CL-RA-Aided Uplink System

Updated 15 January 2026

CL-RA-aided uplink systems are wireless architectures that integrate cross-linked physical designs with resource allocation methods for enhanced performance and reduced hardware complexity.
They combine innovations such as movable and rotatable antenna arrays with joint optimization of beamforming and array configuration to maximize spectral and energy efficiency.
Joint optimization frameworks in these systems yield significant performance gains, robust multi-user interference mitigation, and scalability in massive MIMO and heterogeneous networks.

A CL-RA-aided uplink system refers broadly to uplink wireless communication systems in which the physical, analog, or digital infrastructure is augmented by some form of cross-linking (CL) and/or resource allocation (RA) strategy—often with an explicit connection to cross-linked antenna architectures, cross-cell coordination, or multi-tier network elements. Such systems frequently combine hardware innovations (e.g., cross-linked movable or rotatable antenna arrays) and advanced resource allocation or beamforming mechanisms, to achieve major gains in spectral efficiency, energy efficiency, and scalability in both massive MIMO and heterogeneous network contexts. This entry surveys the principal CL-RA-aided uplink architectures, signal models, optimization frameworks, and performance results provided in the current research literature.

1. Cross-Linked Physical Architectures

The CL-RA paradigm in uplink systems arises from several key physical-level innovations that generalize the concept of cross-linking beyond simple spatial multiplexing:

Cross-Linked Movable Antenna (CL-MA) Arrays: In this design, an $M\times N$ antenna grid is created and the elements are mounted at the intersections of $M$ vertical and $N$ horizontal mechanical tracks. Each row or column of antennas is collectively position-tunable by moving its corresponding track, allowing $M+N$ actuators to control all $MN$ antennas simultaneously—substantially reducing hardware complexity compared to per-element actuators. These architectures enable the array response to be flexibly adapted in both horizontal and vertical dimensions, facilitating powerful multi-user interference suppression (Zhu et al., 6 May 2025).
Cross-Linked Rotatable Antenna (CL-RA) Arrays: Each element in an $M\times N$ array is mounted such that its pointing (orientation) is determined by a combination of row-shared elevation and column-shared azimuth angle. This configuration enables both hardware-efficient collective control of orientations and high-fidelity three-dimensional beam adaptation. Both element-level (per-antenna) and panel-level (per-subarray) cross-linking schemes are defined, where panelization merges several rows or columns for joint orientation control (Zheng et al., 8 Jan 2026).

These hardware structures drastically reduce actuator or control circuit count, while retaining most of the spatial degrees of freedom available to individually controlled elements. Notably, performance gaps between CL-RA and fully flexible architectures are minimal in optimized multi-user settings.

2. System and Channel Models

CL-RA-aided uplink systems can be found across a spectrum of topologies, unified by the structuring of channel access or physical antenna response:

Multiuser Uplink Models: The standard paradigm is $K$ single-antenna users transmitting to a base station equipped with an $M\times N$ CL-MA or CL-RA array. The channel for each user is constructed using field-response or geometry-based line-of-sight (LoS) plus non-LoS components. For CL-MA, the total channel for each user $k$ is a Khatri–Rao product of horizontal and vertical steering vectors as a function of the collectively selected antenna tracks (Zhu et al., 6 May 2025). For CL-RA, the channel includes orientation-induced elementwise gains and phasing (Zheng et al., 8 Jan 2026).
Optimization Variables: Cross-linked architectures introduce block variables (antenna track positions; row/column rotation angles). Resource allocation then proceeds via joint optimization of these structural variables with beamformers or combiners.
Constraints: Mechanically feasible regions for positions or rotation angles, minimal element spacing to preclude physical collision, and per-user SINR or rate requirements under total or per-node power constraints are imposed.

3. Joint Optimization and Algorithmic Frameworks

CL-RA-aided uplink systems are unified by the need to jointly optimize physical array configuration and digital resource allocation:

Sum-Power Minimization (CL-MA): The objective is to minimize the total transmit power required to satisfy user rate constraints, via joint selection of antenna position vectors (APVs) and receive combining at the base station (Zhu et al., 6 May 2025). In the single-path scenario, the globally minimum power admits a closed-form, attained when array positions give orthogonal channel vectors for all users—a condition achievable via a factorization of $M$ , $N$ when $K(K-1)/2 \leq I_M+I_N$ (where $I_M$ / $I_N$ are the total number of prime factors of $M$ , $N$ , respectively).
Alternating Optimization (CL-RA/CL-MA):
- In multi-path or multi-user scenarios, sequential elimination plus refinement low-complexity algorithms select discrete APVs on a grid, converging in $<100$ iterations (Zhu et al., 6 May 2025).
- For CL-RA element-level rotation, an alternating optimization alternates between MMSE combining and a feasible-direction convex local update for row/column angles (Zheng et al., 8 Jan 2026). Discrete actuation is handled via GA-based combinatorial search over quantized angle sets, recovering most sum-rate performance.
Spectral/Energy Efficiency Maximization: In panel- or element-level CL-RA, the sum-rate is maximized over the joint configuration of analog variables and combiners, with performance increasing monotonically with the actuation dimension but exhibiting diminishing returns as hardware constraints are relaxed (Zheng et al., 8 Jan 2026).

