Pith. sign in

REVIEW 3 major objections 3 minor 41 references

Towards Intelligent Edge Sensing for ISCC Network: Joint Multi-Tier DNN Partitioning and Beamforming Design

T0 review · 3 major / 3 minor · reviewed 2026-08-16 · deepseek-v4-flash

Pith's one-line read Splitting a sensing DNN across device, edge, and cloud and jointly optimizing beamforming and computing resources minimizes inference latency and outperforms two-tier offloading.

desk verdict A useful three-tier ISCC/DNN-partitioning setup, undercut by a reversed MM inequality that leaves the convergence and near-optimality claims unsupported. read the letter →

arxiv 2504.21409 v1 pith:TWTKATPD submitted 2025-04-30 eess.SP

classification eess.SP
keywords integratedsensingandcommunicationmobileedgecomputingDNNpartitioningbeamformingdesigncross-entropyoptimizationinferencelatencythree-tiercollaborativeMIMO
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 tries to show that a three-tier cloud–edge–device network can run a pre-trained deep neural network for radar sensing faster than any two-tier offloading scheme by splitting the DNN at two carefully chosen layers and transmitting only intermediate features over the air. It argues that jointly choosing the two partition points, the ISAC beamforming matrices, and the CPU cycles allocated at each device and the MEC server can minimize total sensing-task inference latency under a sensing beampattern constraint. The proposed two-layer algorithm—closed-form resource allocation in the inner layer and a cross-entropy search over partitioning strategies in the outer layer—is reported to match the branch-and-bound optimum while keeping runtime nearly flat as the number of devices grows, and to outperform two-tier baselines across MEC capacity, device capacity, bandwidth, and backhaul settings. If correct, this gives a practical way to push DNN-based sensing tasks into edge networks without sacrificing sensing quality.

What carries the argument

The load-bearing object is the two-point DNN partition encoded by binary variables $\beta_{kl}$: each device's DNN has one or two layers at which execution hands off to the next tier, and the FLOP counts $S_k^L$, $S_k^M$, $S_k^C$ and feature sizes $o_k(l_1)$, $o_k(l_2)$ are derived from that partition. The inner-layer beamforming solution is carried by the MM lower-bound surrogate in (34), by the WMMSE equivalence that turns weighted sum-rate maximization into weighted sum-MSE minimization, and by the orthogonal Procrustes transformation (Lemma 1) that gives a closed-form SVD-based update $\hat{W}_k^{c*}=A(1:d)B^H$ under the covariance constraint $W_k W_k^H=\hat{R}_k^{\mathrm{pre}}$. The outer layer is carried by a Bernoulli-parameterized cross-entropy search that samples partitioning strategies, keeps elite samples, and updates the probability vector $\omega$ via $\omega^{(it+1)}=\rho \upsilon^{(it)}+(1-\rho)\omega^{(it)}$.

What would settle it

Take $o_1=o_2=1$, $R_1^{it}=R_2^{it}=1$, $R_1=2$, and $R_2=1$; the claimed bound in (34) would require $2/3 \ge 3/4$, which is false, so computing either side of (34) for such a configuration settles whether the MM surrogate is a true lower bound and whether the convergence argument can hold.

Watch

Extended reading notes

Core claim

The paper's central claim is that the inference latency of a sensing task executed by a pre-trained DNN can be minimized by allowing the DNN to be split at up to two layers across three computational tiers: the ISAC device runs early layers, the MEC server runs middle layers, and the cloud runs the remainder. Because only the output of the first partition layer ($o_k(l_1)$) is transmitted over the ISAC link and only the output of the second ($o_k(l_2)$) crosses the backhaul, offloading intermediate features rather than raw spectrograms cuts fronthaul load. The authors formulate this as a mixed-integer nonlinear program over binary partition decisions, beamforming matrices, and CPU frequencies, and solve it by decoupling into an inner continuous problem—KKT closed forms for resource allocation and an MM–WMMSE iteration with an orthogonal Procrustes step for beamforming—and an outer discrete search driven by cross-entropy. They report that the resulting scheme achieves lower average inference latency than two-tier partitioning and than non-partitioned three-tier offloading, tracks the branch-and-bound optimal DNN partitioning, and exposes a trade-off between sensing beampattern mainlobe width and latency.

