Near-Field Tapering with Slepian Window: Balancing the Range–Angle Sidelobe Trade-off

Ahmed Hussain, Ahmed Sultan, Asmaa Abdallah · 2025

Read Near-Field Tapering with Slepian Window: Balancing the Range–Angle Sidelobe Trade-off by Ahmed Hussain, Ahmed Sultan, Asmaa Abdallah (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2511-12733.txt.

Introduction

Ahmed Hussain and four co-authors improve near-field beamforming, a technique that focuses radio energy at a specific distance as well as direction. Beams formed this way have sidelobes, stray energy lobes along both the range and angle axes, which invite interference and weaken detection. Taper methods borrowed from far-field design cannot suppress both kinds at once. The authors propose an amplitude taper based on the Slepian window, chosen to concentrate energy in the main lobe.

In simulations, their taper cut peak sidelobes by about twenty-four decibels sideways and ten decibels along the range axis, compared with a plain uniform window. Those are large margins in antenna engineering. The work is a short letter aimed at near-field communications and sensing, where focusing close to the array is the point. Better tapering means cleaner targets and fewer false echoes from sidelobe leakage.

Catalog record

CREATORAhmed Hussain, Ahmed Sultan, Asmaa Abdallah
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE23 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2511-12733

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Near-Field Tapering with Slepian Window: Balancing the Range–Angle Sidelobe Trade-off Thanks: The authors are with Computer, Electrical, and Mathematical Sciences and Engineering (CEMSE) Division, King Abdullah University of Science and Technology (KAUST), Thuwal, 23955-6900, KSA. Ahmed Hussain, Ahmed Sultan, Asmaa Abdallah Abstract Near-field beamforming enables target discrimination in both range (axial) and angle (lateral) dimensions. Elevated sidelobes along either dimension, however, increase susceptibility to interference and degrade detection performance. Conventional amplitude tapering techniques, designed for far-field scenarios, cannot simultaneously suppress axial and lateral sidelobes in near-field. In this letter, we propose a Slepian-based amplitude tapering approach that maximizes mainlobe energy concentration, achieving significant sidelobe reduction in both dimensions. Numerical results show that the proposed taper improves peak sidelobe suppression by approximately 24 dB in the lateral domain and 10 dB in the axial domain compared to a conventional uniform window. I Introduction Sensing, formally recognized as a key service in 3GPP release 20, is envisioned as a core capability of 6G ( 6G ) wireless networks. This paradigm shift is driven by the adoption of mmWave ( mmWave ) bands, coupled with UM ( UM )- MIMO ( MIMO ) architectures, that jointly enable fine spatial resolution through wide bandwidths and large antenna apertures, respectively [ 4 ] . At such scales, electromagnetic propagation occurs in the radiative NF ( NF ) regime, where spherical wavefronts enable the formation of NF beams. These beams exhibit finite beamwidth and beamdepth, thereby facilitating target resolution in both angular (lateral) and range (axial) domains. However, unlike the FF ( FF ), where interference arises solely from LSL s, NF beams exhibit both ASL s and LSL s, rendering them susceptible to interference in both axial and lateral dimensions. Sidelobe suppression is critical for reliable sensing and secure communication, as the mainlobe width determines spatial resolution while SLL s govern susceptibility to