Overview and comparison of V2X technologies

Summary

V2X is the common name for all types of communications between vehicles. Consider and compare the available specifications, mathematical models, see what commercial solutions exist in this area and how to buy them in Russia. Let's start with a short review of the finished and emerging V2X standards (IEEE 802.11p, IEEE 802.11bd, 3GPP LTE-V2X, 3GPP 5G NR-V2X and even a little 3GPP 6G NR-V2X). The second part is a translation of the comparison of the reliability of the mathematical models IEEE 802.11bd and 3GPP 5G NR-V2X. The third part is an overview of commercial products for V2X, processors and OBU.

Introduction

Glossary

V2X - vehicle to everything communications

C-V2X - cellular V2X

DSRC - Dedicated Short-Range Communications. This is WIFI-based V2X

LOS - line of sight - line of sight

NLOS - non line of sight - out of line of sight

ITS - Intelligent Transport Systems

Midamble - used in computer networks to separate the header message from the data, can be a character or word

Use cases

We will evaluate V2X technologies for active safety and autonomous driving scenarios, which makes a short response time in any environmental conditions the main quality criterion.



The main scenarios:

• LOS on a country road
• LOS on the motorway
• LOS in the city
• in conditions of heavy traffic at the intersection in the city
• NLOS on the motorway
• NLOS in the city
• Contactless toll payment
• Crossroads Regulation

Motivation

I am an engineer with 10 years of experience in IT. Of these, 3 (until 2019), I was testing in a automotive startup. My working draft was not related to V2X. Interest in the topic came after the Automotive Testing Expo 2019, in which I discovered the market for V2X devices and applications. According to Bloomberg estimates, by 2022 this market will be valued at 1.2 billion USD.

Part 1. Lyric Overview of V2X Specifications

DSRC

Existing DSRC (Dedicated Short-Range Communications) products are designed using the IEEE 802.11p standard.



This is the first V2X standard released in 2010 and is based on the IEEE 802.11a WLAN. The 802.11p edition, compared to 802.11a, introduces changes to the OSI PHY and MAC levels to improve the performance of the classic WLAN for communicating fast-moving (up to 250 km / h) vehicles. The next version of the IEEE 802.11bd WLAN, based on the IEEE 802.11ac WLAN and backward compatible with 802.11p, is at an early stage. It is planned to improve the work in an environment with a high density of signal sources, increase the throughput to more than 1 Gb / s, the ability to work with weak signals with a power of 3 dB to increase the range, support for positioning, increase the relative maximum speed to 500 km / h

C-v2x

LTE-V2X is the current implementation of the specifications for the C-V2X. The 5G NR-V2X is at an early stage.



The development of cellular technology releases is coordinated by 3GPP, the 3rd Generation Partnership Project, to ensure, whenever possible, the direct and backward compatibility of versions of cellular standards with each other. The backward compatibility of cellular standards is even legally reviewed in accordance with the European Commission Standardization Directive M / 453 and optionally ITS Directive 2010/40 / EU. 5G NR-V2X and LTE-V2X are not backward compatible. When developing the new release, it was considered that LTE-V2X does not have a sufficient penetration level to support backward compatibility. 3GPP release 8, known as LTE, was released on December 3, 2009 and took about 3 years to develop. 3GPP release 10 LTE-Advanced was released in 2011, 3GPP release 14 LTE-V2X - in 2014. Generation 5G starts with 3GPP release 15. In the release schedule of 3GPP there are projects that are divided into 3+ phases, which almost completely coincide with the releases, phases are divided into 3+ stages. In each case, a phase / release with an intermediate index of 1, 2, 2+ is a specification. Index 3 means the implementation of standards at the physical level. Releases are developed partially in parallel - the development of the next begins before the release of the previous one. In some cases, stage 4 may be added, which relates to test specifications. Thus, according to the 3GPP roadmap , the 5G project is still at a very early stage. It is planned to complete the development of release 16 only by March 2020. From the experience of the 3G and 4G releases, it can be assumed that it is realistic to expect a 5G-NR release ready for rollout that will include 3GPP release 17 , which will happen at the end of 2021.