Load-bearing premise

The entire beamforming optimization rests on the inequality in Equation (34) being a valid lower bound on $1/\sum_k o_k/R_k$, so if that inequality fails the inner-loop convergence and near-optimality results do not follow.

Editorial extensions

If this is right

  • If the central claim holds, ISAC devices can offload intermediate DNN features instead of raw sensing data, reducing fronthaul load while still completing the sensing task.
  • Jointly optimizing partition points, beamforming, and compute allocation yields lower inference latency than either two-tier DNN partitioning or non-partitioned three-tier offloading across the simulated ranges of MEC capacity, device capacity, bandwidth, and backhaul rate.
  • The cross-entropy outer loop reaches the same DNN partitioning as branch-and-bound in the tested cases while its runtime grows far more slowly with the number of devices, making near-optimal partitioning feasible for larger networks.
  • There is an explicit trade-off: widening the sensing beampattern mainlobe lowers latency by improving offloading rates, while narrowing it improves sensing gain at the cost of higher latency.
  • The closed-form resource allocation formulas and the SVD-based beamforming update give a low-complexity inner loop that can be embedded in each evaluation of a partitioning candidate.

Reading between the lines

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

  • Editorial extension: The same two-layer decomposition—closed-form continuous allocation under a discrete handoff-point search—applies to other split-inference settings, such as multiple DNN stages or hierarchical federated learning, wherever the cost separates additively across tiers.
  • Editorial extension: Because the model already lets each device have its own number of layers, layer FLOPs, and feature sizes, a mixed-device network with different DNN architectures is directly within the formulation, though the simulations use a shared AlexNet.
  • Editorial extension: Relaxing the equality sensing-covariance constraint to a bounded mismatch would enlarge the beamforming feasible region; the expected effect is lower latency for the same sensing quality, at the cost of a more complex constraint set.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 3 minor

Summary. The paper studies a three-tier cloud-edge-device collaborative inference framework for integrated sensing, communication, and computing (ISCC) networks. Each ISAC device runs part of a pre-trained DNN locally and offloads intermediate features to a MEC server and cloud server, while the same transmitted signal is used for sensing. The authors formulate a mixed-integer nonlinear program that jointly optimizes DNN partition points, ISAC beamforming matrices, and computational resource allocation, subject to a sensing beampattern covariance constraint. The proposed solution is a two-layer algorithm: an inner layer with closed-form CPU allocation and an MM-WMMSE-OPP-based beamforming update, and an outer layer using a cross-entropy (CE) method to search over partition choices. Simulations compare the proposed scheme with two-tier baselines and with a branch-and-bound (BnB) benchmark.

Significance. The paper addresses a timely and relevant problem in the ISCC literature, and the high-level problem decomposition into an inner continuous optimization and an outer discrete search is sensible. The closed-form resource allocation solutions in Eqs. (28) and (31) are correct, and the CE-based outer loop is a standard and reproducible approach. However, the central beamforming algorithm rests on the MM surrogate in Eq. (34), which is not a valid minorizer of the reciprocal-latency objective. Since the convergence of Algorithm 1 and the near-optimality comparison in Fig. 9 both depend on this inequality, the main contribution of the paper is not supported by the provided derivation. The paper cannot be accepted while this load-bearing error remains.

