{"id":"9ff34d00-b937-494b-ad03-9a868d0e7d33","arxiv_id":"1908.08719","paper_version":1,"verdict":"REJECT","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"high","formal_verification":"none","parameter_count":0,"one_line_summary":"A joint SCA-based power and power-splitting allocation for hybrid TDMA-NOMA reportedly beats TDMA in transmit power, but the rate-constraint conversion misses the time-sharing factor.","lead":"This paper designs a transmit-power minimization for hybrid TDMA-NOMA with wireless energy harvesting, splitting received signals into information and energy parts. It claims via simulations that this hybrid scheme consumes less transmit power than conventional TDMA while meeting rate and harvested-energy targets.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Rate surrogate in Eqs. (14a)-(14c) and OP1 drops the time-sharing factor t_i, so the SCA may solve an infeasibly relaxed problem and the reported power comparison is not established.","rationale":"The reader's weakest assumption is exactly the load-bearing defect: the SCA surrogate drops t_i from the rate constraint, so OP1 can return powers whose true rates are below Rmin. This is not a cosmetic typo. The optimization objective is to lower transmit power, so the solver has every incentive to exploit the underestimated SINR target; the resulting P_t values are lower bounds for a relaxed problem, not feasible powers. The paper gives no independent verification that the output of OP1 satisfies the original constraints, and the convergence plot (Fig. 4) only tracks the objective, not feasibility. Without a corrected derivation or a feasibility check of the reported solutions, the headline claim that hybrid TDMA-NOMA outperforms TDMA under identical QoS constraints is unsupported. No change to the reader's REJECT verdict is needed; the attack reinforces it.","tokens_in":10424,"tokens_out":14191,"duration_ms":141041,"concrete_test":"Re-run the Section IV simulations with the corrected rate constraint: replace the OP1 rate link (14c)/r_{j,i}≥Rmin by θ_{j,i} ≥ 2^{Rmin/t_i} (equivalently ϑ_{j,i}≥Rmin/t_i), apply the analogous correction to OP2 with t_i^{TDMA}=T/K, and set the undefined γ in Eq. (21a) to 1 or to its intended defined value. Using Table I parameters and the same 500 channel realizations, regenerate Fig. 3 and verify feasibility of each reported point against the original OPP constraints. If the required transmit power increases or becomes infeasible at some Pmin values, or the TDMA-NOMA curve no longer stays below the TDMA curve, the central claim fails.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim requires every simulation point to be feasible for OPP (12), i.e. R_{j,i}=t_i log2(1+SINR_{j,i})≥Rmin. The SCA chain does not enforce this. OP1 (25) contains r_{j,i}≥Rmin, but it does not carry over Eq. (14a) (R_{j,i}≥ϑ_{j,i}); instead Eqs. (14b)-(14c) force θ_{j,i}≥2^{ϑ_{j,i}} and 1+SINR≥θ, which gives log2(1+SINR)≥Rmin. Since t_i=T/C<1, the true rate can be as low as t_i Rmin, below the required Rmin. The correct surrogate needs θ_{j,i}≥2^{Rmin/t_i} (or ϑ_{j,i}≥Rmin/t_i); for Rmin=0.1 and t_i=0.2 this changes the SINR target from 2^{0.1}-1≈0.072 to 2^{0.5}-1≈0.414, roughly a 7.6 dB gap. Eq. (21a) also contains an undefined multiplier γ, so the denominator constraint is incomplete. The reported transmit-power advantage is therefore a comparison of possibly infeasible relaxed solutions, not a comparison of designs meeting the stated QoS.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a hybrid TDMA-NOMA downlink system with simultaneous wireless power and information transfer (SWIPT). Users are divided into groups, each served in a dedicated time slot using power-domain NOMA, and each user splits its received signal between information decoding and energy harvesting. The authors formulate a non-convex power minimization problem with minimum rate and minimum harvested-power constraints per user, and propose an iterative successive convex approximation (SCA) algorithm to jointly optimize the power allocations and power-splitting ratios. Simulation results compare the required transmit power of the proposed hybrid scheme with