Pinching Antennas for Next-Generation Wireless Communications: A Network Perspective

Yanqing Xu; Shan Shan; Yongxu Zhu; Zhiguo Ding; Mugen Peng; Xiaohu You

Source-linked AI summary

Pinching Antennas for Next-Generation Wireless Communications: A Network Perspective

Yanqing Xu, Shan Shan, Yongxu Zhu, Zhiguo Ding, Mugen Peng, Xiaohu You

arXiv:2609.17332v1Sep 15, 2026eess.SP

TL;DR

Fixed radiation sites limit how wireless service can be spatially provided, motivating architectures with controllable radiation locations. This tutorial develops a network perspective on pinching antennas and shows through analytical and numerical examples how reconfiguration can strengthen desired links, reshape interference, and support network-level adaptation.

Problem
Existing wireless architectures flexibly coordinate transmission over largely fixed radiation sites, limiting control over where wireless service is provided.
Method
The tutorial reviews pinching-antenna architectures and channels, then studies radiation-location reconfigurability across multiple access, multicell transmission, offloading, and region-level design.
Results
Analytical and numerical examples show that pinching-antenna configuration can strengthen desired links, reshape interference, reduce multicell transmit power, and support traffic offloading.
Takeaways & Limitations
Radiation locations can be treated as configurable network resources for adapting wireless service to users, traffic patterns, propagation conditions, and spatial load.
Takeaways & Limitations
Large-scale deployment requires robust and scalable designs under imperfect environment knowledge, mobility, time-varying blockage, and practical reconfiguration constraints.

Abstract

from arXiv · show

Pinching antennas enable radiation points to be created and reconfigured along extended waveguides, making the physical locations of wireless transmission and reception controllable after infrastructure deployment. This capability introduces a spatial degree of freedom beyond conventional beamforming and resource allocation, with implications that extend from individual links to network-wide operation. This tutorial develops a network-oriented perspective on pinching-antenna systems. We first review their system architecture, channel characteristics, and fundamental design opportunities. We then examine how radiation-location reconfigurability can support environment-division multiple access by strengthening desired links and suppressing cross-links, and discuss its roles in multi-cell transmission, traffic offloading, and inter-cell cooperation. Moving from instantaneous users to entire service regions, we introduce traffic-aware and geometry-aware designs under a cell-free-inspired multi-waveguide architecture. Representative system models, optimization problems, analytical illustrations, and numerical results are provided to explain the main design tradeoffs and network-level insights. We further discuss generalized physical realizations and representative deployment modes, including standalone, conventional base-station-integrated, and distributed deployments, together with their main implementation considerations. Finally, we identify future research directions involving mobility management, programmable and virtualized radio access, environment knowledge acquisition, AI-native control, low-altitude networks, space-air-ground integration, and integrated sensing and communication. Overall, this tutorial presents radiation locations as configurable network resources and provides a framework for understanding how pinching antennas may reshape future wireless network.

I. INTRODUCTION … D. Scope and Organization of This Tutorial

Wireless architectures have evolved toward increasingly distributed, coordinated, dense, and user-centric operation, yet radiation locations remain largely fixed after deployment. Pinching antennas address this limitation by making radiation locations configurable network resources and motivating a network-level tutorial spanning access, multi-cell, region-level, deployment, and future-wireless applications.

A. Evolution of Wireless Network Architectures: Wireless networks evolved from fixed macro-cells toward coordinated multi-cell, dense heterogeneous, cloud or fog, and cell-free architectures serving users more flexibly and distributively.This progression aims to reduce communication distance, improve spatial reuse and service uniformity, and adapt resources to heterogeneous users and traffic.
B. The Remaining Spatial Limitation of Existing Wireless Architectures: Despite architectural flexibility in coordination and resource allocation, existing networks retain largely fixed electromagnetic radiation locations after deployment.Consequently, user distribution, traffic demand, blockage, and visibility are mainly handled through transmission and network optimization rather than controllable service locations.
C. Pinching Antennas as a Promising Answer: Pinching antennas convey RF signals through extended guiding structures and generate radiation at configurable locations along them after deployment.Unlike fluid or movable antennas, whose position changes are typically local, pinching antennas enable radiation-location reconfiguration over a much larger spatial domain.
C. Pinching Antennas as a Promising Answer: Radiation-location control creates a network degree of freedom for modifying desired and interfering links, effective serving relationships, user separation, multiple access, and interference management.It extends network design from optimizing operation over fixed radio sites to controlling where wireless service is provided.
D. Scope and Organization of This Tutorial: This tutorial develops a network perspective on pinching antennas, covering architecture and channels, EDMA, and multi-cell functions including association, offloading, interference management, and cooperation.EDMA exploits differences between desired and cross-links, with analytical examples, optimization problems, and numerical results supporting the discussion.
D. Scope and Organization of This Tutorial: The tutorial extends design from instantaneous users to communication regions through traffic-aware and geometry-aware control under a cell-free-inspired multi-waveguide architecture.Long-term spatial demand and blockage information guide configurations for network-average performance and region-wide coverage.
D. Scope and Organization of This Tutorial: It also examines standalone, conventional base-station-integrated, and distributed deployments, then identifies future directions in mobility, programmability, environment knowledge, AI-native control, and emerging wireless integration.The identified integrations include selected technologies such as low-altitude networks, space-air-ground integration, and integrated sensing and communication.