interference. Reducing sidelobes enhances sensing performance by mitigating undesired signal leakage from interference sources and improves multiuser communication performance by increasing the achievable sum rate. The overall lateral beam pattern depends on the product of the element pattern and array factor; hence, reducing element sidelobes directly lowers LSL s. Conversely, the element pattern does not influence the axial beam pattern. In the FF , amplitude weights are applied in the spatial domain to reduce LSL s in the angular domain [ 7 ] . In contrast, within the NF context, it is essential to investigate how a single set of amplitude weights (in the spatial domain) can simultaneously suppress both ASL and LSL . Recent findings reveal a fundamental trade-off between ASL and LSL suppression: optimizing tapering weights for one domain often degrades performance in the other [ 3 , 2 ] . This inherent coupling raises a critical research question: Can we design an amplitude distribution that jointly minimizes ASL and LSL ? Interestingly, the trade-off between ASL and LSL suppression mirrors the classical time–frequency uncertainty, in which a signal cannot be simultaneously time-limited and band-limited. Slepian sequences address this by maximizing the energy concentration of a finite-length sequence within a prescribed bandwidth [ 8 ] . These sequences are mutually orthogonal, and the principal Slepian sequence yields the most effective window function for sidelobe suppression. Drawing on this analogy, the NF problem exhibits a similar duality, as range and angle domains are inherently coupled with unavoidable resolution trade-offs. Motivated by this parallel, in this letter we first characterize the NF beam patterns across the range–angle domain and quantify the underlying trade-offs. We then propose Slepian-based amplitude tapers for NF beamforming that jointly suppress ASL and LSL , yielding superior performance compared to existing windowing methods. II System Model In this section, we present the NF beam pattern and specify the metrics to quantify the SLL s. II-A Channel Model We consider a ULA ( ULA ) equipped with N isotropic antenna elements spaced d=\lambda/2 . Based on the planar wavefront assumption, the normalized FF array response vector \mathbf{a}(\theta)\in\mathbb{C}^{N\times 1} is formulated as \mathbf{a}(\theta)=\tfrac{1}{\sqrt{N}}\Big[1,e^{-j\nu d\sin(\theta)},\dots,e^{-j\nu d(N-1)\sin{(\theta)}}\Big]^{\mathsf{T}}, (1) where \nu=\frac{2\pi}{\lambda} is the wavenumber, and \lambda is the wavelength. On the other hand, the NF array response vector \mathbf{b}(\theta,r)\in\mathbb{C}^{N\times 1} based on the spherical wave model is given by \mathbf{b}(\theta,r)=\tfrac{1}{\sqrt{N}}\Big[e^{-j\nu(r^{(0)}-r)},\dots,e^{-j\nu(r^{(N-1)}-r)}\Big]^{\mathsf{T}}, (2) where r is the distance between the focal point and the center of the ULA , while r^{(n)} is the distance between the focal point and the {n}^{\text{th}} antenna element. The corresponding phase shift \nu(r^{(n)}-r) , obtained via the law of cosines, is expressed as \nu(r^{(n)}-r)=\tfrac{2\pi}{\lambda}\left(\sqrt{r^{2}+n^{2}d^{2}-2rnd\sin(\theta)}-r\right) . Accordingly, {n}^{\text{th}} component of the NF array response vector can be written as {b}_{n}(\theta,r)=\tfrac{1}{\sqrt{N}}e^{-j\nu\sqrt{r^{2}+n^{2}d^{2}-2rnd\sin(\theta)}-r}. (3) The above equation can be simplified by using near field expansion based on the second-order Taylor expansion as \sqrt{1+t}\approx 1+\frac{t}{2}-\frac{1}{8}t^{2} where \sqrt{r^{2}+n^{2}d^{2}-2rnd\sin(\theta)}-r\approx nd\sin(\theta)-\frac{1}{2r}n^{2}d^{2}\cos^{2}(\theta) . Hence, {b}_{n}(\theta,r) in ( 3 ) is approximated as {b}_{n}(\theta,r)\approx\tfrac{1}{\sqrt{N}}e^{-j\nu\{nd\sin(\theta)-\frac{1}{2r}n^{2}d^{2}\cos^{2}(\theta)\}}. (4) II-B Problem Formulation The NF beam pattern is obtained by applying the complex weights \mathbf{g} to the antenna elements as \mathcal{G}(\theta,r)=\left|\mathbf{g}^{\mathsf{H}}(\theta_{\mathrm{\scalebox{0.5}{0}}},r_{\mathrm{\scalebox{0.5}{0}}})\,\mathbf{b}(\theta,r)\right|^{2}, (5) where (\theta_{\mathrm{\scalebox{0.5}{0}}},r_{\mathrm{\scalebox{0.5}{0}}}) denotes the desired beamfocusing location in angle and range, respectively. The weighting vector \mathbf{g} can be decomposed into element-wise amplitude and phase components as \mathbf{g}=\mathbf{w}\odot e^{j\boldsymbol{\phi}}, (6) where \mathbf{w} is the real-valued amplitude vector, e^{j\boldsymbol{\phi}} is the complex phase vector, and \odot denotes the Hadamard (element-wise) product. In general, the phase component focuses the beam toward the desired location (\theta_{\mathrm{\scalebox{0.5}{0}}},r_{\mathrm{\scalebox{0.5}{0}}}) , while the amplitude component is designed to control the SLL . For a given focus point, the phase component e^{j\boldsymbol{\phi}} is obtained using ( 4 ). Substituting the decomposition of \mathbf{g} from ( 6 ) into the beam pattern expression in ( 5 ) yields \mathcal{G}(\theta,r)=\left|\left(\mathbf{w}\odot\mathbf{b}(\theta_{\mathrm{\scalebox{0.5}{0}}},r_{\mathrm{\scalebox{0.5}{0}}})\right)^{\mathsf{H}}\mathbf{b}(\theta,r)\right|^{2}. (7) Given the peak of the mainlobe is \mathcal{G}(\theta_{\mathrm{\scalebox{0.5}{0}}},r_{\mathrm{\scalebox{0.5}{0}}}) , the PSL ( PSL ) and ISL ( ISL ) are defined as follows: a) PSL : The PSL quantifies the ratio of the highest sidelobe magnitude to the mainlobe peak, indicating worst-case sidelobe susceptibility against narrowband interference. \mathrm{PSLL}=10\log_{10}\left(\frac{\max\limits_{(x)\in\mathcal{S}}\mathcal{G}(\theta,r)}{\mathcal{G}(\theta_{\mathrm{\scalebox{0.5}{0}}},r_{\mathrm{\scalebox{0.5}{0}}})}\right), (8) where x denotes either angle ( \theta ) or range ( r ) depending on the domain of interest. \mathcal{S} corresponds to the set of points in the sidelobe region. b) ISL : The ISL quantifies the ratio of the total sidelobe power to the mainlobe power, characterizing the beampattern’s susceptibility against wideband interference. \mathrm{ISLL}=10\log_{10}\left(\frac{\displaystyle\int_{\mathcal{S}}|\mathcal{G}(\theta,r)|^{2}\,dx}{\displaystyle\int_{\mathcal{M}}|\mathcal{G}(\theta,r)|^{2}\,dx}\right), (9) where \mathcal{M} corresponds to the mainlobe. Fig. 1 illustrates the mainlobe and sidelobe regions, as well as the PSL , for a NF beam pattern in the range domain. Conventional window designs entail a trade-off between ASL and LSL suppression, as detailed in Section III-D . To address this limitation, we propose a window that jointly attenuates both ASL and LSL in terms of PSL and ISL . Fig. 1 : NF beam pattern with mainlobe and . III Beam Pattern Analysis In this section, we derive the beamwidth, beamdepth, and PSL of the NF beam in both the angular and range domains, assuming uniform amplitude excitation \mathbf{w}=\mathbf{1} . We then highlight the inherent range–angle trade-off in sidelobe suppression. III-A Angle-Domain To evaluate the beam pattern in the angular domain, we consider a distance ring defined by \frac{\cos^{2}(\theta)}{r}=\frac{\cos^{2}(\theta_{\mathrm{\scalebox{0.5}{0}}})}{r_{\mathrm{\scalebox{0.5}{0}}}} [ 10 ] . This curve corresponds to a set of range-angle pairs over which the beamwidth remains approximately invariant. Accordingly, the beam pattern in the angle domain can be

Read the complete text