Pith. sign in

REVIEW 1 major objections 6 minor 49 references

DFRC Systems Co-existing in Licensed Spectrum: Cognitive Beamforming Designs

T0 review · 1 major / 6 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read A dual-function radar-communication array can operate on licensed spectrum by jointly designing beams that null primary users while tracking targets and serving secondary users.

desk verdict Useful incremental SDR result for cognitive DFRC, but the firefly algorithm enforces the wrong per-antenna power budget and its reported gains may be artifacts. read the letter →

arxiv 2506.01202 v2 pith:KV3SDILI submitted 2025-06-01 cs.IT eess.SPmath.IT

Tuan Anh Le , Ivan Ku , Xin-She Yang , Christos Masouros , Tho Le-Ngoc This is my paper
classification cs.ITeess.SPmath.IT MSC 90C2290C59
keywords cognitiveradiodual-functionradar-communicationbeamformingdesignsemidefiniterelaxationfireflyalgorithmspectrumsharingMIMOradarintegratedsensingandcommunication
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper claims that a dual-function radar-communication (DFRC) base station, which shares spectrum owned by a primary system, can design its transmit beams so that it tracks targets and serves secondary users while keeping interference to each primary user below a preset threshold. The beam design is cast as an optimization problem: minimize the mean-square error between the designed radar beam pattern and a desired pattern, subject to per-user SINR requirements, a per-antenna power budget, and per-primary-user interference caps. Two solvers are proposed: a semidefinite relaxation that turns the non-convex problem into a convex one, and a nature-inspired firefly algorithm that handles constraints via penalty terms. Simulations with ten to sixteen antennas show both methods form deep nulls, below -20 dB, at the primary user's direction while keeping beams on targets and secondary users. The price of this cognitive protection is a higher beam-pattern MSE than a non-cognitive design that ignores the primary system.

What carries the argument

The engine of the paper is the beam-pattern MSE written as a quadratic form $r^H \Omega r$, where $r$ stacks $\omega$ and the vectorized covariance $R = W W^H + V V^H$ (equations (11)–(13)), together with three constraints. The SDR approach lifts each rank-one term $w_i w_i^H$ and $v_t v_t^H$ into positive semidefinite matrices $F_i$ and $V_t$, drops the rank-one constraints, and solves the resulting convex problem (14); if the solution is rank-one it is optimal for (7), otherwise a randomization step gives a sub-optimal design. The firefly approach converts the constrained problem into an unconstrained penalized objective (20)–(21), defines firefly brightness as the inverse of the penalized objective (22), and moves fireflies toward brighter ones with attraction $\beta_0 e^{-\gamma r^2}$ (23)–(25). The quadratic-form structure lets both methods share the same objective and constraints, and it is also what drives the stated complexity estimates (Lemma 1 and Lemma 2).

What would settle it

Take the same $M=10$ setup but add a mismatch between the assumed PU angle ($-40^\circ$) and the actual PU angle (say $-38^\circ$), or increase the angular spread from $2^\circ$ to $8^\circ$; if the interference imposed on the PU then exceeds $I_{\text{thres}} = 0.01$ with either proposed algorithm, the central protection claim fails. A second check is a feasibility scan over $(U,K,L)$ with $U+K+L > M$, which should reveal whether the problem becomes infeasible as the degree-of-freedom condition is violated.

Watch

Extended reading notes

Core claim

The central claim is that the optimization problem (7) — minimize the radar beam-pattern MSE (5) over communication beamforming vectors $\{w_i\}$, radar beamforming vectors $\{v_t\}$, and a scaling factor $\omega$, subject to $\text{SINR}_i \geq \eta_i$, per-antenna power $[W W^H + V V^H]_{i,i} \leq P_m/M$, and per-primary-user interference $\sum_j w_j^H H_{p,l} w_j + \sum_t v_t^H H_{p,l} v_t \leq I_{\text{thres}}$ — can be solved by the SDR formulation (14) and by the firefly Algorithm 1, and that both solutions yield beam patterns with deep nulls at the primary users while maintaining the radar beams and secondary-user SINRs. The paper demonstrates this by comparing against non-cognitive DFRC, radar-only, and cognitive-communication-only baselines, showing that the cognitive designs achieve interference below the threshold while the non-cognitive and radar-only designs do not.