II. PINCHING ANTENNAS: FUNDAMENTALS AND NETWORK SIGNIFICANCE · A. Physical Principle and Basic Realization of Pinching Antennas · B. Signal Propagation and Channel Characteristics

Pinching antennas separate RF delivery from wireless radiation, allowing radiation points to be reconfigured along deployed waveguides. This location control creates a spatial degree of freedom because it changes distance, phase, and LoS availability, with consequences for network architecture and higher-layer functions.

II. PINCHING ANTENNAS: FUNDAMENTALS AND NETWORK SIGNIFICANCE: Configurable radiation locations provide a new spatial degree of freedom that can reshape wireless network architectures and affect higher-layer network functions.The tutorial frames radiation locations as controllable resources extending beyond fixed antenna placement.
II. PINCHING ANTENNAS: FUNDAMENTALS AND NETWORK SIGNIFICANCE: These location-dependent channel effects support network-level mechanisms including EDMA link reconfiguration, multi-cell traffic offloading, adaptive virtual cell boundaries, and coordinated region-wide service.A cell-free-inspired architecture connects distributed waveguides to a common processing unit for coordinated service across regions.
A. Physical Principle and Basic Realization of Pinching Antennas: Radiation locations can be reconfigured along a waveguide without relocating the RF source or underlying infrastructure, making wireless transmission points configurable after deployment.The waveguide transports RF signals, while pinching antennas determine where guided signals couple into free space.
A. Physical Principle and Basic Realization of Pinching Antennas: This guided-then-radiated realization differs from fixed-position antennas by decoupling signal feeding from wireless radiation and producing location-dependent channels.The configurable point where the signal enters the wireless propagation environment is the key physical distinction from conventional systems.
B. Signal Propagation and Channel Characteristics: The equivalent channel combines guided propagation from the feed point to the pinching antenna with free-space propagation from that antenna to the user.The basic illustration uses a single-user, single-waveguide setting with a waveguide parallel to the x-axis and a configurable antenna position.
B. Signal Propagation and Channel Characteristics: Moving a pinching antenna changes free-space distance and propagation phase, while guided propagation contributes a second phase component to the overall channel.The channel amplitude depends on distance from the configurable radiation location to the user, so repositioning can alter large-scale channel gain.
B. Signal Propagation and Channel Characteristics: In blockage-prone environments, radiation-location reconfiguration also changes LoS availability; a suitable location can shorten propagation distance and increase favorable-LoS probability.The probabilistic model uses a LoS indicator and blockage density β, with larger β implying lower LoS probability for a given propagation distance.

C. New Spatial Degrees of Freedom Enabled by Pinching Antennas … 3) Reconfigurable Service Regions and Region-Level Design:

Pinching antennas add configurable radiation locations along extended waveguides, enabling wireless adaptation through spatial control beyond fixed deployment, signal processing, and resource allocation. This flexibility reshapes interference, serving relationships, and region-wide service organization across multiuser, multi-cell, and region levels.

C. New Spatial Degrees of Freedom Enabled by Pinching Antennas: Radiation-location control changes link geometry and propagation, adding spatial adaptation beyond conventional power, beamforming, scheduling, and spectrum allocation.The configurable location can lie along an extended waveguide, with a reconfiguration range potentially much larger than wavelength-scale antenna-location adjustments.
D. Why Pinching Antennas Can Reshape Wireless Network Structure: Overall, radiation-location reconfigurability progressively reshapes user interference, infrastructure-user serving relationships, and the spatial distribution of wireless service.The resulting network adaptation extends beyond transmission and resource allocation performed over fixed radio geometries.
1) Reconfigurable Interference Relationships and Environment-Division Multiple Access:: At the multiuser level, radiation locations can strengthen desired links while weakening interfering links according to user locations and environmental conditions.A favorable line-of-sight link can be created for an intended user while the corresponding interference link experiences less favorable propagation.
1) Reconfigurable Interference Relationships and Environment-Division Multiple Access:: This interference-shaping capability enables environment-division multiple access by exploiting propagation-environment differences as an additional basis for multiuser access.The environment becomes a design resource that can help shape multiuser interference rather than merely a condition to accommodate.
2) Reconfigurable Serving Relationships and Multi-Cell Operation:: At the multi-cell level, effective service points can move closer to users or cell boundaries, adapting serving relationships without relocating base stations or waveguide infrastructure.Radiation-location control therefore adds flexibility to user association and neighboring-cell service organization.
3) Reconfigurable Service Regions and Region-Level Design:: A cell-free-inspired architecture distributes multiple waveguides across infrastructure surfaces, connects them to common processing, and jointly serves regions without rigid cell boundaries.Unlike conventional cell-free systems with fixed access-point locations, this architecture also controls where effective radiation points form along distributed waveguides.
3) Reconfigurable Service Regions and Region-Level Design:: At the region level, radiation locations can be optimized using longer-term traffic and environmental information to strengthen hotspots, balance coverage, and avoid unfavorable blockage.This extends optimization beyond instantaneous-user configurations toward the spatial organization of service across a communication region.

E. Impacts on Higher-Layer Network Functions … 2) Realization of EDMA:

Pinching antennas extend network control from fixed radio topology to reconfigurable radiation locations, enabling cross-layer functions and environment-division multiple access (EDMA). By strengthening desired links and weakening blocked or lossy cross-links, they support simultaneous resource reuse and broader access, mobility, coordination, and service-control decisions.

