1

PLEASE FIND THE FULL EXTENDED ARTICLE OF THIS WHITE PAPER HERE: [1]
(ACCEPTED FOR PUBLICATION AT THE IEEE SIGNAL PROCESSING MAGAZINE - SPECIAL ISSUE ON
“SIGNAL PROCESSING FOR THE INTEGRATED SENSING AND COMMUNICATIONS REVOLUTION”)

AFDM vs OTFS: A Comparative Study of Promising

Waveforms for ISAC in Doubly-Dispersive Channels
Hyeon Seok Rou1 , Giuseppe Thadeu Freitas de Abreu1 , Junil Choi2 ,
David González G.3 , Osvaldo Gonsa3 , Yong Lian Guan4 , Marios Kountouris5 .
1
School of Computer Science and Engineering, Constructor University, 28759 Bremen, Germany
2
School of Electrical Engineering, Korea Advanced Institute of Science and Technology, Daejeon, South Korea
3
Wireless Communications Technologies, Continental AG, 65936 Frankfurt/Main, Germany
4
School of Electrical and Electronic Engineering, Nanyang Technological University, 639798 Singapore
5
Communication Systems Department, EURECOM, Sophia-Antipolis, France
arXiv:2309.04998v3 [eess.SP] 27 Jun 2024

I. AUTHOR B IOGRAPHY
Hyeon Seok Rou (Graduate Student Member, IEEE) [hrou@constructor.university] is a Ph.D. Candidate at Constructor
University, Bremen, Germany, funded as a Research Associate at Continental AG on a researach project on 6G vehicular-
to-everything (V2X) integrated sensing and communications (ISAC). His research interests lie in the fields of ISAC, hyper-
dimensional sparse modulation schemes, B5G/6G V2X wireless communications technology, and Bayesian inference.
Giuseppe Thadeu Freitas de Abreu (Senior Member, IEEE) [gabreu@constructor.university] is a Full Professor of Electrical
Engineering at Constructor University, Bremen, Germany. His research interests include communications theory, estimation the-
ory, statistical modeling, wireless localization, cognitive radio, wireless security, MIMO systems, ultrawideband and millimeter
wave communications, full-duplex and cognitive radio, compressive sensing, energy harvesting networks, random networks,
connected vehicles networks, and many other topics. He has served as an editor for various IEEE Transactions, and currently
serves as an editor to the IEEE Signal Processing Letters and the IEEE Communications Letters.
Junil Choi (Senior Member, IEEE) [junil@kaist.ac.kr] is a (Named) Ewon Associate Professor with the School of Electrical
Engineering, Korea Advanced Institute of Science and Technology (KAIST), South Korea. From 2007 to 2011, he was a member
of the Technical Staff with the Samsung Advanced Institute of Technology (SAIT) and Samsung Electronics Company Ltd.,
South Korea, where he contributed to advanced codebook and feedback framework designs for the 3GPP LTE/LTE-Advanced
and IEEE 802.16m standards. His research interests include the design and analysis of massive MIMO, mmWave communication
systems, distributed reception, and vehicular communication systems. He was the recipient of numerous awards from various
IEEE societies, and is currently an Associate Editor of IEEE Transactions on Wireless Communications, IEEE Transactions
on Communications, IEEE Communications Letters, and IEEE Open Journal of the Communications Society.
David González G. (Senior Member, IEEE) [david.gonzalez.g@ieee.org] is a Senior Research Engineer at Continental AG,
Germany, and has previously served at Panasonic Research and Development Center, Germany. His research interests include
aspects of cellular networks and wireless communications, including interference management, radio access modeling and
optimization, resource allocation, and vehicular communications. Since 2017, he has represented his last two companies as
delegate in the 3GPP for 5G standardization, mainly focused on physical layer aspects and vehicular communications.
Osvaldo Gonsa [osvaldo.gonsa@continental-corporation.com] is the Head of the Wireless Communications Technologies
group by Continental AG in Frankfurt, Germany. He has worked in research and standardization in radio access network,
serving as an advisor to the German Federal Ministry of Economy and Energy for the “PAiCE” projects, and currently as
the GSMA Advisory Board for automotive and the 6GKom project of the German Federal Ministry of Education and Research.
Yong Lian Guan (Senior Member, IEEE) [eylguan@ntu.edu.sg] is a Professor of Communication Engineering with the
School of Electrical and Electronic Engineering, Nanyang Technological University (NTU), Singapore, where he leads the
Continental-NTU Corporate Laboratory and the successful deployment of the campus-wide NTU-NXP V2X test bed. His
research interests broadly include coding and signal processing for communication systems and data storage systems. He is a
Distinguished Lecturer of the IEEE Vehicular Technology Society from 2021 to 2023 and an Editor of the IEEE Transactions
on Vehicular Technology.
Marios Kountouris (Fellow, IEEE) [kountour@eurecom.fr] is a Professor with the Communication Systems Department,
EURECOM, Sophia-Antipolis, France. His research interests include communications theory, machine learning for wireless
communications, low latency networking, and stochastic modeling and performance analysis. He was a recipient of the
Consolidator Grant of the European Research Council (ERC) in 2020 on goal-oriented semantic communication. He is an
AAIA Fellow. He has received several awards and distinctions at various IEEE venues. He has served as an Editor for
IEEE Transactions on Wireless Communications, IEEE Transactions on Signal Processing, and IEEE Wireless Communication
Letters. He is a Chartered Professional Engineer of the Technical Chamber of Greece.
 2