major comments (3)
  1. [III-B-3, Eq. (34)] The claimed lower-bound surrogate in Eq. (34) is mathematically false. Setting x_k = o_k(l_1)/R_k and x_k^{it} = o_k(l_1)/R_k^{it}, the inequality reads 1/(sum x_k) >= (sum (x_k^{it})^2 / x_k) / (sum x_k^{it})^2. Cauchy-Schwarz gives (sum x_k^{it})^2 <= (sum x_k)(sum (x_k^{it})^2 / x_k), which implies the opposite inequality, i.e., the right-hand side is generally an upper bound, not a lower bound, with equality only at x = x^{it}. A concrete counterexample is K=2, o_1=o_2=1, R_1^{it}=R_2^{it}=1, R_1=1, R_2=1.5, for which the left-hand side is 0.6 and the right-hand side is 0.625. Therefore the MM update in Algorithm 1 is not a valid minorization step, and the monotone-convergence argument for the beamforming update is unsupported.
  2. [III-B-4 and IV-D, Fig. 9] Because the surrogate in Eq. (34) is not a lower bound, the convergence of Algorithm 1 is not established, and the claimed near-optimality of the beamforming update is unproven. The comparison against the BnB baseline in Fig. 9 evaluates both algorithms with the same inner-layer solver, so the observed match only shows that the CE search and BnB agree on the same (possibly suboptimal) inner-layer solutions; it does not validate the optimality of the beamforming update itself. The paper's central claim that the proposed algorithm substantially reduces inference latency therefore rests on an invalid algorithmic step.
  3. [III-B-3, Eqs. (36) and (38)] In the MSE expression (36) and the MMSE receiver (38), the interference covariance is written as sum_i U_k H_i \hat{R}_k^{pre} H_i^H U_k^H and sum_i H_i \hat{R}_k^{pre} H_i^H, respectively, but the covariance of the i-th user is \hat{R}_i^{pre}, not \hat{R}_k^{pre}. As written, the formulas are dimensionally inconsistent (\hat{R}_k^{pre} is used for every i) and the receiver is not the correct MMSE receiver. If this is a typesetting error, it must be corrected, because the subsequent WMMSE-OPP derivation relies on these expressions.
minor comments (3)
  1. [III-C, Algorithm 2] The algorithm states that feasible samples satisfying constraints (21a)-(21b) are generated from the Bernoulli distribution in (50), but no rejection or conditioning mechanism is described. Since independent Bernoulli draws can yield sum_{l} beta_{kl} = 0 or >2 for some device, the sampling procedure is not fully specified.
  2. [III-D, complexity analysis] The complexity expression O(it_I it_O V (N_t^3 + M^2 N_t + N_t^2 M)) does not explicitly account for the inner BCD loop in Algorithm 1, which itself requires multiple iterations to converge. The stated complexity should include the number of BCD iterations or clarify that it is absorbed into it_I.
  3. [IV-C, Fig. 5] The sentence 'the proposed scheme ... acting as lower bound for ED-DP and CED-WDP' is ungrammatical and should be rewritten to state that the proposed scheme achieves lower latency than both baselines for the displayed range of F_M.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the derivation chain is self-contained; the disputed MM inequality is a correctness concern, not a circularity.

full rationale

The paper's central claim is that jointly optimizing DNN partitioning, beamforming, and computational resource allocation minimizes the total sensing inference latency. The latency model in (18) is assembled from independent computation and transmission models; the resource-allocation closed forms (28) and (31) are derived from KKT conditions on those models; the beamforming updates (38), (40), (46), and (47) follow from standard WMMSE and OPP transformations of the rate expression (3); and the CE outer loop (49)-(54) evaluates each sampled partition by actually solving the inner problem and updates a Bernoulli distribution from the resulting objective values. No parameter is fitted to the target latency, and no load-bearing premise is imported from the authors' prior work: references [17] and [18] are used as benchmarks and background for the three-tier architecture, not as the justification for the optimization results. The desired beampattern is imposed externally through (21f), and the BnB comparison in Fig. 9 is an independent optimality benchmark rather than a restatement of the method's assumptions. The skeptic's concern about Eq. (34) is a mathematical-correctness issue: the claimed Jensen lower bound may fail as a minorizer, which would undermine the convergence proof of Algorithm 1, but that is a proof error, not circular reasoning, because the surrogate is not equivalent to the objective by construction and the claim is not saved by any fitted parameter or self-citation. Honest finding: no circularity.

