**First
passive radar observations of auroral E region irregularities**

Lind,
F. D., J.D. Sahr, and D.M. Gidner

Geophys.
Res. Lett. Vol. 26 , No. 14 , July 15, 1999; p. 2155

On
August 27, 1998 we observed auroral E-region irregularities using the Manastash
Ridge Radar system. These are the first radio science results from passive
radar observations of commercial FM broadcasts near 100 MHz. These are
also the first reported observations of 1.5 m irregularities in 15 years.
The data show scatter most consistent with type-1 and type-3 ion-acoustic
turbulence as observed by 50 MHz and 140 MHz radars. The instrument is
extraordinarily low in cost and provides data free of range and doppler
velocity aliasing problems. The range and velocity resolution achieved
by this radar is substantially superior to that of other comparable coherent
scatter radars, and the time resolution is similar.

**Optimal
truncation criterion for application of singular value decomposition to
ionospheric tomography**

Zhou,
C., E.J. Fremouw, and J.D. Sahr

Radio
Science. vol.34, no.1; Jan.-Feb. 1999; p.155-66

In
this paper, we present a generic method to solve the subspace-oriented
estimation problem. We have optimized our approach by taking account of
the a priori and a posteriori covariances of both the data and the model
parameters in general linear inverse notations. In our work, singular value
decomposition (SVD) was employed to provide a robust optimal solution.
In computation of the generalized matrix inverse, a very simple truncation
criterion on the singular value (SV) spectrum was set up which guarantees
the minimal variance of the estimate. This algorithm based on SVD produces
an optimal estimate independent of computing resources. Specifically,
we applied this method to studies of ionospheric tomography by inverting
total electron content (TEC), which may be measured by means of satellite
beacons. We processed two simulated cases. The residual variances of the
a posteriori covariances of the model parameters were used as the measure
to evaluate the uncertainties of the estimates. Our examples indicate that
this algorithm can resolve about 60% of the a priori variance while achieving
a significant decrease of the computation time by truncation of the SV
spectrum.

**Recursive
structures and finite impulse response implementations of linear multiuser
detectors for an asynchronous CDMA system**

Uppala,
S.V., and J.D. Sahr

IEEE
Journal on Selected Areas in Communications. vol.16, no.9; Dec. 1998; p.1736-46

We
present order-recursive structures for implementing the linear decorrelating
and linear minimum mean square error (MMSE) detectors in a computationally
efficient manner. These infinite memory length, linear time invariant,
noncausal systems can be approximated by implementing them as K-input K-output
finite impulse response (FIR) filters. We developed a linear multiuser
receiver architecture called a recursive linear multiuser detector which
has lower computational and memory requirements then an equivalent tap
delay line FIR filter implementation. We establish the tradeoff between
window length and bit error rate and find that relatively small window
size yields performance indistinguishable from the ideal decorrelating
detector and the ideal MMSE detector.

**Computing
probability of errors of linear multiuser detectors in white Gaussian noise
and Rayleigh fading channels**

Uppala,
S.V., and J.D. Sahr

ICC
'98. 1998 IEEE International Conference on Communications. Conference Record.
Affiliated with SUPERCOMM'98 (Cat. No.98CH36220). IEEE, New York, NY, USA;
1998; 3 vol. xxxvii+1838 pp. p.360-4 vol.1

We
present a tool for rapidly computing the bit error rate of linear multiuser
detectors. Computing bit error rates of linear multiuser detectors involves
averaging the probability of error over the bits of all the users which
is an exponentially complex operation. The technique presented in this
paper is based on computing the Laplace inversion integral of the moment
generating function of the variate which determines the probability of
error of a user. We apply this technique for computing the probability
of error of an arbitrary linear multiuser detector operating in an additive
white Gaussian noise and in a Rayleigh fading channel.

**Passive
radio remote sensing of the atmosphere using transmitters of opportunity**

Sahr,
J.D., and F. D. Lind

Radio
Science Bulletin. no.284; March 1998; p.4-7

Commercial
FM broadcasts near 100 MHz illuminate the natural environment with continuous,
high power illumination. By fortunate coincidence these broadcasts are
superb for radar applications. With proper signal processing and a bistatic,
synchronous receiver system, target range may be resolved to about one
kilometer, and Doppler spectra resolved to a few meters per second, with
integration times of the order 30 seconds. Such a system suffers no large
ambiguities in range or Doppler shift, an extremely attractive feature
in studies of meter scale irregularities in the ionosphere.