At the stage of R&D and development of specifications, the so-called 6G standard is in parallel, which they plan to roll out in 2030. If 6G is also not backward compatible with 5G NR-V2X like 5G NR-V2X with LTE-V2X, then organizations interested in introducing new C-V2X technologies will probably ignore 5G NR-V2X to wait for 6G release and work right away with him.

Differences

The main difference between DSRC and C-V2X from the point of view of commercial use is that DSRC in production is stable, tested, evolving and predictably evolving, but C-V2X is not. OEMs and hardware developers can plan for DSRC technology on a long-term basis. At the same time, the DSRC does not surpass the C-V2X in technical specifications, and is inferior in some components. The problem with LTE-V2X is that not all the tests necessary for large-scale implementation have been passed, such as cross-border tests and cross-operator tests. At the same time, in 3GPP releases from 16 onwards, support for LTE-V2X is not guaranteed. Thus, the comparison of LTE-V2X and IEEE 802.11p is not relevant due to the cessation of support for the first.



Comparative characteristics:







6G specifications are very optimistic and are more likely to be advertising for hype.

Part 2. Results of comparative reliability tests of NR-V2X and IEEE 802.11bd from the Technical University of Dresden in 2019 [TRANSLATION FROM ENGLISH]

Translator's Note

Translation is not strictly verbatim. Self-repetitions were removed where possible and grammar was simplified where appropriate. Thanks for the constructive feedback.

Authors

Waqar Anwar, Andreas Trasl, Norman Franchi and Gerhard Fettweis, Vodafone Chair Mobile Communications Systems, Technical University of Dresden, Germany

annotation

Ultra-reliable communications enable complex scenarios covering autonomous driving and safety-critical applications. Modern automotive communications technologies such as IEEE 802.11p and LTE-V2V do not meet the reliability requirements for these scenarios. The next generation of these technologies is being developed, capable of approaching these requirements. This paper analyzes the estimated reliability of IEEE 802.bd and NR-V2X technologies. Although the standard for the physical layer is not yet available, we used the available parameters for our study. We used Monte Carlo based simulations to analyze the physical layer performance of these technologies in various V2V scenarios. One of the main challenges for ultra-reliable communication is High Doppler shift in V2V scenarios. NR-V2X is shown to outperform IEEE 802.11bd in reliability due to better Doppler shift processing. In IEEE 802.11bd, a high Doppler offset causes packet errors even with a high SNR - signal to noise ratio. Therefore, various measures to improve the performance of IEEE 802.11bd when exposed to a high Doppler effect are discussed and discussed.



Key terms are IEEE 802.11p, IEEE 802.11bd, LTE-V2X, NR-V2X, Ultra-reliable communication.

Introduction

The introduction is about the following: in the presence of radars, lidars, all types of cameras, we additionally need V2X devices for conditions of limited visibility (obstacles, weather conditions, terrain), work at a greater distance and autonomous driving scenarios.



The first IEEE 802.11p V2X standard was introduced in 2010 and was based on the IEEE 802.11a wireless LAN standard. Revision of the standard 802.11p introduced PHY and MAC levels aimed at improving WLAN performance for cars. An alternative to IEEE 802.11p is the LTE-based cellular V2X standard LTE-V2X, introduced by 3GPP in 2016. Both technologies are suitable for basic user scenarios, for example, notifications of road works, warning of emergency braking, data from traffic lights, notifications of special vehicles. IEEE and 3GPP are working on the next generation of V2X technology with support for more complex scenarios. The next cellular V2X standard, based on the fifth generation of 5G mobile communications systems, is expected to be completed in June 2019 as part of release 16, with the letter designation 5G NR, where NR stands for New Radio. Hence the name NR-V2X.



A working group called “IEEE 802.11 next generation V2X (NGV)” is working on the creation of the IEEE 802.11bd standard, the successor to 802.11p.



Comparison and testing, including field, 802.11p and LTE-V2X, is the subject of many publications [1] - [5]. Recently, a performance comparison was published in terms of the expected throughput, latency and reliability of the NGV technologies 802.11bd and NR-V2X [6] (the same authors as this document). Scenarios such as cooperative adaptive cruise control or safety-critical applications have high demands on gearshift delays and the reliability of communication systems. Achieving communication reliability is particularly difficult in automobiles due to the rapidly changing nature of the wireless communication channel environment, which leads to rapid obsolescence of channel estimation data. In addition to the high Doppler shift in V2X scenarios, bottleneck becomes intercarrier interference (ICI). Since the Doppler shift is significantly different in different scenarios, for example, the intracity Line-of-sight (LOS) scenario is different from the LOS script on the Route, a separate check of the work in these scenarios is required.



