PaperlayerView exact revision on arXiv

Source-linked AI summary

Empirical Analysis of Near-Field Beam Shaping for Blockage Events

Jy-Chin Liao, Volodymyr Vizitiv, Edward W. Knightly

arXiv:2609.28400v1eess.SP

TL;DR

Dynamic blockage makes high-frequency links vulnerable, and the performance of different near-field beams over complete blockage events remains uncertain. The paper compares fixed and per-position optimized beam parameters, finding that focused beams are strongest without adaptation while focused and curved beams perform best when adapted.

  • Problem

    The paper addresses limited evidence on how multiple near-field beam types perform over complete dynamic blockage events, despite the vulnerability of mmWave and sub-THz links to obstruction.

    Source: §I · p. 1

  • Method

    The study compares fixed clear-channel-optimal beam settings with an adaptive upper-bound policy that re-optimizes parameters at each blockage position.

    Source: §I · p. 1

  • Results

    Focused beams perform best without adaptation; with adaptation, focused and curved beams are nearly tied and both outperform Bessel beams.

    Source: C. Experimental Results · p. 11

  • Takeaways & Limitations

    Blockage-aware protocols should jointly select beam type and adapt beam parameters to the operating environment while balancing sensing, estimation, feedback, optimization, and hardware overhead.

    Source: §VI · p. 11

Abstract

from arXiv · show

Millimeter-wave (mmWave) to sub-terahertz (sub-THz) links can realize high data rates, yet are highly vulnerable to dynamic blockages due to high directivity, limited multipath, and dependence on line-of-sight (LoS) propagation. While structured and self-healing beams are promising solutions, it remains unclear how beam types perform under dynamic blockage. In this paper, we present an empirical analysis of structured near-field beams using two transmission policies. At one extreme, we study beam resilience without obstacle adaptation. At the other extreme, we empirically optimize beam parameters to study the limits of multiple beam types under idealized adaptation. We employ numerical analysis based on the angular spectrum method (ASM) and experimental validation using a sub-THz time-domain spectroscopy (TDS) platform to characterize beam performance over complete blockage events. For the scenarios considered, despite lacking self-healing, focused beams provide the strongest resilience when beam parameters are not adapted during a blockage event (i.e., when all beams are optimized only a priori). In contrast, when optimally adapted to obstacle position, curved beams provide the best performance. Lastly, although Bessel beams benefit from self-healing, this property alone can be insufficient to outperform optimized curved and focused beams. These findings provide guidance for blockage-aware beam selection and adaptation protocols.

I. INTRODUCTION

The paper evaluates structured near-field beams over complete blockage events using fixed clear-channel settings and an idealized policy that re-optimizes parameters at each blocker position. Focused beams are most resilient without adaptation, while curved beams achieve the largest adaptation recovery; prior work left multi-beam, full-event comparisons unresolved.

I. INTRODUCTION

  • The analysis defines a blockage event as the interval from blockage entry to exit and evaluates link performance across that interval, addressing a gap left by static-state studies.The paper's motivation is that beam resilience depends on beam structure, obstacle conditions, and adaptation.Source: §I · p. 1

III. ANALYSIS FRAMEWORK FOR BLOCKAGE EVENTS

  • Each beam type has a beam-specific discretized parameter space searched exhaustively; the two policies compare fixed clear-channel-optimal settings with per-position SNR re-optimization.The adaptive upper-bound policy approximates continuous reconfiguration and excludes practical feedback delays, overhead, and imperfect channel estimates.Source: §I · p. 1

I. INTRODUCTION

  • The numerical evaluation uses ASM for efficient repeated parameter searches, followed by experimental validation with a sub-THz TDS platform and passive metasurface beam shapers.The measured trends agree with the numerical analysis.Source: §I · pp. 1–2
  • Focused beams provide the strongest resilience with fixed parameters, while curved beams gain most through trajectory adaptation; Bessel self-healing alone does not ensure the best blocked-link performance.The numerical analysis reports up to 10 dB higher average SNR for focused beams and up to 13 dB adaptation gain for curved beams; experiments confirm the trends.Source: §I · p. 2

II. PRIOR WORK

  • Earlier studies examined moving blockage without comparing multiple near-field beam types, or evaluated structured beams at single blockage states rather than across full events.Prior work also left optimal beam performance unclear when demonstrated parameters were not optimized.Source: §II · p. 2

A. System Geometry and Beam Control

The framework sweeps a blocker laterally at fixed range and represents four beam types through beam-specific controls in a phase-only, fixed-amplitude array model.