**The
Manastash Ridge radar: a passive bistatic radar for upper atmospheric radio
science**

Sahr,
J.D., and F. D. Lind

Radio
Science. vol.32, no.6; Nov.-Dec. 1997; p.2345-58

We
describe a novel method for radar remote sensing of the upper atmosphere
which relies upon commercial FM broadcasts near 100 MHz. These broadcasts
have high average power and excellent radar ambiguity function. With proper
processing we can study the spatial and temporal distribution and Doppler
spectrum with excellent and completely unambiguous resolution. Since this
passive system has no transmitter, there are enormous benefits in safety,
expense, shielding, antenna and receiver design, and licensing issues.
Some new problems are introduced, but these are solved with relatively
little expense. After presenting the technical basis for such a radar,
we describe an instrument that we are building at the University of Washington
to study high-latitude plasma irregularities in the E region.

**Performance
analysis of the decorrelating detector in the presence of synchronization
errors**

Uppala,
S.V., and J.D. Sahr

1997
IEEE Pacific Rim Conference on Communications, Computers and Signal Processing,
PACRIM. 10 Years Networking the Pacific Rim, 1987-1997 (Cat. No.97CH36060).
IEEE, New York, NY, USA; 1997; 2 vol. xxiii+1021 pp. p.18-21 vol.1

We
present a semi-analytic method for computing the bit error rate of the
decorrelating detector in the presence of phase and timing errors (synchronization
errors). These errors in effect introduce multiple access interference.
We first compute the moment generating function of the variate which determines
the probability of error of a user. It can be shown that the probability
of error is obtained by computing the Laplace inverse integral of a modified
version of the moment generating function. Helstrom (1986) used this approach
to analyze a BPSK system corrupted with intersymbol interference. In this
paper are show that this method can be used for computing the probability
of errors of users in the presence of multiple access interference. Due
to the analytic approach used systems with larger packet sizes and larger
user populations can be tackled.

**On
the design of quadratic filters with application to image processing**

Uppala,
S.V., and J.D. Sahr

IEEE
Transactions on Image Processing. vol.6, no.4; April 1997; p.608-14

We
present a systematic design method for quadratic filters, and show an example
for an edge detector. Our design maximizes the output signal-to-noise ratio
(SNR) constrained to detect desired signals while ignoring specified error
signals such as shot noise. This filter compares favorably with Sobel and
morphological detectors of similar complexity.

**Auroral
electrojet plasma irregularity theory and experiment: a critical review
of present understanding and future directions**

Sahr,
J.D., Fejer, B.G.

Journal
of Geophysical Research. vol.101, no.A12; 1 Dec. 1996; p.26893-909

We
review the experimental observations of meter scale plasma irregularities
in the auroral E region and the status of their theoretical understanding.
Most of the experimental data is derived from VHF radar scatter experiments,
but sounding rockets also provide crucial information not obtainable from
radars. Linear theories correctly predict the altitude of occurrence, strong
magnetic aspect sensitivity, marginal instability, and typical phase velocities.
Subsequent nonlinear theories have been developed to account for other
observed features but with less satisfying application. Further understanding
of auroral electrojet irregularities is impeded by precision limitations
of existing instruments, by radar data which may seem incompatible, by
the usually poor knowledge of the ambient conditions during these experiments,
and by some confusion in the nomenclature (e.g. "type 2") used to describe
the irregularities. We hope to clarify some of these experimental and theoretical
issues. We discuss observational "facts" that need to be refined and point
out weaknesses of existing theories or their common interpretations. Finally,
we outline some avenues for future experimental and theoretical pursuit.