that of a conventional TDMA system and report a transmit-power advantage for the hybrid scheme.","tokens_in":10729,"tokens_out":8492,"duration_ms":83437,"significance":"If the proposed algorithm were correct, the paper would provide a useful design framework for energy-harvesting NOMA systems and a quantitative comparison against a TDMA baseline. The system model is clearly presented, the SWIPT formulation is standard, and the comparison with TDMA is a natural evaluation. However, the central claim in the conclusion rests entirely on the correctness of the SCA reformulation, and that reformulation contains several load-bearing errors, detailed below, which mean the numerical results in Section IV do not currently establish the claimed transmit-power advantage.","major_comments":[{"comment":"The rate constraint is not enforced with the time-sharing factor t_i. The chain in (14a)-(14c) together with r_{j,i} >= Rmin in (25b) imposes log2(1+SINR_{j,i}) >= Rmin, but the actual rate is R_{j,i} = t_i log2(1+SINR_{j,i}) with t_i = T/C < 1. Consequently, a point feasible for the surrogate can have R_{j,i} as low as t_i Rmin, which is below the required minimum rate. The correct surrogate requires theta_{j,i} >= 2^{Rmin/t_i} (or equivalently vartheta_{j,i} >= Rmin/t_i). In addition, OP1 omits (14a) and does not link r_{j,i} to theta_{j,i} or vartheta_{j,i}; the variable set Gamma in (25) does not contain theta or vartheta, so the chain is incomplete as printed.","section":"Section III.B, Eqs. (14a)-(14c) and OP1 (25)"},{"comment":"The multiplier gamma in (21a) is undefined. It is not introduced in the text, is not included in the optimization variable set Gamma, and is not listed as a simulation parameter. Since (21) is used in OP1, the feasible set actually implemented in simulations is unspecified; if gamma is not equal to 1, the constraint is not equivalent to (17). The authors must either remove gamma or define it explicitly and justify its value.","section":"Section III.B, Eq. (21a)"},{"comment":"The first-order Taylor linearization of beta p^2 (and similarly of (1-beta)p^2) is not a valid SCA lower bound. The function f(beta,p) = beta p^2 has an indefinite Hessian, so its linearization is not a global underestimator; for example, at (beta0,p0)=(1,1), the linearization at (beta,p)=(4,0.5) equals 3 while f equals 1. Therefore the constraint (19) >= alpha can be satisfied while the original constraint (18) is violated. This means the SCA may return points that are infeasible for OPP, and the claimed equivalence of OP1 and OPP is not established. A different convex underestimator or a proof of conservativeness over the domain is needed.","section":"Section III.B, Eqs. (19) and (23)"}],"minor_comments":[{"comment":"The abstract says 'we employ successive interference cancellation to overcome these non-convexity issues'; this should be 'successive convex approximation' (SCA), not SIC.","section":"Abstract"},{"comment":"The variable set Gamma lists r_{j,i}, while the derivation uses vartheta_{j,i} as the rate slack variable; the relationship between r_{j,i} and vartheta_{j,i} is never defined, which makes OP1 difficult to interpret.","section":"Section III.B, Eq. (25)"},{"comment":"The grouping notation in (13a) mixes two indexing conventions and is difficult to follow; an explicit example for the simulated case K=10, C=5 would clarify the proposed grouping strategy.","section":"Section II.A, Eq. (13a)"},{"comment":"Figure 3 contains an inset whose axes and meaning are not described in the caption; please add an explanation.","section":"Section IV, Fig. 3"},{"comment":"The stopping threshold mu in Algorithm 1 is not given a value in Section IV, so the convergence results in Fig. 4 are not fully reproducible.","section":"Section III, Algorithm 1"},{"comment":"Reference [28] is cited as 'Accepted IEEE Trans. Commun. 2019' without volume, pages, or DOI; please complete the reference.