Load-bearing premise

The design assumes the base station knows the primary users' channel covariance matrices exactly, including their angles and angular spreads; the deep-null protection guarantee has no formal robustness if that knowledge is imperfect, and the paper defers imperfect CSI to future work.

Editorial extensions

If this is right

  • A DFRC system can operate as a secondary user in licensed spectrum, so a single antenna array can provide radar tracking and data service without dedicated spectrum.
  • The SDR approach returns the optimal beamforming solution when the relaxed solution is rank-one; otherwise the randomization step yields a near-optimal design.
  • The firefly algorithm gives beam patterns nearly identical to SDR with 10–12 antennas and can outperform SDR at low SINR requirements, but needs a larger firefly population to match SDR as the array grows to 16 antennas.
  • The beam-pattern MSE increases compared with non-cognitive DFRC and radar-only designs, quantifying the cost of protecting primary users.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Robust extensions could replace exact PU covariance knowledge with uncertainty sets and re-derive the null constraint; this would likely preserve deep nulls under CSI error at the cost of a slightly wider null.
  • The FA's convergence within about 110 generations suggests the same penalty schedule (with penalty constants growing as $k^2$) may transfer to other non-convex beamforming problems, but the population size must scale with problem size.
  • The stated condition $U + K + L \leq M$ is necessary for spatial degrees of freedom, but the SINR and null constraints may tighten feasibility; a numerical feasibility map across $(U,K,L)$ would show whether the design degrades gracefully or fails abruptly near that boundary.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

1 major / 6 minor

Summary. This paper considers a dual-function radar-communication (DFRC) system that operates on spectrum licensed to a primary system. The authors formulate a joint beamforming optimization problem that minimizes the MSE between the designed and desired radar beam patterns, subject to per-SU SINR constraints, per-antenna transmit-power limits, and per-PU interference thresholds. They propose two solution methods: a semidefinite relaxation (SDR) that yields a convex program, and a nature-inspired firefly algorithm (FA) based on a penalty method. Numerical results compare the proposed designs with non-cognitive DFRC, radar-only, and cognitive-communication-only baselines, and claim that both proposed approaches create deep nulls at the PU while maintaining radar and communication beams.

Significance. If the central claims are supported, the paper offers a useful extension of DFRC beamforming to cognitive spectrum sharing, with a clean SDR formulation and a heuristic nature-inspired alternative. The SDR derivation in Section IV appears correct and is validated against external baselines, and the paper provides complexity estimates and simulation comparisons across a range of numbers of antennas, SINR levels, and interference thresholds. However, the FA formulation contains a constraint error that invalidates the claim that Algorithm 1 solves problem (7) as stated, so the numerical advantages reported for FA versus SDR are not trustworthy until the formulation is corrected and the experiments are re-run.

major comments (1)
  1. [V, Eq. (18)] The FA formulation of the per-antenna power constraint is inconsistent with problem (7). In (7) the constraint is [W W^H + V V^H]_{i,i} ≤ P_m/M, and the SDR formulation (14) correctly retains this. However, Eq. (18) defines d_i({w_i},{v_t}) = R_{i,i} - P_m, and the penalty in (21) becomes ρ_i max{0, R_{i,i} - P_m}^2. Thus the FA penalizes a violation only when an antenna exceeds the total power budget P_m, not the per-antenna budget P_m/M. For the simulations with P_m=1 and M=10, the FA is permitted to use up to ten times the total power budget of the SDR, so the FA's lower MSE in Figs. 7-9 and the deep nulls in Fig. 2 may be artifacts of a more permissive problem. Either the implementation used Eq. (18) (and the reported FA results solve a different problem) or the implementation used P_m/M (and the paper's equations do not describe the experiments). In both cases, the claim that Algorithm 1 solves problem (7) is not supported. Please correct Eq. (18) and re-run the simulations with the stated per-antenna budget.
minor comments (6)
  1. [IV, Lemma 1 and Appendix A] The claimed complexity O(G K M^8) for evaluating Ω appears overestimated: Ω is a sum of G K outer products of (M^2+1)-dimensional vectors, so forming it costs O(G K M^4), and the Hessian of the barrier terms for M × M PSD matrices costs O(M^4) per matrix. The stated O(G K M^8.5) therefore seems to overestimate the scaling, and Lemma 2 inherits the same issue. Please verify the complexity derivations.
  2. [VI] The paper does not state whether the optimal F_i and V_t from the SDR solver were rank-one or whether the randomization step of Section IV was used to obtain the beam patterns in Figs. 2-4. Reporting this would improve reproducibility.
  3. [V, Algorithm 1] Lines 9-14 of Algorithm 1 appear redundant: since the fireflies are ranked in descending order before the loop and B* is initialized to the brightest firefly, the conditions B_n > B* and B_m > B* cannot be true for n,m>1. The best solution is only updated again at line 22. Consider simplifying the pseudocode.
  4. [II] The notation is inconsistent: in the system model, t runs from 1 to M for the radar waveforms v_t and x_t^r, while in the radar metric (5) the outer sum runs over t=1..K. Clarify whether there are M radar beams or K targets.
  5. [II-B] The paper assumes perfect CSI and defers imperfect CSI to future work; this is acceptable, but the PU-protection guarantee is only as good as the CSI, so adding a sentence on this limitation in the conclusion would be helpful.
  6. [Appendix A and VI-D] In Appendix A, the sentence 'the complexity is dominated by the complexity is dominated by' contains a duplicated phrase. In Section VI-D, 'coginitve' should be 'cognitive'.