Assumptions & free parameters 3 free parameters · 6 assumptions · 0 invented entities

The paper introduces no new physical entities. The only free parameters are the cross-entropy hyperparameters, which are hand-set rather than fitted to data. The model relies on standard ISAC and MEC assumptions (perfect CSI, uncorrelated radar symbols, known desired beampattern, FLOPs model from prior work). The central flaw is a mathematical error in the MM surrogate, not an invented parameter or entity.

free parameters (3)
  • Number of CE samples V = 1000
    Hand-set in Section IV-A; controls stochastic search size but is not fitted to any data.
  • Elite sample count V_elite = 50
    Hand-set; selects top performers in the cross-entropy update.
  • CE learning rate rho = 0.9
    Smoothing factor in Eq. (54); hand-set, not fitted.
assumptions (6)
  • domain assumption Perfect CSI is available at all nodes.
    Section II-A states channels are block-faded quasi-static and perfect CSI can be obtained [2]; this underpins the rate expression R_k.
  • domain assumption Radar waveforms are uncorrelated with communication symbols.
    Section II-A assumes E{s_r (s_c)^H}=0, which is required for the rate and interference derivations.
  • domain assumption The desired sensing beampattern covariance Rhat_pre is given.
    Footnote 2 and Eq. (8) treat Rhat_pre from [34] as an input; the equality constraint (21f) forces exact matching.
  • domain assumption DNN layer FLOPs and output sizes follow the model of [27].
    Section II-B Eq. (9) adopts the FLOPs formulas from Lee et al.; the correctness of the latency model depends on these formulas.
  • domain assumption The cloud server has a fixed sufficient computation f^C for each device.
    Section II-C5 assumes fixed f^C, which makes the cloud execution latency linear in the cloud workload.
  • standard math The problem reduces to a multi-choice knapsack problem when continuous variables are fixed.
    Section II-D uses this reduction to claim NP-hardness; this is a standard complexity argument.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Towards Intelligent Edge Sensing for ISCC Network: Joint Multi-Tier DNN Partitioning and Beamforming Design." pith.science (2026). https://pith.science/paper/TWTKATPD

@misc{pith2026250421409,
  author       = {Pith},
  title        = {Pith review of: Towards Intelligent Edge Sensing for ISCC Network: Joint Multi-Tier DNN Partitioning and Beamforming Design},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/TWTKATPD}},
  note         = {Machine review of arXiv:2504.21409}
}
read the original abstract

The combination of Integrated Sensing and Communication (ISAC) and Mobile Edge Computing (MEC) enables devices to simultaneously sense the environment and offload data to the base stations (BS) for intelligent processing, thereby reducing local computational burdens. However, transmitting raw sensing data from ISAC devices to the BS often incurs substantial fronthaul overhead and latency. This paper investigates a three-tier collaborative inference framework enabled by Integrated Sensing, Communication, and Computing (ISCC), where cloud servers, MEC servers, and ISAC devices cooperatively execute different segments of a pre-trained deep neural network (DNN) for intelligent sensing. By offloading intermediate DNN features, the proposed framework can significantly reduce fronthaul transmission load. Furthermore, multiple-input multiple-output (MIMO) technology is employed to enhance both sensing quality and offloading efficiency. To minimize the overall sensing task inference latency across all ISAC devices, we jointly optimize the DNN partitioning strategy, ISAC beamforming, and computational resource allocation at the MEC servers and devices, subject to sensing beampattern constraints. We also propose an efficient two-layer optimization algorithm. In the inner layer, we derive closed-form solutions for computational resource allocation using the Karush-Kuhn-Tucker conditions. Moreover, we design the ISAC beamforming vectors via an iterative method based on the majorization-minimization and weighted minimum mean square error techniques. In the outer layer, we develop a cross-entropy based probabilistic learning algorithm to determine an optimal DNN partitioning strategy. Simulation results demonstrate that the proposed framework substantially outperforms existing two-tier schemes in inference latency.

