U.S. patent number 5,191,594 [Application Number 07/798,780] was granted by the patent office on 1993-03-02 for fading channel simulator.
This patent grant is currently assigned to The United States of America as represented by the United States. Invention is credited to Paul E. Argo, T. Joseph Fitzgerald.
United States Patent |
5,191,594 |
Argo , et al. |
March 2, 1993 |
Fading channel simulator
Abstract
Fading channel effects on a transmitted communication signal are
simulated with both frequency and time variations using a channel
scattering function to affect the transmitted signal. A
conventional channel scattering function is converted to a series
of channel realizations by multiplying the square root of the
channel scattering function by a complex number of which the real
and imaginary parts are each independent variables. The
two-dimensional inverse-FFT of this complex-valued channel
realization yields a matrix of channel coefficients that provide a
complete frequency-time description of the channel. The transmitted
radio signal is segmented to provide a series of transmitted signal
and each segment is subject to FFT to generate a series of signal
coefficient matrices. The channel coefficient matrices and signal
coefficient matrices are then multiplied and subjected to
inverse-FFT to output a signal representing the received affected
radio signal. A variety of channel scattering functions can be used
to characterize the response of a transmitter-receiver system to
such atmospheric effects.
Inventors: |
Argo; Paul E. (Los Alamos,
NM), Fitzgerald; T. Joseph (Los Alamos, NM) |
Assignee: |
The United States of America as
represented by the United States (Washington, DC)
|
Family
ID: |
25174257 |
Appl.
No.: |
07/798,780 |
Filed: |
November 27, 1991 |
Current U.S.
Class: |
375/130;
455/506 |
Current CPC
Class: |
H04B
17/0087 (20130101); H04B 17/3912 (20150115) |
Current International
Class: |
H04B
17/00 (20060101); H04L 001/04 (); H04B 001/10 ();
H04B 015/00 () |
Field of
Search: |
;375/1 ;455/52.1,52.3,65
;364/801,802 |
References Cited
[Referenced By]
U.S. Patent Documents
Other References
P A. Bello, "Characterization of Randomly Time-Variant Linear
Channels," IEEE Trans. Communications Systems, vol. CS-11, pp.
360-393 (Dec. 1963). .
P. A. Bello, "Measurement of Random Time-Variant Linear Channels,"
IEEE Trans. Information Theory, vol. IT-15, No. 4, pp. 469-475
(Jul. 1969). .
C. G. Watterson et al., "Experimental Confirmation of an HF Channel
Model," IEEE Trans. Communication Technology, vol. COM-18, No. 6,
pp. 792-803 (Dec. 1970). .
P. A. Bello, "Some Techniques for the Instantaneous Real-Time
Measurement of Multipath and Dopper Spread," IEEE Trans.
Communication Technology, vol. 13, No. 3, pp. 285-291 (Sep. 1965).
.
R. A. Shepard et al., "Frequency Spread in Ionospheric Radio
Propagation," IEEE Trans. Communication Technology, vol. COM-15,
No. 2, pp. 268-275 (Apr. 1967)..
|
Primary Examiner: Cangialosi; Salvatore
Attorney, Agent or Firm: Wilson; Ray G. Gaetjens; Paul D.
Moser; William R.
Claims
What is claimed is:
1. A fading channel simulator for affecting a transmitted radio
signal with a selected channel scattering function for use in
testing fading channel transmission characteristics of a receiver
and transmitter in a communications system, comprising:
first transform means for receiving an analog output from said
transmitter and outputting a plurality of Fast Fourier Transform
(FFT) signal coefficient matrices, each said FFT matrix
corresponding to one of a series of segments in a data string
representing said analog output;
second transform means for receiving said selected scattering
function and outputting a plurality of inverse-FFT channel
realization coefficient matrices, each matrix representing
successive delay-spread and Doppler-spread characteristics;
multiplication means for multiplying one of said signal coefficient
matrices with one of said channel realization coefficient matrices
to form an output matrix; and
output signal means for forming the inverse FFT of said output
matrix and outputting a faded channel signal string representing
said received radio signal for input to said receiver.
