PaperlayerView exact revision on arXiv

Source-linked AI summary

Holographic MIMO Communications: Theoretical Foundations, Enabling Technologies, and Future Directions

Tierui Gong, Panagiotis Gavriilidis, Ran Ji, Chongwen Huang, George C. Alexandropoulos, Li Wei, Zhaoyang Zhang, Mérouane Debbah, H. Vincent Poor, Chau Yuen

arXiv:2212.01257v3eess.SPcs.IT

TL;DR

HMIMO research addresses how future 6G systems can meet extreme wireless requirements despite limitations of conventional architectures and incomplete understanding of HMIMO's fundamental limits. This survey synthesizes HMIMO's physical structures, theoretical foundations, and enabling technologies, and compares it with existing multi-antenna systems. It reports that LoS HMIMO can achieve up to π2 times higher power gain and a corresponding 3.30 bits/s/Hz spectral-efficiency gain than LoS mMIMO in a point-to-point setting.

  • Problem

    HMIMO studies remain at an initial stage, with fundamental limits and critical technical challenges still requiring investigation for extreme 6G requirements.

  • Method

    The survey reviews HMIMO physical aspects, theoretical foundations, enabling technologies, prototypes, comparisons with mMIMO, and extensions.

  • Results

    Up to π2 times higher power gain and a corresponding 3.30 bits/s/Hz spectral-efficiency gain are reported for LoS HMIMO versus LoS mMIMO in a point-to-point setting.

  • Takeaways & Limitations

    HMIMO combines holographic surfaces, EM-domain processing, and communication theory as a potential approach to large spatial multiplexing and programmable wireless environments.

Abstract

from arXiv · show

Future wireless systems are envisioned to create an endogenously holography-capable, intelligent, and programmable radio propagation environment, that will offer unprecedented capabilities for high spectral and energy efficiency, low latency, and massive connectivity. A potential and promising technology for supporting the expected extreme requirements of the sixth-generation (6G) communication systems is the concept of the holographic multiple-input multiple-output (HMIMO), which will actualize holographic radios with reasonable power consumption and fabrication cost. The HMIMO is facilitated by ultra-thin, extremely large, and nearly continuous surfaces that incorporate reconfigurable and sub-wavelength-spaced antennas and/or metamaterials. Such surfaces comprising dense electromagnetic (EM) excited elements are capable of recording and manipulating impinging fields with utmost flexibility and precision, as well as with reduced cost and power consumption, thereby shaping arbitrary-intended EM waves with high energy efficiency. The powerful EM processing capability of HMIMO opens up the possibility of wireless communications of holographic imaging level, paving the way for signal processing techniques realized in the EM-domain, possibly in conjunction with their digital-domain counterparts. However, in spite of the significant potential, the studies on HMIMO communications are still at an initial stage, its fundamental limits remain to be unveiled, and a certain number of critical technical challenges need to be addressed. In this survey, we present a comprehensive overview of the latest advances in the HMIMO communications paradigm, with a special focus on their physical aspects, their theoretical foundations, as well as the enabling technologies for HMIMO systems. We also compare the HMIMO with existing multi-antenna technologies, especially the massive MIMO, present various...

I. INTRODUCTION

6G requirements for higher data rates, massive connectivity, and extreme applications expose practical limits in current architectures, motivating HMIMO as a holographic, programmable approach to electromagnetic wave manipulation. HMIMO combines dense, nearly continuous surfaces with EM-domain processing to support large spatial multiplexing and new wireless-environment paradigms.