In this work, we check the PHY level of the developed V2X technologies 802.11bd and NR-V2X for the ratio of errors to transmitted packets (packet error rate - PER). PER is often used in measuring receiver reliability. Note that reliability is important because in most cases relaying cannot be used due to strict delay requirements. This work demonstrated that high Doppler offsets can lead to packet errors at high Signal to Noise Ratio (SNR) of 802.11bd. This is due to an outdated channel rating coupled with deep fading.



To improve channel estimation, use syncwords or midambles

to prevent saturation effect. It is shown that midambles must be used periodically in proportion to the speed of the vehicle. To further improve 802.11bd in low SNR areas, we suggest using features defined in IEEE 802.11ax, such as the extended range preamble and dual carrier modulation DCM. Finally, the performance of 802.11bd after all these improvements has been evaluated and compared with the performance of the NR-V2X.

Technology Overview

In this section, you can find a discussion and comparison of the most likely improvements to future 802.11bd and NR-V2X standards with their predecessors.

IEEE 802.11bd

IEEE 802.11p was introduced in 2010 as a revision of the 802.11 standard. Since then, several types of PHY layer implementations for WLAN systems have become available that need to be adapted for V2X. It can be expected that the next 802.11bd standard will be based on existing WLAN technologies such as IEEE 802.11ac and will use the available PHY configurations.



In the latest 802.11 revisions, throughput at the PHY level has been increased due to a higher modulation order and coding schemes (MCS - Modulation Coding Scheme), wider bandwidth configurations (more bandwidth) using carrier aggregation and the multiple input multiple output (MIMO) transmission method . A Low Density Parity Check, a low-density parity check method, was introduced, which is more effective for higher payloads, increasing speed and reliability. Reliability was later improved by space time block coding (STBC) or DCM. STBC is a diversity antenna configuration that allows two diversity branches on the transmitter side, where DCM is frequency diversity, use two diversity branches. Numerous cyclic prefixes (CP) timeslots allow you to implement a specific choice scenario to prevent intersymbol interference (ISI), thus making 802.11 standards more suitable for an outdoor environment. Using midambles allows less frequent channel estimations using midambles, and allows you to better cope with high Doppler shifts.



An extended range configuration is also available that speeds up synchronization and channel estimation, repeats specific fields of preamble signals to increase range and reliability.



According to the 802.11bd project authorization report, the following PHY parameters are considered:

Carrier modulation scheme: OFDM

Subcarrier spacing: 156.25 kHz, and 178.125 kHz.

CP durations: 1.6ms, 3.2ms

Channel coding: LDPC

Lowest rate: MCS9 (⅚ 256-QAM)

Target speed = 250 km / h

Doppler recovery method: high density midambles







The derivation of numerical values ​​from the table is shown in a previous work by the same authors.

NR-V2X

The first cellular V2X standard, LTE-V2X, was completed by 3GPP in 2016 in release 14. Significant changes are expected with the future 5G NR standard, which defines new V2X scenarios and requirements. The final version of the specifications is planned to be completed by the end of 2019 in release 16. Based on the possible settings of the physical layer, we assume that the NR-V2X will focus on NR uplink (UL). Since the NR-UL specifications are already available, you can design a framework for simulating the NR-V2X.



The main improvement in the physical layer of NR UL compared to LTE is that DFT-spread-OFDM and OFDM both methods can be used for data transmission. OFDM improves throughput efficiency for broadband operations with lower implementation complexity, and thus is more suitable for applications where high throughput is required. In the case of low-budget devices where high energy efficiency is required, DFT-s-OFDM is the best choice due to its low PARP (peak-to-average power ratio). Another improvement introduced in NR is the scalable OFDM numerologies, which allow you to choose between different subcarrier spacing from 15 kHz to 480 kHz. Together with these numerology, slot duration is in the range of 1ms - 0.031ms. Unlike LTE, the minimum transmission time interval (TTI) in NR is equal to the duration of one slot (one slot duration). In addition to low latency communication, a mini-slot option is provided for data transmission using only 2, 4, or 7 OFDM symbols without any slot boundaries. Scalable numerology along with variable durations (CPs) in NR provide the required applications and adaptations to a specific environment.