2. A fading channel simulator according to claim 1, wherein said
first transform means includes: an analog-to-digital converter for
receiving said input analog output and outputting a series of
digital data strings each having a predetermined duration; and
first FFT means for forming and storing a FFT for each one of said
series of digital data strings.
3. A fading channel simulator according to claim 1, wherein said
second transform means includes: means for generating a time series
of complex channel realizations; and
second FFT means for forming an inverse-FFT for each complex
channel realization in said time series.
4. A fading channel simulator according to claim 2, wherein said
second transform means includes: means for generating a time series
of complex channel realizations; and
second FFT means for forming an inverse-FFT for each complex
channel realization in said time series.
5. A method for simulating fading channel effects on a transmitted
high frequency radio signal, comprising the steps of:
generating a series of complex realizations of said fading channel
from a channel scattering function including both a delay-spread
function and a Doppler-spread function;
segmenting said transmitted radio signal into time length segments
each having a preselected duration;
forming a digital representation of said signal in each
segment;
performing a FFT on each digital representation to obtain a matrix
of signal coefficients;
affecting each one of said segments with a one of said realizations
of said fading channel to generate a received signal; and
combining said received signal segments to recreate a received
radio signal with amplitude and phase fading effects for input to a
receiver.
6. A simulation method according to claim 5, wherein said step of
generating said complex realization of said fading channel further
includes the step of forming an inverse-FFT of each complex
realization of the channel scattering function in said series to
generate a series of channel coefficient matrices.
7. A simulation method according to claim 5, where said step of
combining said segments with said transmitted signals comprises the
steps of:
multiplying one said signal coefficient matrix with one said
channel coefficient matrix to form a combined matrix; and
forming the inverse FFT of said combined matrix to provide the
effected received signal segment.
8. A simulation method according to claim 7, further including the
step of placing said affected received signal segments in an order
functionally related to said transmitted radio signal.
Description
BACKGROUND OF INVENTION
This invention relates to high frequency (HF) radio signal
propagation through fading channels and, more particularly, to
simulation of fading channels in order to characterize HF radio
system performance in transmitting and receiving signals through
such fading channels. This invention is the result of a contract
with the Department of Energy (Contract No. W-7405-ENG-36).
HF radio signals propagate through the ionosphere along a variety
of paths of different length ("multipaths") and with different
propagation characteristics. Interference arises between the
signals on the various paths producing signal fading on time scales
from a fraction of a second to a few seconds, whereby the "fading
channels" produce a degradation of the signal quality. Methods of
modulation, diversity, and coding in the system design are selected
in order to minimize this signal degradation.
Testing of HF communication systems can be done by using
operational systems either through a real circuit, or through a
channel simulator. Real circuit tests can be expensive, and may not
provide actual "worst case" situations. Alternately, one must
ensure that the channel simulator accurately depicts the effects of
a "real" channel.
HF ionospheric channels are nonstationary in both frequency and
time, but for narrow bandwidths (tens of kHz) and short times
(minutes) most channels can be adequately represented by a
stationary model. In addition, except under extreme conditions, the
ionosphere supports propagation over a limited number of discrete
modes representing different average signal levels and delays.
Moreover, each path will show a different fading rate and delay
spread.
A radio channel may be modeled as a randomly time-varying linear
channel that can be characterized by a channel scattering function.
This scattering function is defined as the density of power
scattered by the channel as a function of Doppler shift, time
delay, and spatial angle-of-arrival. A received HF signal is
usually the composite of several signals arriving via different
ionospheric propagation modes, and the signal power is spread in
the three dimensions of time, frequency, and arrival angle.
Time spreading is the result of the signal propagating via two or
more paths having slightly different propagation times. Frequency
spreading is the result of movements of the reflecting ionospheric
layers and of the time variation of the electron density along the
ray paths, both of which cause changes in the phase of the received
signal. The rate of change of phase can be interpreted as a Doppler
shift of the transmitted frequency.
A line-of-sight propagation channel can be characterized by a
linear filter described by a gain G(f,t) and a propagation time
delay t.sub.o, where G(f,t) is likely to be a complex valued
function. The time varying frequency response of such a channel
is:
where f is the frequency and t is the time.