E. Impacts on Higher-Layer Network Functions: At the MAC layer, user grouping, multiple access, scheduling, and pinching-antenna activation can be jointly designed because radiation locations alter desired and interfering channels.At broader RRC and RAN scales, radiation locations can complement association, handover, and fixed-point coordination by reconfiguring serving infrastructures.
E. Impacts on Higher-Layer Network Functions: More reliable and stable links can reduce RLC retransmissions and support higher-layer protocols and demanding services such as URLLC, while radiation location usually acts indirectly through communication conditions.The paper therefore focuses higher-layer integration mainly on MAC functions and RRC-oriented network control.
E. Impacts on Higher-Layer Network Functions: Radiation-location reconfigurability adds a cross-layer control capability linking fast access and scheduling decisions with slower mobility, traffic, coordination, and region-level service adaptation.Its higher-layer effects arise through changed channel conditions, interference, reliability, latency, and service continuity rather than direct control of every protocol function.
A. Introduction to EDMA: EDMA exploits propagation-environment differences, including blockage, path loss, LoS availability, multipath, and directionality, to separate users sharing radio resources.Unlike orthogonal schemes that divide time or frequency, EDMA uses environmental differences to strengthen desired links and suppress interference.
B. EDMA Realization Using Pinching Antennas: Pinching antennas realize EDMA by selecting radiation locations that preserve favorable desired links while increasing cross-link blockage or distance-dependent path loss.This enables multiple users to reuse the same time-frequency resource with reduced interference and less reliance on additional separation processing.
1) System Architecture and Channel Model:: A representative EDMA architecture partitions a rectangular service area into regions, assigns one independently fed waveguide segment per region, and activates one pinching antenna for each scheduled user.The segmented architecture supports independent signal transmission or processing across waveguide segments.
2) Realization of EDMA:: With user-aligned pinching locations, separated users experience additional cross-link distance, causing greater path loss and lower LoS probabilities than on desired links.The resulting SINR tradeoff determines whether simultaneous EDMA reuse exceeds pinching-antenna-assisted TDMA, which avoids interference but gives each user only 1/M of the time resource.

3) A Two-User Illustration: · 4) Pinching-Antenna Location Optimization and Performance Evaluations:

The two-user analysis shows that EDMA benefits from user separation when cross-links are weakened by path loss and blockage, whereas location optimization improves service by jointly balancing desired-link enhancement and interference suppression. Evaluations show optimized EDMA achieves the highest ergodic downlink sum rate among the compared schemes.

3) A Two-User Illustration:: As user separation increases, EDMA benefits because cross-link SNR and LoS probability both decrease, eventually approaching interference-free simultaneous transmission.The two effects arise from path loss and blockage, respectively.
3) A Two-User Illustration:: EDMA is advantageous with sufficient cross-link isolation, while TDMA can achieve a higher sum rate for closely located users whose cross-links are likely to remain LoS.The two-user comparison assumes desired links remain LoS.
4) Pinching-Antenna Location Optimization and Performance Evaluations:: User-aligned pinching antennas are effective for separated users, but closely located neighboring users may require location optimization to reduce strong cross-links.The optimization balances desired-link enhancement against interference suppression.
4) Pinching-Antenna Location Optimization and Performance Evaluations:: The location design maximizes the minimum average downlink rate subject to each antenna’s feasible activation interval, thereby targeting balanced service among users.Average rates are defined over random blockage states.
4) Pinching-Antenna Location Optimization and Performance Evaluations:: Because each antenna location affects its desired link and other users’ interference, the resulting rate-optimization problem is nonconvex and requires joint optimization.The cited approach uses a tractable lower-bound approximation followed by successive convex approximation.
4) Pinching-Antenna Location Optimization and Performance Evaluations:: Optimized EDMA achieves the highest ergodic downlink sum rate over the evaluated transmit-power range, outperforming pinching-antenna-assisted and conventional fixed-antenna TDMA.The comparison considers four users distributed across four service regions and averages results over 100 independent user-location realizations.
4) Pinching-Antenna Location Optimization and Performance Evaluations:: Pinching-antenna-assisted TDMA outperforms conventional fixed-antenna TDMA because radiation locations can be horizontally aligned with scheduled users.Optimized EDMA additionally serves all four users simultaneously while controlling multiuser interference.
4) Pinching-Antenna Location Optimization and Performance Evaluations:: Across broader analytical and numerical scenarios, EDMA gains generally improve with larger service areas and greater cross-link blockage, while many closely located users can reduce those gains.The investigations vary service-area size, blockage, user count, user distribution, transmit power, and user-clustering conditions.

5) Extension of EDMA to NLoS Channels: … 2) Queue-Aware Online Pinching Antenna Scheduling:

The section extends EDMA to unreliable NLoS propagation, AI-native environment control, configuration-aware random access, and queue-aware online scheduling. Across these settings, pinching-antenna configurations become adaptive access and spatial-service resources constrained by reliability, hardware, overhead, and queue dynamics.

