Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 45 of the 46 references it cites were found in a registry or an index; 5 of the 20 sources its claims cite were checked for support.

Reinforcement Learning-Based Resource Allocation Scheme of NR-V2X Sidelink for Joint Communication and Sensing

Open access
Li, Zihan, Wang, Ping, Shen, Yamin, Li, Song · 2025 · Sensors

Abstract

Joint communication and sensing (JCS) is becoming an important trend in 6G, owing to its efficient utilization of spectrums and hardware resources. Utilizing echoes of the same signal can achieve the object location sensing function, in addition to the V2X communication function. There is application potential for JCS systems in the fields of ADAS and unmanned autos. Currently, the NR-V2X sidelink has been standardized by 3GPP to support low-latency high-reliability direct communication. In order to combine the benefits of both direct communication and JCS, it is promising to extend existing NR-V2X sidelink communication toward sidelink JCS. However, conflicting performance requirements arise between radar sensing accuracy and communication reliability with the limited sidelink spectrum. In order to overcome the challenges in the distributed resource allocation of sidelink JCS with a full-duplex, this paper has proposed a novel consecutive-collision mitigation semi-persistent scheduling (CCM-SPS) scheme, including the collision detection and Q-learning training stages to suppress collision probabilities. Theoretical performance analyses on Cramér-Rao Lower Bounds (CRLBs) have been made for the sensing of sidelink JCS. Key performance metrics such as CRLB, PRR and UD have been evaluated. Simulation results show the superior performance of CCM-SPS compared to similar solutions, with promising application prospects.

Overview

CoChat Summary

TL;DR

The paper proposes CCM-SPS, a Q-learning-based sidelink resource allocation scheme for NR-V2X joint communication and sensing that uses full-duplex echo collision detection to mitigate consecutive resource collisions. Simulations show CCM-SPS improves sensing CRLB, communication PRR, and update delay over SB-SPS and an FD-enhanced benchmark across vehicle densities and packet sizes.

Key findings

  • CCM-SPS outperforms SB-SPS and FD-enhanced in CRLB and PRR across density scenarios; in high-density environments it maintains a CRLB accuracy of 0.1 m and a reliable communication range of 50 m.
  • Compared with FD-enhanced, CCM-SPS increases the maximum communication range where PRR exceeds 0.95 by 22.2%, 30%, and 50% in low-, mid-, and high-density scenarios, respectively.
  • Under traditional SB-SPS, increasing packet size improves sensing accuracy (smaller root CRLB range) but reduces communication reliability (lower PRR); increasing vehicle density degrades both sensing and communication performance.
  • CCM-SPS achieves higher PRR than SB-SPS, with PRR decreasing more slowly as packet size increases, and it also reduces update delay compared with SB-SPS.

Methods

  • System-level simulation of a three-lane bidirectional highway with vehicle densities of 50, 150, and 250 veh/km; average speed 70 km/h, speed STD 7 km/h, radar cross-section 10 dBsm, WINNER + B1 channel, 40 MHz bandwidth, 5.9 GHz center frequency, 23 dBm transmit power, 3 dBi antenna gain, and 6 dB noise gap.
  • Traffic uses periodic V2V messages every 100 ms with packet sizes of 350 or 1000 bytes; performance metrics are CRLB, PRR, and UD.
  • The proposed CCM-SPS combines full-duplex echo-power threshold collision detection with Q-learning: state is (COL, RC), action is ΔRC in [0, 15], reward is 1 for no collision and (exp(−N_col)−1)/log10(RC+1) for collision; Q-learning uses ε-greedy with ε from 0.9 to 0.1, 1000 iterations over 100 s, learning rate 0.01, and discount factor 0.9.

Limitations

  • Stated: future work includes practical full-duplex impacts from interference and cross-layer optimization, as well as extending CCM-SPS to other JCS applications such as AIOT and integrating edge computing.
  • Stated assumptions: all vehicles are initially assumed to have in-band full-duplex transceivers with perfect SIC, and only the closest signal to the transmitting vehicle is considered for echo power calculation.
  • Evident from the text: evaluation is simulation-only, uses fixed SCS and MCS sizes, and relies on a simplified three-lane highway scenario; no real-world testbed or field validation is reported.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.3390/s25020302