  • Motivation: Rising data-rate, device-density, and connection requirements are motivating the emergence of 6G wireless networks.The passage connects these demands with envisioned applications requiring extreme wireless performance.
  • Motivation: mMIMO, mmWave, and ultra-densification improve 5G capacity, latency, and connectivity but introduce hardware, power, cost, and interference challenges.Large numbers of RF chains increase power consumption and hardware cost, while tiny-cell deployments create inter-cell interference concerns.
  • HMIMO Concept: Holography records and reconstructs electromagnetic wavefront amplitude and phase, enabling holographic-type radios for manipulating radio waves.Metamaterials and metasurfaces provide feasible technologies for electromagnetic-wave recording and reconstruction.
  • HMIMO Concept: HMIMO uses almost spatially continuous apertures with sub-wavelength-spaced elements, unlike conventional half-wavelength-spaced mMIMO arrays.The denser aperture activates mutual coupling, which may be exploited for superdirectivity and potentially higher received SNR.
  • Capabilities: EM-domain wave manipulation and near-field propagation give HMIMO unprecedented spatial degrees of freedom and holographic imaging-level communications for extremely large spatial multiplexing.HMIMO surfaces may operate as active transceivers or nearly passive smart reflectors within programmable wireless environments.
  • Scope and Vision: The survey positions HMIMO as a blend of antenna technologies and communication and electromagnetic-wave theories, covering foundations, enabling technologies, comparisons, and future applications.Envisioned deployments span smart cities and integrated space-air-ground-sea wireless networks, including active transceivers and passive relays or reflectors.

D. Literature Overview

The survey positions itself as a comprehensive overview of HMIMO communications, extending beyond RIS-focused reviews to cover active HMIMO surfaces, physical foundations, enabling technologies, comparisons, and open challenges.

  • This survey focuses mainly on active HMIMO surfaces that implement RF front ends for holographic-type radios, an area described as still being in its infancy.
  • Existing HMIMO overviews address different physical, theoretical, hardware, and operational perspectives but generally provide limited scope and detail.
  • The survey reviews holographic applications and technology enablers spanning optical holography, computer-generated holography, and EM holography.
  • It systematically covers HMIMO surface hardware, holographic design, tuning mechanisms, aperture shapes, functionalities, prototypes, and communication trials.
  • It surveys HMIMO channel models, degrees of freedom, capacity, EM wave sampling, and electromagnetic information theory under LoS and NLoS conditions.
  • It covers channel estimation, beamforming, beam focusing, comparisons with mMIMO and RISs, and research challenges concerning hardware, fundamental limits, and operation.

F. Organization

This part introduces holography’s applications and technology enablers, while organizing the survey around applications, physical HMIMO aspects, theory, enabling technologies, comparisons, extensions, and future directions.

  • The survey organizes its technical coverage into holographic applications, HMIMO surface physics, theoretical foundations, enabling technologies, comparisons and extensions, and future challenges.
  • Basic holography: Holography records amplitude and phase information to support high-fidelity three-dimensional imaging through recording and reconstruction stages.
  • Applications: Potential 6G holographic applications include entertainment, education, medical healthcare, production, and other domains.
  • Holographic entertainment: Entertainment applications target immersive viewing and interactive experiences combining holograms with multidimensional senses.
  • Holographic education: Education applications aim to make learning and teaching more efficient, immersive, and consistent through shared mixed-reality classes.
  • Medical healthcare: Medical applications include high-resolution body visualization, surgical planning, remote surgery, specialist access, and realistic medical training.
  • Holographic production: Industrial applications support remote supervision, operational management, reduced operating costs, environmental perception, and emergency assistance.
  • Technology enablers: The technology-enabler discussion distinguishes conventional optical holography, computer-generated holography, and EM-based holography.

1) Basic Principle of Holography:

Holography records the interference of object and reference waves, then reconstructs the object wave; optical, computer-generated, and EM implementations realize this principle differently.

