Source-linked AI summary
Multiple Access in Cellular V2X: Performance Analysis in Highly Congested Vehicular Networks
Behrad Toghi, Md Saifuddin, Hossein Nourkhiz Mahjoub, M. O. Mughal, Yaser P. Fallah, Jayanthi Rao, Sushanta Das
TL;DR
High-density C-V2X resource allocation can affect latency and packet delivery, motivating analysis of its configurable parameters and performance-degrading phenomena. The paper studies the allocation procedure with a release-14-compliant simulator; results show reliability degrades when vehicle density exceeds resource capacity, while resource-selection behavior also affects IPG and PER.
Problem
The paper examines how configurable C-V2X resource allocation affects latency and packet delivery in high-density vehicular networks.
Source: §I · p. 1Method
The authors use a release-14-compliant event-based ns-3 simulator to investigate PHY and MAC resource-allocation mechanisms and performance-degrading phenomena.
Source: §I · p. 2, B. C-V2X Resource Allocation · p. 3Results
Above the network saturation point, insufficient radio resources prevent the network from maintaining the required packet delivery rate and PER increases significantly.
Source: §IV · p. 6Takeaways & Limitations
The authors identify limiting transmit range or slowing beacon generation as possible responses to resource shortages, although 3GPP has not specifically mandated such congestion control.
Source: §IV · p. 6
Abstract
from arXiv · showhide
Vehicle-to-everything (V2X) communication enables vehicles, roadside vulnerable users, and infrastructure facilities to communicate in an ad-hoc fashion. Cellular V2X (C-V2X), which was introduced in the 3rd generation partnership project (3GPP) release 14 standard, has recently received significant attention due to its perceived ability to address the scalability and reliability requirements of vehicular safety applications. In this paper, we provide a comprehensive study of the resource allocation of the C-V2X multiple access mechanism for high-density vehicular networks, as it can strongly impact the key performance indicators such as latency and packet delivery rate. Phenomena that can affect the communication performance are investigated and a detailed analysis of the cases that can cause possible performance degradation or system limitations, is provided. The results indicate that a unified system configuration may be necessary for all vehicles, as it is mandated for IEEE 802.11p, in order to obtain the optimum performance. In the end, we show the inter-dependence of different parameters on the resource allocation procedure with the aid of our high fidelity simulator.
I. INTRODUCTION
The introduction frames C-V2X as a release-14 alternative for vehicular safety communication and focuses on mode-4 resource allocation in high-density networks. The paper evaluates configurable parameters and performance-degrading mechanisms using PER and IPG in a standards-compliant simulator.
I. INTRODUCTION
- DSRC based on IEEE 802.11p was described as the primary vehicular communication solution, while automakers, suppliers, and regulators showed interest in LTE as an alternative.The cited applications include sharing position and mobility information through base safety messages or model-based communication.Source: §I · p. 1
- LTE D2D sidelink lets UEs communicate peer-to-peer without the central base station, but does not meet strict vehicular safety reliability and latency requirements.This motivates the release-14 V2X modes described in the introduction.Source: §I · p. 1
- Release 14 specifies modes 3 and 4 for V2X, with enhancements to collision avoidance, channel access, sub-channelization, HARQ, and DMRS for congestion and high-speed mobility.Mode 3 uses eNodeB scheduling in coverage, whereas mode 4 supports distributed, unsupervised allocation out of coverage or in partial coverage.Source: §I · p. 1
- C-V2X research was sparse, with prior work covering system-level errors, tutorials, C-V2X versus DSRC, and effects of reservation periodicity and resource availability.The introduction identifies PER and IPG analysis as absent from the previously mentioned works.Source: §I · pp. 1–2
- The study assesses configurable C-V2X resource-allocation parameters in high-density vehicular scenarios using PER and IPG, with an event-based ns-3 simulator compliant with release 14.The analysis examines PHY and MAC mechanisms and phenomena that can cause performance degradation or deadlock loops.Source: §I · pp. 1–2
II. CELLULAR VEHICLE-TO-EVERYTHING COMMUNICATION
The section situates C-V2X in the ITS spectrum and identifies resource allocation and channel access as its focus.
II. CELLULAR VEHICLE-TO-EVERYTHING COMMUNICATION
- In 1999, the FCC specified 75 MHz in the 5.9 GHz band for ITS and vehicle-safety applications.C-V2X is expected to operate in this band and possibly coexist with DSRC.Source: §II · p. 2
- The section examines C-V2X communication with a focus on resource allocation and channel access mechanisms.Source: §II · p. 2
- C-V2X is expected to operate in the ITS band and possibly coexist with DSRC.Source: §II · p. 2
A. Principles of the Sidelink Communication
The section describes C-V2X sidelink interfaces and the LTE frame and resource structure used for communication. It explains how control information accompanies data transmissions and how resources are organized for allocation.