**Probing
electric fields near sprites and jets using multiparameter radar and chaff**

Seliga,
T.A., J.D. Sahr, and R.H. Holzworth

IGARSS
'96. 1996 International Geoscience and Remote Sensing Symposium. Remote
Sensing for a Sustainable Future (Cat. No.96CH35875). IEEE, New York, NY,
USA; 1996; 4 vol. lxxi+2383 pp. p.569-73 vol.1

The
discovery of electric discharges above thunderstorms has generated intense
scientific interest. Studies of these `sprites' and `jets' have focused
on their characterization by optical and radio techniques, with radar measurements
of the causative storms providing insight into related weather. The authors
describe a method for investigating the electric field structure above
thunderstorms using ground-based radar to observe chaff dispersed by rockets.
Slender conducting or dielectric chaff will generally align itself with
the ambient electric field. This alignment is readily detected by appropriate
configurations of polarimetric radar(s) such as are now used in meteorology
to observe the nature and motion of hydrometeors. This is especially convenient
as it permits the thunderstorms associated with sprites and jets to be
characterized with the same experimental facility. This paper renders a
preliminary examination of factors such experiments would entail and features
that a multiparameter radar might utilize to probe chaff dispersed by small
rockets. Monostatic and bistatic radar measurements of scatter from chaff
provide a powerful tool to study electric fields associated with sprites
and jets as well as other atmospheric electric fields.

**Aperiodic
transmitter waveforms for spectrum estimation of moderately overspread
targets: new codes and a design rule**

Uppala,
S.V., and J.D. Sahr

IEEE
Transactions on Geoscience and Remote Sensing. vol.34, no.5; Sept. 1996;
p.1285-7

Presents
several new classes of aperiodic waveforms suitable for FFT-based spectral
analysis of moderately overspread radar targets. In previous work, the
authors explained the principle of operation with a particular code. They
now provide codes with higher performance as well as a design rule for
the user to apply these codes to real targets.

**Application
of closure phase and self-calibration to radar interferometric imaging
of atmospheric and ionospheric irregularities**

Sahr,
J.D.

Journal
of Atmospheric and Terrestrial Physics. vol.58, no.8-9; June-July 1996;
p.959-64

Describe
extensions to radar interferometric techniques used to study irregularities
in the upper atmosphere, based on techniques developed by radio astronomers.
In particular, the techniques of 'self-calibration' and 'closure phase'
offer great promise to radars that measure properties of stochastic geophysical
targets. The author shows how useful extensions arrive and comments upon
implications of such techniques for the analysis of existing data and the
design of new instruments.

**Correlative
range - Doppler detectors and estimators in bistatic radar using commercial
FM broadcasts**

Hall,
P.W.

MS
thesis, Univ. of Wash., Seattle, 1995

Commercial FM radio broadcasts entertain us with music
and keep us informed of current events. With the help of digital
signal processing and fortunate geography we learn more from FM radio than
today's news - we can use commercial FM broadcasts as a radar waveform
to probe the auroral electrojet. As the earth moves through the Solar
Wind, charged particles set up a current in the ionosphere at about 70
degrees latitude and a height of 100 to 120 km. This current carries
a million amperes and is responsible for the Northern Lights, communications
impairments and sometimes affects power distribution networks. Using
two receivers on either side of the Cascade Mountains in Washington state,
we correlate the FM broadcast scatter off the auroral electrojet to learn
more about the physics of this important part of our ionosphere.
the ambiguity function of FM broadcasts is measured and agrees w9ith our
thoretical model but differences are found between stations. The
"dead-air" between spoken words greatly affects the range and doppler resolution
available. We also propose an unbiased and consistent estimator for
the four product correlation which is computationally efficient.

**Three-wave
coupling in the auroral E-region**

Sahr,
J.D., and D.T. Farley

Annales
Geophysicae. vol.13, no.1; 1 Jan. 1995; p.38-44

Analysis
of a simple model for two-stream weak fluid turbulence suggests a straightforward
mechanism for the excitation of plasma wave modes whose phase velocities
are roughly half the ion-acoustic speed. This slow wave is linearly stable
but is excited through three-wave interaction with two other waves, at
least one of which is an unstable two-stream wave. All three waves satisfy
the linear dispersion relation. Careful consideration of the coupling coefficients
and resonance conditions shows that the driven (linearly stable) modes
have k-vectors that are perpendicular to the local magnetic field and are
at some large angle to the Hall drift direction. These waves have phase
velocities in the neighborhood of half the ion-acoustic speed and they
could be responsible for type-3 radar echoes observed at 50 MHz. This mechanism
cannot, however, generate fast waves moving at type-4 velocities.

**A
new radar technique for remote sensing of atmospheric irrecgularities by
passive observation of the scattering of commercial FM broadcasts**

Hansen,
J.M.