  • Holography consists of recording and reconstruction, using interference to capture wave information and diffraction or illumination to recover the object wave.
  • The recorded intensity I = |O + R|^2 contains cross terms that preserve the phase difference between object and reference waves.
  • Illuminating the recorded hologram with a replica of the reference wave can reconstruct the desired object wave, including amplitude and phase.
  • Optical holography: Conventional optical holography uses coherent laser light, beam splitters, mirrors, and photographic film for recording and reconstruction.
  • Computer-generated holography: Computer-generated holography implements recording and reconstruction computationally, reducing dependence on complicated optical hardware and real objects.
  • EM holography: EM holography transfers the recording-and-reconstruction principle to RF or other EM waves, using antenna apertures and reference-wave feeds.
  • HMIMO realization: HMIMO surfaces realize EM hologram recording and can be fixed or reprogrammable, while configured surfaces produce desired radiation when illuminated by reference waves.

III. PHYSICAL ASPECTS OF HMIMO SURFACES

HMIMO surfaces implement EM holography through feed, substrate, and antenna-element building blocks, with feed placement and propagation mode determining representative configurations.

  • EM holography superimposes object and reference waves to create an interference wave that maps the object wave to a hologram.
  • Reconstruction requires one or more feeds to generate reference waves and a designed HMIMO surface carrying the hologram.
  • Feed configurations: Feeds may be integrated into the surface or placed externally, including surface-fed, bottom-fed, edge-fed, and external-fed arrangements.
  • HMIMO surfaces mainly comprise a feed, substrate, and antenna element, with representative structures shown in four schematics.
  • Propagation modes: Feed hardware and location support different reference-wave propagation modes, including TE, TM, and quasi-TEM configurations.
  • Feed hardware: Available feed implementations include dipoles, Vivaldi feeds, Yagi-Uda feeds, dipole arrays, coaxial probes, and through-glass vias.
  • Feed configurations: Table III summarizes equivalent reference-wave names, propagation modes, feed hardware, and corresponding feed locations.

2) Substrate:

HMIMO surfaces use substrates, antenna elements, and surface patterns to transform reference waves into intended radiated fields. The survey covers material choices, element geometries, sampling strategies, and holographic design methodologies.

  • Substrate: Substrates support reference-wave propagation, either across a plate-like surface mode or along a confined microstrip waveguide mode.The substrate must support leaky-wave-based electromagnetic holography through suitable control of surface permittivity and permeability.
  • Substrate: Dielectric substrates include PCB laminates, silicon dioxide glass, and anisotropic artificial dielectrics, while silicon is highlighted as a low-cost semiconductor option.PCB laminates provide mechanical and electrical stability, low dielectric loss, and suitability for high-frequency or broadband applications.
  • Antenna elements: Antenna elements transform reference waves into radiated object waves and may use metals, dielectrics, or graphene in discrete or approximately continuous forms.Their dimensions and spacing determine how densely the surface samples electromagnetic fields.
  • Antenna elements: Subwavelength conductive patches enable dense field sampling and allow their scattering to be modeled through an effective surface impedance.Square patches, strip gratings, slot-shaped units, and other geometries implement different surface configurations.
  • Design methodologies: The survey distinguishes locally maximum phase lines, macroscopic surface impedance, and geometric polarizable-particle approaches for holographic surface design.The locally maximum phase-line method places strip gratings on interference-wave phase maxima, with fixed element positions and responses for a target radiation pattern.

2) Macroscopic Surface Impedance Based Approach:

The macroscopic surface-impedance approach densely samples reference waves with subwavelength elements and maps holograms to effective impedance and physical geometry. It supports simultaneous multibeam radiation, but element constraints limit reconfigurability and independent weight control.

  • Surface-impedance formulation: Subwavelength conductive patches densely sample the reference wave, with their scattering represented by a macroscopic effective surface impedance.The impedance depends on substrate permittivity and thickness, antenna-element period, and inter-element gap.
  • Surface-impedance formulation: The hologram determines the spatial surface impedance, which maps through substrate and element geometry to a realizable HMIMO surface.The average impedance X and modulation depth M parameterize the impedance variation produced by the interference of reference and object waves.
  • Multibeam radiation: Multiple object waves can be radiated simultaneously by dividing or superposing multiple textured surface-impedance patterns on one shared aperture.In the superposition case, each object wave is excited by one reference wave.
  • Reconfigurability: Once substrate and antenna-element geometries are designed, the radiating properties are fixed, limiting reconfigurable HMIMO surfaces.Graphene can introduce tunability by using external DC control to vary conductivity and surface impedance.
  • Polarizable-particle approach: The polarizable-particle approach models radiation as a weighted sum of dipole patterns, but resonant elements constrain the achievable weights.Amplitude and phase are coupled, while Lorentzian resonators restrict phase to [0, π] and produce unavoidable sidelobes in the directive-beam case.