5) Extension of EDMA to NLoS Channels:: Outage-constrained EDMA jointly optimizes pinching-antenna locations and transmit powers under per-user outage limits, achieving near-optimal performance and retaining an advantage over pinching-antenna baselines.Closed-form outage probabilities and projected-gradient optimization are used for two users, while tractable approximations and SCA address general multiuser systems.
1) AI-Assisted EDMA:: AI-assisted EDMA uses environment sensing and learning-based resource allocation to handle explicit propagation knowledge and mixed discrete-continuous decisions across different protocol timescales.Relevant inputs include user locations, blockage, propagation paths, environmental objects, and mobility patterns; learning can identify groups, initialize optimization, or produce decisions directly.
2) AI-Native EDMA:: AI-native EDMA forms a continuous sensing-learning-prediction-reconfiguration loop that treats the propagation environment as a multiple-access resource adapted to network demand.Each reconfiguration is sensed again, updating the learned environment model and subsequent decisions.
D. EDMA-Assisted Access Protocol and Scheduling: EDMA-assisted access and scheduling incorporate environment configuration into procedures for sporadic users and time-varying service demand, adding configurable access opportunities and queue-dependent spatial actions.During random access, configurations can separate uncoordinated users; during dynamic scheduling, activation patterns are selected according to evolving queue states.
1) Configuration-Aware Random Access:: Configuration-aware random access exploits configuration-dependent channel diversity, using sparse probing and channel-oracle inference to identify favorable opportunities without exhaustively probing all configurations.A segmented waveguide divides a long waveguide into independently controlled segments, each supporting one active pinching antenna.
1) Configuration-Aware Random Access:: Segment aggregation offers randomized low-complexity access but coupled observations limit collision resolution, whereas segment multiplexing improves distinction and resolution using dedicated RF chains at added hardware and processing cost.Sparse channel-oracle acquisition balances environment knowledge against pilot and switching overhead.
2) Queue-Aware Online Pinching Antenna Scheduling:: Queue-aware online scheduling learns service rates from arrivals and prior decisions, maps them to feasible activation patterns, and converts them into slot-by-slot spatial service actions.The approach addresses discrete constraints such as at most one active pinching antenna on each waveguide without requiring arrival and service statistics in advance.

3) Protocol-Level Interpretation: … 2) Problem Formulation and Optimization:

Pinching antennas turn radiation locations into configurable network resources, enabling cross-layer EDMA and multi-cell control over links, interference, serving relationships, traffic, and transmission power. The resulting optimization jointly configures radiation geometry and radio resources, while practical deployment requires coordination, robustness, and scalable control.

3) Protocol-Level Interpretation:: EDMA extends contention resolution and scheduling by jointly configuring propagation environments, access, RF complexity, queue states, and pinching-antenna activation.This cross-layer design supports environment-dependent random access and queue-aware spatial service actions.
3) Protocol-Level Interpretation:: Architecture choices trade configuration knowledge, training overhead, RF complexity, access reliability, collision resolution, adaptation speed, queue stability, and delay.Sparse channel-oracle acquisition, segment aggregation or multiplexing, and online activation scheduling expose these tradeoffs.
E. Summary and Outlook: EDMA strengthens desired links and weakens cross-links through radiation placement, favorable LoS connectivity, propagation distance, blockage, and directional radiation.AI-assisted sensing, prediction, and resource allocation can extend EDMA toward an environment-aware multiple-access framework under changing propagation, locations, and traffic.
E. Summary and Outlook: Large-scale EDMA still requires scalable grouping, region partitioning, configuration, power allocation, and interference management under imperfect knowledge, mobility, blockage, and reconfiguration constraints.These challenges span multi-antenna and multi-waveguide systems and motivate robust protocol-level designs.
A. Motivation for Multi-Cell Pinching-Antenna Systems: In multi-cell networks, radiation-location reconfigurability changes desired links, inter-cell interference, serving relationships, effective service ranges, cell boundaries, and traffic offloading.Coordinated configurations can support local service, joint assistance, neighboring-cell offloading, and reshaped large-scale network geometry.
A. Motivation for Multi-Cell Pinching-Antenna Systems: Pinching-antenna multi-cell design jointly configures radio resources, serving relationships, radiation geometry, transmit power, traffic offloading, and inter-cell cooperation within deployment and coordination constraints.This extends conventional optimization beyond a fixed radio topology.
1) Multi-Cell Transmission Model:: The modeled network assigns each cell one waveguide and configurable active antenna, with coordinated locations and powers but no joint user-data transmission across cells.Each base station serves its in-cell user while creating inter-cell interference to other users.
2) Problem Formulation and Optimization:: The optimization minimizes total transmit power subject to every user meeting a common target rate and each antenna location remaining within its feasible waveguide segment.For fixed locations, power allocation is explicit; location optimization remains nonconvex because channel gains depend nonlinearly on radiation positions, motivating cross-entropy search.

3) Numerical Illustration and Design Insights: · C. Cell Reconfiguration and Traffic Offloading · 1) System and Offloading Models:

The section shows that pinching-antenna positioning supports cell reconfiguration and traffic offloading by jointly managing radiation locations, serving relationships, and resources. Numerical results indicate that desired-link shortening dominates under moderate QoS, whereas cross-link control becomes essential at stringent rates.