A. Principles of the Sidelink Communication
- PC5 supports one-to-many broadcasts among neighboring UEs; mode 4 carries both data and control over PC5, while mode 3 uses PC5 for data and Uu for control exchange with the base station.The text notes that LTE sidelink uses four communication channels, listed in Table I.Source: A. Principles of the Sidelink… · p. 2, Table I · p. 2
- A 10 MHz frame contains 50 frequency RBs and 20 time RBs; each RB spans 12 subcarriers and one slot, and two time-consecutive RBs form a scheduling block.Frames contain 10 subframes, each 1 ms wide, with two slots; each slot has seven SC-FDMA symbols under normal cyclic prefix.Source: A. Principles of the Sidelink… · p. 2
- Release 14 adds two DMRS symbols to the two LTE D2D pilot symbols; four DMRSs occupy symbols three, six, nine, and twelve.The first symbol is reserved for AGC and the last for guard band; the paper assumes nine SC-FDMA symbols per subcarrier are available for data.Source: A. Principles of the Sidelink… · p. 2
- A UE sends BSM data as a transport block on PSSCH and sends its associated SCI on PSCCH; both must occupy the same subframe.SCI occupies two contiguous RBs per slot, and the TB and SCI may be adjacent or non-adjacent.Source: A. Principles of the Sidelink… · p. 2
- The PSSCH resource pool comprises subchannels, while each PSCCH resource uses two contiguous PRBs; the standard allows adjacent or separated TB and SCI placement.For a 10 MHz channel with a 1 MHz guard band, the text gives five subchannels of size 10 as an example.Source: A. Principles of the Sidelink… · p. 2
- TB capacity per PRB depends on modulation order and code rate; C-V2X supports QPSK and 16-QAM, with higher modulation orders unsupported by the current standard.Their combination is called the modulation and coding scheme (MCS) index.Source: A. Principles of the Sidelink… · p. 3
B. C-V2X Resource Allocation
Mode-4 uses sensing-based semi-persistent scheduling (SB-SPS) to narrow resource choices, reserve periodic transmission opportunities, and transmit data and control information.
B. C-V2X Resource Allocation
- SB-SPS combines channel sensing, semi-persistent MAC reservation, and PHY transmission of data and control information.The sensing record narrows the available resources, while randomness is used in resource allocation.Source: B. C-V2X Resource Allocation · p. 3
- The sensing window uses the prior 1 second of candidate single-subframe resources, while the report window covers future resources subject to application latency and processing-time settings.The sensing window ends 100 ms before packet arrival; the report window starts at an offset of up to four subframes.Source: B. C-V2X Resource Allocation · p. 3
- PHY excludes likely-to-collide CSRs from the report-window set if a sensing-window subframe is unmonitored and reservation horizons overlap, or a received SCI/TB has PSSCH-RSRP ≥ the preset threshold and horizons overlap.Source: B. C-V2X Resource Allocation · p. 3
- PHY ranks remaining candidates by linear-average sidelink RSSI and reports the 20% with the lowest metric values to MAC for reservation.The metric is identified as D in the text.Source: B. C-V2X Resource Allocation · p. 4
- MAC reselects when a random counter expires or transmission, missed-opportunity, latency, or resource-sufficiency triggers occur, then reserves periodic opportunities from the reported set.The reservation interval is set by higher layers; the procedure also includes a probability of retaining previously selected resources when the counter reaches zero.Source: B. C-V2X Resource Allocation · p. 4, Figure 2 · p. 4
- When two redundant versions are configured and resources permit, MAC allocates a second periodic opportunity set within 15 subframes of the initially selected resources.PHY uses reserved opportunities to send the data transport block and its corresponding sidelink control information.Source: B. C-V2X Resource Allocation · p. 4
III. SYSTEM CONFIGURATION
The study uses an event-based link-level simulator in ns-3, with the LENA module supporting LTE functionalities.
III. SYSTEM CONFIGURATION
- The authors implemented a link-level event-based simulator to evaluate vehicular scenarios involving large numbers of UEs.The simulator uses the ns-3 environment.Source: §III · p. 4
- The simulation uses the LENA ns-3 module to support the main LTE functionalities.The module was developed by Baldo et al..Source: §III · p. 4
- The simulator implementation is intended for system-level performance evaluation in large-UE vehicular scenarios.The passage describes it as a link-level event-based simulator.Source: §III · p. 4
A. Air Interface
The air-interface model combines a field-test-derived propagation channel with separate models for large- and small-scale effects.
A. Air Interface
- The propagation model was derived from RSSI field-test data collected at the FTT-A Fowlerville Proving Ground and includes large- and small-scale effects.The field test was conducted by the CAMP consortium in collaboration with USDOT.Source: A. Air Interface · p. 4
- Large-scale attenuation is modeled using a two-ray model whose path-loss exponent varies across transmitter–receiver distance ranges.The passage states that the derived exponent values are listed in Table II.Source: A. Air Interface · p. 4
- Small-scale propagation uses a Nakagami model up to 639 m and a Weibull distribution with characteristic coefficient 1.4 beyond that distance.The Nakagami parameter values are given in Table II.Source: A. Air Interface · p. 4
B. Transceiver Model
The transceiver model specifies device assumptions and maps received SINR to packet decoding outcomes using a parameter-dependent receiver model.