Circularity Check

0 steps flagged · score 2.0 of 10

Central SDR and FA derivations are self-contained and benchmarked externally; no circular reduction found. Minor self-citations to the authors' own FA framework are not load-bearing, and the Eq. (18) power-budget mismatch is an internal correctness risk, not a circularity.

full rationale

Walking the derivation chain: problem (7) is an explicit extension of the external baseline in [5] obtained by adding the PU-interference constraint; the SDR reformulation (14) follows by setting Fi = wi wi^H, Vt = vt vt^H and relaxing rank-one constraints, so no fitted quantity is renamed as a prediction. Algorithm 1 is a fully specified penalty-method FA with update equations (23)-(25), and its MSE is compared directly with CVX-SDR and with external baselines [5], [30], [49]; therefore the numerical claims are independently checkable. The self-citations [22], [24], [26] motivate the framework and the conference version but do not supply the numerical comparisons, and the convergence claim "The fireflies can theoretically attain the global optimum if k >> 1 and N -> infinity [20]-[22]" is heuristic and is supported in-paper by Fig. 5, so it is not load-bearing. The paper also states "Imperfect CSI is beyond the scope of this work", which limits robustness but is an acknowledged assumption, not a circular step. One internal inconsistency must be flagged for correctness: Eq. (18) in Section V writes d_i = R_{i,i} - P_m, whereas problem (7) and SDR (14) require R_{i,i} <= P_m/M; if Algorithm 1 was run as written, the FA solves a more permissive per-antenna power budget by a factor of M, which can explain its lower MSE at M=10 and must be corrected for the claim that Algorithm 1 solves (7) to hold. This is a reproducibility/correctness defect, not a circularity, because it does not make any output equal to an input by construction. Overall circularity score 2 due only to minor, non-load-bearing self-citation.

Assumptions & free parameters 5 free parameters · 5 assumptions · 0 invented entities

The central claim rests on standard convex optimization, the SDR rank-one relaxation, and the firefly heuristic. No new physical entities are introduced. The main free parameters are firefly hyperparameters chosen by hand. The strongest domain assumptions are perfect CSI including PU channels, plus the spatial degree-of-freedom condition U+K+L<=M. These are reasonable for a first design study but not for deployment without robustness analysis.

