Pith. sign in

REVIEW 3 major objections 5 minor 1 cited by

Resilience of Task-Oriented V2X Networks to Incomplete Information Sharing

T0 review · 3 major / 5 minor · reviewed 2026-08-02 · deepseek-v4-flash

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

Pith's one-line read Task-oriented V2X networks can tolerate frequent content-selection errors because neighboring vehicles spontaneously retransmit the omitted relevant information.

desk verdict Useful quantification of content-selection-error resilience in task-oriented V2X, but the 'guarantee' is overbroad and Eq. (1) needs a sign fix before the overlap-based mechanism is reproducible. read the letter →

arxiv 2602.18620 v2 pith:CX3ITLLU submitted 2026-02-20 cs.NI

Luca Lusvarghi , Javier Gozalvez This is my paper · ORCID
classification cs.NI
keywords task-orientedV2Xcontent-selectionerrorrelevanceestimationredundancycooperativeperceptionnetworkresilienceC-V2Xcongestioncontrol
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 asks whether relevance-based content selection in vehicle-to-everything (V2X) networks breaks when transmitters often misjudge what information matters to receivers. It claims the networks are inherently resilient: when one vehicle wrongly omits a relevant object from its broadcast, another vehicle that detects the same object tends to include it, so the receiver still gets the information. In the paper's cooperative-perception simulations, more than half of content-selection errors are successfully recovered, and the fraction of relevant objects available to receivers stays close to the error-free case even when the relevance-estimation error probability reaches 0.4. The resilience grows with vehicle density until congestion control starts dropping packets, and it weakens only at very high error probabilities around 0.6–0.8. A sympathetic reader would take away that perfect relevance estimation is not a prerequisite for task-oriented V2X to work.

What carries the argument

The load-bearing mechanism is multi-vehicle detection overlap. The scenario is built so that relevant objects are typically seen by more than one vehicle: perception range 150 m, relevance range 400 m, and an average of 20 locally detected objects per vehicle from a Poisson field of 750 objects per square kilometer, with 40 relevant variables per receiver. Detection overlap means a transmission omission is not a loss; a neighboring vehicle that detects the same object can independently select and broadcast it. The paper's quantitative instruments are the CSE recovery-attempt probability Patt(CSE), the successful-recovery probability Psucc(CSE), the congestion-control drop probability P(drop)

What would settle it

Rerun the same simulation with the perception range reduced to about 50 m while keeping the 400 m relevance range, so that most relevant objects are seen by exactly one vehicle; the paper's mechanism predicts Patt(CSE) and Psucc(CSE) should collapse and ARR should fall steeply as beta grows. If ARR remains stable under single-vehicle detection, the claimed compensation mechanism would not be the source of resilience. Equivalently, plot Psucc(CSE) against the mean number of vehicles per relevant object; the claimed mechanism requires a positive monotone relation before congestion control sets i

Watch

Extended reading notes

Core claim

The central claim is that task-oriented V2X networks possess a network-level, 'inherent' resilience to content-selection errors (CSEs). A CSE is defined as a correctly delivered message that nonetheless omits at least one object relevant to an intended receiver, caused either by an error in estimating whether the object is redundant or in estimating its relevance. The paper's numerical study of a 2 km by 2 km cooperative-perception scenario with 20 locally detected objects per vehicle shows that omission events are frequently repaired by other transmitters: at least 50% of CSEs are subject to recovery attempts and more than 50% are successfully recovered across densities. As a result, the Av

Load-bearing premise

The resilience rests on the assumption that most relevant objects are detected by more than one vehicle, which is guaranteed by the paper's specific scenario parameters (750 objects per square kilometer, 150 m perception range, 400 m relevance range, 40 relevant variables per receiver) and is not tested under reduced detection overlap; at N > 200 congestion control already reduces recovery, so the 'inherent' resilience is bounded.