A multipath channel can be described as a linear sum of several
such channels, or "modes":
where i labels the individual modes.
Signal fading is simply the constructive and destructive
interferences generated by the vector addition of the signal
propagated through these several channels.
An exemplary channel scattering function is described in Proakis,
Digital Communications, McGraw-Hill, New York 1989. See also Bello,
"Characterization of Randomly Time-Variant Linear Channels," IEEE
Trans. Commun. Systems, pp. 360-393 (December 1963). Both teachings
are incorporated herein by reference. A channel correlation
function, .phi..sub.c (.DELTA.f,.DELTA.t), describes the
correlation in frequency and time of the channel response, H(f,t),
and is given by:
where the square brackets denote the expectation value. The channel
correlation function describes the coherence bandwidth and
coherence time of the channel. The two-dimensional Fourier
transform, S(.tau.,.lambda.), is called the channel scattering
function and describes the channel response in delay, .tau., and
Doppler frequency, .lambda., where ##EQU1## A form of scattering
function is illustrated by FIG. 1, which graphically depicts power
density as a function of delay, .tau., and Doppler frequency,
.lambda..
Given the Fourier transform relation between the channel
correlation function, .phi..sub.c, and the channel scattering
function, S, the Wiener-Khintchine theorem may be applied to
interpret the channel scattering function, S(.tau.,.lambda.), as
the average power spectral density of the random process, H(f,t).
Then a realization of H(f,t) may be generated with the known
technique of inverse Fourier transforming the random complex
process, h(.tau.,.lambda.), whose real and imaginary parts are
independent, Gaussian random variables, each with zero mean and
variance of S(.tau.,.lambda.)/2.
In the usual characterization of a channel, the gain function (G)
is not treated as a function of frequency (f), so that the channel
is in reality a "nonselective or multiplicative" fading channel. In
this case, all frequencies fade together. In most cases using this
characterization, the delay is treated as a fixed delta function in
time, with Rayleigh fading imposed upon each modal gain function.
Further, in order to generate a statistical model for the short
term fading channel, one must assume stationary statistics, i.e.,
that the mean values of the model parameters are constant. Then the
channel model presents a particular realization of a stochastic
process.
In one prior art representation, Watterson et al., Experimental
Confirmation of an "HF Channel Model," COM-18 IEEE Trans. Commun.
Technol., No. 6, pp. 792-803 (1970), showed that the channel can be
modeled as a "tapped delay line" with a limited number of taps with
adjustable delays. The signal at each tap is modulated in phase and
amplitude by a suitable tap-gain function, and the several delayed
and modulated signals are summed to form the output signal. The
Watterson model uses "independent zero-mean complex-Gaussian
functions with Rayleigh amplitude and uniform phase density" to
modulate the incoming signal.
It would be desirable to represent the channel transmissions as a
function of frequency and to provide nonstationary channel
statistics to more accurately model fading channels for HF radio
transmission. These problems are addressed by the present invention
wherein a channel scattering function is used to represent the
channel delay-spread and Doppler-spread functions.
Accordingly, it is an object of the present invention to provide
the channel gain function as a function of frequency.
It is another object of the present invention to provide a time
varying representation of channel transmissions.
Additional objects, advantages and novel features of the invention
will be set forth in part in the description which follows, and in
part will become apparent to those skilled in the art upon
examination of the following or may be learned by practice of the
invention. The objects and advantages of the invention may be
realized and attained by means of the instrumentalities and
combinations particularly pointed out in the appended claims.
SUMMARY OF INVENTION
To achieve the foregoing and other objects, and in accordance with
the purposes of the present invention, as embodied and broadly
described herein, the apparatus of this invention may comprise a
fading channel simulator for use in testing fading channel
transmission characteristics of a receiver and transmitter in a
communications system. A first transform circuit receives an input
analog output from the transmitter and outputs a plurality of Fast
Fourier Transform (FFT) signal coefficient matrices. Each FFT
matrix corresponds to one of a series of segments in a data string
representing the transmitter analog output. A second transform
circuit receives a selected scattering function and outputs a
plurality of inverse-FFT channel realization coefficient matrices
to represent successive channel delay-spread and Doppler-spread
characteristics. Each one of the stored signal coefficient matrices
is then multiplied with one of the channel realization coefficient
matrices to form an output matrix. An output converter then
performs an inverse FFT on each output matrix to output a faded
channel signal string representing the transmitted radio signal for
input to the receiver.