NR also provides a variety of de-modulation reference signaling (DMRS) options for better channel recovery for frequency and time selective channels. Since channel coding has a significant impact on the reliability and throughput of wireless communications, more efficient and reliable coding techniques have been applied (originally adopted), for example, LTE turbo codes have been replaced by LDPC codes for data channels and LTE convolutional codes replaced by a cyclic redundancy check (CRC) supplemented by polar codes for channel monitoring. Moreover, NR has the ability to use the millimeter wave spectrum with frequencies above 24 GHz also with frequencies below 6 GHz. The maximum channel width available to the user in NR is 100 MHz for frequencies below 6 GHz and 400 MHz for data transmission in the millimeter spectrum, which is much higher than the channel width of 20 MHz available for LTE. A larger channel width allows either higher peak data rates or a higher density of transmitters in NR. All of the above features make NR more reliable, channel bandwidth more efficient and flexible.

V2V scenarios and channel models

A set of V2V channel models presented by the 802.11 dedicated short range communications DSRC testing and performance assessment team. Channel models were derived from three measurement campaigns conducted by several organizations for five common V2V scenarios, as shown in fig 1. These channel models were used by the 802.11bd research team to evaluate performance and are the base reference ) for further improvements. The measured RMS delay profile and Doppler of these channel models were generally shown in Table 2

Los in the village

This model implements communication between two vehicles in an open space. Outside the city, LOS communications generally occur in the absence of other cars, large fences and buildings. Thus, the obtained delay profile demonstrates the strong influence of the LOS components in several weak multipath components and a maximum Doppler shift of 490 Hz.

LOS approaching vehicles in the city

Due to buildings and the high density of cars, strong reflections and multi-path fading are observed. It can be seen from the measurements that the high power reflected components have a strong influence on the signal, in contrast to the out-of-town LOS scenario.

NLOS City Crossroads

Consider the communication between two approaching cars at a city intersection with limited visibility in a moving stream. There will be buildings and fences at the corners of the intersection, which will lead to reflections and many multi-path components. Due to the absence of a dominant LOS component and small differences in the power of the reflected components, one can expect strong multi-pass attenuation.

LOS Motorway

This scenario simulates communication between two machines following each other on a multi-lane trunk. Despite the high traffic density, signs, hills, overpasses, LOS communication is still possible, as there are no physical obstacles between the communicating machines. Compared to other scenarios, a higher Doppler shift can be expected due to the high relative speed between oncoming machines.

NLOS Motorway

This scenario is similar to the LOS scenario except that the truck blocks visibility between the communicating machines. Strong degradation and changes in the quality of the compound can occur, because, due to the high flow rate, there are no strongly reflective objects for a long period of time. This is the most difficult scenario among all of the above, since we are dealing with NLOS communication, multi-pass attenuation and fast attenuation due to the high Doppler shift.

Performance rating

The theoretical calculations shown in Table 1 can be used to compare technologies within the available data rates and delays. These values ​​can be achieved only if all packets are delivered, which is impossible in the physical world. Thus, in this section, the performance of both technologies is evaluated in terms of PER for the previously described V2V channel models. PER is defined as the ratio of transmission errors to the total number of transmitted packets. This is a generic metric used to evaluate recipient performance and reliability. We want to know which technology works more reliably in different conditions. Reliability is critical, among other things, because in most low-latency applications, data transfer is not supported. Despite this, NR describes the possibility of sending a hybrid automatic repeat request (HARQ) to improve reliability. We do not consider this possibility in this work because of the low latency requirement.



To compare the performance of these technologies, a full-fledged implementation of the PHY functionality of the level of these technologies in MATLAB was performed. In order to simulate the fast attenuation and multipath effects in addition to additive white Gaussian noise (AWGN), the above models were constructed using the Rician distribution. In NR-V2X DMRS, type A mapping with 3 additional reference symbols is used with a maximum length = 1. Using this scheme, 24 DMRS symbols are used within the time interval (in each 3-m OFDM symbol while in the alternative subcarrier in frequency). Due to the high density of the DMRS, a better channel estimation is possible. Other relevant parameters used in the simulation are shown in Table 3.