free parameters (5)
  • FA attractiveness variation gamma = 1
    Set by hand in Section VI-A; controls firefly visibility and strongly affects convergence and solution quality; no sensitivity analysis is given.
  • FA randomization factor alpha0 = 0.9 with decay 0.93^k
    Chosen in Section VI-A to control exploration; affects the stochastic search and the reported MSE values.
  • FA attractiveness at zero distance beta0 = 1
    Set in Section VI-A to maintain the scale of firefly moves; a standard but hand-chosen value.
  • FA population size and generation count = N = 40 or 60, Theta = 140
    Section VI-A; results in Figs. 5 to 9 depend on these values, and the authors note that N and Theta must grow with the number of antennas.
  • FA penalty constants = lambda_i = rho_i = zeta_l = k^2
    Dynamic penalty weights in Section VI-A; they control how strictly the SINR, power, and interference constraints are enforced and are not fitted to data.
assumptions (5)
  • domain assumption The BS has perfect instantaneous or statistical CSI for all secondary and primary user channels.
    Invoked in Section II-B and used in the covariance model (28); without it, the PU null guarantee is not established.
  • domain assumption U + K + L <= M, providing enough spatial degrees of freedom.
    Stated in Section II before Eq. (1); if violated, problem (7) may be infeasible and the proposed designs may not exist.
  • domain assumption Communication symbols and radar waveforms are uncorrelated, E[x_i^c x_t^r] = 0.
    Used to write R = W W^H + V V^H in Eq. (3); in practice radar and communication signals may be correlated.
  • ad hoc to paper The firefly algorithm converges to the global optimum as iterations grow and population size tends to infinity.
    Invoked in Section V without a formal proof; it is a heuristic property cited from the authors' own prior work [20]-[22], not a certified guarantee.
  • standard math Standard SDR rank-one relaxation and convex optimization results.
    Used in Section IV to transform (7) into (14), following [15]-[19].

how reviews work

0 comments
Cite this review

Pith. "Pith review of DFRC Systems Co-existing in Licensed Spectrum: Cognitive Beamforming Designs." pith.science (2026). https://pith.science/paper/KV3SDILI

@misc{pith2026250601202,
  author       = {Pith},
  title        = {Pith review of: DFRC Systems Co-existing in Licensed Spectrum: Cognitive Beamforming Designs},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KV3SDILI}},
  note         = {Machine review of arXiv:2506.01202}
}
read the original abstract

This paper introduces a dual-function radar-communication (DFRC) system with cognitive radio capability to tackle the spectral scarcity problem in wireless communications. Particularly, a cognitive DFRC system operates on a spectrum owned by a primary system to simultaneously perform data communication and target tracking with the condition that its interference to the primary users (PUs) is below a certain threshold. To achieve this, an optimization problem is formulated to jointly design the beamforming vectors for both the radar and communication functions in such a way that the mean square error (MSE) of the beam pattern between the designed and desired waveforms is minimized. The optimization problem has the following three constraints: i) the signal-to-interference-plus-noise ratio (SINR) at each data communication user is above a predetermined level; ii) the per-antenna transmit power is maintained at a given level; iii) the interference imposed on each PU is below a certain threshold. Both the semidefinite relaxation and nature-inspired firefly algorithms are proposed in order to search for the optimal solutions to the optimization problem. The simulation results indicate that our proposed algorithms can enable the DFRC system to protect the PUs while simultaneously performing its communication and radar functions.

Figures

Figures reproduced from arXiv: 2506.01202 by the authors.

Figure 1
Figure 1. A cognitive dual-function radar-communications (D [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Radiation beam patterns of the BS with M = 10 antennas for different approaches. The interference tolerance level at the PU is Ithres = 0.01. The required SINR at each SU is ηi = 10 dB. The FA population size is N = 40. The number of FA generations is Θ = 140. Pm = 1. towards 3 targets without caring for the communication users, i.e., the primary and secondary users. The cognitive-comm￾only approach serves 2 seconda… view at source ↗
Figure 6
Figure 6. Beam pattern MSE of the proposed FA approach versus th [PITH_FULL_IMAGE:figures/full_fig_p009_6.png] view at source ↗
Figures from the paper (4 more)
Figure 5
Figure 5. Figure 5: Beam pattern MSE of the proposed FA approach versus th [PITH_FULL_IMAGE:figures/full_fig_p009_5.png]
Figure 7
Figure 7. Figure 7: Beam pattern MSE versus the target SINR level at secon [PITH_FULL_IMAGE:figures/full_fig_p010_7.png]
Figure 8
Figure 8. Figure 8: Beam pattern MSE versus the interference tolerance l [PITH_FULL_IMAGE:figures/full_fig_p010_8.png]
Figure 10
Figure 10. Figure 10: Beam pattern MSE versus transmit power for di [PITH_FULL_IMAGE:figures/full_fig_p011_10.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

49 extracted references · 42 canonical work pages

  1. [5]