B. Transceiver Model
- Each device is assumed to have a 9 dB noise figure, two receive and one transmit omnidirectional antennas, 3 dBi antenna gain, and 1.5 m height.The stated height represents a typical sedan car.Source: B. Transceiver Model · pp. 4–5
- The receiver model maps each packet’s SINR to TBLER and compares the error rate with a uniform random variable to determine reception failure.A packet is considered failed when its BLER exceeds the random variable.Source: B. Transceiver Model · p. 5
- The simulation uses Huawei’s receiver model, selected for its fit to the assumptions, and notes that separate models are needed for different parameter sets.The listed factors include retransmission count, MCS, message size, and vehicle speed.Source: B. Transceiver Model · p. 5
C. Application Layer
The simulations model a moving platoon on a four-lane, 2 km highway and analyze digested BSM packets under fixed transmission settings.
C. Application Layer
- Vehicles were uniformly arranged across four lanes on a 2 km straight highway, with results taken from the middle 1,000 m to avoid edge effects.The modeled network was fully connected, and transceivers operated in half-duplex mode.Source: C. Application Layer · p. 5
- Each vehicle was unable to receive data while transmitting because the transceivers operated in half-duplex mode.The setup also assumed a fully connected network.Source: C. Application Layer · p. 5
- Vehicles transmitted BSMs containing 190 bytes of digested data, using fixed power and message rate.The study used adjacent PSSCH and PSCCH sub-channelization and set transmit power to 23 dBm.Source: C. Application Layer · p. 5
IV. ANALYSIS AND RESULTS
The analysis evaluates C-V2X resource allocation under varying congestion and parameter settings using packet error rate and inter-packet gap. It finds that congestion, resource-reselection behavior, and packet-generation timing shape reliability and latency outcomes.
MCS and effective code rate
- Using the MCS-index mapping to modulation order and transport-block size, the study calculates effective code rates for safety-beacon transmissions; under QPSK, the maximum achievable code rate is approximately 0.8.Figure 3 shows results for 190-byte packets (top) and 300-byte packets (bottom). The standard does not support modulation orders higher than 16-QAM.Source: §IV · p. 5, Figure 3 · p. 5
Congestion performance
- In scenario #1, the network maintains 90% reliability across almost the whole communication range, while performance degrades in more congested scenarios.PER is evaluated over transmitter-receiver distance in four scenarios.Source: §IV · p. 6, Figure 4 · p. 6
- Above the network saturation point, insufficient radio resources prevent the required packet delivery rate and significantly increase PER.The paper identifies reducing transmit range through lower power or slowing beacon generation as possible responses.Source: §IV · p. 6, Figure 4 · p. 6
- Higher vehicle density spreads IPG distributions and can produce gaps exceeding 1 s between consecutive packet receptions.Below saturation, the network almost maintains the approximately 100 ms IPG; wider distributions in scenarios 3 and 4 indicate greater unpredictability.Source: §IV · p. 6, Figure 5 · p. 6
SB-SPS parameter effects
- The SB-SPS exclusion threshold and its 20% candidate-resource rule can each constrain resource selection, depending on network density.At low density the 20% rule plays the main role; under high congestion, the threshold is raised until the 20% condition is met.Source: §IV · p. 6
Scope and limitations
- The parameter analysis assumes a fully connected, low-dynamics network, unlike real-world conditions that may include hidden nodes, near-far effects, and high dynamics.The sensing threshold defines the maximum range a vehicle should sense.Source: §IV · p. 6
SB-SPS parameter effects
- A higher probability of keeping previously selected resources is associated with a more predictable, stable network under medium congestion.At higher probabilities of selecting new resources, IPG distributions spread and their mean and median increase; the study cautions that tuning may be less straightforward in highly congested networks.Source: §IV · pp. 6–7, Figure 6 · p. 7, Figure 7 · p. 7
Packet-generation synchronization
- Limiting packet-generation offsets to 49 ms yields lower PER than synchronized generation or offsets up to 99 ms.The paper explains that greater overlap between vehicles’ candidate resource sets increases collision probability, while time offsets reduce that overlap.Source: §IV · p. 7
V. CONCLUDING REMARKS
The conclusion argues that C-V2X reliability in congested scenarios depends on parameter tuning and a unified configuration across vehicles. It identifies application-layer standards, congestion control, and further parameter study as priorities.
V. CONCLUDING REMARKS
- A unified configuration for all vehicles is needed to ensure communication reliability, with C-V2X parameters regulated through application-layer standards.The paper compares this approach with the standardization defined for DSRC in SAE J2945/1.Source: §V · p. 7
- Geo-zoning limits a UE’s resource pool according to location and can mitigate packet collisions, especially the near-far problem.Source: §V · p. 7
- Future work will examine congestion control and additional configurable parameters, including MCS index, report window size, and reservation interval.The stated aim is to assist rule-makers in specifying mandatory C-V2X system requirements.Source: §V · p. 8