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The paper appears to be an early draft with multiple technical errors in the core SCA derivation. The topic is suitable for the journal, and the errors are identifiable and in principle fixable, so I recommend major revision rather than rejection. However, the authors must correct the rate constraint, define or remove gamma, and replace the invalid linearizations with proper convex underestimators, then rerun the simulations before the central claim can be evaluated."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: the paper takes on a sensible problem — jointly allocating power and power-splitting ratios for a hybrid TDMA-NOMA system with SWIPT to minimize transmit power under rate and harvested-power constraints — and the broad SCA approach is a reasonable template. The system model is clearly set out, the SIC ordering is conventional, and the simulation setup is standard. But the way the rate constraints are written does not actually enforce the minimum rate, so the comparison against TDMA cannot be trusted as printed.\n\nWhat is genuinely new is the specific combination: hybrid TDMA-NOMA plus SWIPT plus a P-Min formulation solved with SCA. That is a legitimate extension of prior NOMA resource-allocation work, including the authors' own. The problem statement OPP is well posed, and the grouping heuristic is sensible. The paper is not circular; it is solving a new combination of known ingredients.\n\nWhere it falls apart: the rate constraint chain in Eqs. (14a)–(14c) and OP1. Because R_j,i = t_i log2(1+SINR_j,i), enforcing R_j,i ≥ Rmin requires log2(1+SINR) ≥ Rmin/t_i. The chain only enforces log2(1+SINR) ≥ ϑ and then, at best, ϑ ≥ Rmin. With t_i < 1, the actual rate can be as low as t_i Rmin, below the stated requirement. For the simulations' t_i = 0.2 and Rmin = 0.1, the SINR target is understated by about 7.6 dB. That is a load-bearing flaw, not a typo.\n\nThere are also internal inconsistencies in OP1 as printed. The variable set Γ includes p, r, β, ϱ, ρ, α, but the constraints include (14c), (20), and (21a), which use ϑ, θ, χ, and an undefined multiplier γ. Even allowing for typos, a reader cannot reproduce the algorithm as stated. The TDMA comparison uses the same SCA chain, so the relative power gap is also suspect.\n\nCredit where due: the paper is clearly written, the literature is cited appropriately, and the authors are not overclaiming beyond the reported simulation. The central idea is salvageable. With the rate constraint corrected to θ ≥ 2^(Rmin/t_i) and OP1 fully specified, the result could be a modest but useful incremental contribution to NOMA-SWIPT resource allocation. As it stands, the main claim is not supported.\n\nI would reject the current version. The flaw is central but fixable, so a serious editor could send it to peer review rather than desk reject — a reviewer will likely catch the same issue and the authors can repair it. But my own verdict on this manuscript is clear: not acceptable without major revision.","headline":"A reasonable problem and a standard SCA template, but the printed rate constraints do not enforce the minimum rate, so the headline power savings over TDMA are not established.","tokens_in":11274,"tokens_out":6542,"would_cite":false,"duration_ms":64261,"reading_group":"no","serious_thinker":"no","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"By having users split each received signal into information and energy branches and serving grouped users with NOMA inside TDMA slots, the paper claims minimum transmit power falls below TDMA's while meeting the same rate and…","keywords":["hybrid TDMA-NOMA","energy harvesting","SWIPT","power splitting","power minimization","successive convex approximation","non-orthogonal multiple access","wireless power transfer"],"falsifier":"Run OP1 for a single two-user group with $C=1$ and fixed channel gains, then evaluate the returned allocation against the original formulas: $R_{j,1}=t_1\\log_2(1+\\mathrm{SINR}_{j,1})$ using (6) and $P_{j,1}=\\eta(1-\\beta_{j,1})|h_{j,1}|^2(p_{1,1}^2+p_{2,1}^2)$. If the slack variables satisfy the printed constraints while the true rate or harvested power drops below $R_{\\min}$ or $P_{\\min}$, the power-saving comparison is settled by the surrogate rather than by the original problem.","tokens_in":10245,"feed_emoji":"⚡","tokens_out":9536,"duration_ms":86487,"temperature":0.7,"pith_summary":"This