II. H ISTORY, MOTIVATION , AND SIGNIFICANCE OF THE TOPIC

Abstract: This white paper aims to briefly describe a proposed article that will provide a thorough comparative study
of waveforms designed to exploit the features of doubly-dispersive channels arising in heterogeneous high-mobility
scenarios as expected in the beyond fifth generation (B5G) and sixth generation (6G), in relation to their suitability to
integrated sensing and communications (ISAC) systems. In particular, the full article will compare the well-established
delay-Doppler domain-based orthognal time frequency space (OTFS) and the recently proposed chirp domain-based
affine frequency division multiplexing (AFDM) waveforms. Both these waveforms are designed based on a full delay-
Doppler representation of the time variant (TV) multipath channel, yielding not only robustness and orthogonality
of information symbols in high-mobility scenarios, but also a beneficial implication for environment target detection
through the inherent capability of estimating the path delay and Doppler shifts, which are standard radar parameters.
These modulation schemes are distinct candidates for ISAC in B5G/6G systems, such that a thorough study of their
advantages, shortcomings, implications to signal processing, and performance of communication and sensing functions
are well in order. In light of the above, a sample of the intended contribution (Special Issue paper) is provided below.

B5G and 6G wireless systems rely on extremely high-frequency (EHF) technology developed for the millimeter-wave
(mmWave) and Terahertz (THz) bands [2]–[4], to provide a variety of enhanced applications such as Internet-of-things (IoT),
edge computing, and smart cities [5], [6], with particular interest in heterogeneous high-mobility environments found in vehicle-
to-everything (V2X), cell-free/cooperative multi-cell, and non-terrestrial network (NTN) communication scenarios [7]–[9].
A promising enabling technology to satisfy the demands of B5G and 6G is ISAC, which combines sensing and communi-
cations functions under a single wireless system, with unified hardware and signal processing techniques [10]–[13].
Besides the greater support to services such as environment mapping and node localization, which is of fundamental impor-
tance to the aforementioned high-mobility scenarios, various enhancements are expected to be brought by ISAC technology,
including increased efficiency in spectrum, energy, and hardware costs.
High-mobility environments are a great challenge to wireless communications systems due to the resulting doubly-dispersive
wireless channel [14], [15], also referred to as TV multipath, or time-frequency selectivity. Although the delays and Doppler
shifts can be estimated [16]–[18], the effect of such a scattering environment onto received signals is severe inter-carrier
interference (ICI) [19], [20], which can drastically decrease the communication performance under conventional of even
highly effective modulation schemes such as orthogonal frequency division multiplexing (OFDM). And despite various clever
contributions that have been proposed to mitigate the issue [21]–[23], hefty consequences in overhead and increased complexity
are inevitable.
Motivated by the above challenge, as well as the potential of ISAC technology within B5G and 6G systems, novel waveforms
have recently been proposed which are inherently robust to high-mobility environments, thanks to the orthogonality they