Editorial extensions

If this is right

  • Transmitters do not need highly accurate relevance estimates; error probabilities up to 0.4 leave the delivery of relevant information nearly unchanged at medium-to-high vehicle density.
  • The resilience is a density effect: more communicating vehicles mean more potential rescuers, up to the point where congestion control starts dropping messages, after which recovery worsens.
  • Hard redundancy estimation, despite producing more content-selection errors, can deliver more relevant information than soft estimation because it wastes less capacity on redundant rebroadcasts.
  • The same resilience should appear in any task-oriented network where multiple sensing devices with overlapping fields of view serve a common set of receivers, such as distributed cameras feeding an edge server.
  • Redundancy-CSEs are inherently easier to recover than relevance-CSEs, because a redundancy error can happen only when at least two vehicles already detect the same object.

Reading between the lines

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

  • A direct test of the boundary condition would vary detection overlap, for example by shrinking the perception range or lowering object density; the paper does not sweep these parameters, so 'inherent' should be read as conditional on sufficient overlap.
  • The error model assumes each transmitter errs independently; if relevance misestimation is correlated across vehicles, for instance from a shared context-prediction error, neighboring vehicles might omit the same object simultaneously and the claimed resilience would weaken.
  • Because the ARR metric counts an object as available even if a recovery arrives late, the paper does not measure the timeliness of recovered information; a latency-aware metric such as age of information per relevant object would show whether situational awareness is preserved in real time.
  • The generalization to other task-oriented networks is plausible but not demonstrated; the paper's evidence is a single V2X cooperative-perception setup, so the cross-domain claim is a hypothesis awaiting similar overlap conditions elsewhere.
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 / 5 minor

Summary. The paper studies content-selection errors (CSEs) in task-oriented V2X cooperative perception networks. In these networks, vehicles choose which locally detected objects to broadcast based on estimated relevance to intended receivers; relevance estimation errors (probability β) or redundancy estimation errors can cause a vehicle to omit relevant objects, i.e., a CSE. Using a C-V2X sidelink simulation with ETSI/SAE-compatible parameters, the authors evaluate the probability of CSEs, the Available Relevant variables Ratio (ARR), and the probability that a CSE is recovered by neighboring vehicles. The central claim is that task-oriented V2X networks exhibit an inherent network-level resilience to CSEs: when one vehicle omits a relevant object, other vehicles that detect the same object spontaneously retransmit it, so ARR remains stable even at non-negligible β. The authors report that this resilience holds up to β ≈ 0.4, degrades at higher β, and identify the overlapping detection of multiple transmitters as the underpinning condition, which they argue extends to other task-oriented multi-transmitter networks.

Significance. If the result holds as stated, it would be practically useful: it would relax the accuracy requirements on relevance estimation for task-oriented V2X, reduce the need for perfect context estimation, and support the scalability of cooperative perception. The paper also provides a plausible mechanism—multi-vehicle detection overlap leading to spontaneous CSE recovery—and quantifies it with recovery-attempt and recovery-success probabilities. Strengths include the use of standard parameter values (C-V2X analytical models, ETSI/SAE settings, congestion control from ETSI TS 103 574) and the distinction between hard and soft redundancy estimation. However, the paper is a simulation study with no analytical derivation, no code release, and no sensitivity analysis over the scenario parameters that generate the overlap on which the entire claim rests. The abstract and conclusions state a guarantee that is contradicted by the paper's own Fig. 6(b) at β = 0.8. Moreover, the printed detection model in Eq. (1) is internally inconsistent with the stated scenario statistics, making the simulation currently non-reproducible. The significance is therefore moderate: the idea is credible and worth re