C. Tuning Mechanisms of HMIMO Surfaces

HMIMO surfaces are moving from fixed designs toward dynamically tunable apertures, using electrical, liquid-crystal, graphene, optical, mechanical, and lithographic technologies. Their physical implementations span multiple aperture shapes and support holographically controlled wave manipulation.

  • Reconfigurability lets HMIMO surfaces dynamically adjust element responses to realize required holograms for changing propagation environments.
  • PIN diodes provide electrically controlled ON/OFF states through external DC bias voltages for reconfigurable HMIMO surfaces.
  • Liquid crystals tune antenna frequency response by changing permittivity and capacitance under external stimuli, with low power consumption but relatively slow response.
  • Graphene enables reconfigurable antenna elements by electrically controlling conductivity and realizing arbitrary surface impedance through DC bias voltages.
  • HMIMO surfaces use lithography-based fabrication and can adopt 1D, planar 2D, circular, hexagonal, or conformal apertures for installation-specific designs.
  • HMIMO transceiver surfaces focus on holographically guided wave manipulation, extending beyond the primarily passive-reflector operation of RIS.

1) EM Wave Polarization:

HMIMO surfaces support polarization control, beam steering, focusing, scattering, and multiplexing, with prototypes demonstrating increasingly integrated electromagnetic functionalities. Experimental and commercial developments indicate progress toward practical communication systems.

  • EM Wave Polarization: Polarization control transforms the oscillation orientation of EM waves through tensor impedance designs or geometrically tuned antenna elements.
  • EM Wave Steering: Beam steering includes transmission or reflection of single and multiple beams, focal spots, and scattering through constructive and destructive phase combination.
  • EM Wave Steering: HMIMO surfaces support multibeam scanning, multiwavelength multiplexing, Bessel and vortex beams, and distinct orbital angular momentum modes.
  • EM Wave Steering: Programmable metasurfaces have realized reconfigurable polarization conversion and dynamic steering on the same surface using PIN diodes.
  • HMIMO Surfaces Prototypes: More than 20 dBi maximum gain was achieved by representative 60 GHz HMIMO apertures using locally maximum phase-line hologram designs.
  • HMIMO Surfaces Prototypes: 3.93 dB antenna-gain and 7.7 dB sidelobe improvements were achieved at 10 GHz with a specifically designed decoupling structure.
  • HMIMO-Aided Communication Prototypes: HMIMO communication prototypes demonstrated 2 × 2 MIMO-16QAM transmission at 20 megabit-per-second data rate and multiple modulation and multiplexing systems.

1) LoS Channel Modeling:

HMIMO channel modeling must account for dense apertures, strong spatial correlations, mutual coupling, and near-field propagation. The survey presents spherical-wave and tensor-Green’s-function approaches for LoS channels while also covering spatial-correlation and spectral methods for NLoS conditions.

  • LoS Channel Modeling: Future HMIMO links are expected to be LoS dominated because higher frequencies and extremely large apertures shift communications toward near-field propagation.
  • Spherical-Wave Propagation Channel Model: Spherical-wave models represent element-specific path-amplitude and phase variations caused by different transmit-receive distances in near-field HMIMO systems.
  • Tensor Green’s Function Based Channel Model: Tensor Green’s-function models connect transmit current distributions to receive electric fields while obeying Maxwell-equation-based wave propagation.
  • Tensor Green’s Function Based Channel Model: Generalized EM-domain channel models support arbitrary HMIMO surface placements and have been used to study point-to-point capacity limits.
  • Tensor Green’s Function Based Channel Model: Tensor Green’s-function modeling captures three polarization states and may provide performance benefits from cross-polarization, especially in near-field regions.
  • NLoS Channel Modeling: NLoS modeling includes spatial-correlation approaches and wavenumber-domain expansions, while near-field spatial correlation remains comparatively underexplored.