maintain in the doubly-dispersive channel. One of the best-known and investigated methods with such features is the OTFS
waveform [24], [25], which leverages the inverse symplectic finite Fourier transform (ISFFT) [26] to modulate a two-dimensional
(2D) grid of information symbols directly onto the delay-Doppler domain. The OTFS scheme gained great attention as a
promising candidate for high-mobility B5G systems, thanks to the demonstrated superior performance it achieves in comparison
to other proposed waveforms [27]–[29].
Alternatively, chirp domain-based multicarrier waveforms have also been investigated, which are attractive due to their
inherent spread-spectrum property and potential for full-duplex operations [12], [30], [31]. This effort led to the proposal of
several novel modulation schemes for doubly-dispersive wireless channel [32]–[34], including the recently proposed AFDM
waveform [35], [36], which leverages the inverse discrete affine Fourier transform (IDAFT) [37] to modulate information
symbols into a “twisted” time-frequency domain in order to achieve the desired delay-Doppler orthogonality. This particular
feature of AFDM, as well as other properties such as the full-diversity guarantee, optimizable parametrization, increased
throughput, and reduced computational dimension [36], promotes the AFDM approach as a strong contender to OTFS.
It was quickly noticed that the full delay-Doppler representation of the channel in OTFS and AFDM inherently conveys the
velocity and range information of the scatterers in the form of the respective multipath delays and Doppler-shifts, such to
imply significant benefits in terms of ISAC. This was followed by a plethora of OTFS-based ISAC techniques being proposed
to extract the delay and Doppler parameters of the resolvable paths directly from the channel state information (CSI) [38]–[40],
which have been shown to approach the sensing performances of the full OFDM and frequency modulated continuous wave
(FMCW) radars, with a higher robustness to mobility and achievable capacity [41].
Naturally, the AFDM waveform is also expected to be a very promising candidate for ISAC in doubly-dispersive environments
[12] with similarities and advantages over OTFS, in addition to the natural relationship between chirp waveforms and radar
signal processing. However, very few works have been published so far on AFDM-based ISAC [42], which is likely due to the
fact that the AFDM technique was only very recently proposed. In view of all the above, the full article here described aims to
offer a thorough analysis on the future of ISAC in heterogeneous high-mobility environments, in the form of a comprehensive
comparison of AFDM as a rising competitor, and OTFS as its leading state-of-the-art (SotA) alternative.
 3

The proposed contributions of the full article are:

• An introduction to ISAC, especially in high-mobility environments, highlighting the motivation, challenges, and implica-
tions to the consequent signal processing.
• A SotA survey of the novel waveforms designed for high-performance communication in doubly-dispersive channels,
exemplified by the AFDM and OTFS waveforms.
• A thorough comparison and analysis of the SotA ISAC techniques leveraging the novel waveforms and other relevant
methods such as FMCW radar and OFDM, highlighting various factors such as computational complexity, radar resolution,
spectral efficiency, pilot overhead, etc.
• A conclusive summary with the potential research directions and remaining challenges of ISAC in high-mobility environ-
ments with doubly-dispersive channels.

