Bridging the MIMO Scalability Gap in Wireless Networks

95 concepts101 papers1 proposal

For decades, wireless network optimization enjoyed a convenient mathematical simplification: the single-antenna assumption. In this paradigm, the channel between a transmitter and a receiver is modeled as a complex scalar or, at worst, a vector. This keeps the underlying mathematics highly tractable, allowing complex problems like load balancing, user scheduling, and interference management to be reduced to well-behaved, convex power-allocation strategies.

However, this tractability comes at a steep cost. As H. H. M. Tam et al. noted in Joint Load Balancing and Interference Management for Small-Cell Heterogeneous Networks With Limited Backhaul Capacity, while single-antenna models allow for efficient computational solutions, they fail to capture the physical realities of modern multi-antenna deployments. By forcing nodes into single-antenna constraints to preserve mathematical tractability, traditional frameworks blind themselves to critical spatial multiplexing and beamforming trade-offs, a limitation highlighted by Xidong Mu et al. in Capacity and Optimal Resource Allocation for IRS-Assisted Multi-User Communication Systems.

The Tyranny of the Scalar Channel

To understand why the single-antenna constraint is so pervasive—and so difficult to escape—we must look at how interference is modeled. In a single-antenna system, interference is purely an isotropic volume knob. It is a scalar value added to the noise floor, and the primary way to mitigate it is to turn down transmit power, shift frequencies, or schedule users at different times.

In modern Multiple-Input Multiple-Output (MIMO) systems, however, interference has a direction. An antenna array can shape signals into narrow beams, directing energy toward a desired user while placing "nulls" in the direction of unintended receivers.

Why can't we simply plug multi-antenna parameters into existing single-antenna optimization frameworks? The answer lies in the transition from scalar algebra to high-dimensional matrix calculus. When a receiver possesses an antenna array, the channel becomes a matrix ($\mathbf{H}$). Managing this matrix introduces complex, non-convex constraints.

For example, in physical layer security, optimizing beamforming to keep data secret from an eavesdropper is straightforward if the legitimate user has one antenna. However, as Hoang Duong Tuan et al. demonstrate in Beamforming Optimization for Physical Layer Security in MISO Wireless Networks, moving to multi-antenna receiver architectures introduces intricate matrix inequalities that standard secrecy throughput optimization frameworks cannot easily ingest.

Similarly, fundamental information-theoretic boundaries do not scale linearly from single to multiple antennas. As Mahtab Mirmohseni et al. showed in Capacity bounds for multiple access-cognitive interference channel, capacity bounds derived for single-input single-output (SISO) architectures fundamentally fail to map to MIMO cognitive networks due to the high-dimensional interactions of spatial diversity.

The Multi-User Scaling Bottleneck

The scalability gap worsens when moving from point-to-point links to multi-user environments. Consider emerging 6G paradigms like Reconfigurable Intelligent Surfaces (RIS) and Non-Orthogonal Multiple Access (NOMA). In a single-antenna setup, an RIS simply acts as a passive phase-shifter to boost a scalar channel's amplitude. Jiakuo Zuo et al. in Reconfigurable Intelligent Surface Enhanced NOMA Assisted Backscatter Communication System point out that while single-antenna configurations simplify phase-shift optimization, they completely miss out on spatial multiplexing.

When we introduce multi-antenna base stations and user equipment, the optimization problem shifts from finding an optimal phase angle to jointly optimizing high-dimensional active beamforming vectors at the base station and passive beamforming matrices at the surface. As Wanli Ni et al. illustrate in Resource Allocation for Multi-Cell IRS-Aided NOMA Networks, this joint optimization introduces severe non-convexity and tight coupling between variables, making real-time multi-user MIMO coordination an open algorithmic challenge.

Single-Antenna Model (Scalar)
[Tx] -----( h: Complex Number )-----> [Rx]
* Interference is just noise volume. Easy math, low capacity.

Multi-Antenna Model (Matrix)
[Tx] =====( H: High-Dimensional Matrix )=====> [Rx]
* Interference has spatial direction. Hard math (non-convex), high capacity.

From Point-to-Point to Spatial Coordination

To close this scalability gap, researchers are shifting from isolating single links to designing frameworks that explicitly handle multi-antenna spatial dimensions. For wideband Terahertz massive MIMO, Linglong Dai et al. proposed Delay-Phase Precoding in Delay-Phase Precoding for Wideband THz Massive MIMO to resolve the "beam split" effect. The authors noted that while single-user beam alignment is well-understood, scaling these architectures to align beams for multiple users simultaneously across massive bandwidths requires a significant leap in hardware and mathematical coordination.

Furthermore, the single-antenna assumption limits our understanding of network density scaling laws. Suhas S Kowshik et al. proved in Fundamental Limits of Many-User MAC With Finite Payloads and Fading that near-perfect interference cancellation is theoretically possible in dense networks, but their bounds remain restricted to single-antenna receivers.

To unlock the true capacity of next-generation networks, the wireless community is actively dismantling these single-antenna modeling constraints. By replacing scalar approximations with robust, high-dimensional matrix optimization frameworks, modern research treats space not as a mathematical complication to be avoided, but as the primary resource to be exploited.

Key Foundational Work

Opportunistic Relay Selection with Limited Feedback (2007) — contributed 1 concept elements to this direction.
History & citationVersion 1 · 2026-10-04

Current · Permanent link

View frozen version 1

Cite this direction

Latest version (a living page):

@misc{lacuna_single_antenna_constraint_7a405f54,
  title = {single-antenna constraint},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/single-antenna-constraint/rdr_7a405f54e5fb4275bf05256630410aa7},
  note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-19},
}

This exact version (version 1, frozen text):

@misc{lacuna_single_antenna_constraint_7a405f54_v1,
  title = {single-antenna constraint},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/single-antenna-constraint/rdr_7a405f54e5fb4275bf05256630410aa7/v/1},
  note = {Version 1, frozen 2026-10-04. Accessed 2026-11-19},
}
About

These directions overlap with this one through shared papers.

See something wrong? Report a mistake