{"id":"30641ccb-b6e4-4c28-85bc-0ce0816a61eb","arxiv_id":"2506.17189","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":3.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Simulations show that RIS-assisted CoMP-NOMA with phase-shift designs that combine signal enhancement and interference cancellation improves energy efficiency, peaking at moderate cooperation levels.","lead":"This paper simulates a wireless network where base stations cooperate and use non-orthogonal multiple access while bouncing signals off reconfigurable intelligent surfaces to serve users. It finds that adjusting the surfaces to both strengthen the desired signal and cancel interference improves energy efficiency, with peak gains at moderate numbers of cooperating base stations and surface elements.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Eq. (11) is not the stated energy efficiency: it is a sum of per-link ratios that double-counts cooperative BS power, so the J-dependent EE curves in Fig. 2 are artifacts of the denominator, not of the system.","rationale":"The reader's conditional verdict focuses on the cell-center isolation assumption and the unproven optimality of Eqs. (13)-(14). I agree those are weaknesses, but the most load-bearing issue is more basic: the objective in Eq. (11) does not implement the energy efficiency it defines. Because Eq. (12) optimizes this same objective and Figs. 2-5 report values of it, the double-counting propagates into every quantitative result. The flaw is demonstrable analytically without simulations: under the stated global definition, EE cannot fall as J grows, yet Fig. 2 shows a decline after J=4. The 'optimal number of cooperative BSs' is thus an artifact. This is not an ad hominem or a disagreement with consensus; it is an internal inconsistency between the definition and the equation. The phase-shift rules themselves may survive a correction, which is why the paper is not beyond repair, but as written the central energy-efficiency claims are not supported. I therefore move the verdict from CONDITIONAL to REJECT pending recomputation with the correct metric.","tokens_in":9014,"tokens_out":13041,"duration_ms":144900,"concrete_test":"Recompute Figs. 2 and 3 using the global ratio stated in the text: eta_EE = (sum_i R^out_i_c + sum_j R^out_e_f) / (sum_i (P_i/lambda + P_Q) + total RIS power), with the same channel realizations and all RIS elements active. In parallel, verify the analytic monotonicity: with zeta=0.7, adding a cooperative BS changes gamma_e^f by moving 0.7P|H|^2 into the numerator and reducing the denominator by the same amount, while total power is unchanged, so the outage sum rate and hence the corrected EE are nondecreasing in J. If the corrected Fig. 2 no longer declines after J=4, the reported optimum is an artifact of Eq. (11); if the EC/EO ordering changes, the comparative claim is also affected.","verdict_should_be":"REJECT","load_bearing_attack":"Section III-B defines eta_EE as the ratio of the achievable outage sum rate to the total power expended, but Eq. (11) is a sum of I cell-center ratios plus J edge-user ratios. Each cooperative BS's power term (1/lambda)P_j + P_Q appears twice: once in the cell-center denominator and once in the edge-user denominator, while the model's total transmit and static power is independent of J because all BSs transmit simultaneously and all RIS are active. Under the stated definition, moving any BS from the non-cooperative set into the cooperative set strictly increases the edge SINR in Eq. (4): its channel term leaves the inter-cell interference sum Y_f with weight P|H|^2 and enters the CoMP numerator with weight zeta P|H|^2 and the intra-cluster denominator with weight (1-zeta)P|H|^2, with total power fixed. Therefore the true global EE cannot decrease with J, so the peak at J=4 in Fig. 2 is produced by the additive denominator. The same structural issue contaminates the K-dependence and absolute EE values in Figs. 3 and 5 and the optimization objective in Eq. (12). This invalidates the quantitative basis for the paper's central energy-efficiency claims as written.","agreement_with_reader":"disagree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript studies a downlink RIS-assisted CoMP-NOMA network with I cells, one cell-center user per cell, one shared edge user, and J cooperating base stations. It proposes two RIS phase-shift policies: enhancement-only PBF (EO), which aligns RIS reflections to maximize the effective channels from cooperative BSs to the edge user, and enhancement-and-cancellation PBF (EC), which additionally steers non-cooperative BS reflections to suppress inter-cell interference. The authors define an outage-aware energy-efficiency metric and use Monte Carlo simulations to report that EC outperforms EO except when all BSs cooperate, that energy efficiency peaks at J=4 and K=90, and that RIS-assisted CoMP-NOMA outperforms no-RIS and OMA baselines. The abstract