III. O UTLINE OF THE PROPOSED S PECIAL I SSUE PAPER

A. Background on Communications over Doubly-Dispersive (TV Multipath) Channels
Consider a wireless channel in a high-mobility scattering environment with P resolvable propagation paths, where each p-th
path with p ∈ {1, · · ·, P } is respectively described by a corresponding channel fading coefficient hp ∈ C, delay τp ∈ [0, τ max ],
and Doppler shift νp ∈ [−ν max , +ν max ]. The consequent channel impulse response (CIR) of a TV multipath channel is typically
described by a time-variant impulse response function in the time-delay domain,
PP
g(t, τ ) ≜ p=1 hp ·ej2πνp t ·δ(τ − τp ), (1)
where t and τ respectively denote the instantaneous time and delay, and δ(x) is the unit impulse function.
Alternatively, the CIR g(t, τ ) in the time-delay domain can be represented in the dual domains [15, Ch.2] by leveraging the
Fourier transform (FT) F{f (x)}. For example, the CIR can be represented as the time-variant transfer function g TF (t, f ) or
the delay-Doppler spread function g DD (τ, ν), by performing a FT over the delay domain or the time domain, respectively, i.e.,
PP
g TF (t, f ) ≜ F g(t, τ ) = p=1 hp·ej2πνp t ·e−j2πτp f, (2)
 PP
g DD (τ, ν) ≜ F g(t, τ ) = p=1 hp ·δ(τ − τp ) · δ(ν − νp ), (3)

τ →f t→ν

where f and ν respectively denote the instantaneous frequency and Doppler shift.
The two representations of the doubly-dispersive CIR given by eq. (2) and eq. (3) are respectively visualized in Fig. 1,
highlight the continuously fast-varying nature in the time-frequency domain, in contrast to the sparse and impulsive property
in the delay-Doppler domain, with the implication of an inherently more efficient signal processing in the latter.
Following the above, given a discrete sequence of the transmit signal s[n] sampled at a rate of T with discrete time indices
n ∈ {0, · · ·, N − 1}, the received signal over the CIR in eq. (1) is given by
P∞ PP n τ

r[n] = ℓ=0 s[n − ℓ] · p=1 hp · ej2πνp N · δ( ∆τp − ℓ) + w[n], (4)
τ
where w[n] is the additive white Gaussian noise (AWGN) signal, and ℓ ≜ ∆τ is the normalized delay with resolution ∆τ
assumed to be sufficient such the normalized delays can be rounded to the nearest integer with negligible error.
Omitting the full derivations at this point, eq. (4) can also be described with as circular convolution by
PP
· s + w ∈ CN ×1 ,
ℓp


r ≜ Hs + w = p=1 hp ·Cp ·Np ·Π (5)
where r ∈ CN ×1 is the received signal vector; and Cp ∈ CN ×N is a diagonal matrix arising from a cyclic prefix, Np ∈ CN ×N
is a diagonal Doppler shift matrix, and Π ∈ CN ×N is the forward cyclic-shift matrix, respectively for the p-th path.
In other words, the resultant delay-Doppler representation of the doubly-dispersive channel is a superposition of P different
shifted diagonal matrices.

1 1 1
3

2.5 0.8 0.8 0.8

2
0.6 0.6 0.6
jg DD (=; 8)j
jg TF (t; f )j

jHAFDM j
jHOTFS j

1.5
0.4 0.4 0.4
1

0.2 0.2 0.2

0.5

0 0 0 0
0 0 0 0
0 0.5 0 2 0 0
20 20
2 1 #10!5 2
4 4 20 20
#106 1.5 4 40 40
6 40 40
2 6
8 Time 6 Delay
Frequency Doppler 60 60 60 60

(a) Time-frequency transfer function. (b) Delay-doppler spread function. (a) OTFS effective channel. (b) AFDM effective channel.
Fig. 1. Different representations of a doubly-dispersive CIR with P = 3 Fig. 2. Illustration of the OTFS and AFDM effective channels of the
resolvable paths (illustrated by unique colors) of random integer delays and exemplary CIR in Fig. 1, which is a simple case with integer multiple delays
Doppler-shifts, with carrier frequency of 5.9GHz and signal bandwidth of and Doppler shifts. In case of fractional Doppler shifts, the diagonals of the
10MHz (following the IEEE 802.11p vehicular environment [43]). effective matrices would become a thicker band, retaining other properties.
 4

