Keywords

1 Introduction

Recent research and advancements in detection, automation, computing, communication and networking vehicle technologies promise improved and more efficient road safety and traffic efficiency and fuel consumption and emissions, by exploiting detection and communication capabilities, vehicles can cooperate and extend their awareness of context beyond the visual field. Cooperative vehicles share their driving intentions with other traffic actors, thus accurately predicting which others traffic participants will make and optimize their own decisions and maneuvers [1], in fact there are 5 levels of automation which leads to an autonomous vehicle or car which requires no human intervention using several technologies: 5G, V2X communication.

Dsrc and one of the technologies used by V2X communication which suffers suffers from the quality of the links degradation with the presence of buildings and vehicles, especially in urban areas, where canal collisions become serious when the density of vehicles is high on the other hand the DSRC is still to be implemented, the cellular V2X (C-V2X) catches up thanks to the advancement of radio access technologies such as well as the well-maintained infrastructure [2].

We understand that the autonomous vehicle needs to capture billions of data to know what is happening (and especially what will happen) around it, and with a latency time of the order of a thousandth. Seconds, so that it can translate this into a safety benefit, to capture all this data, manufacturers have therefore planned sensors placed on cars capable of detecting what is happening up to 250 m. To “see” beyond, arrives V2X (Vehicle To Everything), which means that cars and other 5G connected objects communicate with each other to transmit information that each has received, and which therefore becomes useful to everyone.

In fact many researchers are looking for ways to improve the performance of the DSRC; many researchers are looking for alternative technologies that could be used in V2X system. LTE-V is considered to be one of the most promising communications technologies that could replace DSRC. LTE-V is a recent version of LTE, which can provide low latency mobile communication speed. LTE is now predominant in UMTS and the ubiquitous deployed LTE base stations make building the V2X system much easier. 3GPP actively conducts studies and specifications work on V2X based on LTE. An element of study on LTE V2X Services has been approved by 3GPP, in which PC5- V2V based had been given the highest priority.

This radio Access Network Feasibility Study (RAN) has completed part of the PC5 transport for V2V services [3], VANET (Vehicle Ad Hoc Network) applications can present their own unique requirements and challenges of wireless communication technology, although considered the first standard for VANETs, IEEE 802.11p is still in the field testing stage. Recently, the LTE V2X (Long-Term Evolution Vehicular to X) protocol appeared as a systematic V2X solution based on 4G TD-LTE (Time Division Long-Term Evolution).

In this article we first we will present a comparison between the DSRC and LTE V2X technologies based on several criteria and the principle and architecture of each technology and finally a simulation to assess the success rate of package delivery in case of congestion.

2 Related Work

Several researchers are exploiting complementary or alternative vehicle communication technologies because of the limitations of DSRC. Lately, researchers became interested in exploiting and using cellular communication networks as well as 5G for V2X, this technology is also known under the Cellular name-V2X (C-V2X), This technology has been standardized by the 3rd Generation Partnership (3GPP) Cellular networks are expected to develop and improve the performance of vehicle communication (V2X).

3GPP released in version 14 C-V2X (also called LTE-V or LTE-V2X) which uses the LTE PC5 dedicated interface for V2V (Vehicle-to-Vehicle) communications, This standard was designed to support load cooperative traffic efficiency and security applications, and it is composed of two modes of operation. In C-V2X Mode 3, vehicles communicate directly with each other, however communications are handled by the cellular infrastructure which selects the radio resource or sub-channels for each V2V transmission. On the other hand, the CV2X Mode 4 does not require the support of the cellular network infrastructure and the vehicles autonomously select the sub-channels or the radio resources for their V2V transmission. This is the reason why the 3GPP standard defines a semi-persistent distribute scheduling program that all vehicles must implement. C-V2X Mode 4 is very powerful and efficient because it can support V2V security applications in the absence of cellular infrastructure coverage. Therefore, careful configuration of C-V2X Mode 4 is necessary to increase its communication range, efficiency and capacity [4].

