Signal Processing - September 2017 - 123

TOA measurements from cellular CDMA and LTE signals.
Moreover, a framework for navigating exclusively with cellular signals in the case of GNSS unavailability was discussed.
Experimental results were presented demonstrating a UAV navigating with cellular CDMA signals using the proposed framework and the presented CDMA SDR. An RMS position error
of 9.39 m over a 512-m trajectory was achieved using only two
cellular CDMA BTSs. Experimental results were presented demonstrating a ground vehicle navigating exclusively with LTE signals using the proposed LTE SDR in an environment in which
the LTE signals experienced multipath. The LTE's CRS signal
was utilized to estimate the channel impulse response, and subsequently alleviate the multipath. The LTE experimental results
demonstrated the robustness of the proposed LTE SDR in a
multipath environment. An RMS position error of 9.32 m over
a 2-km trajectory was achieved using only two LTE eNodeBs.
While these experimental results do not seem impressive
compared to GPS, one needs to consider several factors affecting the accuracy achieved with the experimental results presented in this article. First, due to hardware or UAV payload
limitations, only two CDMA BTSs (two LTE eNodeBs) were
used in the CDMA (LTE) experiments, respectively, compared to eight GPS SVs. Second, the BTS and eNodeB layout
offered poor geometric diversity. Third, the vehicle dynamics
(UAV and ground vehicle) were assumed to evolve according
to a fixed dynamical model. Also, the statistics of the process
noise driving the clock states of the vehicle-mounted SDRs,
CDMA BTSs, and LTE eNodeBs were chosen according to
assumed oscillator qualities, leading to dynamical and statistical model mismatches in the EKF. The navigation accuracy with cellular signals could be significantly improved by
listening to more BTSs and eNodeBs, which would inherently offer better geometric diversity. Furthermore, to reduce
dynamical model mismatches in the EKF, an INS could be
used to propagate the position and velocity states of the UAV
and ground vehicle. To reduce statistical model mismatches
in the EKF, the statistics of the process noise driving the
BTSs' and eNodeBs' oscillators could be characterized a
priori over a long period of time or estimated on the fly via
an adaptive filter [27].
On the other hand, when GNSS signals were available, the
article demonstrated how exploiting the abundance and geometric diversity of cellular transmitters could significantly
improve the navigation solution over that of a standalone GPS.
To this end, a UAV fused pseudoranges from CDMA BTSs,
LTE eNodeBs, and GPS SVs, achieving a superior navigation
solution when compared to a standalone GPS, particularly in
the vertical direction.
While the potential of exploiting cellular signals for accurate navigation via SDRs was demonstrated in this article,
hardware and payload limitations remain a major challenge.
This prevented hearing more CDMA BTSs and LTE eNodeBs
in the receiver's environment. Future work could focus on
optimizing the receiver to increase the number of processed
signals and to reduce the size, weight, and power issues of the
cellular navigation SDRs, making them embeddable on mobile

devices. Moreover, this article discussed that, compared to
GNSS signals, cellular signals are received at significantly
higher C/N 0 and are available and usable indoors. Future work
could study enabling indoor navigation via cellular signals.
Finally, by diversifying the portfolio of signals used in
producing a navigation solution beyond GNSS signals, one
achieves security against malicious GNSS jamming and
spoofing attacks. If GNSS signals are jammed, the receiver
could continue navigating with non-GNSS signals. If GNSS
signals are spoofed, the receiver could detect such spoofing
by cross-checking against its portfolio of signals of opportunity (e.g., cellular signals). In the future, the pursuit of GNSS
spoofing detection and mitigation via cellular signals can
be explored.

Acknowledgment
This work was supported in part by the Office of Naval
Research under grant N00014-16-1-2305.

Authors
Zaher (Zak) M. Kassas (zkassas@ieee.org) received his B.E.
degree in electrical engineering from the Lebanese American
University, Beirut, in 2001, M.S. degree in electrical and computer engineering (ECE) from The Ohio State University in
2003, M.S.E. degree in aerospace engineering from the
University of Texas at Austin in 2010, and Ph.D. degree in ECE
from the University of Texas at Austin in 2014. From 2004 to
2010, he was a research and development engineer with the
Control Design and Dynamical Systems Simulation Group at
National Instruments Corp. He is an assistant professor at the
University of California, Riverside, and director of the
Autonomous Systems Perception, Intelligence, and Navigation
Laboratory. His research interests include optimal estimation,
navigation, software-defined receivers, autonomous vehicles,
and intelligent transportation systems.
Joe Khalife (jkhal001@ucr.edu) received his B.E. degree in
electrical engineering and M.S. degree in computer engineering
from the Lebanese American University, Beirut, in 2011 and
2014, respectively. He is a Ph.D. degree student at the University
of California, Riverside, and a member of the Autonomous
Systems Perception, Intelligence, and Navigation Laboratory.
His research interests include opportunistic navigation, autonomous vehicles, and software-defined receivers.
Kimia Shamaei (ksham002@ucr.edu) received her B.S.
and M.S. degrees in electrical engineering from the University
of Tehran, Iran, in 2010 and 2013, respectively. She is a Ph.D.
degree candidate at the University of California, Riverside,
and a member of the Autonomous Systems Perception,
Intelligence, and Navigation Laboratory. Her current research
interests include analysis and modeling of signals of opportunity and software-defined receivers.
Joshua Morales (jmora047@ucr.edu) received his B.S.
degree in electrical and computer engineering with high honors from the University of California, Riverside, in 2014
where he is currently a Ph.D. degree student and a member of
the Autonomous Systems Perception, Intelligence, and

IEEE SIGNAL PROCESSING MAGAZINE

September 2017

123

Table of Contents for the Digital Edition of Signal Processing - September 2017