B. HMIMO Performance Analysis

HMIMO performance analysis centers on degrees of freedom and capacity across near-field, far-field, LoS, and scattering environments. Existing studies derive geometry- and wavelength-dependent limits, while mutual coupling, evanescent waves, hardware impairments, and quantization motivate further work.

  • DoF: The DoF quantifies independent simultaneous EM communication modes and is essential for identifying HMIMO communication limits.Continuous surfaces and near-field operation require DoF analysis beyond conventional discrete-array models.
  • DoF: 2/λ per meter applies to 1D UE layouts, while π/λ2 per square meter applies to 2D-plane and 3D-volume layouts in LIS-based HMIMO.These results use spherical-wave propagation for near-field LoS channels.
  • DoF: LoS point-to-point DoF depends only on geometry normalized by wavelength and can exceed 1; closed-form basis sets support near-optimal near-field communication.The corresponding EM eigenfunction problem is approximated by accurate closed-form expressions and beamspace models.
  • DoF: 2Lx/λ for 1D linear apertures and πLxLy/λ2 for 2D planar apertures provide spatial-DoF-based spacing guidelines in far-field isotropic scattering.The analysis uses a 4D Fourier plane-wave expansion of the channel response.
  • DoF: Evanescent waves can add spatial DoF and increase system capacity in near-field HMIMO communications.The result is reported for isotropic scattering and can extend to non-isotropic cases.
  • Capacity: HMIMO capacity studies cover LoS and NLoS channels, with open limits caused by strong mutual coupling and differing near- and far-field behavior.Reported analyses include asymptotic LoS capacity, receive power, spectral efficiency, SNR, and EM-domain capacity bounds.

C. EM Wave Sampling

EM wave sampling addresses how continuous HMIMO fields can be discretized for digital processing while preserving electromagnetic information. Nyquist-based spatial sampling captures the available DoF, and an elongated hexagonal structure reduces sampling density under isotropic propagation.

  • Sampling motivation: Continuous HMIMO EM waves must be sampled and discretized for digital processing while retaining maximum information with minimum samples.This requirement motivates spatial EM-wave sampling methods.
  • Nyquist sampling: Nyquist-rate sampling fully captures EM-wave DoF with the minimum number of samples, whereas conventional half-wavelength sampling is redundant.The result was established for LoS uplink LIS-based HMIMO serving multiple single-antenna users.
  • Sampling efficiency: 13% fewer samples per square meter are achieved with elongated hexagonal sampling than with half-wavelength sampling for isotropic propagation.More reduction is expected for non-isotropic propagation when scattering knowledge is used.
  • EM information theory: EM information theory combines information theory, circuit theory, and EM wave theory to evaluate wireless communication limits with greater physical consistency.The framework responds to the abstraction of wireless channels as conditional probability distributions and supports EM-domain signal processing.
  • EM wave theory: Maxwell’s equations describe the coupled electric and magnetic fields, while tensor Green’s functions solve the vector wave equation for point sources.The resulting current-to-electric-field relation provides a physically consistent wireless channel model.

2) Circuit Theory:

Circuit theory models HMIMO communication systems as multi-port networks whose voltages, currents, self-impedances, and mutual impedances describe signal transfer and coupling. This framework supports physical analyses and requires near-field mutual coupling to be retained.

