Transmit Antenna Selection (TAS)

• Transmit Antenna Selection (TAS) is a spatial processing technique that selects optimal antennas based on metrics such as SNR, shadowing, or security criteria to enhance wireless performance.
• It employs methods like main-channel SNR maximization, shadowing-based selection, and optimization-driven algorithms to reduce RF-chain complexity and feedback overhead.
• TAS is applied in MIMO, NOMA, and IRS-assisted systems to achieve diversity gains, mitigate interference, and improve secrecy capacity under varying channel conditions.

Transmit antenna selection (TAS) is a fundamental spatial processing technique for multi-antenna wireless systems, whereby a subset of transmit antennas is selected for communication based on certain channel metrics or system constraints. TAS enables spatial diversity, reliability, power efficiency, and security, while reducing hardware complexity by minimizing the number of active RF chains required at the transmitter. The methodology, theoretical analyses, and practical implications of TAS depend strongly on system architecture, channel conditions, available channel state information (CSI), optimization metrics, and additional functionalities such as space-time coding, multiple-access, or physical-layer security enhancements.

1. Core Algorithms and Selection Criteria

The canonical TAS procedure involves measuring the instantaneous or statistical channel quality associated with each transmit antenna, and selecting one or more antennas to optimize a specific metric; typically, instantaneous SNR at the receiver, secrecy capacity, or outage probability. The selection may be based on:

• Main channel CSI-driven selection: The transmitter selects the antenna(s) yielding the maximum main-channel SNR; e.g., for MIMO wiretap channels, the two antennas maximizing $\|\mathbf{f}_\alpha\|$ are chosen when applying an Alamouti code (Yan et al., 2013).
• Shadowing side information (SSI)-based selection: Selection is performed based solely on slowly-varying shadowing coefficients $g_\ell$, significantly reducing feedback rate and channel estimation overhead; the antenna with the highest $g_\ell$ is chosen (Yilmaz et al., 2012).
• Majority-based voting: In multi-user contexts (e.g., downlink NOMA), each user votes for the transmit antenna maximizing its estimated channel, and the antenna with the highest vote count is selected (Aldababsa et al., 2019).
• Optimization-driven selection: Advanced frameworks employ sequential quadratic programming or combinatorial search to select antennas that, in conjunction with optimal power allocation, minimize total transmit/reception power while meeting rate constraints (Zhu et al., 2022).
• Security-oriented selection: To maximize secrecy rates, selection may be performed based on the minimization of eavesdropper SNR or by leakage-based criteria, especially when the eavesdropper's channel is strong (Anaya-López et al., 2021, Shu et al., 2018).

2. TAS in Coded and Diversity-Enhancing Structures

TAS is often integrated with space-time coding (STBC), MIMO diversity schemes, and physical-layer security coding for simultaneous diversity and reliability:

• TAS + Alamouti/OSTBC: Selecting two strongest antennas, followed by Alamouti (2×2 OSTBC) transmission, achieves order-2 spatial diversity and improves secrecy for wiretap channels; a closed-form secrecy outage probability for arbitrary $N_A$, $N_B$, $N_E$ is available (Yan et al., 2013). Similar constructs apply to IRS-assisted massive MIMO links with hybrid beamforming (Elganimi et al., 2022).
• TAS/STBC under Feedback Errors: Cross-analysis of joint transmit-receive antenna selection (TRAS/STBC) and TAS/STBC reveals that diversity is lower-bounded by STBC order even with feedback link errors. Analytical outage and BER/SER performance is quantified using Lauricella and hypergeometric functions for Nakagami-m fading (Coskun et al., 2012).
• Physical-Layer Network Coding: For two-way relay channels, antenna selection based on maximizing Euclidean inter-cluster distance yields diversity order $\min\{N_A, N_B\}$, outperforming strongest-SNR selection which can suffer diversity degradation in higher-order modulations (Kumar et al., 2017).