    Joint transmit beamforming for multiuser MIMO communications an d MIMO radar,

    X. Liu, T. Huang, N. Shlezinger, Y . Liu, J. Zhou, and Y . C. E ldar, “Joint transmit beamforming for multiuser MIMO communications an d MIMO radar,” IEEE Trans. Signal Process. , vol. 68, pp. 3929–3944, Jul. 2020

    work page 2020

  2. [1]

    Acc ess strate- gies for spectrum sharing in fading environment: Overlay, u nderlay and mixed,

    M. G. Khoshkholgh, K. Navaie, and H. Y anikomeroglu, “Acc ess strate- gies for spectrum sharing in fading environment: Overlay, u nderlay and mixed,” IEEE Trans. Mobile Comput. , vol. 9, no. 12, pp. 1780–1793, Dec. 2010

    work page 2010

  3. [2]

    Downlink beamforming in underlay cognitive cellular networks,

    T. A. Le and K. Navaie, “Downlink beamforming in underlay cognitive cellular networks,” IEEE Trans. Commun., vol. 62, no. 7, pp. 2212–2223, Jul. 2014

    work page 2014

  4. [3]

    ISAC - accelerated edge intelligence: Framework, optimization, and analysis,

    T. Zhang, G. Li, S. Wang, G. Zhu, G. Chen, and R. Wang, “ISAC - accelerated edge intelligence: Framework, optimization, and analysis,” IEEE Trans. Green Commun. and Netw. , vol. 7, no. 1, pp. 455–468, Mar. 2023

    work page 2023

  5. [4]

    To- ward dual-functional radar-communication systems: Optim al waveform design,

    F. Liu, L. Zhou, C. Masouros, A. Li, W. Luo, and A. Petropul u, “To- ward dual-functional radar-communication systems: Optim al waveform design,” IEEE Trans. Signal Process. , vol. 66, no. 16, pp. 4264–4279, Aug. 2018

    2018

  6. [6]

    Integrated sensing and communications: Toward dual-func tional wire- less networks for 6G and Beyond,

    F. Liu, Y . Cui, C. Masouros, J. Xu, T. X. Han, Y . C. Eldar, an d S. Buzzi, “Integrated sensing and communications: Toward dual-func tional wire- less networks for 6G and Beyond,” IEEE J. Sel. Areas in Commun. , vol. 40, no. 6, pp. 1728–1767, Jun. 2022

    work page 2022

  7. [7]

    A method for calculating error probabilit ies in a radar communication system,

    R. M. Mealey, “A method for calculating error probabilit ies in a radar communication system,” IEEE Trans. Space Electron. and Telemetry , vol. 9, no. 2, pp. 37–42, Jun. 1963

    work page 1963

  8. [8]

    Phase - modulation based dual-function radar-communications,

    A. Hassanien, M. G. Amin, Y . D. Zhang, and F. Ahmad, “Phase - modulation based dual-function radar-communications,” IEEE Radar , Sonar & Navigation, vol. 10, no. 8, p. 1411–1421, 2016

    work page 2016

Show all 49 references
  1. [9]

    Dual-function radar-communications: Informatio n embedding using sidelobe control and waveform diversity,

    ——, “Dual-function radar-communications: Informatio n embedding using sidelobe control and waveform diversity,” IEEE Trans. Signal Process., vol. 64, no. 8, pp. 2168–2181, Apr. 2016

    2016

  2. [10]

    Dual-function MI MO radar communications system design via sparse array optimizatio n,

    X. Wang, A. Hassanien, and M. G. Amin, “Dual-function MI MO radar communications system design via sparse array optimizatio n,” IEEE Trans. on Aerosp. Electron. Syst. , vol. 55, no. 3, pp. 1213–1226, Jun. 2019

    2019

  3. [11]

    Waveform design and signal pr ocessing aspects for fusion of wireless communications and radar sen sing,

    C. Sturm and W. Wiesbeck, “Waveform design and signal pr ocessing aspects for fusion of wireless communications and radar sen sing,” Proc. IEEE, vol. 99, no. 7, pp. 1236–1259, 2011

    2011

  4. [12]

    Robust transmit bea mforming for secure integrated sensing and communication,

    Z. Ren, L. Qiu, J. Xu, and D. W. K. Ng, “Robust transmit bea mforming for secure integrated sensing and communication,” IEEE Trans. Com- mun., vol. 71, no. 9, pp. 5549–5564, Sep. 2023