Figures

Figures reproduced from arXiv: 2504.21409 by the authors.

Figure 1
Figure 1. The ISAC-enabled cloud-edge-device three-tier net [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Three-Tier Collaborative Inference Framework. [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. The AlexNet Architecture. 3) Overall Algorithm for the Outer-Layer Optimization Problem: To make it clearer, the CE-based algorithm for solving the outer-layer optimization problem is summarized in Algorithm 2. Specifically, we first generate V feasible samples through random sampling based on the probability distribution parameters ω it. For each sample, the inner-layer optimization problem is solved using Algorith… view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: (a) The convergence performance of Algorithm 1. (b) [PITH_FULL_IMAGE:figures/full_fig_p011_4.png]
Figure 5
Figure 5. Figure 5: Average inference latency versus different MEC [PITH_FULL_IMAGE:figures/full_fig_p011_5.png]
Figure 7
Figure 7. Figure 7: Beampattern versus the mainlobe width ∆θ. 10 12 14 16 18 20 22 24 26 28 30 Beampattern Mainlobe Width (°) 0.5 0.55 0.6 0.65 0.7 0.75 0.8 0.85 0.9 0.95 Average Inference Latency (s) [PITH_FULL_IMAGE:figures/full_fig_p012_7.png]
Figure 8
Figure 8. Figure 8: Average inference latency versus beampattern mainl [PITH_FULL_IMAGE:figures/full_fig_p012_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

41 extracted references · 24 canonical work pages

  1. [29]

    Task- oriented sensing, computation, and communication integra tion for multi- device edge AI,

    D. Wen, P . Liu, G. Zhu, Y . Shi, J. Xu, Y . C. Eldar, and S. Cui , “Task- oriented sensing, computation, and communication integra tion for multi- device edge AI,” IEEE Trans. Wireless Commun. , vol. 23, no. 3, pp. 2486–2502, Mar. 2024

  2. [32]

    Integrated sensing, comm unication, and computation with adaptive DNN splitting in multi-UA V ne tworks,

    C. Deng, X. Fang, and X. Wang, “Integrated sensing, comm unication, and computation with adaptive DNN splitting in multi-UA V ne tworks,” IEEE Trans. Wireless Commun., vol. 23, no. 11, pp. 17 429–17 445, Nov. 2024

  3. [1]

    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 Commun. , vol. 40, no. 6, pp. 1728–1767, Jun. 2022

  4. [2]

    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

  5. [3]

    Joint waveform and di screte phase shift design for RIS-assisted integrated sensing and communi- cation system under cramer-rao bound constraint,

    X. Wang, Z. Fei, J. Huang, and H. Y u, “Joint waveform and di screte phase shift design for RIS-assisted integrated sensing and communi- cation system under cramer-rao bound constraint,” IEEE Trans. V eh. Technol., vol. 71, no. 1, pp. 1004–1009, Jan. 2022

  6. [4]

    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, Jun. 2020

  7. [5]

    Bidirectional integrated sen sing and communication: Full-duplex or half-duplex?