3) Numerical Illustration and Design Insights:: The numerical illustration uses two cells occupying the left and right halves of a 20 m by 100 m service region, with users clustered near their common boundary.Results are averaged over 100 independent user realizations, and the cross-entropy method uses 1000 candidates per iteration, retains 10, and runs for at most 10 iterations.
3) Numerical Illustration and Design Insights:: The cross-entropy method essentially matches exhaustive location search across target rates, while optimized positioning enables lower-power multi-cell transmission than conventional fixed-antenna operation.Fig. 8 compares average total transmit power against the common target rate; the optimized strategy balances desired-link shortening and cross-link suppression.
3) Numerical Illustration and Design Insights:: As the target rate approaches 1 bit/s/Hz, the user-aligned scheme’s required power rises sharply because direct user alignment does not sufficiently account for cross-links.At low and moderate rates, user alignment remains close to optimized schemes because shortening desired links provides the dominant power-saving gain.
3) Numerical Illustration and Design Insights:: Desired-link shortening provides most power savings under moderate QoS, whereas coordinated cross-link control becomes essential under stringent requirements.Balancing both effects moves multi-cell transmission from a high-power “shouting” strategy toward low-power “whispering.”
C. Cell Reconfiguration and Traffic Offloading: Traffic offloading addresses spatial imbalance by transferring users from overloaded cells to underutilized cells, while radiation-location reconfigurability provides additional flexibility.The subsection jointly designs pinching-antenna configuration, traffic offloading, and resource allocation.
1) System and Offloading Models:: In the two-cell offloading model, underutilized BS 0 can move an activated pinching antenna toward BS 1’s user U1, extending its effective service region and potentially lowering service power.U0 is originally served by BS 0, U1 by overloaded BS 1, and BS 1 uses a conventional fixed antenna.
1) System and Offloading Models:: The pinching-antenna location has a waveguide-determined y-coordinate, while its x-coordinate is adjustable within the feasible interval X = [exmin, exmax].The model represents user and base-station locations explicitly and treats the radiation point as the configurable variable.
1) System and Offloading Models:: The two offloading modes differ in resource treatment: one preserves U1’s original Cell 1 resource, while complete offloading releases it and requires joint positioning, power allocation, and multiple-access design with U0.In both cases, BS 0 serves U1; only the second mode makes U1 share a resource in Cell 0 with U0.

2) Traffic-Offloading and Pinching-Antenna Location Design: · 3) Numerical Illustration and Design Insights:

Pinching-antenna location design enables traffic offloading by extending a cell’s effective service region and balancing service links. Numerical results show that both offloading modes reduce transmit power, while complete traffic-and-resource offloading additionally releases overloaded-cell resources at moderate extra power cost.

2) Traffic-Offloading and Pinching-Antenna Location Design:: For complete offloading, the optimal antenna location is the midpoint of the users’ projections when feasible, or the closest feasible waveguide endpoint otherwise.This positioning balances the two service links and extends BS 0’s effective coverage toward Cell 1.
2) Traffic-Offloading and Pinching-Antenna Location Design:: With NOMA-based resource sharing, the stronger pinching-antenna-link user performs successive interference cancellation, enabling complete offloading while releasing Cell 1’s occupied resource.The stronger and weaker users are determined by their pinching-antenna links.
2) Traffic-Offloading and Pinching-Antenna Location Design:: In the complete-offloading design, users’ projections onto the waveguide determine the optimal antenna location, while perpendicular distances determine the stronger user and decoding order.Compared with offloading without resource release, this mode requires more involved coordination but can reduce load and release radio resources in Cell 1.
3) Numerical Illustration and Design Insights:: The numerical illustration uses two adjacent cells with BS 0 connected to a 40-m waveguide, 28-GHz carrier frequency, −90 dBm noise power, and averages over 100 user realizations.Users are generated near the cell boundary, with BS 1 using a conventional fixed antenna.
3) Numerical Illustration and Design Insights:: Both offloading schemes substantially reduce average total transmit power versus no offloading by moving the pinching antenna toward the cell boundary and creating short links to both users.The gains remain pronounced across the considered target-rate range and are particularly evident at high target rates.
3) Numerical Illustration and Design Insights:: Offloading without resource release requires slightly less power than complete traffic-and-resource offloading because the offloaded user retains its original resource.Complete offloading incurs moderate extra power from resource sharing and successive interference cancellation.
2) Traffic-Offloading and Pinching-Antenna Location Design:: Complete traffic-and-resource offloading frees the original resource of the offloaded user, providing a direct mechanism for relieving congestion in the overloaded cell.This benefit comes at the stated moderate additional power cost relative to offloading without resource release.

D. Large-Scale Cell-Boundary Reconfiguration … E. Summary and Outlook

Pinching antennas extend reconfigurability from individual links and serving relationships to cell boundaries and the statistical geometry of large cellular networks. The results show gains under interference, while practical deployment requires joint adaptation to mobility, traffic, heterogeneity, and coordination conditions.

1) One-Dimensional Cell-Boundary Analysis:: In one-dimensional Poisson–Voronoi cells, the expected longer boundary distance is three times the shorter, creating asymmetric serving distances around a fixed seed BS.The expected distances are rmin = 1/(4λb) and rmax = 3/(4λb).
1) One-Dimensional Cell-Boundary Analysis:: Moving the pinching antenna to the representative cell center compensates for random-deployment asymmetry by shifting radiation toward the longer side and regularizing effective cell geometry.The physical BS remains fixed while the effective radiation location changes.
1) One-Dimensional Cell-Boundary Analysis:: Optimized one-dimensional positioning continues to provide rate gains after inter-cell interference is included because it changes both desired serving distances and neighboring interference relationships.The SINR analysis separates interference from conventional and pinching-antenna-enabled BSs.
2) Extension to Two-Dimensional Networks:: In two-dimensional irregular Voronoi networks, a conservative lower-bound location rule is slightly below the numerical optimum but closely approaches it, especially for dense BS deployments.The rule assumes the waveguide is sufficiently long to contain the selected location.
2) Extension to Two-Dimensional Networks:: Even in the two-dimensional interference-limited case, location optimization clearly outperforms fixed antennas, while inter-cell interference lowers average achievable rate and links required waveguide length to BS density.Interference from conventional and pinching-antenna-enabled BSs is characterized separately.
D. Large-Scale Cell-Boundary Reconfiguration: Large-scale deployment adjusts effective radiation points after infrastructure installation, reducing excessive serving distances and regularizing cell geometry created by random BS placement.Fixed BS locations impose irregular Voronoi cells, whereas pinching antennas provide an additional network-scale degree of freedom.
3) Network-Level Insights / E. Summary and Outlook: Across increasing scopes, positioning reshapes desired and interfering links, changes serving relationships and cell boundaries for offloading, and ultimately modifies the statistical geometry of many-cell networks.The summary also identifies reduced QoS-constrained transmit power under fixed serving relationships and the need to adapt configurations jointly as network conditions evolve.