major comments (3)
  1. [Section V, Eq. (1)] The printed coefficients c1=0.08, c2=-0.08, c3=-60 do not reproduce the stated scenario. For these values P(detect)=0.093 at D=0, 0.020 at D=20 m, and 0.004 at D=40 m; the integral over the 2 km × 2 km area with density 750 vars/km² gives a mean of ~0.01–0.02 detected variables per vehicle, not 20, and the curve does not define a 150 m perception range. The intended curve appears to require c3=+60 (with c2=-0.08), which yields ~20 detected variables and a ~150 m effective range, but as printed the simulation is not reproducible. Because the resilience mechanism is exactly multi-vehicle detection overlap, this error is load-bearing; please correct the equation or the constants and report the detection-multiplicity statistics.
  2. [Abstract and Section VI-B (Fig. 6(b))] The abstract and Section VI claim 'inherent resilience ... guarantees a consistent delivery of relevant information even under high relevance estimation error probability conditions.' The paper's own results contradict this: Fig. 6(b) shows ARR drops substantially at β=0.8, and the text states 'the task-oriented V2X network resilience can be compromised if the relevance estimation error becomes excessively large.' The guarantee holds, at best, for β≤0.4 in the considered scenario. Please qualify the claim and state the threshold/regime explicitly.
  3. [Sections IV–VI] The 'fundamental conditions' underpinning resilience are not quantified. The mechanism depends on the likelihood that a relevant object is detected by more than one vehicle, which in this setup follows from the exogenous-variable density (750/km²), perception range (150 m), relevance range (400 m), and M=40. The paper never reports the distribution of detection multiplicity per relevant object, nor does it vary overlap (e.g., density, perception range, M) to show where resilience breaks. Given the general claim in the abstract and conclusions, this is a load-bearing gap; a sensitivity analysis over detection overlap (or an explicit analytic characterization) is needed.
minor comments (5)
  1. [Section VI-A] The text says 'As long as N < 200, P(drop)=0 ... Conversely, when N < 200, the network load increases beyond the predefined limits and congestion control intervenes.' The second inequality should read N > 200; as written it makes the discussion of Figs. 4 and 5 confusing.
  2. [Section VI-B, footnote] The footnote references 'Fig. 8(a)' and 'Fig. 7(a)'; there is no Fig. 8 and the ARR plot is Fig. 6(b). Please correct the cross-references.
  3. [Section VI-B] The text 'as long as β ≤ 0.41' is oddly precise; the scenarios use β = 0, 0.2, 0.4, 0.6, 0.8. This should likely read 'β ≤ 0.4'.
  4. [Section V] Eq. (1) is a logistic curve that only asymptotically approaches zero; the statement that the perception range is 'the distance at which the probability ... is equal to zero' is technically incorrect. Consider defining the range via a small threshold (e.g., P = 0.01) or reporting the effective detection radius used in the simulation.
  5. [Section IV, ARR definition] The ARR definition counts locally detected variables as available even if they are never transmitted; this is reasonable for cooperative perception, but the paper should state explicitly that ARR is a network-level availability metric, not a measure of what the receiver obtains strictly via V2X. This also explains why ARR cannot fall below the local-detection baseline.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the resilience metrics are simulation outputs, not fitted inputs or self-citation-derived conclusions.

full rationale

The paper is a numerical simulation study. The central quantities (P(CSE), ARR, Patt(CSE), Psucc(CSE)) are measured outputs of a simulator whose inputs (detection probability Eq. (1), exogenous-variable density, perception range, relevance range, M, beta, channel and congestion models from [23]-[24]) are stated independently and are not tuned to reproduce the resilience claim. The 'inherent resilience' mechanism is explained after the fact as multi-vehicle redundant detection of relevant variables; the paper does not define ARR as equivalent to CSE recovery, and the reported Psucc(CSE) > 0.5 is an emergent numerical result, not an identity. The only self-references ([3], [17]) are background/motivational and are not load-bearing evidence for the resilience conclusion; the paper's own simulations establish the result. A separate concern—the printed signs in Eq. (1) appear not to reproduce the stated average of 20 local detections—is a correctness/reproducibility issue, not a circularity, because that detection curve is an input assumption rather than a prediction derived from the conclusion. The overlap regime is a conditioning assumption of the model, and an assumption is not circular unless the claimed result is defined into the metric; here it is not.

Assumptions & free parameters 6 free parameters · 5 assumptions · 0 invented entities

The central claim is generated by, and in part defined by, the scenario's free parameters and domain assumptions. The most important are the overlap-producing choices (object density, perception range, relevance range, number of relevant objects) and the binary beta model of relevance error. No new physical entities are introduced; 'content-selection error', 'ARR', and 'recovery probability' are metrics rather than entities.