  • Circuit model: A circuit-theoretic HMIMO model represents transmitters and receivers as multi-port networks with circuit voltages and currents for each signal.The framework facilitates analysis of impedance matching, antenna mutual coupling, and related physical factors.
  • Impedance matrices: Ztt and Zrr model transmitter and receiver self-impedances, while Ztr and Zrt model transmitter–receiver mutual coupling.The matrices respectively capture BS-element, inter-UE, BS–UE, and UE–BS coupling effects.
  • Impedance determination: Free-space mutual and off-diagonal self-impedances can be obtained through EM-domain analysis, while diagonal self-impedances use energy conservation, zero-distance limits, or Chu-antenna expressions.In far-field regions Ztr can be neglected, but near-field analyses must retain it.
  • Circuit laws: The multi-port framework relies on Ohm’s law and Kirchhoff’s laws to connect voltages, currents, impedances, and node currents.These laws provide the basic circuit relationships used in the network analysis.
  • Implications for HMIMO: HMIMO enabling technologies must account for continuous apertures, EM-domain processing, massive-element complexity, and mutual coupling absent from conventional mMIMO designs.These differences motivate new channel-estimation and beamforming or beam-focusing methods.

A. HMIMO Channel Estimation

HMIMO channel estimation is complicated by near-field propagation, EM-domain processing, non-stationarity, wideband effects, and hybrid near-/far-field channels. Existing studies use compressed sensing, model-based learning, deep learning, geometric estimation, and beam training, but a unified practical framework remains unavailable.

  • Motivation and scope: HMIMO channel estimation requires new communication and channel models because EM-domain and near-field processing differ from conventional discrete-array estimation.A practical framework covering all HMIMO features has not yet been developed.
  • Compressed sensing: Polar-domain compressed sensing represents both near- and far-field sparsity, replacing angle-domain models that rely on far-field assumptions.Two-phase angular and cascaded angular–polar estimation reduces pilot use and computational complexity.
  • Deep learning: Model-based deep learning uses spherical-wave dictionaries and learned iterative thresholding to improve near-field estimation accuracy while reducing computational complexity.The dictionary is embedded as a neural-network layer within the iterative algorithm.
  • Wideband estimation: Wideband THz estimation accounts for beam split through beam-split-aware dictionaries, while federated learning reduces complexity and training overhead in model-free estimation.The approaches use angular and distance deviations caused by near-field beam split.
  • Non-stationary channels: Subarray-wise and scatterer-wise compressed-sensing methods address non-stationary channels modeled with spherical-wave last-hop scatterers.The large aperture is divided into multiple subarrays for estimation.
  • Hybrid near-/far-field estimation: Hybrid-field estimation separates LoS and NLoS components or exploits their distinct sparsity structures when scatterers occupy both near- and far-field regions.Deep-learning estimators have also been developed for this hybrid setting.

1) DMA Input-Output Response Based Work:

DMA-based HMIMO work models microstrip propagation and configures tunable element weights for beamforming, beam focusing, and rate optimization. Related studies incorporate mutual coupling and holographic principles, extending designs across near-field, wideband, multiuser, and satellite scenarios.

  • DMA input-output response: Each DMA element uses a tunable weight, while microstrip propagation is modeled as a linear multi-tap filter.Multiple microstrip lines jointly contain N = N_mN_e metamaterial elements.
  • DMA input-output response: The established input-output response enables HMIMO beamforming and beam focusing through appropriate configuration of DMA weights.AO algorithms optimize practical DMA weights for uplink and downlink sum-rate maximization.
  • Mutual coupling based work: Coupling-aware precoding models mutual coupling through a coupling matrix, allowing symbols to be transmitted using optimized voltage or current vectors.Subsequent studies validated coupling-aware superdirectivity through full-wave simulations and experiments.
  • Holographic principle guided work: Holographic beamforming configures analog amplitude weights from interference patterns between reference waves and intended object waves.Studies apply this framework to wideband OFDM, multiuser, satellite, and integrated sensing-communication scenarios.

4) EM Level Model Based Work:

EM-level HMIMO models represent transmission and reception through dedicated continuous-surface patterns, current distributions, fields, and receive-pattern decoding. This framework supports electromagnetic pattern design while motivating comparisons with RIS and related surface technologies.