V. REGION-LEVEL DESIGN FOR CELL-FREE-INSPIRED PINCHING-ANTENNA SYSTEMS … 1) System, Signal, and Traffic Models:

Region-level pinching-antenna design uses distributed, centrally coordinated waveguides whose radiation locations can be reconfigured according to spatial traffic demand, propagation, and geometry. The traffic-aware model represents hotspot activity probabilistically and evaluates long-term service with a traffic-weighted network-average SNR.

V. REGION-LEVEL DESIGN FOR CELL-FREE-INSPIRED PINCHING-ANTENNA SYSTEMS: Multiple waveguides connected to a common processing unit jointly serve a region without rigid cell boundaries, combining centralized coordination with distributed transmission.The architecture is cell-free-inspired and is intended to retain macro-diversity, more uniform coverage, and flexible serving relationships.
A. Motivation for Pinching-Antenna-Enabled Region-Level Design: Radiation-location reconfigurability extends cell-free operation by adapting both participating waveguides and the physical locations providing wireless service.This adds a spatial control dimension beyond AP selection, beamforming, power allocation, and user association.
A. Motivation for Pinching-Antenna-Enabled Region-Level Design: Region-level configurations can be optimized over longer time scales using persistent traffic patterns and propagation conditions rather than repeatedly repositioning radiation points for instantaneous users.This can direct radiation resources toward high-demand areas and improve coverage balance.
B. Traffic-Aware Region-Level Design: Traffic-aware design adapts radiation locations to spatially nonuniform demand, while geometry-aware design uses environmental geometry and blockage information to improve coverage and robustness.The framework selects the design approach according to the spatial information available to the network.
1) System, Signal, and Traffic Models:: The traffic-aware system deploys N parallel waveguides across a rectangular region, with one continuously adjustable pinching antenna per waveguide and joint service from a common processor.The region is R = [0, DL]×[−DW /2, DW /2], and the waveguides are distributed along the y-axis at height dv.
1) System, Signal, and Traffic Models:: Long-term user activity is modeled by a weighted mixture of two-dimensional Gaussian hotspots, capturing different hotspot locations, traffic intensities, and spatial spreads.The hotspot weights satisfy Σℓ αℓ = 1, while µℓ specifies each center and Σℓ its spatial spread.
1) System, Signal, and Traffic Models:: The region is discretized into grids, and all N antennas jointly serve each active user with maximum-ratio transmission under probabilistic LoS and residual NLoS channel modeling.The traffic-weighted network-average SNR gives greater importance to improvements at grids where users are more likely to appear.

2) Pinching-Antenna Location Optimization: · 3) Performance Illustration and Design Insights:

The section formulates traffic-aware pinching-antenna location optimization as a long-term, region-level design and illustrates its benefits over simpler placement benchmarks. Its separable structure enables independent one-dimensional searches, while simulations show optimized placement best serves traffic-weighted network performance.

2) Pinching-Antenna Location Optimization:: The location problem jointly optimizes radiation points for representative user locations weighted by their traffic probabilities, rather than serving only an instantaneous user.This produces a long-term radiation configuration for region-level service.
2) Pinching-Antenna Location Optimization:: Rearranging the traffic-weighted objective decomposes the generally nonconvex problem into N independent one-dimensional location-optimization problems.The separable structure is the key simplification used by the proposed design.
2) Pinching-Antenna Location Optimization:: Each one-dimensional objective combines hotspot contributions with traffic weights and may contain multiple stationary points.The method brackets stationary points coarsely, refines them by bisection, and selects the candidate with the largest objective value.
3) Performance Illustration and Design Insights:: The representative evaluation uses a 60 m × 200 m region, three randomly located hotspots, and averages results over 100 independently generated traffic maps.The simulation uses 28 GHz carrier frequency, 40 dBm total transmit power, and 400 × 120 spatial grids.
3) Performance Illustration and Design Insights:: Optimized traffic-aware placement achieves the highest traffic-weighted network average SNR across all tested numbers of waveguides and pinching antennas.The comparison is reported against hotspot-center and fixed-antenna schemes in Fig. 11.
3) Performance Illustration and Design Insights:: Hotspot-center placement substantially outperforms fixed half-wavelength antenna placement, confirming the benefit of positioning radiation points near high-demand areas.However, hotspot-center placement does not jointly model hotspot intensity, spatial spread, propagation distance, and LoS availability.
3) Performance Illustration and Design Insights:: Additional evaluations compare exhaustive position search and projected-gradient optimization, analyze computational complexity and region size, and consider fairness-oriented worst-average-SNR design.These studies are identified as extensions in the cited work.