paper tries to establish that a hybrid TDMA-NOMA downlink with simultaneous wireless power and information transfer can serve users at guaranteed data rates and harvested power using less transmit power than conventional TDMA. The reason this would matter is that wireless networks could harvest part of their energy from the same radio signal that carries data, easing battery and environmental pressures in dense deployments. The paper formulates the joint choice of transmit powers and receiver power-splitting ratios as a nonconvex minimization of total base-station power, solves it with successive convex approximation, and uses simulations over 500 channel realizations to support the claim that the hybrid scheme is more power-efficient.","feed_headline":"Hybrid NOMA cuts transmit power needed for rate and energy targets","feed_subtitle":"By grouping users and splitting each signal, interference feeds energy harvesting while SIC preserves the data rate.","key_machinery":"The central object is the power-splitting SIC SINR model combined with a successive convex approximation reformulation. Each nonconvex rate constraint $t_i\\log_2(1+\\mathrm{SINR}_{j,i})\\ge R_{\\min}$ is replaced in OP1 by slack variables $\\vartheta_{j,i}$, $\\theta_{j,i}$, $\\alpha^d_{j,i}$, and $\\chi_{j,i}$ linked by constraints (14a)--(14c) and (19)--(21), with bilinear products linearized by first-order Taylor expansions around the previous iterate. This machinery turns the joint power-allocation and power-splitting design into a sequence of convex subproblems; the paper treats the converged point of that sequence as the solution of the original minimization.","core_discovery":"The paper's central claim is that in a single-antenna downlink, dividing users into groups, serving each group in its own time slot with power-domain NOMA, and letting every receiver split the received signal between information decoding and energy harvesting yields a lower minimum transmit power than the TDMA-only schedule under the same constraints. The decisive mechanism is that the NOMA superposition signal contributes to the harvested power at each user, $P_{j,i}=\\eta(1-\\beta_{j,i})|h_{j,i}|^2\\sum_s p_{s,i}^2$, while SIC keeps the information branch decodable. The paper reports simulations in which the hybrid scheme's required transmit power stays below TDMA's over the tested range of minimum harvested-power requirements.","pith_inferences":["The paper fixes two users per group; because harvested power scales with total received power, larger groups would likely make the energy constraint easier to satisfy, though at additional SIC decoding burden, and testing $K_i>2$ would show whether the power advantage grows.","The time slots are fixed at $T/C$; allowing unequal slot lengths would add a degree of freedom that could trade per-user rate against harvesting time, a direct extension of the same problem.","The rate surrogate in (14c) would need to be scaled by the slot duration to strictly enforce $R_{\\min}$; rerunning the reported simulations with $\\theta \\ge 2^{R_{\\min}/t_i}$ would test how much of the power saving is robust to the surrogate gap.","Before the algorithm can be reimplemented, a value or rule for the multiplier $\\gamma$ in (21a) must be supplied, since the printed linearized bound leaves it unspecified."],"forward_implications":["For the simulated system, the hybrid TDMA-NOMA design satisfies the same minimum rate per user and minimum harvested power with lower base-station transmit power than TDMA.","As the minimum harvested-power requirement grows, both schemes need more transmit power, but the hybrid scheme's interference-assisted harvesting keeps its power lower.","The SCA algorithm converges in a small number of iterations for the tested parameters, so the design is computationally practical for the ten-user, five-group setup.","The grouping strategy pairs the strongest user with the weakest so that SIC differences are large, which the paper argues is needed for practical NOMA."],"supporting_citations":[{"why":"Supplies the power-splitting SWIPT model and the harvested-power formula used in constraint (12c).","marker":"[16]"},{"why":"Introduces