free parameters (6)
  • Detection logistic coefficients c1=0.08, c2=-0.08, c3=-60 = c1=0.08, c2=-0.08, c3=-60
    Chosen so the perception range is 150 m; no calibration against sensor data is shown. Eq. (1), Sec. V.
  • Exogenous variable density = 750 vars/km^2
    Set to produce an average of 20 locally detected objects, citing highway/suburban references. This directly controls the amount of multi-vehicle overlap and hence resilience.
  • Relevance range D_max and number of relevant variables M = D_max=400 m, M=40
    D_max is tied to ETSI V2X requirements; M=40 is assumed. M matters because a larger M with fixed detection range reduces the fraction of relevant objects seen by multiple vehicles.
  • Relevance estimation error probability beta = 0, 0.2, 0.4, 0.6, 0.8 (swept)
    Beta is varied rather than fitted, but the boundary between 'resilient' and 'compromised' (around beta=0.4) is an output shaped by all other scenario choices.
  • Soft redundancy probability p_red = link-level message reception probability; hard p_red=1
    Assigned from the C-V2X analytical model; the hard/soft comparison is central to the paper's protocol conclusion.
  • Transmission period, object size, channel configuration = T=100 ms, 52 bytes, 10 MHz, QPSK-0.7
    Taken from cooperative perception standards and prior studies; reasonable but nevertheless chosen values that affect the load and congestion-control behavior.
assumptions (5)
  • domain assumption Exogenous variables are distributed as a homogeneous Poisson Point Process.
    Sec. IV. The spatial distribution and density drive how often multiple vehicles detect the same relevant object, which is the core resilience mechanism.
  • domain assumption Vehicles include only locally collected variables in transmitted messages.
    Sec. III/IV. This prevents multi-hop forwarding of detected objects and makes compensation depend on other vehicles' own local detections.
  • domain assumption C-V2X sidelink reception and sensing are modeled by [23] and congestion control by [24].
    Sec. V. All P(CSE), P(drop), and recovery curves inherit the assumptions of these external models (half-duplex, interference, shadowing, CBR).
  • ad hoc to paper Relevance is modeled as a weight w in [0,1] and relevance estimation error beta flips relevance independently with probability beta.
    Sec. IV. This binary i.i.d. error model is a simplification; real relevance estimation errors are likely correlated with context, distance, and object dynamics, which could change recovery behavior.
  • domain assumption An intended receiver is defined as a vehicle that correctly decoded at least one of the transmitter's last two messages.
    Sec. IV. This definition affects which receivers the transmitter considers when filtering content, and therefore influences CSE prevalence.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Resilience of Task-Oriented V2X Networks to Incomplete Information Sharing." pith.science (2026). https://pith.science/paper/CX3ITLLU

@misc{pith2026260218620,
  author       = {Pith},
  title        = {Pith review of: Resilience of Task-Oriented V2X Networks to Incomplete Information Sharing},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CX3ITLLU}},
  note         = {Machine review of arXiv:2602.18620}
}
read the original abstract

Task-oriented Vehicle-to-Everything (V2X) networks have recently been proposed to scalably support the large-scale deployment of connected vehicles. In task-oriented V2X networks, vehicles select the content of the transmitted messages based on its relevance to the intended receivers. However, estimating relevance can be challenging, especially in highly dynamic and complex driving scenarios. Relevance estimation errors may cause a transmitting vehicle to share incomplete information, omitting relevant data that is critical for the intended receivers' situational awareness. This work numerically demonstrates that task-oriented V2X networks exhibit an inherent resilience to incomplete information sharing. We show that such resilience guarantees a consistent delivery of relevant information even under high relevance estimation error probability conditions. Furthermore, we show that the fundamental conditions underpinning such inherent resilience can also be encountered outside of the V2X domain - in particular, in other task-oriented networks where multiple transmitters select the content of their messages based on the task-related requirements of a common set of intended receivers.

Figures

Figures reproduced from arXiv: 2602.18620 by the authors.

Figure 2
Figure 2. depicts a cooperative perception example to analyse and illustrate the conditions that can lead to a Redundancy-CSE [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 1
Figure 1. ). The relevance of non-redundant exogenous variables for an intended receiver 𝑉𝑉𝑅𝑅 is modelled through a contextual relevance function 𝑓𝑓𝑅𝑅(𝒦𝒦) that assigns a non-uniform weight 𝑤𝑤𝑅𝑅,𝑘𝑘 ∈ [0,1] to each non-redundant exogenous variable 𝑥𝑥𝑘𝑘. The weight 𝑤𝑤𝑅𝑅,𝑘𝑘 is a measure of the context-dependent relevance of each 𝑥𝑥𝑘𝑘 for 𝑉𝑉𝑅𝑅. Relevant variables have a weight larger than zero (𝑤𝑤𝑅𝑅,𝑘𝑘 > 0), whereas irrelevant var… view at source ↗

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Semantic and Task-Oriented V2X Communications: Pushing the Limits of V2X Networks Scalability

    cs.NI 2026-06 unverdicted novelty 5.0 of 10

    Semantic and task-oriented V2X communications increase supported vehicles by 4.1x, reduce inter-reception time by 67%, and double successful relevant information delivery in high-density simulations.