    Z. Wang, X. Mu, and Y . Liu, “Bidirectional integrated sen sing and communication: Full-duplex or half-duplex?” IEEE Trans. Wireless Commun., vol. 23, no. 8, pp. 8184–8199, Aug. 2024

  8. [6]

    Optimal beamformin g for MIMO DFRC systems with transmit covariance constraints,

    C. Y ang, X. Wang, W. Ni, and Y . Jiang, “Optimal beamformin g for MIMO DFRC systems with transmit covariance constraints,” IEEE Trans. Signal Process. , vol. 73, pp. 601–616, Jan. 2025

Show all 41 references
  1. [7]

    NOMA for integrating sensing a nd communications toward 6G: A multiple access perspective,

    X. Mu, Z. Wang, and Y . Liu, “NOMA for integrating sensing a nd communications toward 6G: A multiple access perspective,” IEEE Wireless Commun., vol. 31, no. 3, pp. 316–323, Jun. 2024

  2. [8]

    Next- generation multiple access for integrated sensing and comm unications,

    Y . Liu, T. Huang, F. Liu, D. Ma, W. Huangfu, and Y . C. Eldar, “Next- generation multiple access for integrated sensing and comm unications,” Proc. IEEE , vol. 112, no. 9, pp. 1467–1496, Sept. 2024

  3. [9]

    Dynamic power allocation for integrate d sensing and communication-enabled vehicular networks,

    H. Y ang, L. Wang, Z. Feng, Z. Wei, J. Peng, X. Y uan, T. Q. S. Quek, and P . Zhang, “Dynamic power allocation for integrate d sensing and communication-enabled vehicular networks,” IEEE Trans. Wireless Commun., vol. 23, no. 9, pp. 12 313–12 330, Sept. 2024

  4. [10]

    Resource allocation for integrated sensing and communication in digital twin en abled internet of vehicles,

    Y . Gong, Y . Wei, Z. Feng, F. R. Y u, and Y . Zhang, “Resource allocation for integrated sensing and communication in digital twin en abled internet of vehicles,” IEEE Trans. V eh. Technol., vol. 72, no. 4, pp. 4510–4524, Apr. 2023

  5. [11]

    User association and resource allocation for MEC-enabled IoT networks,

    Y . Sun, J. Xu, and S. Cui, “User association and resource allocation for MEC-enabled IoT networks,” IEEE Trans. Wireless Commun. , vol. 21, no. 10, pp. 8051–8062, Oct. 2022

  6. [12]

    Pushing AI to wireless network edge: An overview on integrated sensing, communication, and computation towar ds 6G,

    G. Zhu et al. , “Pushing AI to wireless network edge: An overview on integrated sensing, communication, and computation towar ds 6G,” Sci. China Inf. Sci , vol. 66, no. 3, pp. 130–301, Feb. 2023

  7. [13]

    Integr ated sensing- communication-computation for edge artificial intelligen ce,

    D. Wen, X. Li, Y . Zhou, Y . Shi, S. Wu, and C. Jiang, “Integr ated sensing- communication-computation for edge artificial intelligen ce,” IEEE Inter- net Things Mag. , vol. 7, no. 4, pp. 14–20, July 2024

  8. [14]

    Semantic-aware vision-assisted integrated sensing and c ommunication: Architecture and resource allocation,

    Y . Lu, W. Mao, H. Du, O. A. Dobre, D. Niyato, and Z. Ding, “Semantic-aware vision-assisted integrated sensing and c ommunication: Architecture and resource allocation,” IEEE Wirel. Commun. , vol. 31, no. 3, pp. 302–308, June 2024

  9. [15]

    A survey on integrated sensing, communication, a nd com- putation,

    D. Wen, Y . Zhou, X. Li, Y . Shi, K. Huang, and K. B. Letaief, “A survey on integrated sensing, communication, a nd com- putation,” IEEE Commun. Surv. Tutor . , early Access, 2024, doi: 10.1109/COMST.2024.3521498

  10. [16]

    Joint MIMO precoding and computation resource allocation for dual-function radar and communication syst ems with mobile edge computing,

    C. Ding et al. , “Joint MIMO precoding and computation resource allocation for dual-function radar and communication syst ems with mobile edge computing,” IEEE J. Sel. Areas Commun. , vol. 40, no. 7, pp. 2085–2102, Mar. 2022

  11. [17]

    Joint offloading and beamforming design in in- tegrating sensing, communication, and computing systems: A dis- tributed approach,