    2023

  5. [13]

    Optimal transmit beamformi ng for integrated sensing and communication,

    H. Hua, J. Xu, and T. X. Han, “Optimal transmit beamformi ng for integrated sensing and communication,” IEEE Trans. V eh. Technol. , vol. 72, no. 8, pp. 10 588–10 603, 2023

    2023

  6. [14]

    Transmit design for joint M IMO radar and multiuser communications with transmit covariance constr aint,

    X. Liu, T. Huang, and Y . Liu, “Transmit design for joint M IMO radar and multiuser communications with transmit covariance constr aint,” IEEE J. Sel. Areas Commun. , vol. 40, no. 6, pp. 1932–1950, Jun. 2022

    1932

  7. [15]

    Sem idefinite relaxation of quadratic optimization problems,

    Z.-Q. Luo, W.-K. Ma, A. M.-C. So, Y . Y e, and S. Zhang, “Sem idefinite relaxation of quadratic optimization problems,” IEEE Signal Process. Mag., vol. 27, no. 3, pp. 20–34, May 2010

    2010

  8. [16]

    Rank-constrained separabl e semidefinite programming with applications to optimal beamforming,

    Y . Huang and D. P . Palomar, “Rank-constrained separabl e semidefinite programming with applications to optimal beamforming,” IEEE Trans. Signal Process. , vol. 58, no. 2, pp. 664– 678, Feb. 2010

    2010

  9. [17]

    A rank-one optimiza tion framework and its applications to transmit beamforming,

    T. A. Le, D. W. K. Ng, and X.-S. Y ang, “A rank-one optimiza tion framework and its applications to transmit beamforming,” IEEE Trans. V eh. Technol., vol. 73, no. 1, pp. 620–636, Jan. 2024

    2024

  10. [18]

    Boyd and L

    S. Boyd and L. V andenberghe, Convex Optimization . Cambridge University Press, 2004

    2004

  11. [19]

    Multicell interference mitigati on with joint beamforming and common message decoding,

    H. Dahrouj and W. Y u, “Multicell interference mitigati on with joint beamforming and common message decoding,” IEEE Trans. Commun. , vol. 59, no. 8, pp. 2264–2273, Aug. 2011

    2011

  12. [20]

    Y ang, Nature-Inspired Metaheuristic Algorithms

    X.-S. Y ang, Nature-Inspired Metaheuristic Algorithms . Luniver Press, 2008

    2008

  13. [21]

    Wiley, 2009

    ——, Engineering optimisation: an introduction with metaheuri stic applications. Wiley, 2009

    2009

  14. [22]

    Generalized firefly algorithm fo r optimal transmit beamforming,

    T. A. Le and X.-S. Y ang, “Generalized firefly algorithm fo r optimal transmit beamforming,” IEEE Trans. Wireless Commun. , vol. 23, no. 6, pp. 5863–5877, Jun. 2024

    2024

  15. [24]

    Firefly algorithm for movable antenna arrays,

    M. K. Hoang, T. A. Le, K.-X. Thuc, T. V . Luyen, X.-S. Y ang, and D. W. K. Ng, “Firefly algorithm for movable antenna arrays,” IEEE Wireless Commun. Letters , vol. 13, no. 11, pp. 3157–3161, 2024

    2024

  16. [25]

    Opportunistic sharing between rotating radar and cellular,

    R. Saruthirathanaworakun, J. M. Peha, and L. M. Correia , “Opportunistic sharing between rotating radar and cellular,” IEEE J. Sel. Areas Com- mun., vol. 30, no. 10, pp. 1900–1910, Nov. 2012

    1900

  17. [26]

    Cognitive beamforming design for dual-function radar-communicatio ns,

    T. A. Le, I. Ku, X.-S. Y ang, C. Masouros, and T. Le-Ngoc, “ Cognitive beamforming design for dual-function radar-communicatio ns,” in 2024 IEEE 99th V eh. Technol. Conf. (VTC2024-Spring) , 2024, pp. 1–5

    2024

  18. [27]