For comparison, two combinations of modulation and coding rate were used. We compare the lowest available MCS option (MCS0) in both standards (for example, QPSK with 0.12 coding rate in NR-V2X and ½ BPSK in 802.11bd), which also determines the range of the technology. For a fair comparison, we also compare them for ½ 16QAM which applies to the MCS13 in NR-V2X and MCS3 in 802.11bd. We limit our analysis to only the two MCSs that are most relevant for achieving ultra-reliable connectivity. To verify compliance with ultra-reliability requirements, it is necessary to measure PER <10 ^ -5, which is difficult to do in simulation for a limited time. Therefore, our estimates are limited to PER = 10 ^ -3. However, for lower PERs, performance can be predicted by extending the graphs obtained.

PER comparison with the lowest MCS

In Figure 2 (a), you can see the PER graph for the lowest MCS (MCS0) for both technologies and the V2X channel models described above. Since NR V2X and 802.11bd define different combinations of coding rate modulation, the maximum data rate will differ from that shown in Table 1.



It can be observed that the NR-V2X behaves equally well in all V2V channel models and its PER has a negligible effect with various delay profiles and Doppler shifts. For 802.11bd, PER changes are highly dependent on the type of channel. The NR-V2X has an advantage of 9dB over 802.11bd if you compare the similar PER in the scenario with suburban LOS and about 10dB in the case of the scenario with urban oncoming traffic. In all other scenarios, the PER for 802.11bd gets gets as soon as the channel estimation becomes outdated during the transmission of the packet. (over the course of packet). This saturation occurs when the ratio of the packet duration to the coherence time is greater than or equal to 1. If the packet duration is longer than the channel coherence time, the channel estimation is no longer valid for the characters at the end of the packet. The effect of this outdated channel estimate depends on the current attenuation depth. If deep fade does not occur, a small absolute deviation of the channel estimate from its reference value can be observed. In combination with deep attenuation, this absolute deviation leads to a large relative error, which then causes packet transmission errors. To improve performance in fast-paced channels, you can use midambles, which are discussed below. Other techniques can be used to reduce the likelihood of deep attenuation, such as diversification.



Unlike 802.11bd, DMRSs are embedded in the channel estimation data in the NR-V2X on the receiver side. Moreover, the NR-V2X provides various DMRS configurations, depending on the time and frequency of channel selection, which leads to a better channel estimate. Another reason for the excellent performance of the NR-V2X is its lower coding rate compared to 802.11bd - 0.12 <0.5. Thus, when using MCS0, the NR-V2X can achieve greater range and reliability compared to 802.11bd.

PER comparison with ½ 16QAM

For the sake of an objective comparison of technologies, equal modulation and coding rates are used here. Figure 2 (b) shows the PER achieved by technologies with ½ 16QAM in different V2X channel models. Although the difference between NR-V2X and 802.11bd is significantly reduced compared to MCS0, the NR-V2X has an advantage of about 3dB compared to 802.11bd for scenarios of suburban LOS, highway LOS, and NLOS city intersection. The advantage in the case of a scenario of an urban oncoming LOS channel model is only 1 dB, and when the NLOS highway model is used, the PER for 802.11bd becomes saturated due to poor channel estimation, as shown earlier. The performance of 802.11bd with ½ 16QAM is much better compared to MCS0 for the LOS trunk, NLOS trunk, and NLOS city junction channel models.The reason for the improvements is a 4-fold reduction in packet duration, since the peak data transfer rate is 4 times higher when using ½ 16QAM compared to MCS0 (½ BPSK). Reduced packet lengths lead to improved performance for preamble based channel estimations of 802.11bd channels, as the relationship between packet duration and coherence time is reduced. If this ratio is << 1, saturation will not occur. (no saturation will occur). For a NLOS scenario on a motorway, this ratio is greater than 1, since 50% of the consistency time (often roughly estimated as 9 / (16pi * f_d)) = 202ms, which is less than the duration of the packets = 236ms.since the peak data transfer rate is 4 times higher when using ½ 16QAM compared to MCS0 (½ BPSK). Reduced packet lengths lead to improved performance for preamble based channel estimations of 802.11bd channels, as the relationship between packet duration and coherence time is reduced. If this ratio is << 1, saturation will not occur. (no saturation will occur). For a NLOS scenario on a motorway, this ratio is greater than 1, since 50% of the consistency time (often roughly estimated as 9 / (16pi * f_d)) = 202ms, which is less than the duration of the packets = 236ms.since the peak data transfer rate is 4 times higher when using ½ 16QAM compared to MCS0 (½ BPSK). Reduced packet lengths lead to improved performance for preamble based channel estimations of 802.11bd channels, as the relationship between packet duration and coherence time is reduced. If this ratio is << 1, saturation will not occur. (no saturation will occur). For a NLOS scenario on a motorway, this ratio is greater than 1, since 50% of the consistency time (often roughly estimated as 9 / (16pi * f_d)) = 202ms, which is less than the duration of the packets = 236ms.as the relationship between packet duration and coherence time is reduced. If this ratio is << 1, saturation will not occur. (no saturation will occur). For a NLOS scenario on a motorway, this ratio is greater than 1, since 50% of the consistency time (often roughly estimated as 9 / (16pi * f_d)) = 202ms, which is less than the duration of the packets = 236ms.as the relationship between packet duration and coherence time is reduced. If this ratio is << 1, saturation will not occur. (no saturation will occur). For a NLOS scenario on a motorway, this ratio is greater than 1, since 50% of the consistency time (often roughly estimated as 9 / (16pi * f_d)) = 202ms, which is less than the duration of the packets = 236ms.