    P . Liu et al. , “Joint offloading and beamforming design in in- tegrating sensing, communication, and computing systems: A dis- tributed approach,” IEEE Trans. Commun. , early Access, 2024, doi: 10.1109/TCOMM.2024.3516488

  12. [18]

    Joint beamform- ing and offloading design for integrated sensing, communica tion, and computation system,

    P . Liu, Z. Fei, X. Wang, Y . Zhou, Y . Zhang, and F. Liu, “Joint beamform- ing and offloading design for integrated sensing, communica tion, and computation system,” IEEE Trans. V eh. Technol. , early Access, 2025, doi: 10.1109/TVT.2025.3561430

  13. [19]

    Mo- bile edge computing aided integrated sensing and communica tion with short-packet transmissions,

    N. Huang, C. Dou, Y . Wu, L. Qian, B. Lin, H. Zhou, and X. She n, “Mo- bile edge computing aided integrated sensing and communica tion with short-packet transmissions,” IEEE Trans. Wireless Commun. , vol. 23, no. 7, pp. 7759–7774, Jul. 2024

  14. [20]

    Over-the-air integ rated sens- ing, communication, and computation in IoT networks,

    X. Li, Y . Gong, K. Huang, and Z. Niu, “Over-the-air integ rated sens- ing, communication, and computation in IoT networks,” IEEE Wirel. Commun., vol. 30, no. 1, pp. 32–38, Feb. 2023

  15. [21]

    In- tegrated sensing, communication, and computation over-th e-air: MIMO beamforming design,

    X. Li, F. Liu, Z. Zhou, G. Zhu, S. Wang, K. Huang, and Y . Gon g, “In- tegrated sensing, communication, and computation over-th e-air: MIMO beamforming design,” IEEE Trans. Wireless Commun. , vol. 22, no. 8, pp. 5383–5398, Aug. 2023

  16. [22]

    Integrated sensing, c ommunication, and computation over the air: Beampattern design for wirele ss sensor networks,

    S. Wang, Y . Gong, X. Li, and Q. Li, “Integrated sensing, c ommunication, and computation over the air: Beampattern design for wirele ss sensor networks,” IEEE Internet Things J. , vol. 11, no. 6, pp. 9681–9692, Mar. 2024

  17. [23]

    NOMA-aided jo int communication, sensing, and multi-tier computing systems ,

    Z. Wang, X. Mu, Y . Liu, X. Xu, and P . Zhang, “NOMA-aided jo int communication, sensing, and multi-tier computing systems ,” IEEE J. Sel. Areas Commun. , vol. 41, no. 3, pp. 574–588, Mar. 2023

  18. [24]

    Integrated sensing and two-tier task offloading via non-or thogonal multiple access: An energy-minimization design,

    C. Dou, M. Dai, N. Huang, Y . Wu, L. Qian, and T. Q. S. Quek, “Integrated sensing and two-tier task offloading via non-or thogonal multiple access: An energy-minimization design,” IEEE Trans. Wireless Commun., vol. 23, no. 12, pp. 19 157–19 171, Dec. 2024

  19. [25]

    Neurosurgeon: Collaborative intelligence betwe en the cloud and mobile edge,

    Y . Kang, J. Hauswald, C. Gao, A. Rovinski, T. Mudge, J. Ma rs, and L. Tang, “Neurosurgeon: Collaborative intelligence betwe en the cloud and mobile edge,” SIGPLAN Not., vol. 52, no. 4, p. 615–629, Apr. 2017

  20. [26]

    Split com puting and early exiting for deep learning applications: Survey and researc h challenges,

    Y . Matsubara, M. Levorato, and F. Restuccia, “Split com puting and early exiting for deep learning applications: Survey and researc h challenges,” ACM Comput. Surv. , vol. 55, no. 5, DEC. 2022

  21. [27]

    Wireless channel adaptive D NN split inference for resource-constrained edge devices,