B. Signal Models of OTFS and AFDM Waveforms

1) Orthogonal Time Frequency Space (OTFS) Modulation:
In OTFS, the information symbols are first placed directly on a K × L grid in the delay-Doppler domain, which is first
transformed into the time-frequency domain via the ISFFT [26], then pulse-shaped into the continuous time signal via the
Heisenberg transform (HT) [44]. Mathematically, the OTFS modulation process described by
Pulse-shaped HT ISFFT ≜x
z }| { z }| {
z }| {
OTFS
s ≜ vec G FK · FK XFL = (FL ⊗ G ) · vec(X) ∈ CKL×1 ,
tx H H H tx
(6)
where X ∈ CK×L is the information symbol matrix, Gtx ∈ CK×K is the diagonal transmit pulse-shaping filter matrix, and
FK ∈ CK×K and FL ∈ CL×L are respectively the K-point and L-point normalized discrete Fourier transform (DFT) matrix.
Given KL = N , the filtered and demodulated signal yOTFS after the convolution channel in eq. (4) is given by
yOTFS ≜ (FL ⊗ Grx )(HsOTFS + w) = HOTFS x + w̃ ∈ CN ×1 , (7)
rx K×K N ×1
where G ∈ C is the diagonal matched filter matrix, and w̃ ∈ C is the AWGN after a unitary transform.
The effective OTFS channel on the vectorized information symbol matrix x ∈ CN ×1 is therefore given by
PP
HOTFS ≜ p=1 (FL ⊗ Grx ) · hp Cp Np Πℓp · (FH N ×N
tx


L⊗G ) ∈C (8)
which is a block-wise pulse-shaped DFT and IDFT, respectively on the rows and columns of the channel from eq. (5).
2) Affine Frequency Division Multiplexing (AFDM):
On the other hand, the AFDM unlike the OTFS, directly multiplexes a one-dimensional vector of symbols unto a twisted
time-frequency chirp domain using the IDAFT [45], which is described by
IDAFT
z }| {
AFDM −1 −1 N ×1
s ≜ (ΛH F H H
c1 N c2 ) x = (Λc2 FN Λc1) x = A x ∈ C
Λ , (9)
where A ≜ Λc2 FN Λc1 ∈ CN ×N is the forward N -point discrete affine Fourier transform (DAFT) matrix with the N -point
2 2
normalized DFT matrix FN ∈ CN ×N , and Λci ≜ diag[e−j2πci (0) , · · · , e−j2πci (N −1) ] ∈ CN ×N is a diagonal discrete
quadratic-chirp matrix of central digital frequency ci .
Consequently, the demodulated AFDM signal over the convolution channel in eq. (5) is given by
PP
yAFDM = A + w̃ = HAFDM x + w̃ ∈ CN ×1 ,
ℓp

AFDM
p=1 hp Cp Np Π ·s (10)
where the effective AFDM channel on the information symbol vector x ∈ CN ×1 correspondingly defined as
PP
HAFDM ≜ p=1 A(hp Cp Np Πℓp )A−1 ∈ CN ×N .


(11)
The consequent effect of the DAFT/IDAFT and the block-wise DFT/IDFT unto the channel in eq. (5) can be observed in
Fig. 2, where the simple exemplary effective channels of the OTFS and AFDM schemes have been illustrated.