A. System Geometry and Beam Control

  • A blockage event is formed by sweeping blocker center x_b laterally at fixed z_b; each x_b specifies a blockage state.The receiver is at broadside and the transmit aperture is centered at x = 0.Source: A. System Geometry and Beam Control · p. 3
  • A phase-only, fixed-amplitude array model assigns each beam type a control vector, discretizes its parameter space, and selects the best-performing configuration at each blockage position.The focused controls are steering angle and focal distance; curved controls include trajectory length and curvature; Bessel controls include transverse energy spread.Source: A. System Geometry and Beam Control · p. 3
  • The four representative beams are focused, curved, Bessel, and far-field-optimized steered beams, with distinct spatial concentration, trajectory, self-healing, and baseline characteristics.Focused beams target a location; curved beams form a parabolic trajectory; Bessel beams have a central core and side lobes supporting self-healing,,,.Source: A. System Geometry and Beam Control · p. 3

B. Received Field and Link-Quality Model

The link-quality model initializes a normalized transmit field, propagates it through the near field with ASM and a blockage mask, then computes received power and SNR.

B. Received Field and Link-Quality Model

  • The transmit field uses a(x) = 1 and the beam-specific phase profile, with normalization keeping transmit power equal across beam types and configurations.ASM evaluates near-field propagation and diffraction efficiently for exhaustive optimization over blockage positions.Source: B. Received Field and Link-Quality Model · pp. 3–4
  • The blocker is modeled as a spatial attenuation mask applied during iterative propagation at planes intersecting its thickness; α_B = 0 represents a fully opaque blocker.The attenuation factor ranges from 0 to 1.Source: B. Received Field and Link-Quality Model · p. 4
  • Received power is integrated over the receiver aperture and converted to SNR using receiver noise power; fixed aperture, bandwidth, and noise make power and SNR maximization equivalent.The receiver aperture, bandwidth, and noise power are fixed across beam types, parameter configurations, and blockage states.Source: B. Received Field and Link-Quality Model · p. 4

C. Transmission Policies

The two transmission policies distinguish a beam’s resilience when its clear-channel configuration stays fixed from the performance recoverable through ideal parameter reconfiguration. Each policy evaluates candidate beam configurations for each beam type.

C. Transmission Policies

  • Adaptation-free focused beams show inherent blockage resilience, while adaptive upper-bound re-optimization measures the performance recoverable at each blockage position.The adaptive policy assumes perfect knowledge of blockage state and ignores sensing, feedback, computation, and hardware-reconfiguration overheads, so it is an empirical upper bound over the evaluated parameter set.Source: C. Transmission Policies · pp. 4–5
  • Each beam type is evaluated over a discrete candidate parameter set spanning configurations that can affect power at the receiver.The grids include beam-specific choices such as steering angle, focal depth, curved trajectory, or transverse energy distribution.Source: C. Transmission Policies · p. 4
  • The adaptation-free policy selects the clear-channel SNR-maximizing configuration before blockage and keeps it fixed as the blocker moves.It characterizes the resilience of a nominal clear-channel configuration without real-time beam adaptation.Source: C. Transmission Policies · p. 4
  • The adaptive upper-bound policy re-selects beam parameters for each blockage state to maximize instantaneous SNR within the evaluated parameter set.It separates the potential benefit of reconfiguration from the resilience of a fixed beam.Source: C. Transmission Policies · pp. 4–5

IV. NUMERICAL ANALYSIS

The numerical analysis evaluates near-field beam types under representative dynamic blockage scenarios using the proposed transmission policies.

IV. NUMERICAL ANALYSIS

  • The evaluation studies representative dynamic blockage scenarios under the proposed transmission policies.It considers both instantaneous SNR over the event and event-level metrics for the full blockage event.Source: §IV · p. 5
  • Instantaneous SNR and event-level metrics capture both the evolving signal and the impact of the complete blockage event.Source: §IV · p. 5
  • Table I summarizes the beam-parameter search spaces used in the numerical analysis.Source: Table I · p. 5

A. Numerical Setup

The analysis combines ASM-based numerical propagation with a lateral blockage sweep to compare instantaneous SNR across beam types. Focused beams are strongest overall without adaptation, while curved beams achieve the largest adaptation gain when parameters can be re-optimized.