and Section III-C frame the phase-shift rules as the solution of an energy-efficiency maximization problem.","tokens_in":9278,"tokens_out":8986,"duration_ms":96582,"significance":"If the claimed results were fully supported, the paper would offer a useful design guideline for configuring RIS phases in multi-cell CoMP-NOMA systems and would quantify energy-efficiency/outage trade-offs. The system model is clearly described and the parameter sweeps are reasonably extensive. However, the two central claims—the optimality of the proposed PBF design and the quantitative energy-efficiency comparisons—rest on an energy-efficiency metric that is not the one stated in the text and on phase-shift rules adopted from prior work without a proof of optimality for this objective. No analytical outage expressions, proofs, or reproducible code are included; the contribution is essentially a numerical study. With the metric corrected and the optimality claims softened, the paper could become a useful simulation-based design study, but in its current form the quantitative conclusions are not supported.","major_comments":[{"comment":"The energy-efficiency metric in Eq. (11) is not the ratio of the achievable outage sum rate to the total power expended, as stated in the text. It is instead a sum of per-link rates divided by per-BS power terms. Because the edge-user outage rate is summed over all cooperative BSs, the same edge rate is counted J times. Moreover, the power of a cooperative BS appears both in a cell-center denominator and in an edge-user denominator, so the metric implicitly makes total power consumption increase with J, even though the system model has every BS transmitting at the same power P_t and every RIS active regardless of J. Consequently, the peak at J=4 in Fig. 2 and the J-dependence of the energy-efficiency curves in Figs. 2, 3, and 5 are artifacts of the metric rather than properties of the system. A corrected global energy efficiency with a single total-power denominator would behave differently, and the reported quantitative comparisons and the claimed optimal number of cooperating BSs would need to be re-established.","section":"Section III-B, Eq. (11)"},{"comment":"The optimization problem in Eq. (12) is stated to maximize the energy efficiency η_EE, but the proposed phase-shift rules in Eqs. (13) and (14) only maximize or minimize the individual effective channel gains |H_{j,f}^e|^2 and |H_{m,f}^e|^2. No argument is given that these rules optimize the energy-efficiency objective, which is a non-convex function of SINRs, outage probabilities, and power allocations. The rules are standard single-link phase-alignment and phase-cancellation designs taken from reference [21], and applying them to the sum-EE objective is a heuristic. Therefore the abstract's claim that the 'optimal PBF design' is contingent on system parameters is not supported by the analysis. The authors should either replace the optimality language with 'proposed heuristic phase-shift designs' or provide a rigorous derivation, such as an alternating-optimization or upper-bound argument, linking the phase rules to the energy-efficiency objective.","section":"Section III-C, Eqs. (12)-(14)"},{"comment":"The assumption that 'the impact of RIS on the channels experienced by U_c^i is negligible' is load-bearing but unsupported. The cell-center SINR expressions in Eqs. (6)-(7) exclude all RIS contributions, and those SINRs affect the outage probabilities and hence the energy-efficiency metric in Eq. (11). Since the RISs are placed at the cell edge and reflect signals from all base stations, there is no physical justification in the manuscript for assuming their phase settings do not affect cell-center users, particularly the inter-cell interference terms. A change in cell-center SINRs could alter the SIC decoding order and the outage rates, which could in turn change the relative ordering of the EO and EC designs. The authors should justify this assumption with a quantitative argument, such as a path-loss and angle-of-arrival analysis showing that the reflected power at U_c^i is below the noise floor, or they should include the RIS contributions in the cell-center SINR model and check the sensitivity of their conclusions.","section":"Section III-A, after Eq. (7)"}],"minor_comments":[{"comment":"The section title promises 'Rate and Outage Probability Analysis', but Eqs. (8)-(10) only define rates and outage events; no closed-form outage probability expressions, approximations, or asymptotic results are provided. All numerical results are Monte Carlo estimates with N_mc=10^4. The authors should state explicitly that the contribution is simulation-based rather than analytical.","section":"Section III-A"},{"comment":"In the definition of the effective channel for a non-cooperative BS after Eq. (3), the indices are inconsistent: the expression uses h_{R_i,f}^T Θ_i h_{m,R_i} with i instead of m. It should be