C. Radar Parameter Estimation via AFDM and OTFS Waveforms

In light of the above, to enable ISAC through the aforementioned waveforms, the extraction of the radar parameters of the
scattering targets from the inherent channel information is necessary.
As one example in this white paper, we provide the widely-utilized SotA method estimate the channel parameters, which
aims to solve the maximum likelihood (ML) problem on the channel parameters given by
PP 2
arg min y − p=1 H̃(hp , τp , νp )·x 2 , (12)
θ

where θ = {h1 , · · ·, hP , τ1 , · · ·, τP , ν1 , · · ·, νP } is the set of all 3P channel parameters corresponding to the channel gain,
delay, and Doppler shift of the P paths, respectively, and H̃(hp , τp , νp ) is the conditional channel matrix given the parameters
hp , τp , νp , defined as the channel in eq. (5).
The joint domain of 3P parameters, which are continuously complex on hp and bounded but still continuously real on τp
and nup , for all P , the above minimization problem is extremely challenging. One efficient widely utilized framework is the
iterative grid search or reduced-search methods [39], [46] which are known to exhibit a trade-off in complexity and resolution.
On the other hand, other methods have also investigated leveraged subspace analysis and radar-like signal processing methods
[47]–[49] to directly approximate the continuous parameters of eq. (12) from the effective channel information.
The various approaches and the advantages of the available methods will be discussed in the full article - specifically aiming
to provide relevant results and evaluation for the AFDM-based sensing which is currently lacking in the relevant literature.
 5

D. Comparative Analysis of OTFS and AFDM Schemes

Finally, in light of the above sections including the signal models and ISAC methods of the two waveforms, few differences
in the inherent signal processing between the two schemes can already be highlighted, such as the nature of the underlying
transforms and its effect on the effective channel. Such differences between the OTFS and the AFDM waveforms are briefly
provided in the below Table I, where a few other properties which have not been described in this overview white paper also
provided, including modulation complexity, asymptotic transmit diversity, and required pilot overhead.
We emphasize that the short comparative table below is only an anticipation of the thorough investigation to be expected in
the full article, which will additionaly compare sensing capacity, total throughput, channel/data estimation methods, etc., and
also include other relevant methods such as the OFDM and orthogonal chirp division multiplexing (OCDM).
In hand of the final comparative analysis, a conclusive summary and the identification of the remaining challenges and future
directions of research will be provided, to constructively close the full loop of the proposed article.
TABLE I
B RIEF COMPARISON OF WAVEFORM PROPERTIES OF AFDM AND OTFS.
Property AFDM OTFS
Transform Domain Multicarrier chirp Delay-Doppler
Eff. Channel Structure Shifted band Scattered diagonal
Modulation Complexity N log2 N + 2N (3/2)N log2 N
Asymptotic Diversity Full diversity [36] Order-1 [50]
Pilot Guard Overhead 1D zero-pad 2D zero-pad