In another characterization of the present invention, fading
channel effects on a transmitted high frequency radio signal are
simulated. A series of complex realizations of the fading channel
are generated from a channel scattering function including both a
delay-spread function and a Doppler-spread function. A series of
segments are formed from the transmitted radio signal. Each segment
is then affected by one of the realizations of the fading channel
to generate a received signal. The received signal segments are
then combined to recreate a received signal with amplitude and
phase fading effects for input to a receiver.
BRIEF DESCRIPTION OF THE DRAWINGS
The accompanying drawings, which are incorporated in and form a
part of the specification, illustrate an embodiment of the present
invention and, together with the description, serve to explain the
principles of the invention. In the drawings:
FIG. 1 graphically depicts a form of a channel scattering
function.
FIG. 2 is a block diagram of a fading channel simulator system
according to the present invention.
FIG. 3 is a flow diagram for the operation of the system shown in
FIG. 1.
FIG. 4 is a schematic in block diagram form of a system for
performing the method shown in FIG. 2.
DETAILED DESCRIPTION OF THE INVENTION
Referring first to FIG. 2, there is shown a block diagram of a
fading channel simulation system according to the present
invention. Simulator 10 affects fading channel performance with
delay-spread functions and Doppler-spread characteristics
controlled by computer 12. A channel scattering function is
implemented through computer 12, where the Doppler-spread function
is related through a Fourier transform to the signal decorrelation
(fading) time and is, in fact, the function modeled in the tap-gain
simulators of Watterson. The delay-spread function similarly is
related to signal decorrelation in frequency and modeled in the
tap-gain simulators as a delta function.
Transmitter 14 outputs a signal controlled by computer 12 in
response to the signal received by receiver 16 in order to analyze
the system overall transmission characteristics. According to the
present invention, simulator 10 acts to obtain both the input radio
signal and the channel scattering function, as described below, and
then convolves the two inputs to obtain an output transmission that
simulates an actual radio frequency transmission through the
ionosphere with both amplitude and phase modulation effects of a
fading channel. The convolution is implemented by multiplying the
inverse FFT of the scattering function and the FFT of the input
radio signal.
The tap-gain channel simulation embodies the notion that
measurements of channel fading statistics lead to Rayleigh fade
characteristics. However, channel probe measurements of the high
frequency radio channels have led to the concept of a "channel
scattering function," as depicted in Proakis, for example. This
channel scattering function is comprised of both a delay-spread
function and a Doppler-spread function, which affect signal
decorrelation as noted above. The present simulator recognizes that
the channel scattering function gives a complete description of the
frequency-time effects of the channel so that the fading, both in
time and frequency, can be determined from the inverse Fourier
transforms of the scattering function.
The present simulator provides a scattering function in a manner
that can be meaningfully applied to a signal propagating through
the channel. This is done by generating a series of "realizations"
of the channel that can be sequentially applied to the signal. The
channel realizations are derived by generating a two-dimensional
power spectral density function, obtained by taking the square root
of the scattering function, e.g. as defined by Proakis or Bello,
and multiplying each point by a complex number of which the real
and imaginary parts are each independent Gaussian random variables
with zero mean and unit variance. The two-dimensional inverse-FFT
of this complex-valued channel realization yields a matrix of
channel coefficients to provide a complete frequency-time
description of the channel that can now be multiplied with the
matrix representing the FFT of the signal to realize signal
decorrelations in both time and frequency.
It should be recognized that a given fading channel characteristic
is constant only over short time periods. Thus, the signal stream
is broken into time segments, each having a duration selected to
match the length of the delay axis of the scattering function so
that the segments of transformed signal match the frequency
resolution of the scattering function. Each segment of the input
signal is then Fourier decomposed into a set of coefficients
representing the signal frequency content as a function of time.