simultaneous wireless information and power transfer, motivating the signal-splitting receiver.","marker":"[15]"},{"why":"Provides the distance-based path-loss channel model and simulation parameter values.","marker":"[17]"},{"why":"Defines the decode-SINR as a minimum over SIC detection stages, which underlies the rate expression.","marker":"[24]"},{"why":"Contributes the SCA linearization strategy for NOMA power allocation that the proposed algorithm adapts.","marker":"[14]"},{"why":"Establishes the hybrid NOMA-with-OMA concept that the TDMA-NOMA grouping builds on.","marker":"[8]"}],"fun_headline_variants":["Hybrid TDMA-NOMA uses less power for same rate and energy goals","NOMA's superposed signal powers receivers, cutting required transmit power","Hybrid NOMA beats TDMA: same rates, more energy harvesting, less power","Energy-harvesting NOMA lowers transmit power vs TDMA in hybrid scheme","NOMA's interference becomes power: hybrid TDMA-NOMA cuts transmit power"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the convex chain that replaces the rate constraint actually enforces the promised minimum rate rather than just a slack bound; in the printed version the step $\\theta \\ge 2^{\\vartheta}$ omits the time-slot duration $t_i$, and if that omission is not harmless the reported transmit powers are below what the original problem truly requires.","fun_headline_variants_meta":{"raw":{"variants":["Hybrid TDMA-NOMA uses less power for same rate and energy goals","NOMA's superposed signal powers receivers, cutting required transmit power","Hybrid NOMA beats TDMA: same rates, more energy harvesting, less power","Energy-harvesting NOMA lowers transmit power vs TDMA in hybrid scheme","NOMA's interference becomes power: hybrid TDMA-NOMA cuts transmit power"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000662,"raw_usage":{"total_tokens":3017,"prompt_tokens":932,"completion_tokens":2085,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":548,"completion_tokens_details":{"reasoning_tokens":1997}},"tokens_in":548,"tokens_out":2085,"duration_ms":16081,"temperature":1.0,"reasoning_tokens":1997,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:30:56.959096+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run OP1 for a single two-user group with $C=1$ and fixed channel gains, then evaluate the returned allocation against the original formulas: $R_{j,1}=t_1\\log_2(1+\\mathrm{SINR}_{j,1})$ using (6) and $P_{j,1}=\\eta(1-\\beta_{j,1})|h_{j,1}|^2(p_{1,1}^2+p_{2,1}^2)$. If the slack variables satisfy the printed constraints while the true rate or harvested power drops below $R_{\\min}$ or $P_{\\min}$, the power-saving comparison is settled by the surrogate rather than by the original problem.","supporting_citations":[{"cited_title":"Simultaneous wireless information and power transfer in modern communication systems,","cited_arxiv_id":null,"evidence_quote":"Supplies the power-splitting SWIPT model and the harvested-power formula used in constraint (12c)."},{"cited_title":"Transporting information and energy simultaneously,","cited_arxiv_id":null,"evidence_quote":"Introduces simultaneous wireless information and power transfer, motivating the signal-splitting receiver."},{"cited_title":"Optimal sum-throughput analysis for downlink cooperative SWIPT NOMA systems,","cited_arxiv_id":null,"evidence_quote":"Provides the distance-based path-loss channel model and simulation parameter values."},{"cited_title":"A minorization-maximization method for optimizing sum rate in the downlink of non-orthogonal multiple access systems,","cited_arxiv_id":null,"evidence_quote":"Defines the decode-SINR as a minimum over SIC detection stages, which underlies the rate expression."},{"cited_title":"Energy efficient beamforming design for MISO non-orthogonal multiple access systems,","cited_arxiv_id":null,"evidence_quote":"Contributes the SCA linearization strategy for NOMA power allocation that the proposed algorithm adapts."},{"cited_title":"Non-orthogonal multiple access for 5G and beyond,","cited_arxiv_id":null,"evidence_quote":"Establishes the hybrid NOMA-with-OMA concept that the TDMA-NOMA grouping builds on."}],"review_version":1}