  • EM-level modeling: An EM-level point-to-point model carries data through dedicated transmit patterns that generate a weighted transmit-current distribution.The resulting electric field is measured at the receiver and modeled with EM noise or interference.
  • EM-level modeling: Received signals are decoded using receive patterns defined over the receiving surface.Receive patterns can serve as basis functions for continuous-surface processing.
  • Pattern design: The EM model enables dedicated patterns for generating current distributions and maximizing communication performance from an electromagnetic perspective.Reported designs include Fourier-basis wavenumber-division multiplexing and pattern-division multiplexing.
  • Relation to RIS: Existing RIS studies mostly use conventional information theory, whereas dense and large surfaces motivate EM information theory combining wave, circuit, and information theory.The paper presents this EM-consistent framework as more applicable to future RIS analysis and design.

2) Comparison with XL-MIMO:

The paper treats XL-MIMO as HMIMO’s large-aperture special case, while distinguishing HMIMO through densely packed, nearly continuous apertures. This density supports EM-domain processing and additional physical phenomena, but several performance limits remain to be established.

  • Dense-aperture distinction: HMIMO uses nearly continuous sub-wavelength element spacing, whereas XL-MIMO remains discrete with approximately half-wavelength spacing.This density distinction separates HMIMO from XL-MIMO beyond aperture size.
  • Dense-aperture distinction: HMIMO’s dense aperture enables EM-domain modeling and processing, mutual-coupling exploitation, massive OAM modes, superdirectivity, and massive mode multiplexing.Its analysis extends Shannon information theory toward a blend of electromagnetic, circuit, and information theory.
  • Large-aperture relation: XL-MIMO is considered a special case of HMIMO from the large-aperture perspective.Near-field channel modeling and design ideas are shared between the two technologies.
  • Hardware comparison: HMIMO surfaces can reduce hardware cost and power consumption relative to phased arrays while using low-cost, energy-efficient tuning elements.A cited comparison reports cost and power consumption of 1/10 and 1/3 of phased-array levels, respectively.
  • Performance comparison: Mutual coupling may provide HMIMO superdirectivity and higher array gain than conventional mMIMO, potentially expanding coverage at equal transmit power.The paper notes that the coverage increase and fundamental energy-efficiency limit remain insufficiently unveiled.

3) Sensing, Localization, Positioning, and Tracking:

HMIMO research spans sensing-related applications, physical and theoretical foundations, enabling technologies, and unresolved implementation challenges. The survey also identifies open directions needed to make HMIMO practical.

  • Sensing, Localization, Positioning, and Tracking: HMIMO surfaces may assist 6G sensing, localization, and tracking, including integrated sensing and communication applications.An amplitude-controllable holographic beamformer has been optimized for integrated sensing and communication, with a theoretical lower bound on maximal beampattern gain.
  • Sensing, Localization, Positioning, and Tracking: An HMIMO-based sensing application achieved more than 50% higher beamforming gain at reduced cost than same-size MIMO arrays.
  • Extensions and Applications: HMIMO surfaces are being explored for satellite, UAV, vehicular, and air-to-ground communications under power, path-loss, safety, and data-rate constraints.One air-to-ground design jointly optimized UAV trajectory, active beamforming, and passive beamforming using reinforcement-learning methods.
  • Research Challenges and Future Directions: Practical HMIMO deployment remains constrained by waveguide signal loss, mutual coupling, incomplete channel models, unresolved EM sampling theory, and prohibitive pilot overheads.Existing channel-estimation approaches trade expensive pilot costs against performance loss, while beamforming schemes may not handle mutual coupling or closely separated user angles.
  • Survey Scope: The survey covers HMIMO hardware, holographic design, theoretical foundations, EM-field sampling, channel estimation, beamforming, beam focusing, technology comparisons, and extensions.
Loading 2212.01257v3…