The reason for the higher performance for NR-V2X with the same combination of environmental parameters and coding rate is again in a better channel estimation and the use of DFT-s-OFDM. DFT-s-OFDM provides better PER performance than OFDM for frequency selective fading since the data symbols are distributed over the entire channel width.

Channel estimation for varying Doppler shifts

We saw earlier that 802.11bd performance is heavily influenced by Doppler shift and with preamble-based channel estimates. Thus, in order to improve performance, you can consider the use of midamble, where the channel estimate symbol (known as midamble) is repeated inside the data to get the current channel estimate, as shown in Figure 4. The frequency of midambles should be adapted according to the relative speed of vehicles. Low-frequency midambles will lead to channel estimation errors, and high-frequency midambles will increase the duration of the packets (inversely with the data rate). To evaluate the technology under the influence of a changing Doppler shift, we chose the scenario with the NLOS motorway, which is the worst case.The Doppler channel profile is scaled for three maximum Doppler shifts equal to 250 Hz, 500 Hz, 1000 Hz. For 802.11bd, suppose two sets of settings - one set without midambles, the second with adaptive midambles.



Frequency of midambles = 10, 5, and 3 OFDM symbols for 250, 500, and 1000 Hz, respectively. With these parameters, the frequency of midambles is approximately 90% of the connectivity time for all three Doppler shifts.



Figure 3 (a) shows the PER for MCS0 for different Doppler shifts. It can be seen that the NR-V2X is very reliable under the influence of Doppler shifts due to the high density of DMRS used for channel estimation and very low coding rate. Although high Doppler offsets also cause ICI (Inter-carrier interference), for slower SNRs the noise values ​​are more dominant than ICI, therefore it does not matter in this case. As previously shown, 802.11bd performance degrades with increasing Doppler offsets due to outdated channel estimates. Using adaptive midambles greatly improves the performance of 802.11bd and helps to overcome the minimum level of channel estimation errors, given that the frequency of midambles is significantly longer than the channel connectivity time. Although adding midambles removes PER saturation at 802.11bd, there is still a 10dB difference compared to the NR-V2X.



A similar comparison is presented in Figure 3 (b) for the case with ½ 16QAM. Again, from the presented results it follows that high reliability of 802.11bd is possible only when using midambles. Even with equivalent coding modulation rates, the NR-V2X has more than 1 dB advantage for all Doppler offsets. It can be seen that ICI is becoming increasingly visible at high SNRs. Thus, the difference in PER between technologies increases for different values ​​of Doppler offsets with increasing SNR.