MS
thesis, Univ. of Wash., Seattle, 1994

This
paper investigates the properties of public broadcast FM radio and how
it can be used as a technique for remote sensing. In particular we
study how three different receiving systems can make use of FM broadcasts
to achieve high range and Doppler velocity resolution of scatter from atmospheric
irregularities. Calculations of system performance are provided based
upon other observations of auroral electrojet irregularities. The
ambiguity function of FM radio signals is measured using simulations and
from this further simulations are performed to model the three methods
of detecting targets. The results presented here are promising, indicating
that further study and actual experimental implemention is warranted.

**Spectrum
estimation of moderately overspread radar targets using aperiodic transmitter
coding**

Uppala,
S.V., and J.D. Sahr

Radio
Science. vol.29, no.3; May-June 1994; p.611-23

Presents
a new method of spectrum estimation of moderately overspread targets using
a pulse-to-pulse method, with nonuniform pulse spacing. This method is
especially applicable to VHF observations of the auroral electrojet irregularities,
for which the ratio of the target range extent and the autocorrelation
time exceeds the speed of light by a modest amount. For a suitable choice
of pulse spacings, the range and frequency aliasing can be uniquely unravelled,
and the estimation algorithm can be expressed in terms of the fast Fourier
transform, for excellent computational efficiency. This technique should
find application to other moderately overspread targets, such as Doppler
weather radar and equatorial spread F.

**Simulated
annealing searches for long binary phase codes with application to radar
remote sensing**

Sahr,
J.D., E.R. Grannan

Radio
Science. vol.28, no.6; Nov.-Dec. 1993; p.1053-5

The
authors apply the technique of simulated annealing to search for long discrete
codes with special autocorrelation and ambiguity function properties, suitable
for radar applications. This technique provides access to codes which are
far too long for exhaustive search, and to codes whose correlation properties
do not lend themselves to algebraic investigation. The authors illustrate
the process with two new kinds of binary phase codes.

**Observations
of 3-m auroral irregularities during the ERRRIS campaigns**

Sahr,
J.D., D.T. Farley, W.E. Swartz, J.F. Providakes, and R.F. Pfaff

Journal
of Atmospheric and Terrestrial Physics. vol.54, no.6; June 1992; p.809-18

In
the late winter of 1988 and 1989, three NASA sounding rockets were flown
through the auroral electrojet from ESRANGE (Sweden) as part of the E-region
Rocket-Radar Instability Study (ERRRIS). Many ground-based instruments
supported these flights, including the EISCAT, STARE, and CUPRI radars,
as well as all-sky cameras, riometers, and magnetometers. The authors summarize
the observations of the Cornell University Portable Radar Interferometer
(CUPRI), which detected coherent backscatter from 3-m irregularities in
the auroral E-region. Twenty hours of power spectra and interferometry
data are available, and, during the 1989 campaign, three weeks of nearly
continuous Range-Time-Intensity (RTI) and first moment data were recorded.

**The
E-region Rocket/Radar Instability Study (ERRRIS): scientific objectives
and campaign overview**

Pfaff,R.F.,
J.Sahr, J.F.Providakes, W.E.Swartz, D.T.Farley, P.M.Kintner, I.Haggstrom,
A. Hedberg, H.Opgenoorth, G.Holmgren, A. McNamara, D.Wallis, B. Whalen,
A.Yau, S. Watanabe, F.Creutzberg, P. Williams, E.Nielsen, K Schlegel,
and T.R. Robinson

Journal
of Atmospheric and Terrestrial Physics. vol.54, no.6; June 1992; p.779-808

The
E-region Rocket/Radar Instability Study (Project ERRRIS) investigated in
detail the plasma instabilities in the low altitude (E-region) auroral
ionosphere and the sources of free energy that drive these waves. Three
independent sets of experiments were launched on NASA sounding rockets
from Esrange, Sweden, in 1988 and 1989, attaining apogees of 124, 129 and
176 km. The rocket instrumentation included electric field instruments
(d.c. and wave), plasma density fluctuation ( delta n/n) receivers, d.c.
fluxgate magnetometers, energetic particle detectors (ions and electrons)
ion drift meters, and swept Langmuir probes to determine absolute plasma
density and temperature. The wave experiments included spatially separated
sensors to provide wave vector and phase velocity information. All three
rockets were flown in conjunction with radar backscatter measurements taken
by the 50 MHz CUPRI system, which was the primary tool used to determine
the launch conditions. Two of the rockets were flown in conjunction with
measurements taken by the EISCAT incoherent scatter radar.