h_{R_m,f}^T Θ_m h_{m,R_m}.","section":"Eq. (3)"},{"comment":"The notation P_R is used in Eq. (11) before it is defined; the definition P_R = K P_ele appears only after the equation. In addition, the notation Y_f in Eq. (4) and Y_i in Eqs. (6)-(7) should be defined consistently and with matching subscripts.","section":"Eq. (11) and surrounding text"},{"comment":"The text says EC 'consistently outperforms other scenarios ∀J, except when J=I', which is internally contradictory. It should read 'for all J < I' or otherwise quantify the exception.","section":"Section IV-B, Fig. 2"},{"comment":"Reference [13] already proposes a simultaneous signal enhancement and cancellation (SSECB) design for STAR-RIS-enhanced NOMA-CoMP networks, and the EC design in this manuscript is conceptually similar. The authors should explicitly discuss the differences between their EC scheme and the SSECB design, and position the novelty accordingly.","section":"Reference [13]"},{"comment":"The x-axis label in Fig. 3 appears as 'NumberofRISelementsK' without spaces; please fix the typographical error.","section":"Fig. 3"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is within the scope of the journal, but the current framing significantly overclaims. In my view the most serious issue is the energy-efficiency metric in Eq. (11), which is not the ratio of total outage sum rate to total power and double-counts the edge-user rate; this directly undermines the quantitative conclusions in Figs. 2, 3, and 5. The optimality claims in the abstract and Section III-C are also not supported by the analysis, since Eqs. (13)-(14) are standard single-link heuristics. A revision that corrects the metric, reruns the simulations, and softens the optimality language to 'heuristic' could result in a publishable simulation study, but the changes are substantial. I would not recommend acceptance without seeing the revised results."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear [name],\n\nQuick take: this is a reasonably careful Monte Carlo evaluation of two existing phase-shift heuristics (enhancement-only and enhancement-plus-cancellation) in a RIS-assisted CoMP-NOMA setup. The writing is clear, the simulation parameters are stated, and the authors don't overclaim the analytics—the outage analysis is just definitions. But the central energy-efficiency metric in Eq. (11) is not what they say it is, and that undermines the headline result that EE peaks at J=4.\n\nWhat's actually new: not much. The phase-shift rules in Eqs. (13)-(14) are the standard alignment and pi-shift rules from Wu & Zhang [21] and Hou et al. [13]; the EC scheme is essentially the SSECB design. The paper's own contribution is the EE evaluation for a specific power model with static power and RIS element power. That kind of parameter sweep can be useful to practitioners choosing RIS configurations, and the setup is reproducible in principle.\n\nThe soft spots, in decreasing order of importance:\n\n1. Eq. (11) is a sum of per-link efficiency ratios, not a ratio of total outage sum rate to total power. Cooperative BS power enters the denominator once for the cell-center term and once for the edge-user term, so the denominator grows with J even though total network power doesn't. That means the peak at J=4 in Fig. 2 can be a denominator artifact rather than a real system property. The same issue contaminates the K-dependence in Fig. 3 and the objective in Eq. (12). This needs a rewrite of the metric or a reinterpretation of the curves.\n\n2. Calling Eqs. (13)-(14) 'optimal PBF design' is not supported. They optimize individual channel gains, not the EE objective in Eq. (12). The paper would be more accurate to say they are well-known heuristics.\n\n3. The assumption that RIS does not affect cell-center users is asserted, not shown. It's probably acceptable in a first-order model, but because SIC ordering and outage rates for cell-center users drive the numerator of Eq. (11), the authors should at least state that the RIS is positioned and steered so its effect on cell-center channels is negligible.\n\nThe math is otherwise straightforward, the citation pattern looks reasonable, and there's no fitted-parameter circularity.