DCM Effects and Extended Range Option

In the previous subsection, we showed that adding midambles can significantly improve 802.11bd performance. He still needs at least a 10dB SNR to achieve a PER of 10 ^ -3 while an NR-V2X can achieve the same goal with an SNR of less than 0dB. In the future, to improve 802.11bd performance in low SNR areas and achieve the goal of increasing the range by 2 times compared to 802.11p, you can adapt the extended range mode and DCM from 802.11ax. Using the range extension mode, the signaling field is repeated twice and the power of certain preambula fields is greatly increased, which improves the receiver sensitivity by 3 dB. In DCM, data is duplicated on the lower and upper half of the available subcarriers to improve frequency diversification and frequency selective channels. IEEE 802.11ax provides these options for lower MCS orders to improve cell edge performance. The performance improvement achieved by including these options in 802.11bd is the subject of discussion in this section.







The performance gain for 802.11bd when using extended range mode and DCM is shown in Figure 5. 802.11bd shows a ~ 5dB gain after enabling these options. The spectral efficiency of NR-V2X with MCS0 is 2 times less compared to 802.11bd MCS0. After switching on DCM, the spectral efficiency of 802.11bd decreased by 2 times and became equal to the NR-V2X MCS0. Even though DCM and increased range improve 802.11bd by 5dB, the NR-V2X requires 5dB less SNR to achieve the same PER results with the same spectral efficiency. Since when using DCM, the spectral efficiency is reduced by 2 times, then other diversified options can be used to achieve a similar increase without the cost of reduced spectral efficiency, such as STBC or diversification of signal reception (receive diversity).In the diversification of reception, the antennas are placed far enough from each other, so the received signals on both antennas show unrelated attenuation. In this way, further improvements can be made to use these options.

Conclusion

In this work, we compared the performance of the developed V2X technologies in various V2X scenarios for reliability. Based on the results, it was shown that the NR-V2X is better than 802.11bd, largely because 802.11bd is highly prone to Doppler shifts. Moreover, we have shown that using midambles significantly improves 802.11bd performance under the influence of high Doppler shifts, provided that the midambles period is much shorter than the channel connectivity period. We have shown that DCM and Extended Range Mode enhance 802.11bd performance. Although midambles, extended range preambles and DCMs have shown the possibility of improving the reliability of 802.11bd, it still cannot surpass the NR-V2X due to better channel estimation due to the high density of DMRS.lower coding rates and DFT-s-OFDM. In the future, work will focus on designing analytical methods, such as PHY abstraction, to expand comparisons for other targeted applications and scenarios.

3. V2X

Summary

Firstly, I note that it is practically impossible to order a device in Russia for any money. If you are planning to buy a V2X 802.11p or LTE-V2X-compatible device (from 2000 Euros), firstly, developers and distributors almost never buy a license to import their products in Russia, so no one knows what can happen at customs, how much money they will want to take from you at customs, how much clarification, execution and other bureaucracy can last. Thus, it is expensive, it can go from 3 months without a guarantee that it will be allowed across the border, all the risks are on you. There is an option to buy such a device from a Russian importer company, but as a rule such companies do not order single copies as it is not profitable, then if you are not going to purchase in bulk, this is not your option either.It remains either to order test devices at an address in the European Union and carry it in their hand luggage to their homeland, or go directly to the warehouse and again carry on luggage.



I did a little research on the available V2X devices on the market. Failed to test due to lack of funding for the acquisition thereof. And they cost a little. Below is more detailed how much.

Product

OBU



V2X communication processors



Semiconductor components are easier and cheaper to buy than finished products. But in this case, you will have to assemble the OBU with your own hands, which is difficult.

Conclusion

We started with a brief overview of existing technologies, with a summary of specifications. After him, we concluded that 802.11p and LTE-V2X are not of research interest, since the next generations are already at a late stage in the development of specifications and the 5G NR-V2X is not compatible with LTE-V2X. The next part was a translation of a comparison of models based on the available specifications of 802.11bd and 5G NR-V2X standards. The last part provides a brief overview of existing products based on 802.11p. At the time of publication, there were no commercial products created on the basis of 802.11bd or 5G NR-V2X, respectively, they were not included in the review.



Thus, it is clear that the market for V2X devices is still very young and small. The modest penetration of V2X technologies in my opinion is due to the fact that affordable commercial products do not meet the high requirements of low-latency applications. In the event of solutions that can significantly increase the autonomy of vehicles, we are waiting for the explosive growth of the market and the penetration of V2X technology.