A. Numerical Setup

  • ASM models propagation at 150 GHz with a 0.5-m transmit aperture and a receiver 1 m away, sweeping the blockage laterally in 0.02-m steps.Source: A. Numerical Setup · p. 5

B. Instantaneous SNR Analysis

  • Without adaptation, focused beams provide the strongest overall resilience; with adaptation, curved beams achieve the largest adaptation gain by reshaping around the moving blockage.Source: Figure 3 · p. 6, B. Instantaneous SNR Analysis · p. 6
  • The instantaneous-SNR comparison uses a 0.25-m-wide blocker placed midway between transmitter and receiver and swept laterally across the propagation region.Source: B. Instantaneous SNR Analysis · p. 6, Figure 3 · p. 6
  • Without adaptation, curved beams can outperform focused beams for some obstacle positions but perform poorly when the blocker moves right; the steered baseline performs worst across positions.The clear-channel optimization selected a right-curving trajectory, which avoids blockers on the left but curves the wrong way for blockers on the right.Source: B. Instantaneous SNR Analysis · p. 6
  • All beam types benefit from adaptation: curved beams reshape their trajectories, focused beams adjust focal distance, and Bessel beams jointly adjust steering and deflection angles.Despite self-healing, Bessel beams remain outperformed by focused and curved beams in this scenario.Source: B. Instantaneous SNR Analysis · p. 6

C. Event-Level Metrics

Event-level results vary with blockage size and position: curved beams lead for small obstacles, while focused beams are strongest against larger blockages and across most blockage locations. Adaptation gain is distinct from final SNR, and the event metrics evaluate performance over the beam-dependent region meaningfully degraded by blockage.

Event-Level Metrics

  • Event-level metrics average performance over each beam’s blockage zone, defined where its adaptation-free signal is meaningfully degraded; adaptive upper-bound SNR measures ideal adapted performance.The adaptation gain metric captures average SNR recovery from adaptation.Source: C. Event-Level Metrics · pp. 6–7

Effect of Blockage Size

  • At Wb/D = 1/2 and 1, focused beams outperform curved beams by about 2.5 dB and 3 dB, while curved beams lead for small blockages.Bessel beams outperform steered beams under smaller blockages but degrade more rapidly as blockage grows.Source: Figure 4 · p. 7, C. Event-Level Metrics · p. 7
  • At Wb/D = 1/2, the steered beam gains about 13 dB from adaptation versus about 4 dB for the focused beam, although higher gain does not mean higher average SNR.The focused beam is already resilient without adaptation; the steered beam gains more from its weaker fixed-beam baseline.Source: C. Event-Level Metrics · p. 7
  • Adaptation uses beam-specific controls: focused beams shorten focal length, Bessel beams increase deflection for self-healing, and curved beams alter trajectory to bypass obstacles.The dominant mitigation mechanism changes with blockage size: trajectory control helps with smaller obstacles, while focused beams become more effective as blockage grows.Source: C. Event-Level Metrics · p. 7

Effect of Blockage Distance

  • Average SNR generally falls as blockage moves toward the receiver; focused beams remain strong across most locations, while curved beams drop sharply at zb = 0.75 m.Bessel beams generally outperform steered beams under blockage.Source: Figure 5 · p. 8, C. Event-Level Metrics · p. 8
  • Across blockage locations, the steered beam gains about 12–13 dB, while the focused beam gains about 1 dB at zb = 0.25 m and 4–5 dB at zb = 0.75 m.The focused beam maintains the strongest performance across locations, especially near the receiver.Source: C. Event-Level Metrics · p. 8

D. Adaptive Beam-Parameter Selection

Under the adaptive upper-bound policy, beam parameters are selected to maximize received SNR at each blockage position. The resulting adjustments differ by beam type, exploiting distinct parameter choices to redirect or redistribute energy.

Adaptive parameter selection

Steering-angle selection

  • As blockage intersects the dominant propagation region, steering angles redirect energy; the steered beam varies most because steering is its only degree of freedom.Focused and Bessel beams need smaller steering changes because they also adapt beam-specific parameters; the curved beam’s steering pattern is approximately symmetric.Source: D. Adaptive Beam-Parameter Selection · pp. 8–9

Beam-specific parameter selection

  • Focused beams shorten focal length, Bessel beams increase deflection angle, and curved beams jointly adjust focal length and curvature to reroute energy around blockage.The Bessel deflection rises from 3° to approximately 8°–10° when blockage overlaps the beam core; curved-beam parameter patterns reflect switching among discrete combinations.Source: D. Adaptive Beam-Parameter Selection · p. 9, Figure 7 · p. 9

V. EXPERIMENTAL EVALUATION

The experimental evaluation measures candidate beam types under dynamic blockage, first introducing the setup and beam realization, then presenting measured blockage responses and adaptation results.