    Power-e fficient time- domain scheduling for ISAC beamforming,

    E. Memisoglu, M. B. Janjua, and H. Arslan, “Power-e fficient time- domain scheduling for ISAC beamforming,” IEEE Wireless Com- mun.etters, vol. 13, no. 10, pp. 2837–2841, Oct. 2024

    2024

  19. [28]

    Joint resource allocation for user -centric cell- free integrated sensing and communication systems,

    Y . Cao and Q.-Y . Y u, “Joint resource allocation for user -centric cell- free integrated sensing and communication systems,” IEEE Commun. Letters, vol. 27, no. 9, pp. 2338–2342, Sep. 2023

    2023

  20. [29]

    Optimal user and tar get schedul- ing, user-target pairing, and low-resolution phase-only b eamforming for ISAC systems,

    L. F. Abanto-Leon and S. Maghsudi, “Optimal user and tar get schedul- ing, user-target pairing, and low-resolution phase-only b eamforming for ISAC systems,” IEEE Trans. V eh. Technol., pp. 1–6, 2025, early Access

    2025

  21. [30]

    On probing signal design fo r MIMO radar,

    P . Stoica, J. Li, and Y . Xie, “On probing signal design fo r MIMO radar,” IEEE Trans. Signal Process. , vol. 55, no. 8, pp. 4151–4161, Aug. 2007

    2007

  22. [31]

    Deep learning-based joint transmit beamforming for dual- functional radar-communication system,

    R. Y ang, Z. Zhu, J. Zhang, S. Xu, C. Li, Y . Huang, and L. Y an g, “Deep learning-based joint transmit beamforming for dual- functional radar-communication system,” IEEE Trans. Wireless Commun. , vol. 23, no. 10, pp. 15 198–15 211, 2024

    2024

  23. [32]

    A s uccessive optimization approach to pilot design for multi-cell massi ve MIMO systems,

    H. Al-Salihi, T. V an Chien, T. A. Le, and M. R. Nakhai, “A s uccessive optimization approach to pilot design for multi-cell massi ve MIMO systems,” IEEE Commun. Letters , vol. 22, no. 5, pp. 1086–1089, May 2018

    2018

  24. [33]

    Pare to-optimal pilot design for cellular massive MIMO systems,

    T. A. Le, T. V an Chien, M. R. Nakhai, and T. Le-Ngoc, “Pare to-optimal pilot design for cellular massive MIMO systems,” IEEE Trans. V eh. Technol., vol. 69, no. 11, pp. 13 206–13 215, Nov. 2020

    2020

  25. [34]

    Adaptive jo int sparse Bayesian approaches for near-field channel estimation,

    Z. Zhu, R. Y ang, C. Li, Y . Huang, and L. Y ang, “Adaptive jo int sparse Bayesian approaches for near-field channel estimation,” IEEE Trans. Wireless Commun., vol. 24, no. 3, pp. 2590–2605, Mar. 2025

    2025

  26. [35]

    Knowl edge- driven channel estimation for asymmetrical massive MIMO sy stems,

    R. Y ang, S. Xu, Z. Zhu, C. Li, Y . Huang, and L. Y ang, “Knowl edge- driven channel estimation for asymmetrical massive MIMO sy stems,” IEEE Trans. V eh. Technol., vol. 74, no. 1, pp. 911–924, Jan. 2025

    2025

  27. [36]

    HDnGAN: a cha nnel estimation method for time-varying mmWave massive MIMO,

    J. Zhang, S. Xu, R. Y ang, C. Li, and L. Y ang, “HDnGAN: a cha nnel estimation method for time-varying mmWave massive MIMO,” IEEE Trans. Wireless Commun., vol. 23, no. 11, pp. 17 189–17 204, Nov. 2024

    2024

  28. [37]

    Channel est imation for FDD massive MIMO with complex residual denoising network,

    Q. Zhao, X. Zeng, Z. Fan, Q. Zhang, and W. Li, “Channel est imation for FDD massive MIMO with complex residual denoising network,” IEEE Wireless Commun. Letters , vol. 13, no. 8, pp. 2070–2074, Aug. 2024

    2024

  29. [38]

    Structured c ompressive sensing-based spatio-temporal joint channel estimation f or FDD massive MIMO,

    Z. Gao, L. Dai, W. Dai, B. Shim, and Z. Wang, “Structured c ompressive sensing-based spatio-temporal joint channel estimation f or FDD massive MIMO,” IEEE Trans. Commun. , vol. 64, no. 2, pp. 601–617, Feb. 2016