Cites

45 of 45 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Figure 1 NG-RAN architecture supporting the PC5 interface.
Figure 1 NG-RAN architecture supporting the PC5 interface.CC BY · PubMed Central
Figure 2 Process flow of sensing-based semi-persistent scheduling (SB-SPS).
Figure 2 Process flow of sensing-based semi-persistent scheduling (SB-SPS).CC BY · PubMed Central
Figure 3 Markov chain for state transition of SPS.
Figure 3 Markov chain for state transition of SPS.CC BY · PubMed Central
Figure 4 Reinforcement learning framework.
Figure 4 Reinforcement learning framework.CC BY · PubMed Central
Figure 5 CCM-SPS accelerate reselection.
Figure 5 CCM-SPS accelerate reselection.CC BY · PubMed Central
Figure 6 Empirical CDF of the root CRLB for a range using SB-SPS, FD-enhanced and CCM-SPS.
Figure 6 Empirical CDF of the root CRLB for a range using SB-SPS, FD-enhanced and CCM-SPS.CC BY · PubMed Central
Figure 7 Bar graph of root CRLB (at CCDF = 95-percentile) for a range using different schemes with varying vehicle densities.
Figure 7 Bar graph of root CRLB (at CCDF = 95-percentile) for a range using different schemes with varying vehicle densities.CC BY · PubMed Central
Figure 8 PRR over distance using SB-SPS, FD-enhanced and CCM-SPS.
Figure 8 PRR over distance using SB-SPS, FD-enhanced and CCM-SPS.CC BY · PubMed Central
Figure 9 The maximum distance allowing PRR larger than 0.95 is evaluated using conventional SB-SPS, FD-enhanced methods and CCM-SPS.
Figure 9 The maximum distance allowing PRR larger than 0.95 is evaluated using conventional SB-SPS, FD-enhanced methods and CCM-SPS.CC BY · PubMed Central
Figure 10 Empirical CDF of root CRLB for range estimation.
Figure 10 Empirical CDF of root CRLB for range estimation.CC BY · PubMed Central
Figure 11 PRR vs. distance performance of SB-SPS with different pack sizes in case of density = 50, 150, 250 veh/km.
Figure 11 PRR vs. distance performance of SB-SPS with different pack sizes in case of density = 50, 150, 250 veh/km.CC BY · PubMed Central
Figure 12 CCM-SPS’s range sensing performance evaluation on empirical CDF of root CRLB with different pack sizes in the case of density = 50, 150, 250 veh/km.
Figure 12 CCM-SPS’s range sensing performance evaluation on empirical CDF of root CRLB with different pack sizes in the case of density = 50, 150, 250 veh/km.CC BY · PubMed Central
Figure 13 PRR vs. distance performance of CCM-SPS with different pack sizes in case of density = 50, 150, 250 veh/km.
Figure 13 PRR vs. distance performance of CCM-SPS with different pack sizes in case of density = 50, 150, 250 veh/km.CC BY · PubMed Central
Figure 14 CCM-SPS’s communication performance evaluation on empirical CDF of update delay with different pack sizes in the case of density = 50, 150, 250 veh/km.
Figure 14 CCM-SPS’s communication performance evaluation on empirical CDF of update delay with different pack sizes in the case of density = 50, 150, 250 veh/km.CC BY · PubMed Central

Verification

Claims the paper makes

5 Valid · 2 Not valid · 11 Not checked · 56 Own results and interpretations

23 more claims

Create a free account to see all 74 claims and to verify the 4 references still open.

See all claims — free account
  • Probably not a substantial claim. 14
  • Borderline: one of the tests for a substantial claim is unclear. 1
  • It may state two separate claims as one. 33
  • The quote only partly states this claim. 2

Create a free account to see these claims.

Citations

Cites

Show all 46 references

Research works printed but not linked: 1

  1. Yan, 2021, Integration of communication, sensing and computing: The vision and key technologies of 6G, J. Beijing Univ. Posts Telecommun., 44, 1

Cited by

See all 15

Metadata

Authors
Li, Z., Wang, P., Shen, Y., & Li, S.
Affiliations
Donghua University (CN)
Corresponding author
Wang, Ping
Venue
Sensors
Publisher
MDPI AG
Published
2025-01-07
Volume
25
Issue
2
Pages
302
ISSN
1424-8220 (ISSN-L)
OpenAlex
W4406126183
Cited by
15
References
46
Citations by year
2026: 12 · 2025: 3
FWCI
3.35
Type
journal article
Language
en
Open access
Locations
  • Sensors (publishedVersion, cc-by)
  • Preprints.org (acceptedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Topics
  • Physical Sciences › Engineering › Electrical and Electronic Engineering › Vehicular Ad Hoc Networks (VANETs)
  • Physical Sciences › Engineering › Aerospace Engineering › Radar Systems and Signal Processing
  • Physical Sciences › Engineering › Electrical and Electronic Engineering › Advanced Wireless Communication Technologies
Keywords
integrated sensing and communication, resource allocation, reinforcement learning, Q-learning, collision mitigation, semi-persistent scheduling, full-duplex communication, Cramer-Rao lower bound, communication reliability, 6G, vehicle-to-vehicle communication

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Reinforcement Learning-Based Resource Allocation Scheme of NR-V2X Sidelink for Joint Communication and Sensing. https://app.cochat.ai/papers/10.3390/s25020302
The paper
Li, Z., Wang, P., Shen, Y., & Li, S. (2025). Reinforcement Learning-Based Resource Allocation Scheme of NR-V2X Sidelink for Joint Communication and Sensing. Sensors. https://doi.org/10.3390/s25020302

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.