4. Performance Analysis and Empirical Results

Comprehensive simulation studies confirm pronounced gains from CL-RA-aware uplink schemes versus conventional fixed-array or uncoupled architectures:

Architecture	Power/Rate Gain over Baseline	Hardware Reduc.	Movement/Orientation Savings
CL-MA Array	Up to 30 dB in sum-power (Zhu et al., 6 May 2025)	$O(M+N)$ actuators (vs $O(MN)$ )	Orders of magnitude (statistical APV)
CL-RA Element	128% throughput improvement (Zheng et al., 8 Jan 2026)	$M+N$ actuators	Significant (GA/partitioning)
CL-RA Panel	25% lower sum-rate than element-level	Fewer actuators	Some losses at high panelization
CL-RA Discrete	84% of element-level sum-rate (L=15)	Quantized control	15% gain over naive quantization

Diminishing Returns: As the number of structural degrees of freedom (tracks/angles/panels) increases, further rate or power savings saturate, especially when channel orthogonality becomes saturated or physical aperture bounds are tight.
Statistical vs. Instantaneous Optimization: APV/angle optimization using long-term channel statistics, as opposed to per-fading realignment, incurs minor per-user rate loss ( $\sim$ 2–4 dB) but reduces repositioning overhead by orders of magnitude (Zhu et al., 6 May 2025).
Discrete Actuation: Optimization using quantized position or angle sets converges to performance close to the continuous case with moderate ( $L_{\text{angle}} \sim 15$ ) control levels (Zheng et al., 8 Jan 2026).

5. Comparison with Other CL-RA Paradigms

CL-RA-aided uplink systems encompass but are not limited to cross-linked antenna hardware:

Resource Allocation Approaches:
- In cellular uplinks (e.g., LTE), CL-RA refers to joint multi-cell resource assignment using pure BIP, where cross-cell assignment of contiguous RB patterns mitigates inter-cell interference and yields 14–22% improvements in spectral efficiency and 5th percentile user rate (Zhang et al., 2014).
- In cloud-RAN and coordinated beamforming, CL-RA methods jointly optimize RRH clustering, beam selection, user scheduling, and power allocation, sometimes using hierarchical or decentralized algorithms but often under the umbrella of globally coordinated resource assignment.
Physical Layer CL-RA: In mmWave uplink systems, "CL-RA" concepts extend to robust precoding, with hierarchical, location-aided, decentralized beam coordination mitigating interference under position uncertainty (Maschietti et al., 2017).

6. Implementation Challenges and Design Guidelines

A variety of practicalities and design insights emerge across the literature:

Hardware Scaling: The pivot from $MN$ to $M+N$ (CL-MA/CL-RA) actuator scaling is essential for enabling massive or ultra-high-dimensional arrays without commensurately exploding hardware cost and wiring complexity (Zhu et al., 6 May 2025, Zheng et al., 8 Jan 2026).
Actuator Resolution: Fine angle or position quantization at moderate resolution suffices for most of the performance benefits; further refinement gives only marginal improvement.
Panelization: Partitioning CL-RA arrays into subarrays ("panels") can offer intermediate tradeoffs between performance and complexity, but over-partitioning can degrade spatial coverage due to angular constraints (Zheng et al., 8 Jan 2026).
Optimization Tradeoffs: Statistical, long-term optimization is critical for practical deployment to amortize movement/rotation cost; offline/discrete search methods like the greedy and GA approaches are favored.
Robustness: Physical cross-linking naturally yields more robust, less fragile spatial processing, and cross-linked resource allocation mitigates both self-interference (same cell) and multi-user/cell interference (cooperation).

7. Research Directions and Open Problems

Several open issues and themes emerge:

Scaling and Centralization: While CL-RA yields significant hardware reduction, fully centralized optimization can become computationally or signaling intensive in ultra-dense scenarios, motivating distributed/localized algorithmic variants.
Joint Physical-Digital CL-RA: There is strong potential for combining cross-linked physical architectures and resource allocation with digital processing (e.g., joint CL-RA and fronthaul compression, or CL-RA and quantization-aware combining) for large wireless networks (Liu et al., 2015, Chen et al., 2024).
Heterogeneous Networks: Extension of CL-RA concepts to relayed, RIS-assisted, or cell-free architectures—especially with hardware constraints, multi-tier fronthaul, and statistical CSI—remains a promising research direction.
Massive User/Track Scaling: Algorithmic frameworks for CL-RA hardware with very large $M$ , $N$ , or user counts, particularly under strict latency and movement budgets, are required to capitalize on emerging dense-array paradigms.

Collectively, CL-RA-aided uplink systems integrate hardware cost reduction, high spatial flexibility, and advanced resource allocation to realize superior performance in multi-user, multi-antenna scenarios, with practical algorithms that bridge the gap between theoretical limits and realistic system constraints (Zhu et al., 6 May 2025, Zheng et al., 8 Jan 2026, Zhang et al., 2014).