\n\nBottom line: the paper is a borderline incremental study with a fixable flaw in its headline metric. Worth sending to peer review with a clear request to correct Eq. (11) and soften the optimality language. For a reading group, it's a decent example of how subtle definitions of EE can change conclusions.","headline":"A competent but incremental simulation study of known RIS phase-shift heuristics, whose headline energy-efficiency metric is mis-specified and likely creates the reported J-dependence as an artifact.","tokens_in":9819,"tokens_out":2682,"would_cite":false,"duration_ms":26396,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper claims that a reconfigurable surface can raise the energy efficiency of coordinated NOMA networks, and that an enhance-and-cancel phase rule beats enhancement alone except when all base stations cooperate.","keywords":["energy efficiency","passive beamforming","reconfigurable intelligent surface","NOMA","CoMP","phase-shift optimization","outage probability","interference cancellation"],"falsifier":"One concrete falsifier is to rerun the simulation with RIS contributions included in cell-center user signals and compare EO versus EC energy efficiency; a ranking change, or a live test where cell-center signal quality varies as the RIS phases toggle, would contradict the paper's central premise.","tokens_in":8803,"feed_emoji":"📡","tokens_out":7789,"duration_ms":71977,"temperature":0.7,"pith_summary":"The paper is trying to establish a simple, low-complexity rule for setting the phase shifts of reconfigurable intelligent surfaces in a coordinated multi-cell NOMA network, with the goal of maximizing energy efficiency rather than raw rate. It proposes two configurations—enhancement-only (EO), which aligns reflected paths with the edge user's direct channel, and enhancement-and-cancellation (EC), which additionally flips the phases of non-cooperating base stations' RISs by 180° to suppress interference. Using simulation, it argues that EC generally yields the best energy efficiency, except when all base stations cooperate, and that there is a sweet spot in the number of cooperating base stations and in the number of RIS elements. This matters because RIS is proposed as a cheap, passive way to improve wireless networks, and the paper gives a concrete design rule for when the surface helps or hurts the energy budget.","feed_headline":"Enhance-and-cancel RIS design beats enhance-only for efficiency","feed_subtitle":"Phase tuning boosts NOMA energy efficiency; cancellation helps unless all BSs cooperate.","key_machinery":"The argument is carried by a linear phase-tuning identity. For a cooperative base station's RIS, each element's phase is set to $\\theta_j^k = \\arg(h^e_{j,f}) - \\arg(h^{(k)}_{R_j,f} h^{(k)}_{j,R_j})$ (Eq. (13)), which aligns the reflected cascade with the direct channel and maximizes $|H^e_{j,f}|^2$. For a non-cooperating base station's RIS, Eq. (14) sets $\\theta_m^k = \\bmod[\\phi_m^k+\\pi, 2\\pi]-\\pi$, flipping the reflected component by 180° to cancel interference. These two rules are inserted into the energy-efficiency objective $\\eta_{\\text{EE}}$ (Eq. (11)), the ratio of outage-weighted sum rate to total consumed power including per-element RIS power $KP_{\\text{ele}}$, and their effects are evaluated by Monte Carlo simulation.","core_discovery":"The paper claims that in a downlink multi-cell CoMP-NOMA network, a set of passive RISs can improve energy efficiency if each surface's phase shifts are chosen deliberately rather than randomly, and that the best choice depends on how many base stations cooperate and how many reflecting elements are available. Specifically, the enhancement-and-cancellation (EC) design—cooperative-base-station RISs aligned to strengthen the edge user's signal, non-cooperative-base-station RISs phase-flipped to suppress their interference—consistently gives higher energy efficiency than enhancement-only (EO) or random phase shifts in the simulated scenarios, with one exception: when all base stations cooperate, interference cancellation is unnecessary and EC equals EO. The paper also reports that energy efficiency peaks at an intermediate level of cooperation ($J=4$) and an intermediate number of RIS elements ($K=90$), after which the power drawn by extra elements outweighs the rate gains.","pith_inferences":["Because the energy-efficiency metric charges $P_{\\text{ele}}$ per element, the reported optimum at $K=90$ is tied to the assumed hardware power model; cheaper or more efficient elements would push the optimum toward larger arrays.","The phase rules assume perfect channel knowledge; under phase quantization or estimation error, the 180° cancellation rule in Eq. (14) may degrade faster than the enhancement rule, which could narrow EC's advantage in practice.","If the cell-edge RISs turn out to affect cell-center users, the EC design would face a trade-off between boosting edge users and disturbing the interference-cancellation order at cell-center users, suggesting a joint optimization across both user classes.","The CO/EO split-ratio result implies a resource-allocation subproblem: with a fixed RIS budget and partial cooperation, the fraction of elements assigned to cancellation versus enhancement can itself be optimized."],"forward_implications":["With EC phase shifts, simulated energy efficiency rises as cooperation grows to $J=4$ base stations and then declines, so partial cooperation beats full cooperation.","Energy efficiency peaks around $K=90$ RIS elements; beyond that, the outage-rate gain of extra elements does not pay for their power consumption.","When all base stations cooperate, every interferer