Reference graph

Works this paper leans on

24 extracted references · 2 linked inside Pith · cited by 1 Pith paper

  1. [3]

    The Search for Relevance: A Context -Aware Paradigm Shift in Semantic and Task -Oriented V2X Communications,

    L. Lusvarghi et al. , “The Search for Relevance: A Context -Aware Paradigm Shift in Semantic and Task -Oriented V2X Communications,” arXiv, 2025, arXiv:2508.07394

    arXiv 2025

  2. [17]

    Redundancy Estimation in Semantic and Task-Oriented V2X Communications: Role and Challenges,

    L. Lusvarghi and J. Gozalvez, “Redundancy Estimation in Semantic and Task-Oriented V2X Communications: Role and Challenges,” 2025 IEEE/CIC ICCC, Shanghai, China, 2025, pp. 1-6

    2025

  3. [1]

    LTE-V2X Scalability and Spectrum Requirements to support Multiple V2X Services

    M. Sepulcre et al., “LTE-V2X Scalability and Spectrum Requirements to support Multiple V2X Services”, in Proc. VTC2024 -Fall, Washington DC, USA, 2024, pp. 1-6

    2024

  4. [2]

    Beyond Transmitting Bits: Context, Semantics, and Task-Oriented Communications,

    D. Gündüz et al., “Beyond Transmitting Bits: Context, Semantics, and Task-Oriented Communications,” IEEE Journal on Selected Areas in Communications, vol. 41, no. 1, pp. 5-41, Jan. 2023

    2023

  5. [4]

    Collective Perception Service; Release 2,

    ETSI, “Collective Perception Service; Release 2,” ETSI TS 103 324 V2.1.1, June 2023

    2023

  6. [5]

    Resource Allocation and Common Message Selection for Task -Oriented Semantic Information Transmission With RSMA,

    Y. Cheng et al. , “Resource Allocation and Common Message Selection for Task -Oriented Semantic Information Transmission With RSMA,” IEEE Trans. on Wireless Comm., vol. 23, no.6, pp. 5557-5570, June 2024

    2024

  7. [6]

    Task -Oriented Com munication for Multidevice Cooperative Edge Inference,

    J. Shao et al. , “Task -Oriented Com munication for Multidevice Cooperative Edge Inference,” IEEE Trans. on Wireless Communications, vol. 22, no. 1, pp. 73-87, Jan. 2023

    2023

  8. [7]

    Task -Oriented Sensing, Computation, and Communication Integration for Multi-Device Edge AI,

    D. Wen et al. , “Task -Oriented Sensing, Computation, and Communication Integration for Multi-Device Edge AI,” IEEE Trans. o n Wireless Communications, vol. 23, no. 3, pp. 2486-2502, March 2024

    2024

Show all 24 references
  1. [8]

    Who2com: Collaborative Perception via Learnable Handshake Communication,

    Y. -C. Liu et al. , “Who2com: Collaborative Perception via Learnable Handshake Communication,” in Proc. IEEE ICRA 2020, Paris, France, 2020, pp. 6876-6883,

    2020

  2. [9]

    When2com: M ulti-Agent Perception via Communication Graph Grouping,

    Y. -C. Liu et al. , “When2com: M ulti-Agent Perception via Communication Graph Grouping,” in Proc. IEEE CVPR 2020, Seattle, WA, USA, 2020, pp. 4105-4114

    2020

  3. [10]

    Task -Oriented Wireless Communications for Collaborative Perception in Intelligent Unmanned Systems,

    S. Zhou et al. , “Task -Oriented Wireless Communications for Collaborative Perception in Intelligent Unmanned Systems,” IEEE Network, vol. 38, no. 6, pp. 21-28, Nov. 2024