Experimental evaluation

  • The experimental evaluation measures candidate beam types under dynamic blockage and presents their measured blockage responses and adaptation results.The section introduces the measurement setup and metasurface-based beam realization before reporting the results.Source: §V · p. 9

A. Sub-THz Measurement Testbed

The testbed uses time-domain spectroscopy to measure received power at the 150 GHz design frequency while a metal obstacle moves laterally across the beam. Its small-scale geometry reflects the transmitter’s submicrowatt power and resulting link-budget constraint.

Measurement procedure

  • A Toptica TeraFlash time-domain spectroscopy system measures received power at 150 GHz after transforming the measured time-domain signal into the frequency domain.The setup enables controlled comparison of beam responses to the same moving blockage and of adapted configurations.Source: A. Sub-THz Measurement Testbed · p. 9

Testbed geometry

  • A 100 mm × 20 mm metal blockage moves laterally across the beam, 75 mm from the beam shaper; submicrowatt transmitter power constrains the link budget and spatial scale.The transmitter illuminates a 200 mm × 200 mm aperture from 450 mm away, and the receiver is positioned 150 mm from the beam shaper.Source: A. Sub-THz Measurement Testbed · p. 9

B. Metasurface-Based Beam Realization

The paper uses passive transmissive metasurfaces to realize sub-THz beam-shaping profiles. It compensates for the transmitter’s spherical incident wavefront before quantizing the phase profile into discrete states.

B. Metasurface-Based Beam Realization

  • The metasurface uses C-shaped split-ring resonators to provide controllable sub-THz phase responses with relatively stable transmission amplitude, at a 150 GHz center frequency.The passive beam shapers provide a practical platform while reconfigurable hardware remains at an early development stage.Source: B. Metasurface-Based Beam Realization · pp. 9–10
  • Spherical-wave pre-compensation cancels the transmitter’s nonplanar incident wavefront so the transmitted field follows the intended beam-shaping profile.The compensation term is based on the path-length difference from the transmitter to each beam-shaper location; Fig. 9 illustrates Bessel-beam pre-compensation.Source: B. Metasurface-Based Beam Realization · p. 10, Figure 9 · p. 10
  • The continuous phase profile is quantized into eight CSRR states, spaced by approximately 45° to cover 2π while maintaining near-constant transmission amplitude.Each phase state uses a distinct CSRR geometry, and the patterns are fabricated by hot-stamping.Source: B. Metasurface-Based Beam Realization · p. 10

C. Experimental Results

In the measured blockage scenario, focused beams have the strongest resilience without adaptation, while adapted focused and curved beams outperform Bessel beams. Simulations reproduce the qualitative ordering, but practical implementation losses limit absolute performance.

C. Experimental Results

  • The experiment compares clear-channel beams, fixed clear-channel configurations under blockage, and blockage-specific adapted configurations realized with separately fabricated passive metasurfaces.The selected blockage position corresponds to the fixed curved beam’s largest SNR degradation.Source: C. Experimental Results · p. 10
  • Curved beams gain 6.5 dB with adaptation, compared with 1 dB for focused, 3 dB for Bessel, and 5 dB for steered beams.All four beam types partially recover blockage loss under the adaptive upper-bound policy.Source: C. Experimental Results · p. 10
  • Focused beams retain the highest blocked SNR without adaptation (15 dB); with adaptation, focused and curved beams reach 16.0 dB and 15.9 dB, both above Bessel beams at 13.7 dB.The adapted Bessel beam also exceeds the steered beam at 11.5 dB; the results indicate that self-healing alone does not guarantee the highest blocked-link SNR.Source: C. Experimental Results · pp. 10–11, Figure 10 · p. 11
  • For the same geometry, simulations reproduce the measured qualitative ordering but show a larger dynamic range, including larger clear-channel SNRs and adaptation gains.Focused beams have the smallest fixed-beam degradation in simulation, while curved beams have the largest fixed-beam loss and greatest recovery after adaptation.Source: C. Experimental Results · p. 11
  • Practical phase quantization, fabrication tolerances, finite apertures, alignment errors, and measurement nonidealities reduce experimental SNR and adaptation gains, so simulations may overestimate absolute performance.The reported simulation–experiment agreement supports comparisons of relative beam behavior for the considered geometry.Source: C. Experimental Results · p. 11

VI. CONCLUSION

  • Blockage-aware protocols should jointly optimize beam type and parameters for the operating environment while balancing sensing, estimation, feedback, optimization, and reconfiguration overhead.The paper proposes its analysis as a benchmark for practical blockage-aware near-field adaptation strategies.Source: §VI · p. 11
Loading 2609.28400v1…