Performance Analysis of Sensing-Based Semi-Persistent Scheduling in C-V2X Networks

Abstract

The 3 rd Generation Partnership Project released the cellular vehicular-to-everything (C-V2X) specifications as part of the LTE framework in Release 14. C-V2X is the alternative to dedicated short range communications and both are specifically designed for V2X control signaling. C-V2X extends the device-to-device specifications by adding two more modes of operation targeting the vehicular environment in coverage and out of coverage of LTE base stations. Vehicle-to-vehicle communications (V2V) is established with Mode 4, where the devices schedule their transmissions in a distributed way employing sensing-based semi-persistent scheduling (SPS). Research is needed to assess the performance of SPS, especially in congested radio environments. This paper presents the first open-source C-V2X simulator that enables such research. The simulator is implemented in ns-3. We analyze the effect of the Mode 4 resource pool configuration and some of the key SPS parameters on the scheduling performance and find that the resource reservation interval significantly influences packet data rate performance, whereas resource reselection probability has little effect in dense vehicular highway scenarios. Our results show that proper configuration of scheduling parameters can significantly improve performance. We conclude that research on congestion control mechanisms is needed to further enhance the SPS performance for many practical use cases.

Multiple Access in Cellular V2X: Performance Analysis in Highly Congested Vehicular Networks

“…In [9], after an introductory overview of the applications and use cases of the V2X communication, a link level performance comparison of C-V2X and DSRC is presented which demonstrates that C-V2X out-performs DSRC in most of the vehicular scenarios, e.g., highway and urban scenarios. Recent results in [10], study how the resource reservation periodicity and the number of available radio resources can affect the performance of the C-V2X communication. Results demonstrate that the communication reliability can be improved utilizing more radio resources.…”

“…We demonstrate that an optimum configuration can heavily affect the system performance. This perspective is in contrast with that of [7]- [10], where simplistic assumptions are made for the tunable parameters and the baseline performance in medium or low-density scenarios is under consideration. Moreover, the packet error rate (PER) and inter-packet gap (IPG) metrics are chosen to assess the system performance in terms of both network reliability and latency; such analysis is absent in the previously mentioned works.…”

Reducing Message Collisions in Sensing-Based Semi-Persistent Scheduling (SPS) by Using Reselection Lookaheads in Cellular V2X

“…To facilitate analysis, we apply some simplifying assumptions. First, we set probResourceKeep = 0, namely, we do not keep the current resource upon reselection [35]. Second, we assume that the vehicle population is static, namely no churn in the membership.…”

“…The work noticed the differences in the BSM messages as one source of such poor performance by LTE-V. As a solution to this problem, the authors suggested that no subchannels are reserved for a streak when transmitting the larger and less frequent packets (e.g., 300 bytes). Nabil et al analyzed the effect of the Mode 4 resource pool configuration and some of the key SPS parameters on the scheduling performance and found that the resource reservation interval (RRI) significantly influences packet data rate performance, whereas the resource keeping probability has little effect in dense vehicular highway scenarios [35]. The reason is that when the congestion is severe, there remain only a limited amount of available resources.…”

“…We notice that one reason the current SPS algorithm produces significant packet collisions [33,35] is the uncertainty about the next resource location a UE will (re)select. In fact, there is no provision in the current specification of the sensing-based SPS with which a UE can convey the information about its selected location for the next streak in the time-frequency plane.…”