C. Geometry-Aware Region-Level Design · 1) System, Channel, and Signal Models: · 2) Pinching-Antenna Activation Optimization:

The geometry-aware design uses obstacle-derived propagation maps to select one radiation location per waveguide and maximize region-wide coverage under an average-SNR requirement. Its binary coverage optimization admits an exact mixed-integer formulation but can be computationally expensive at scale.

1) System, Channel, and Signal Models:: The system divides the valid service region into grids and provides each waveguide with M predefined candidate pinching-antenna locations.Obstacle footprints are excluded from the service region, while each waveguide activates exactly one candidate location.
1) System, Channel, and Signal Models:: The geometry-aware channel combines deterministic LoS states with corresponding channel coefficients for every candidate-location and valid-grid pair.The effective channels from all waveguides depend on the selected activation matrix and feed the MRT signal and SNR model.
1) System, Channel, and Signal Models:: With a fixed activation matrix, activated antennas serve one scheduled user using MRT, and coverage is defined by whether average SNR reaches Γth.The coverage probability is the fraction of valid service-region grids whose average SNR satisfies the prescribed threshold.
1) System, Channel, and Signal Models:: Deterministic obstacle geometry specifies LoS conditions and supports a geometry-dependent average-SNR map for selecting radiation locations online.LoS is deterministic once obstacle layouts and candidate locations are known; the resulting SNR terms can be computed offline and stored.
2) Pinching-Antenna Activation Optimization:: The representative activation problem maximizes the number of covered grids by selecting one binary candidate location per waveguide.Unlike traffic-weighted design, all valid grids receive equal importance, so the objective maximizes the fraction meeting the prescribed SNR requirement.
2) Pinching-Antenna Activation Optimization:: The coverage optimization is difficult because its objective contains discontinuous indicators and binary activation variables.An equivalent mixed-integer linear program introduces binary coverage variables linked to per-grid SNR constraints.
2) Pinching-Antenna Activation Optimization:: The mixed-integer linear formulation is exact for moderate problem sizes, but its computational cost can increase with larger grids or denser candidate sets.A lower-complexity coordinate-ascent method is also introduced for reducing solution complexity.

3) Performance Illustration and Design Insights: … 1) Leaky-Coaxial-Cable-Based Realization:

Geometry-aware radiation-location activation improves region-wide coverage by exploiting propagation maps, while generalized pinching-antenna realizations extend this flexibility across deployment technologies such as dielectric waveguides and LCXs. LCX channels combine in-cable propagation, wireless propagation, attenuation, and directional leakage, enabling favorable desired links and weaker cross-links.

3) Performance Illustration and Design Insights:: Geometry-aware activation achieves the highest coverage probability over most SNR-threshold values, exploiting LoS and distance maps to avoid severe blockage and long propagation distances.Coverage decreases as the required SNR threshold increases because fewer valid grids satisfy the stricter requirement.
D. Summary and Outlook: Traffic-aware design uses long-term demand distributions, whereas geometry-aware design uses obstacle information to improve expected user service quality and region-wide coverage, respectively.These designs extend optimization from individual links to entire communication regions.
D. Summary and Outlook: Region-level optimization can jointly configure waveguide infrastructure, radiation locations, and radio resources, including power, beamforming, bandwidth, scheduling, and user association across different time scales.Relevant infrastructure variables include waveguide number, positions, orientations, heights, and lengths.
D. Summary and Outlook: Combining traffic maps with geometry maps can guide service-region configuration, while time-varying conditions, imperfect information, mobility, coordination, and reconfiguration constraints remain important challenges.This broadens region-level design toward planning, configuring, and operating reconfigurable wireless service regions.
A. Generalized Pinching-Antenna Systems: Generalized pinching-antenna systems distribute signals through extended structures and provide configurable radiation or reception locations, while allowing implementation choices based on frequency, geometry, infrastructure, flexibility, and cost.Representative realizations include dielectric waveguides, leaky coaxial cables, radio stripes, and other conductive or guiding structures.
1) Leaky-Coaxial-Cable-Based Realization:: Leaky coaxial cables radiate through periodic slots, traditionally providing reliable coverage in confined or elongated environments, with controllable slot-blocking structures enabling more flexible operation than fixed manufactured slots.Conventional slot locations and operating states are generally determined during cable manufacture and installation.
1) Leaky-Coaxial-Cable-Based Realization:: An LCX channel comprises two propagation stages: feed-to-slot transmission inside the cable followed by wireless slot-to-user propagation, with attenuation and phase accumulation governed by their respective distances.The model includes longitudinal attenuation, relative permittivity, and the slot-to-user elevation angle.
1) Leaky-Coaxial-Cable-Based Realization:: LCX directional leakage makes channel strength depend on distance, in-cable attenuation, and radiation direction: users beneath an active slot obtain shorter paths and larger gains than laterally separated users.Selecting slots aligned with desired users can strengthen desired links and make oblique cross-links weaker, supporting EDMA.

2) Radio-Stripe-Based Realization: … D. Summary and Outlook

The paper generalizes pinching-antenna realization from active radio stripes to other extended structures and organizes deployments from standalone systems to integrated and distributed networks. It emphasizes joint hardware–communication design, practical implementation tradeoffs, and experimentally validated models for scalable deployment.