**The
altitude of type 3 auroral irregularities: radar interferometer observations
and implications**

Sahr,
J.D., D.T. Farley, W.E. Swartz, and J.F. Providakes

Journal
of Geophysical Research. vol.96, no.A10; 1 Oct. 1991; p.17805-11

VHF
coherent scatter radars at auroral latitudes have observed scatterers with
narrow power spectra at sub-ion acoustic mean Doppler shifts. These spectra
have been designated 'type 3'. The mean Doppler shift of these waves is
often near the atomic or, less frequently, the molecular gyro frequencies.
These type 3 echoes have been attributed to an electrostatic ion-cyclotron
instability in the upper E region (h>140 km), where the ion collision frequency
becomes low enough to permit ion gyromotion. Interferometric analysis
of recent coherent radar observations with the Cornell University portable
radar interferometer shows that type 3 events occur at typical electrojet
altitudes (100-120 km), however. The cause of the observed type 3 echoes
hence remains a mystery.

**Ion
cyclotron harmonics in auroral radar echoes: real effect or analysis artifact?**

Farley,
D.T., J.D. Sahr, and J.F. Providakes

Geophysical
Research Letters. vol.18, no.4; April 1991; p.709-12

Certain
analyses of Canadian radar studies of auroral electrojet echoes have apparently
shown spectra with multiple ion cyclotron harmonic peaks. On the basis
of these, it has been argued that ion cyclotron effects must be important
for all echo types seen by VHF radars, not just type 3. The authors reject
this conclusion for two reasons: (1) at the altitude of auroral backscatter
(approximately 100-120 km) the ion-neutral collision frequency is much
higher than the gyro frequency-ions are essentially unmagnetized; (2) the
authors show that the 'high resolution' signal processing algorithm employed
does not distinguish between echoes with and without harmonic structure.
With a simulation, they show that the algorithm itself can impose apparent
quasi harmonic structure onto the spectrum of random noise.

**Removal
of aliasing in pulse-to-pulse Doppler radar measurements**

Sahr,
J.D., D.T. Farley, and W.E. Swartz

Radio-Science.
vol.24, no.6; Nov.-Dec. 1989; p.697-704

Frequency
aliasing is a problem often encountered in pulsed radar studies of some
deep targets, such as the radar aurora. It is possible to resolve the aliasing
ambiguities by transmitting closely spaced pulse pairs, rather than single
pulses. Instead of the usual power spectrum, the authors compute a cross
spectrum of the received signals. The amplitude of this cross spectrum
is just the normal, possibly aliased, power spectrum, and the phase determines
the aliasing, if any, at each frequency bin. The technique works as long
as the spectrum is not seriously wrapped around upon itself, that is, as
long as each frequency bin is dominated by signals with a single Doppler
shift, no matter what that Doppler shift may be. The price paid for this
extra information is the addition of radar clutter, uncorrelated signals
from unwanted ranges, but this price may be acceptable.

**Observations
of auroral E-region plasma waves and electron heating with EISCAT and a
VHF radar interferometer**

Providakes,
J., D.T. Farley, B.G. Fejer, J. Sahr, W.E. Swartz, I. Haggstrom,
A. Hedberg, and J.A. Nordling

Journal
of Atmospheric and Terrestrial Physics. vol.50, no.4-5; April-May 1988;
p.339-56

Two
radars were used simultaneously to study naturally occurring electron heating
events in the auroral E-region ionosphere. During a joint campaign
in March 1986 the Cornell University Portable Radar Interferometer (CUPRI)
was positioned to look perpendicular to the magnetic field to observe unstable
plasma waves over Tromso, Norway, while EISCAT measured the ambient conditions
in the unstable region. On two nights EISCAT detected intense but short
lived (<1 min) electron heating events during which the temperature
suddenly increased by a factor of 2-4 at altitudes near 108 km and the
electron densities were less than 7*10/sup 4/ cm/sup -3/. On the second
of these nights CUPRI was operating and detected strong plasma waves with
very large phase velocities at precisely the altitudes and times at which
the heating was observed. The altitudes, as well as one component of the
irregularity drift velocity, were determined by interferometric techniques.
The authors conclude that the electron temperature increases were caused
by plasma wave heating and not by either Joule heating or particle precipitation.