3. TAS for Secure and NOMA Communications

Security, spectral efficiency, and multi-user fairness are advanced with TAS in several ways:

• PLS-Enhanced Wiretap/TAS Schemes: In MIMO wiretap channels, TAS-Alamouti performs selection using only main-channel CSI, boosting Bob's SNR more than Eve's when $\overline\gamma_B/\overline\gamma_E$ exceeds a threshold; secrecy outage probabilities and $\epsilon$-outage capacities are derived (Yan et al., 2013).
• Strong Eavesdropper Regime: When eavesdropper SNR is comparable/superior, eavesdropper-based TAS (minimizing Eve's channel gain) substantially outperforms conventional Bob-centric selection; closed-form average secrecy capacities for both criteria are provided (Anaya-López et al., 2021).
• Secure MIMO-NOMA Networks: For near/far users in NOMA, selection rules maximizing received signal power for either user deliver closed-form and asymptotic secrecy outage probabilities. Near-user diversity scales with transmit antennas only if selection is near-user-centric, while far-user diversity order saturates (Tran et al., 2018, García et al., 29 Nov 2025). Majority-based TAS further optimizes fairness for most users under imperfect CSI and feedback delay (Aldababsa et al., 2019).

4. Impact of TAS in System-Level and Hardware Design

TAS enables marked improvements in several hardware and system aspects:

• Reduced RF-chain and ADC Complexity: TAS allows system operation with only one (or a few) RF chains even in large antenna arrays, considerably reducing cost and power (Choi et al., 2018). In low-resolution ADC systems, TAS increases ergodic rate substantially, saturating at a closed-form limit determined by quantization parameter $\alpha$.
• Cognitive Radio and Interference Control: Underlay cognitive radio systems deploy TAS to strictly bound instantaneous interference to primary receivers; closed-form outage and ergodic capacity calculations incorporate the selection and guarantee primary protection (Hanif et al., 2014).
• Intelligent Reflecting Surface (IRS) Integration: In mmWave IoT designs, combining TAS, OSTBC, hybrid analog/digital beamforming, and IRS yields full diversity and high SNR, with error-rate performance scaling strongly with the number of reflecting elements and subarray/antenna sizes (Elganimi et al., 2022).
• Channel Estimation and Feedback: Shadowing-based TAS radically decreases feedback channel usage and channel estimation overhead, especially beneficial in slow-fading or massive antenna settings (Yilmaz et al., 2012).

5. Performance Analysis, Diversity, and Asymptotics

Theoretical analyses provide closed-form and asymptotic expressions for most relevant performance metrics:

• Diversity Orders: Variable across protocols: TAS-Alamouti provides linear scaling with $N_A$ for secrecy capacity at mid-high SNR (Yan et al., 2013), while some schemes for finite-alphabet modulations yield zero secrecy diversity order due to MI saturation (Ouyang et al., 2019). TAS/STBC achieves full diversity under ideal feedback, falling to the STBC order under feedback error (Coskun et al., 2012).
• Outage/Capacity Expressions: Closed-form formulas for outage probability, moment generating functions, ergodic rates, and symbol error probabilities under various fading models (Rayleigh, Nakagami-m, Generalized-K, $\alpha$–$\mu$) are derived for standard and security-enhanced TAS protocols (Yan et al., 2013, Tran et al., 2018, Yilmaz et al., 2012, García et al., 29 Nov 2025).
• Zero Diversity Floor and Practical Saturation: In NOMA and finite-alphabet wiretap channels, overall system SOP or ergodic secrecy rates saturate at high SNR; the diversity order can collapse to zero if AN-induced noise or imperfect SIC dominates, or due to MI bounds in BPSK/QPSK (Ouyang et al., 2019, García et al., 29 Nov 2025).

6. Advanced Optimization and Machine Learning for TAS

Modern research applies advanced optimization and learning techniques:

• Sequential Quadratic Programming (SQP) and Boolean Programming: Power-minimization with discrete antenna switching is solved via AD-based SQP with Boolean QP substeps; near-exact Boolean solutions are obtained in a handful of iterations (Zhu et al., 2022).
• Subarray Selection in Hybrid Precoding Architectures: Secure spatial modulation systems employ three main TASS methods—Max-ASR (approximate secrecy rate maximization via exhaustive search), Max-EV (largest singular value eigen-selection), and Max-P-SINR-ANSNR (product SINR/AN-SNR)—to balance complexity with secrecy performance (Shu et al., 2020).
• Deep Learning-Based Selection: In secure relay networks, deep neural networks can learn highly nonlinear mappings from channel features to optimal antenna subsets, achieving near-exhaustive performance at negligible online complexity (Yao et al., 2019).