    J. Lee, H. Lee, and W. Choi, “Wireless channel adaptive D NN split inference for resource-constrained edge devices,” IEEE Commun. Lett. , vol. 27, no. 6, pp. 1520–1524, Jun. 2023

  22. [28]

    Optimal AI model splitting and resource allocation for device-edge co-inference in multi-user wireless sensing s ystems,

    X. Li and S. Bi, “Optimal AI model splitting and resource allocation for device-edge co-inference in multi-user wireless sensing s ystems,” IEEE Trans. Wireless Commun., vol. 23, no. 9, pp. 11 094–11 108, Sept. 2024

  23. [30]

    Task -oriented over-the-air computation for multi-device edge AI,

    D. Wen, X. Jiao, P . Liu, G. Zhu, Y . Shi, and K. Huang, “Task -oriented over-the-air computation for multi-device edge AI,” IEEE Trans. Wire- less Commun. , vol. 23, no. 3, pp. 2039–2053, Mar. 2024

  24. [31]

    In tegrated sensing-communication-computation for over-the-air edge AI inference,

    Z. Zhuang, D. Wen, Y . Shi, G. Zhu, S. Wu, and D. Niyato, “In tegrated sensing-communication-computation for over-the-air edge AI inference,” IEEE Trans. Wireless Commun. , vol. 23, no. 4, pp. 3205–3220, Apr. 2024

  25. [33]

    Wire less sens- ing with deep spectrogram network and primitive based autor egressive hybrid channel model,

    G. Li, S. Wang, J. Li, R. Wang, X. Peng, and T. X. Han, “Wire less sens- ing with deep spectrogram network and primitive based autor egressive hybrid channel model,” in 2021 IEEE SPAWC , 2021, pp. 481–485

  26. [34]

    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

  27. [35]

    Boyd and L

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

  28. [36]

    Majorization-minim ization algo- rithms in signal processing, communications, and machine l earning,

    Y . Sun, P . Babu, and D. P . Palomar, “Majorization-minim ization algo- rithms in signal processing, communications, and machine l earning,” IEEE Trans. Signal Process. , vol. 65, no. 3, pp. 794–816, Feb. 2017

  29. [37]

    An iterativ ely weighted MMSE approach to distributed sum-utility maximization for a MIMO interfering broadcast channel,

    Q. Shi, M. Razaviyayn, Z.-Q. Luo, and C. He, “An iterativ ely weighted MMSE approach to distributed sum-utility maximization for a MIMO interfering broadcast channel,” IEEE Trans. Signal Process. , vol. 59, no. 9, pp. 4331–4340, Sept. 2011

  30. [38]

    Optimization algorithms exploiting unita ry constraints,

    J. Manton, “Optimization algorithms exploiting unita ry constraints,” IEEE Trans. Signal Process. , vol. 50, no. 3, pp. 635–650, Mar. 2002

  31. [39]

    A novel cross entropy approach for offloading learning in mobile edge comp uting,

    S. Zhu, W. Xu, L. Fan, K. Wang, and G. K. Karagiannidis, “A novel cross entropy approach for offloading learning in mobile edge comp uting,” IEEE Wireless Commun. Lett. , vol. 9, no. 3, pp. 402–405, Mar. 2020

  32. [40]

    Classification of automotive ta rgets using inverse synthetic aperture radar images,

    N. Pandey and S. S. Ram, “Classification of automotive ta rgets using inverse synthetic aperture radar images,” IEEE Trans. Intell. V eh., vol. 7, no. 3, pp. 675–689, Sept.. 2022

  33. [41]

    Tran sfer learning with convolutional neural networks for moving target class ification with micro-doppler radar spectrograms,

    E. A. Hadhrami, M. A. Mufti, B. Taha, and N. Werghi, “Tran sfer learning with convolutional neural networks for moving target class ification with micro-doppler radar spectrograms,” in 2018 ICAIBD, 2018, pp. 148–154

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.