    2024

  4. [11]

    Where2comm: Communication -Efficient Collaborative Perception via Spatial Confidence Maps,

    Y. Hu et al. , “Where2comm: Communication -Efficient Collaborative Perception via Spatial Confidence Maps,” in Proc. NeurIPS 2022, New Orleans, Louisiana, USA, 2022, pp. 4874-4886

    2022

  5. [12]

    Which2comm: An Efficient Collaborative Perception Framework for 3D Object,

    D. Yu et al. , “Which2comm: An Efficient Collaborative Perception Framework for 3D Object,” arXiv preprint, 2025, arXiv:2503.17175

    2025 arXiv

  6. [13]

    Value- Anticipating V2V Communications for Cooperative Perception,

    T. Higuchi et al. , “Value- Anticipating V2V Communications for Cooperative Perception,” in Proc. 2019 IEEE IV Symp. , Paris, France, 2019, pp. 1947-1952

    2019

  7. [14]

    Redundancy Mitigation in Cooperative Perception for Connected and Automated Vehicles,

    G. Thandavarayan et al. , “Redundancy Mitigation in Cooperative Perception for Connected and Automated Vehicles,” in Proc. IEEE VTC2020-Spring, Antwerp, Belgium, 2020, pp. 1-5

    2020

  8. [15]

    Uncertainty and Prioritization: Empirical Evaluation of a VoI-Based CPM Generation Pipeline Using Real-World Data,

    V. A. Wolff et al., “Uncertainty and Prioritization: Empirical Evaluation of a VoI-Based CPM Generation Pipeline Using Real-World Data,” Proc. 2025 IEEE VNC Conf., Porto, Portugal, 2025, pp. 1-8

    2025

  9. [16]

    Accuracy and Relevance: A Value of Information based Prioritisation of Perceived Objects for the ETSI Collective Perception Service,

    T. Lyu et al. , “Accuracy and Relevance: A Value of Information based Prioritisation of Perceived Objects for the ETSI Collective Perception Service,” 2025 IEEE VNC Conf., Porto, Portugal, 2025, pp. 1-8

    2025

  10. [18]

    V2X Sensor-Sharing for Cooperative and Automated Driving,

    SAE International, “V2X Sensor-Sharing for Cooperative and Automated Driving,” SAE J3224, Aug. 2022

    2022

  11. [19]

    Object Detection Probability for Highly Automated Vehicles: An Analytical Sensor Model,

    F. Schiegg, I. Llatser and T. Michalke, “Object Detection Probability for Highly Automated Vehicles: An Analytical Sensor Model,” in Proc. 5th VEHITS Conf., Heraklion, Crete, Greece, 2019, pp. 223-231, 2019

    2019

  12. [20]

    nuScenes: A multimodal dataset for autonomous driving,

    H. Caesar et al. , “nuScenes: A multimodal dataset for autonomous driving,” in Proc. IEEE CVPR 2020, June 2020, pp. 11618-11628

    2020

  13. [21]

    5G; Service requirements for enhanced V2X scenarios,

    ETSI, “5G; Service requirements for enhanced V2X scenarios,” ETSI TS 122 186 V17.0.0, April 2024

    2024

  14. [22]

    A First Study on the Spectrum Needs for Release 2 V2X Services,

    E. Xhoxhi et al., “A First Study on the Spectrum Needs for Release 2 V2X Services,” in Proc. VTC2023-Fall, Hong Kong, HK, 2023, pp. 1-6

    2023

  15. [23]

    Analytical Models of the Performance of C- V2X Mode 4 Vehicular Communications,

    M. Gonzalez-Martín et al., “Analytical Models of the Performance of C- V2X Mode 4 Vehicular Communications,” IEEE Transactions on Vehicular Technology, vol. 68, no. 2, pp. 1155-1166, Feb. 2019

    2019

  16. [24]

    Congestion Control Mechanisms for the C -V2X PC5 interface; Access layer part,

    ETSI, “Congestion Control Mechanisms for the C -V2X PC5 interface; Access layer part,” ETSI TS 103 574, V1.1.1, Nov. 2018

    2018

Pith tools

Reviewed August 2, 2026 · model on record in the stance chip above.