band on-channel (IBOC) approach in
the AM broadcasting band to provide enhanced fidelity through
digital signalling without affecting reception by existing
analog AM receivers.
An AM compatible digital broadcast waveform which
satisfies the requirement of substantial orthogonality between

CA 02233063 1998-02-16
W O 97/08877 PCT~US96/13526
a conventional analog AM signal and a digitally modulated
signal set has been developed. The waveform is described in
U.S. Patent Application Serial No. 08/206,368 filed March 7,
1994, entitled METHOD AND APPARATUS FOR AM COMPATIBLE DIGITAL
BROADCASTING. The waveform spectrum consists of in-phase and
quadrature components. An in-phase radio fre~uency (RF)
carrier is ~n~lllated by an analog audio signal and the in-phase
component of a digital signal. The quadrature-phase RF carrier
is ~AIll ated b~ the quadrature component of the digital signal.
The digital signal has an orthogonal frequency division
multiplexed (OFDM) ~ormat. The in-phase signal consists of the
conventional analog AM signal and selected digital carriers.
The in-phase digital carriers are placed outside o~ the
spectral region occupied by the analog AM signal. The
quadrature-phase carriers are situated both within and outside
the spectral region occupied by the analog AM signal (although
not at the center frequency occupied by the unmodulated analog
carrier). The quadrature digital carriers situated within the
same spectral region as the analog AM signal are called
complementary carriers. The above described arrangement is
further described in U.S. Patent Application Serial No.
08/368,061 filed January 3, 1995, entitled METHOD AND APPARATUS
FOR IMP~OVING AM COMPATIBLE DIGITAL BROADCAST ANALOG FIDELITY.
The context of the present invention is a need to
demodulate the composite waveform with mini~l crosstalk. The
modulated composite waveform is produced by a modulation method
in which an analog amplitude modulated (AM) signal and a
digital signal which may be a representation of the analog
audio signal (or it may be any other digital signal) are
encoded together and transmitted simultaneously in the same
frequency channel. This approach places some of the digital
carriers in quadrature with the analog AM, thereby enabling the
AM DAB data to be extracted and decoded with high ~idelity and
without crosstalk, assuming the receiver is capable of proper
signal separation.
A receiver (which is not necessarily prior art) has
been considered which converts the signal to baseband using
conventional I and Q mixers, with the I channel signal being

CA 02233063 1998-02-16
W O 97/08877 PCT~US96/13526
passed through a digital high pass filter to separate the
digital signal from the analog signal, as hereinafter
explained, a problem has been discovered related to crosstalk
between the analog signal and the digital signal. In the
receiver under consideration, the analog signal interferes with
the demodulation of the complementary carriers if the
demodulation of the I and Q component samples is carried out in
a single common FFT processor. What is needed is a
demodulation technique which minimizes the undesired crosstalk
between the analog signal and the digital signals.
8UNMARY OF T~E lNV ~:~.,lON
According to the invention, in an AM compatible
digital audio broadcasting (AM DAB) system using an orthogonal
frequency division multiplexed (OFDM) modulation format, a
radio frequency receiving and demodulating method and apparatus
employs dual fast Fourier transform processes on separate
respective in-phase and quadrature-phase components of a
received OFDM digital signal, the output of the quadrature
channel being used to recover the complementary data, and the
resultant processed component signals being summed to recover
the non-complementary data.
The apparatus includes a mixer for converting a
received signal into two signals, the first of the two signals
representing an in-phase component and the second o~ the two
signals representing a quadrature component; two analog-to-
digital converters for converting the two signals into digital
signals; and two fast Fourier transform processors for
extracting data from the two signals.
The present invention provides better and higher
resolution data extraction than heretofore known. Moreover,
the invention can be useful in providing an in-band, on-channel
(IBOC) solution to digital audio broadcasting (DAB) in the AM
frequency band.
- 35 The advantages of the waveform of the present
invention are that (1) the existing analog AM broadcast
channels can be upgraded to digital without requiring a new FCC
~requency allocation, (2) the AM broadcast stations can be

CA 02233063 1998-02-16
W O 97/08877 PCTAUS96/13S26
upgraded to a digital broadcast format with only limited
capital expenditure, (3) the composite waveform at appropriate
digital power levels yields a coverage area that is essentially
equivalent to the existing analog AM station, ( 4) the existing
AM receivers can recover the analog portion of the composite
signal without modification, and (5) interference between the
digital signal and the analog signal is ;nim;zed.
A description of the AM DAB waveform involved is
presented in U.S. Serial No. 08/206,368, filed on March 7,
1994, which is herein incorporated by reference.
A further underst~n~;ng of the nature and advantages
of the invention will become apparent by reference to the
remaining portions of the specification and drawings.
BRIEF DE8CRIPTION OF T~E DRAWING8
The invention will be more readily apparent to those
skilled in the art by reference to the ac~omr~nying drawings
wherein:
Fig. l is a spectral representation of the in-phase
component of a composite analog AM and digital broadcasting
signal;
Fig. 2 is a spectral representation of the quadrature
component of a digital broadcasting signal;
Fig. 3 is a block diagram of a demodulator which has
been considered for the subject waveform;
Fig. 4 is a block diagram of a demodulator in
accordance with the present invention;
Fig. 5 is a representation of the signal-to-noise
ratio (SNR) for carriers of the receiver shown in Fig. 3; and
Fig. 6 is a representation of the signal-to-noise
ratio (SNR) for carriers of the receiver in Fig. 4.
DE~TT~n DE8CRIPTION OF 8PECIFIC EMBODIMENT8
This invention provides a method o~ simultaneously
receiving both an analog amplitude modulated signal and a
digital signal on the same channel assiynment as the existing
analog AM broadcasting allocation.