becomes a desired signal, so the cancellation branch of EC is redundant and EC matches EO.","With only one cooperating base station, dedicating all RIS elements to cancellation gives the highest outage sum rate, showing that interference suppression carries the benefit when cooperation is thin.","At the tested transmit powers, RIS-assisted CoMP-NOMA with EC phase shifts yields a higher outage sum rate than CoMP-OMA and than CoMP-NOMA without a RIS."],"supporting_citations":[{"why":"Supplies the phase-alignment rule used in Eq. (13) for maximizing the effective edge-user channel gain.","marker":"[21]"},{"why":"Provides the non-coherent joint-transmission model that justifies combining cooperative BS signals without CSI exchange.","marker":"[20]"},{"why":"Introduces the simultaneous signal-enhancement-and-cancellation concept on which the EC configuration is based.","marker":"[13]"},{"why":"Establishes the RIS-assisted CoMP-NOMA setting and motivates placing the RIS at the cell edge.","marker":"[12]"},{"why":"Provides the energy-efficiency formulation for RIS-aided systems that motivates the $\\eta_{\\text{EE}}$ metric.","marker":"[4]"},{"why":"Supports the two-user NOMA pairing and the power-allocation constraint $\\zeta_i > 0.5$.","marker":"[14]"},{"why":"Underpins the decoding-order assumption used to build the outage probability expressions for NOMA users.","marker":"[19]"}],"fun_headline_variants":["Enhance-and-cancel RIS tops enhance-only in NOMA efficiency","EC RIS beamforming beats EO unless all base stations cooperate","RIS-NOMA efficiency peaks at 4 BSs and 90 elements","Deliberate RIS phases beat random; EC best unless full cooperation"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The paper assumes that because the RISs sit at the cell edge, their phase settings do not meaningfully affect the signals received by cell-center users, so cell-center performance is computed without the RIS. If the surface does reach those users, the decoding order and the energy-efficiency ranking between designs would need to be recalculated.","fun_headline_variants_meta":{"raw":{"variants":["Enhance-and-cancel RIS tops enhance-only in NOMA efficiency","EC RIS beamforming beats EO unless all base stations cooperate","RIS-NOMA efficiency peaks at 4 BSs and 90 elements","Deliberate RIS phases beat random; EC best unless full cooperation"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001724,"raw_usage":{"total_tokens":6802,"prompt_tokens":910,"completion_tokens":5892,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":526,"completion_tokens_details":{"reasoning_tokens":5826}},"tokens_in":526,"tokens_out":5892,"duration_ms":34351,"temperature":1.0,"reasoning_tokens":5826,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T19:10:10.072861+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"One concrete falsifier is to rerun the simulation with RIS contributions included in cell-center user signals and compare EO versus EC energy efficiency; a ranking change, or a live test where cell-center signal quality varies as the RIS phases toggle, would contradict the paper's central premise.","supporting_citations":[{"cited_title":"Intelligent reflecting surface enhanced wireless network via joint active and passive beamforming,","cited_arxiv_id":null,"evidence_quote":"Supplies the phase-alignment rule used in Eq. (13) for maximizing the effective edge-user channel gain."},{"cited_title":"A tractable model for noncoherent joint-transmission base station cooperation,","cited_arxiv_id":null,"evidence_quote":"Provides the non-coherent joint-transmission model that justifies combining cooperative BS signals without CSI exchange."},{"cited_title":"A joint design for STAR-RIS enhanced NOMA-CoMP networks: A simultaneous-signal- enhancement-and-cancellation-based (SSECB) design,","cited_arxiv_id":null,"evidence_quote":"Introduces the simultaneous signal-enhancement-and-cancellation concept on which the EC configuration is based."},{"cited_title":"Reconfigurable intelligent surface assisted coordinated multipoint in downlink NOMA networks,","cited_arxiv_id":null,"evidence_quote":"Establishes the RIS-assisted CoMP-NOMA setting and motivates placing the RIS at the cell edge."},{"cited_title":"Reconfigurable intelligent surfaces for energy efficiency in wireless communication,","cited_arxiv_id":null,"evidence_quote":"Provides the energy-efficiency formulation for RIS-aided systems that motivates the $\\eta_{\\text{EE}}$ metric."},{"cited_title":"A simple design of IRS-NOMA transmission,","cited_arxiv_id":null,"evidence_quote":"Supports the two-user NOMA pairing and the power-allocation constraint $\\zeta_i > 0.5$."},{"cited_title":"User pairing, link selection, and power allocation for cooperative NOMA hybrid VLC/RF systems,","cited_arxiv_id":null,"evidence_quote":"Underpins the decoding-order assumption used to build the outage probability expressions for NOMA users."}],"review_version":1}