R EFERENCES
[1] H. S. Rou, G. T. F. de Abreu, J. Choi, D. González G., M. Kountouris, Y. L. Guan, and O. Gonsa, “From OTFS to AFDM: A Comparative Study
of Next-Generation Waveforms for ISAC in Doubly-Dispersive Channels”, Accepted for publication at the IEEE Signal Processing Magazine:
Special Issue on “Signal Processing for the Integrated Sensing and Communications Revolution”, 2024.
[2] T. S. Rappaport et al., “Wireless communications and applications above 100 GHz: Opportunities and challenges for 6G and beyond,” IEEE Access,
vol. 7, pp. 78 729–78 757, 2019.
[3] Z. Chen et al., “A survey on Terahertz communications,” China Communications, vol. 16, no. 2, pp. 1–35, 2019.
[4] A. V et al., “Opportunities and challenges of mmWave NR,” IEEE Wireless Communications, vol. 26, no. 2, pp. 4–6, 2019.
[5] N.-S. Vo et al., “The key trends in B5G technologies, services and applications,” Mobile Net. and Appli., vol. 27, no. 4, pp. 1716–1718, 2022.
[6] W. Saad et al., “A vision of 6G wireless systems: Applications, trends, technologies, and open research problems,” IEEE Network, vol. 34, no. 3, 2020.
[7] J. Wu and P. Fan, “A survey on high mobility wireless communications: Challenges, opportunities and solutions,” IEEE Access, vol. 4, 2016.
[8] H. Zhou et al., “Evolutionary V2X technologies toward the internet of vehicles” Proceedings of the IEEE, vol. 108, no. 2, pp. 308–323, 2020.
[9] J. Wang et al., “High-mobility satellite-UAV communications: Challenges, solutions, and future research trends,” Commun. Mag., vol. 60, no. 5, 2022.
[10] J. A. Zhang, et al., “Enabling joint communication and radar sensing in mobile networks,” IEEE Commun. Surv. Tuts., vol. 24, no. 1, pp. 306–345, 2022.
[11] T. Wild et al., “Joint design of communication and sensing for beyond 5G and 6G systems,” IEEE Access, vol. 9, pp. 30 845–30 857, 2021.
[12] Q. Wang et al., “Towards integrated sensing and communications for 6G,” in International Symposium on Joint Communications & Sensing, 2022.
[13] D. K. Pin Tan et al., “Integrated sensing and communication in 6G: Motivations, use cases, requirements, challenges and future directions,” JC&S, 2021.
[14] D. W. Bliss and S. Govindasamy, Dispersive and doubly dispersive channels. Cambridge University Press, 2013, p. 341–364.
[15] G. A. Vitetta, D. P. Taylor, G. Colavolpe, F. Pancaldi, and P. A. Martin, Wireless communications: algorithmic techniques. John Wiley & Sons, 2013.
[16] W. Siebert, “A radar detection philosophy,” IRE Transactions on Information Theory, vol. 2, no. 3, pp. 204–221, 1956.
[17] M. A. Richards, Fundamentals of Radar Signal Processing. ¡country¿US¡/country¿: McGraw-Hill Professional, 2005.
[18] S. M. Patole et al., “Automotive radars: A review of signal processing techniques,” IEEE Sig. Proc. Mag. Proc., vol. 34, no. 2, pp. 22–35, 2017.
[19] T. Wang et al., “Performance degradation of OFDM systems due to doppler spreading,” IEEE Trans. Wireless Commun., vol. 5, no. 6, 2006.
[20] G. Matz et al., “Time-varying communication channels: Fundamentals, recent developments, and open problems,” in European Sig. Proc. Conf., 2006.
[21] J. Lu, X. Chen, S. Liu, and P. Fan, “Location-aware ICI reduction in MIMO-OFDM downlinks for high-speed railway communication systems,” IEEE
Transactions on Vehicular Technology, vol. 67, no. 4, pp. 2958–2972, 2018.
[22] L. Ma, H. Jia, S. Liu, and I. U. Khan, “Low-complexity doppler compensation algorithm for underwater acoustic OFDM systems with non-uniform
doppler shifts,” IEEE Communications Letters, vol. 24, no. 9, pp. 2051–2054, 2020.
[23] M. F. Keskin et al., “MIMO-OFDM joint radar-communications: Is ICI friend or foe?” IEEE JSTSP, vol. 15, no. 6, pp. 1393–1408, 2021.