CA 02233063 1998-02-16
W O 97/08877 PCTrUS96/13526
The AM DAB waveform of the present invention has been
previously explained. Fig. 1 is a spectral representation of
the in-phase component of a composite analog AM and digital
broadcasting signal. The in-phase component contains the
conventional analog AM signal 100 and non-complementary digital
carriers 102. The in-phase component does not have any digital
carriers 102 in the spectral region occupied by the analog AM
signal 100.
Fig. 2 is a spectral representation of the quadrature
component of a digital broadcasting signal. The quadrature
portion of the spectrum as shown in Fig. 2 contains only
digital carriers 110 and 112. The digital carriers that lie
outside the spectral region of the analog AM signal 100 are
non-complementary signals 112, and the digital carriers that
lie in the same frequency region as the analog AM signal 100
are complementary signals 110.
Fig. 3 illustrates a demodulator which was considered
in connection with the waveform herein described. The
demodulation t~h~ique converts the signal 140 to bas~h~n~
using conventional I and Q mixer 150. Mixer 150 separates the
in-phase and quadrature components of the received signal. The
I and Q channels are then digitized in analog-to-digital (A/D)
converters 154 and 156. Following A/D converter 154, the I
channel is passed through high pass filter 152 which is
designed to eliminate the analog AM signal. The high pass
filter 152 is in the real world less than ideal, which gives
rise to some of the problems overcome by the present invention.
The I and Q channels are then input to a fast Fourier transform
(FFT) processor 158 in order to recover and obtain the received
data.
While the device revealed in Fig. 3 will demodulate
the signal involved, there is room for improvement. In
~ particular, the receiver of Fig. 3 has several problems. For
example, some of the analog AM signal leaks through highpass
filter stopband 152 and interferes with the demodulation of the
complementary carriers.
Fig. 4 illustrates how demodulation is performed in
the present invention. This new demodulation technique avoids

CA 02233063 1998-02-16
W O 97/08877 PCT~US96/13526
the problem of the demodulation technique illustrated in Fig. 3
by using two FFTs which operate separately on the I and Q
channels. The received signal 170 is converted to baseband by
conventional I and Q mixer 180. As with mixer 150, mixer 180
separates the in-phase and quadrature components of the
received signal. The I and Q ~hi~nnels are then separately sent
to A/D converters 182 and 184 where they are digitized. The I
channel is passed through high pass filter 186 which attempts
to eliminate the analog AM signal. The I and Q channels are
then processed separately with dual FFT processors 188 and 190.
The output from the Q channel is used to recover the
complementary carriers, and the sum of the I and Q channels is
used to recover the non-complementary carriers.
The demodulation t~hnique illustrated in Fig. 4
stops leakage of the AM signal through the highpass filter from
inter~ering with the demodulation of the complementary
carriers. This tPchnique also reduces the effects of noise and
compensates for non-ideal operation o~ the I and Q mixer.
The demodulation t~chn;ques illustrated in Fig. 3 and
Fig. 4 are implemented using common, commercially-available RF
hardware, such as mixers and general purpose Digital Signal
Processors (DSPs) with software to provide the various features
of the demodulator (e.g., FFTs).
Fig. 5 shows the SNR for carriers when the receiver
25 shown in Fig. 3 is used. The low SNR in FFT bins +16 to +19
and -19 to -1~ causes the bit error rate to increase. The
carriers in these FFT bins are near the edges of the
complementary band, and the corresponding SNR is approximately
16 dB.
Fig. 6 is a plot of SNR for carriers when the
demodulator illustrated in Fig 4 is used. The SNR for the
complementary carriers in Fig. 6 is approximately 33 dB. This
is a significant improvement over the SNR for the complementary
carriers for the demodulator illustrated in Fig. 3 and a lower
bit error rate results. With the improved demodulator of Fig.
4, the signal can be received further away from the system's
transmitter.

CA 02233063 1998-02-16
W O 97/08877 PCT~US96/13526
Although the foregoing invention has been described
in some detail by way of illustration and example, for purposes
of clarity of understanding, it will be obvious that certain
changes and modifications may be practiced within the scope of
the appended claims.

Dessin représentatif