2) Radio-Stripe-Based Realization:: Radio stripes distribute data, control, synchronization, and power to active APUs rather than carrying a common RF wave, enabling independently controlled radiation points without in-medium RF attenuation or phase terms.Their large aperture and user-proximate activation locations motivate spherical-wave channel models and reconfigurable access geometry.
2) Radio-Stripe-Based Realization:: Generalized radio stripes treat APU activation as a design variable, jointly controlling effective access locations, beamforming, and power or receive combining.Unlike conventional radio stripes with fixed distributed antennas, activation creates or removes radiation or reception points but incurs circuit-power, fronthaul, synchronization, and hardware costs.
3) Other Pinching-Inspired Realizations:: Other pinching-inspired realizations use metallic pipes, wires, engineered surfaces, or conductive structures to transport signals and establish wireless links at selected locations through passive or active radiators.Their performance depends on the supporting structure, transport loss, coupling and radiation efficiency, and available activation and control hardware.
B. Representative Network Deployment Modes: Pinching-antenna networks support standalone, conventional-BS-integrated, and distributed deployments according to service area, infrastructure, coverage, and coordination requirements.These modes determine whether systems operate independently, complement conventional coverage, or form coordinated access infrastructure.
1) Standalone deployment:: Standalone deployments connect waveguides to a dedicated BS or processing unit, configuring radiation or reception points for users and propagation conditions in relatively self-contained environments.Representative environments include factories, warehouses, shopping malls, exhibition halls, tunnels, and transportation facilities.
2) Integrated Deployment with Conventional BSs:: Integrated deployments complement conventional BS antennas with waveguide-mounted pinching antennas for coverage extension, improved LoS connectivity, traffic offloading, spatial cooperation, or remotely fed access.Architectures include cooperative BS–pinching transmission, wireless feeding, and full-duplex amplify-and-forward relaying.
3) Distributed or Cell-Free-Inspired Deployment:: Distributed or cell-free-inspired deployments coordinate multiple pinching-antenna-equipped waveguides or BSs through common processing, serving users by selected or jointly transmitting points without rigid cell boundaries.Centralized, partially coordinated, and user-centric designs trade cooperation gain against processing complexity, information exchange, and scalability.
C. Implementation and Deployment Considerations; D. Summary and Outlook: Practical deployment requires co-designing waveguide layout, segmentation, feed configuration, radiator geometry, activation, synchronization, calibration, channel acquisition, and control with wireless operation.Longer waveguides extend configurable access but increase attenuation and maintenance difficulty, while segmentation can reduce propagation loss and leakage; future progress requires validated models, scalable control, prototypes, and field trials.

VII. FUTURE DIRECTIONS … VIII. CONCLUSIONS

The paper identifies future research directions enabled by radiation-location reconfigurability, spanning mobility, environment knowledge, AI-native control, learning services, programmable access, and emerging wireless systems. It concludes that configurable radiation locations extend network design beyond deployment-fixed infrastructure while requiring advances across physical realization, channel acquisition, synchronization, and control.

VII. FUTURE DIRECTIONS; A. Pinching Antennas for Future Wireless Networks; VIII. CONCLUSIONS: Radiation-location reconfigurability can dynamically adapt service points, serving relationships, and radio topology to mobility, traffic demand, and propagation conditions.The conclusion frames configurable radiation locations as a network resource available after infrastructure deployment.
1) Mobility-Aware Network Operation:: Mobility-aware operation should coordinate radiation-point repositioning with trajectory prediction, environment knowledge, robust control, and handover decisions across antennas, waveguides, and base stations.This coordination must accommodate prediction errors and unexpected blockage.
2) Environment Knowledge Acquisition and Map-Assisted Operation:: Channel and traffic maps can form a closed loop in which configurable antennas collect spatially diverse observations and updated knowledge guides positioning, association, scheduling, and resource allocation.Future work includes adaptive measurement-location selection, joint map updating, and uncertainty-aware configuration.
3) Network Control and AI-Native Operation:: Hierarchical and AI-native control could jointly configure radiation topology and communication resources while continuously acquiring environment and traffic information and predicting their evolution.Existing learning studies address placement, activation, beamforming, and power allocation, but future systems must manage decisions across spatial and temporal scales.
4) Learning-Oriented Network Services:: In synchronous federated learning, joint radiation-location and client-participation optimization can reshape uplink latency, reduce severe stragglers, and balance statistical importance against tail latency.Large fast–slow client latency gaps amplify the penalty of selecting slow clients, while distributed extensions must adapt across access points or waveguides and learning rounds.
5) Programmable and Virtualized Radio Access Networks:: Programmable and virtualized radio access can group waveguides and antennas into logical access points, cells, or service zones according to traffic and service requirements.Implementing this vision requires abstractions and orchestration that account for environment, interference, neighboring configurations, and physical controllability.
B. Integration with Emerging Wireless Systems; 1) Space–Air–Ground Integrated Networks:: In space–air–ground networks, pinching antennas can provide configurable terrestrial access or relay points, and joint satellite precoding and positioning improved energy efficiency over direct transmission and conventional relays.The studied LEO relay forwards satellite signals through a dielectric waveguide to users affected by blockage.
2) Integrated Sensing and Communication; 3) Wireless Networks for the Low-Altitude Economy:: Pinching antennas can jointly shape communication and sensing geometry, enlarge the communication–sensing rate region over fixed antennas, and support adaptive sensing of moving or insufficiently observed targets.Low-altitude networking introduces three-dimensional UAV mobility, motivating vertical deployments, route-aware activation, and future trajectory-aware association and handover.