7. Comparative Summary of TAS Schemes

TAS Scheme	Selection Metric	Complexity	Diversity Scaling / Practical Impact
Max-SNR (main channel)	$\max_\alpha \|\mathbf{f}_\alpha\|$	O(N) – O(N^2)	Linear at mid-high SNR; limited for BPSK/QPSK (Yan et al., 2013, Ouyang et al., 2019)
Shadowing-based	$\max_\ell g_\ell$	O(L), infrequent	High feedback/estimation savings; SSI gain (Yilmaz et al., 2012)
Majority-voting (multi-user)	Per-user vote tally	O(N^2)	Optimizes for majority fairness, full diversity under ideal CSI (Aldababsa et al., 2019)
Sequential QP / AD-SBQP	Power + rate constraints	O(N^3) per pass	Near-exact Boolean selection, fast convergence (Zhu et al., 2022)
Leakage/Min-Eve based	Minimize Eve SNR/SLNR	O(N log N)	Superior secrecy in strong-eavesdropper regimes (Anaya-López et al., 2021, Shu et al., 2018)
Secure-NOMA, TASS-II	Max-Far-user SNR	O(N)	Far-user outage minimized, overall SOP improved (Tran et al., 2018, García et al., 29 Nov 2025)
Deep learning (DNN)	Learned mapping	O(1) online	Near-exhaustive optimal subset, nonlinear mapping (Yao et al., 2019)

• "Transmit Antenna Selection with Alamouti Scheme in MIMO Wiretap Channels" (Yan et al., 2013)
• "Secure Transmit Antenna Selection Protocol for MIMO NOMA Networks over Nakagami-m Channels" (Tran et al., 2018)
• "On the Performance of Transmit Antenna Selection Based on Shadowing Side Information" (Yilmaz et al., 2012)
• "Alternating Direction Based Sequential Boolean Quadratic Programming Method for Transmit Antenna Selection" (Zhu et al., 2022)
• "Transmit Antenna Selection in Underlay Cognitive Radio Environment" (Hanif et al., 2014)
• "Analysis of Ergodic Rate for Transmit Antenna Selection in Low-Resolution ADC Systems" (Choi et al., 2018)
• "Majority Based TAS/MRC Scheme in Downlink NOMA Network with Channel Estimation Errors and Feedback Delay" (Aldababsa et al., 2019)
• "Two High-performance Schemes of Transmit Antenna Selection for Secure Spatial Modulation" (Shu et al., 2018)
• "Transmit Antenna Selection for Physical-Layer Network Coding Based on Euclidean Distance" (Kumar et al., 2017)
• "Outage Analysis of TAS-NOMA Systems With Multi-Antenna Users Over α-μ Fading" (García et al., 29 Nov 2025)
• "Deep Learning Assisted Antenna Selection in Untrusted Relay Networks" (Yao et al., 2019)
• "A New Transmit Antenna Selection Technique for Physical Layer Security with Strong Eavesdropping" (Anaya-López et al., 2021)
• "Characterization of Effective Capacity in Antenna Selection MIMO Systems" (Lari et al., 2016)
• "Error Performance of Various QAM Schemes for Nonrenegerative Cooperative MIMO Network with Transmit Antenna Selection" (Shaik et al., 2018)
• "Secrecy Performance of Antenna-Selection-Aided MIMOME Channels with BPSK/QPSK Modulations" (Ouyang et al., 2019)
• "IRS-Assisted Millimeter-wave Massive MIMO with Transmit Antenna Selection for IoT Networks" (Elganimi et al., 2022)
• "Precoding and Transmit Antenna Subarray Selection for Secure Hybrid Spatial Modulation" (Shu et al., 2020)
• "Unified Analysis of Transmit Antenna Selection/Space-Time Block Coding with Receive Selection and Combining over Nakagami-m Fading Channels in the Presence of Feedback Errors" (Coskun et al., 2012)