The matrix of channel coefficients is multiplied with a
corresponding matrix of signal coefficients to provide an output
matrix with coefficients functionally related to the received
signal with imposed fading channel effects The inverse FFT of the
output matrix is obtained to generate an output signal with a
sequence of fades consistent with the channel described by the
channel scattering function.
Thus, successive signal segments are processed with successive
channel realizations. By reassembling the output signal segments in
time, a complete received signal is obtained having both frequency
and time dependent characteristics, a result that is not available
from tapped delay-line models of the fading channels.
Referring now to FIG. 3, there is shown a flow diagram of the
process for simulating a fading channel according to the present
invention. Computer 12 (FIG. 2) calls the channel model function
which inputs 24 a Doppler-spread function and a delay-spread
function, similar to the square root of the function illustrated in
Figure with delay and Doppler values selected to be representative
of specified channel characteristics for which a system is to be
designed. A complex-valued channel realization is generated 26 by
multiplying the square root of the real channel scattering function
by a complex number of which the real and imaginary parts are each
independent Gaussian random variables with zero mean and unit
variance. The 2D inverse-FFT 28 of the complex channel then
provides a matrix of coefficients functionally related to the
fading vs. frequency characteristics of the channel.
The signal function call 32 first samples 34 the output high
frequency radio signal from transmitter 14 (FIG. 2) at a selected
rate for digitizing the signal, e.g., at 10 kHz, for a
representative period of time, e.g., about 4 seconds, and segments
36 the signal into time lengths, e.g., of about 0.2 seconds. The
FFT of each signal segment is then formed 38 to yield a matrix of
signal coefficients for each segment to represent the Fourier
amplitude vs. frequency.
The matrix of channel coefficients multiplies 42 the matrix of
signal coefficients and the inverse FFT of the resulting matrix is
then obtained 44 to represent a segment of received signal. The
successive channel realizations and signal segments are similarly
processed and the received signal segments are reassembled 46 in
time to form the degraded output signal 48 for input to receiver 16
(FIG. 2). The output from receiver 16 may be used by computer 12 to
formulate a revised channel model for further testing.
FIG. 4 depicts a schematic in block diagram form of one embodiment
of hardware for implementing the flow diagram shown in FIG. 3.
Transmitter 50 outputs a signal for input to A/D converter 52 where
the signal is digitized and formed into segments each having a
selected length. Each segment is input to FFT converters 54. which
generate parallel outputs to multipliers 62. Complex scattering
function generator 56 provides an output to inverse-FFT converters
58 which generate parallel outputs to multipliers 62. Multipliers
62 multiply the matrix coefficients formed in converters 54 and 58
to provide parallel outputs to inverse FFT converter 64. Each
resulting output from converter 64 represents a segment of the
received signal and the outputs are provided to multiplexer 66.
Multiplexer 66 sequentially outputs the input segments to assemble
the received radio signal. The output signal is processed through
digital-to-analog converter 68 for input to receiver 72 as the
degraded high frequency received radio signal.
The fading channel simulator herein described is extremely
versatile with the capability to simulate fading channels with
continuously variable characteristics. Dispersive effects can be
applied easily to the simulator by forcing a frequency dependent
phase shift across the transformed channel scattering function.
Quadratic phase changes might be used to simulate HF ionospheric
channels, or inverse-frequency-cubed changes to simulate
transionospheric satellite channels. Further, both white noise and
frequency dependent noise might be digitally applied to the input
signal stream either in the transformed or untransformed domain.
Although the above description is directed to HF radio
transmissions, the method and apparatus are generally applicable to
communication systems affected by fading multipath channels, e.g.,
cellular phone systems, VHF line-of-sight systems, and the
like.
The foregoing description of preferred embodiments of the invention
have been presented for purposes of illustration and description.
It is not intended to be exhaustive or to limit the invention to
the precise form disclosed, and obviously many modifications and
variations are possible in light of the above teaching. The
embodiments were chosen and described in order to best explain the
principles of the invention and its practical application to
thereby enable others skilled in the art to best utilize the
invention in various embodiments and with various modifications as
are suited to the particular use contemplated. It is intended that
the scope of the invention be defined by the claims appended
hereto.
* * * * *