[24] R. Hadani et al., “Orthogonal time frequency space modulation,” in IEEE Wireless Communications and Networking Conference, 2017, pp. 1–6.
[25] Z. Wei et al., “Orthogonal time-frequency space modulation: A promising next-generation waveform,” IEEE Wireless Comm., vol. 28, no. 4, 2021.
[26] J. M. Gracı́a-Bondı́a, “On homogeneous symplectic spaces,”, vol. 134. American Mathematical Soc., 1992, p. 93.
[27] K. R. Murali and A. Chockalingam, “On OTFS modulation for high-doppler fading channels,” in Inf. Theo. and Appli. Wkshop. (ITA), 2018, pp. 1–10.
[28] W. Anwar et al., “Performance analysis of various waveforms and coding schemes in V2X communication scenarios,” in IEEE WCNC, 2020, pp. 1–8.
[29] F. Wiffen et al., “Comparison of OTFS and OFDM in ray launched sub-6 GHz and mmWave line-of-sight mobility channels,” in IEEE PIMRC, 2018.
[30] J. R. Klauder et al., “The theory and design of chirp radars,” Bell System Technical Journal, vol. 39, no. 4, pp. 745–808, 1960.
[31] C. Capus et al., “The analysis of multiple linear chirp signals,” in IEEE Seminar on Time-scale and Time-Frequency Analysis and Appli., 2000.
[32] X. Ouyang and J. Zhao, “Orthogonal chirp division multiplexing,” IEEE Transactions on Communications, vol. 64, no. 9, pp. 3946–3957, 2016.
[33] T. Erseghe et al., “A multicarrier architecture based upon the affine fourier transform,” IEEE Trans. Commun., vol. 53, no. 5, pp. 853–862, 2005.
[34] M. Martone, “A multicarrier system based on the f-FT for time-frequency-selective channels,” IEEE Trans. Commun., vol. 49, no. 6, 2001.
[35] A. Bemani et al., “AFDM: A full diversity next generation waveform for high mobility communications,” in 2021 IEEE ICC Workshops, 2021, pp. 1–6.
[36] ——, “Affine frequency division multiplexing for next generation wireless communications,” IEEE Trans. Wireless Commun., pp. 1–1, 2023.
[37] J. J. Healy et al., Linear canonical transforms: Theory and applications. Springer, 2015, vol. 198.
[38] P. Raviteja et al., “Orthogonal time frequency space (OTFS) modulation based radar system,” in IEEE RadarConf, 2019, pp. 1–6.
[39] L. Gaudio et al., “Joint radar target detection and parameter estimation with MIMO-OTFS,” in IEEE RadarConf, 2020, pp. 1–6.
[40] M. F. Keskin et al., “Radar sensing with OTFS: Embracing ISI and ICI to surpass the ambiguity barrier,” in 2021 IEEE ICC Workshops, 2021, pp. 1–6.
[41] L. Gaudio et al., “On the effectiveness of OTFS for joint radar parameter estimation and communication,” IEEE Trans. Wireless Commun., vol. 19,
no. 9, pp. 5951–5965, 2020.
[42] Y. Ni et al., “An AFDM-based integrated sensing and communications,” in 2022 ISWCS, 2022, pp. 1–6.
[43] IEEE 1609 - Family of Standards for Wireless Access in Vehicular Environments (WAVE). U.S. Department of Transportation, 2013.
[44] W. F. G. Mecklenbräuker, “A tutorial on non-parametric bilinear time-frequency signal representations,” 1989.
[45] S.-C. Pei and J.-J. Ding, “Closed-form discrete fractional and affine fourier transforms,” IEEE Trans. Sig. Proc., vol. 48, no. 5, pp. 1338–1353, 2000.
[46] Z. Tang, Z. Jiang, W. Pan, and L. Zeng, “The estimation method of sensing parameters based on OTFS,” IEEE Access, vol. 11, pp. 66 035–66 049, 2023.
[47] H. Qu et al., “Low-dimensional subspace estimation of continuous-doppler-spread channel in OTFS systems,” IEEE Trans. Comm., vol. 69, no. 7, 2021.
[48] Y. Liu et al., “Near-optimal BEM OTFS receiver with low pilot overhead for high-mobility communications,” IEEE Trans. Comm., vol. 70, no. 5 2022.
[49] O. Zacharia et al., “Target parameter estimation for OTFS radar using sparse signal processing,” Physical Communication, vol. 58, p. 102040, 2023.
[50] G. D. Surabhi et al., “On the diversity of uncoded OTFS modulation in doubly-dispersive channels,” IEEE Trans. Wireless Comm., vol. 18, no. 6, 2019.