    2016

  30. [39]

    Channel Training for An alog FDD Repeaters: Optimal Estimators and Cram´ er–Rao Bounds,

    S. Wesemann and T. L. Marzetta, “Channel Training for An alog FDD Repeaters: Optimal Estimators and Cram´ er–Rao Bounds,” IEEE Trans. Signal Process. , vol. 65, no. 23, pp. 6158–6170, Dec. 2017

    2017

  31. [40]

    Deep learning-based downlin k channel estimation for FDD massive MIMO systems,

    B. Xiang, D. Hu, and J. Wu, “Deep learning-based downlin k channel estimation for FDD massive MIMO systems,” IEEE Wireless Commun. Letters, vol. 12, no. 4, pp. 699–702, Apr. 2023

    2023

  32. [41]

    CVX: Matlab software for disciplin ed convex programming, academic users,

    CVX Research Inc., “CVX: Matlab software for disciplin ed convex programming, academic users,” http: //cvxr.com/cvx, 2015

    2015

  33. [42]

    Downlink optimization with in terference pricing and statistical CSI,

    T. A. Le and M. R. Nakhai, “Downlink optimization with in terference pricing and statistical CSI,” IEEE Trans. Commun. , vol. 61, no. 6, pp. 2339–2349, Jun 2013

    2013

  34. [43]

    Firefly algorithms for multimodal optimiz ation,

    X.-S. Y ang, “Firefly algorithms for multimodal optimiz ation,” in Stochastic Algorithms: F oundations and Applications , O. Watanabe and T. Zeugmann, Eds. Berlin, Heidelberg: Springer Berlin Heid elberg, 2009, pp. 169–178

    2009

  35. [44]

    An tenna selection with beam squint compensation for integrated sen sing and communications,

    A. M. Elbir, A. Abdallah, A. Celik, and A. M. Eltawil, “An tenna selection with beam squint compensation for integrated sen sing and communications,” IEEE Journal of Selected Topics in Signal Processing , vol. 18, no. 5, pp. 857–870, Jul. 2024

    2024

  36. [45]

    Dual- layer metamaterial rectangular antenna arrays for in-band full- duplex massive mimo,

    Y . Gong, M. Mahmood, R. Morawski, and T. Le-Ngoc, “Dual- layer metamaterial rectangular antenna arrays for in-band full- duplex massive mimo,” IEEE Access , vol. 11, pp. 135 708–135 727, 2023

    2023

  37. [46]

    Joint antenna selecti on and beamforming for area surveillance with spatially distribu ted array radar,

    C. Y ang, W. Yi, and B. Champagne, “Joint antenna selecti on and beamforming for area surveillance with spatially distribu ted array radar,” IEEE Trans. Aerospace and Electronic Systems , pp. 1–15, 2024

    2024

  38. [47]

    Optimal downlink beamf orming using Semidefinite optimization,

    M. Bengtsson and B. Ottersten, “Optimal downlink beamf orming using Semidefinite optimization,” in Proc. 37th Annu. Allerton Conf. Com- mun., Control, and Comput. , 1999, pp. 987 – 996

    1999

  39. [48]

    An iterative algorithm for dow nlink multi- cell beam-forming,

    T. A. Le and M. R. Nakhai, “An iterative algorithm for dow nlink multi- cell beam-forming,” in Proc. IEEE Global Telecommun. Conf. , Dec. 2011, pp. 1–6

    2011

  40. [49]

    Rank-constrained separabl e Semidefinite programming with applications to optimal beamforming,

    Y . Huang and D. P . Palomar, “Rank-constrained separabl e Semidefinite programming with applications to optimal beamforming,” IEEE Trans. Signal Process. , vol. 58, no. 2, pp. 644–678, Feb. 2010. Tuan Anh Le (Senior , IEEE) received the Ph.D. degree in telecommunications resea...

    2010

  41. [2012]

    His Ph.D. research contributed to the Green Radio project funded by the Core 5 joint research program of the U.K.’s Engineering and Physical Sciences Research Council (EPSRC) and the Vir- tual Center of Excellence in Mobile and Personal Communications (Mobile VCE). He was a Po...

    2019

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.