Simulation research shows that LTE latency and message delivery rate outperform DSRC, on a number of parameters such as range, vehicle speed and number of vehicles on a channel [5]. Several simulation experiments that have been carried out prove that the latency is always less than 100 ms. As the number of vehicles increases, the latency increases, but when there are 150 vehicles in the same channel it does not exceed 60 ms [5]. The research literature also shows that the packet delivery rate of LTE is better than that of DSRC, which at its absolute best is only 80% and decreasing rapidly, while that of LTE is 95% or more. The communication range for LTE is greater than that of DSRC, where the range of DSRC is considered to be between 300m and 1km, while cellular radios, depending on the power and type of cell tower, and might have coverage up to about 10 miles [6].

3 Comparison between LTE V2X and DSRC4

In This section, we mainly focus on a comparison between LTE V2X and DSRC at the physical level, we present to you a vision allowing to know the principle of DSRC and LTE V2X, Architecture, Physical Layer, Frame Structure, and frequency offset estimation algorithm.

3.1 DSRC (Dedicated short-range communications)

  • Definition

DSRC (Dedicated Short Range Communications) is one of the research hotspot and has already become the V2X communication standard in some regions, such as America and Europe. In America, the Federal Communications Commission (FCC) has allocated 75 MHz bandwidth for DSRC, from 5.850 GHz to 5.925 GHz, which is divided into 7 channels, 6 service channels (SCH) and 1 control channel (CCH).

Indeed On the basis of the allocated spectrum, the IEEE has published a series of communication standards for the entire protocol stack. The standards are collectively referred to as Wireless Access for Vehicle Environments (WAVE). The WAVE stack includes IEEE 802.11p and IEEE 1609.x, the former defines the physical layer and part of the middle access control (MAC) layer, which is mostly changed from 802.11a. Compared to 802.11a, some procedures such as authentication and acknowledgment are omitted to speed up the access process, and other changes are made to suit the transport environment. The IEEE 1609.x family defines security services, architecture, resource management, networking services, multi-channel operations, and physical access for short-range, low-latency communications in vehicular environments [7].

Fig. 1.
figure 1

DSRC protocol suite [8].

Full size image

Dedicated Short Range Communication (DSRC), it is often used in Wireless Access in Vehicle Environment (WAVE), is a suite of dedicated protocols for low latency networks in vehicular environments.

This group of protocols as can be seen in Fig. 1 looks a lot like TCP / IP over Wifi. In fact, it supports the IPv6 stack in parallel with a network and transport layer protocol called Wave Short Message Protocol (WSMP) dedicated to the DSRC suite. The WSMP branch of the protocol suite allows faster configuration and more space-saving transmissions [9].

3.1.1 DSRC Architecture

The DSRC protocol for vehicle-to-beacon communications has been defined as a lightweight OS1 communication stack ([10]).

Fig. 2.
figure 2

DSRC protocol stack

Full size image

In fact it consists of three layers: L1: the physical layer, L2 the data link layer and L7: the application layer, see Fig. 2. This architecture is popular for real-time systems because it reduces protocol overhead and meets the challenges and time constraints. The system was dedicated to support different physical media, multi-application, scenarios and an environment containing several channels. This will ensure a wide variety of possible fields of application for this technology [10].

3.1.2 Physical Layer Architecture

The layers shown in Fig. 1 allow you to examine the different layers of the DSRC Protocol Stack in detail, from bottom to top and starting with the physical layer. The DSRC PHY protocol is defined in IEEE 802.11, specifically, as modified by IEEE 802.11p [11]. In fact it is divided into two sublayers: the dependent physical medium (PMD) sublayer and physical layer convergence procedure (PLCP) sublayer. As the name suggests, PMD interfaces directly with wireless support. It uses the familiar orthogonal frequency division multiplexing (OFDM) technique, originally added to 802.11 in the 802.11a amendment. PLCP represents the mapping between the MAC frame and the basic PHY layer data unit, the OFDM symbol.

In 2003, an earlier version of DSRC PHY was published under the auspices of ASTM International in ASTM E2213–03 [11], which was also based on IEEE 802.11. In 2004, interested parties obtained approval to create the WAVE IEEE 802.11p amendment for DSRC within the IEEE 802.11 (WG) working group.

The amendment was released in 2010. Deviations from the main 802.11 standard has been minimized to encourage 802.11 silicon vendors to add support for 802.11p, which would help reduce costs by taking advantage of the large volume of 802.11 chips produced annually. There are orders of more WiFi equipped cell phones sold each year than new vehicles. The automotive industry considers the PHY and MAC parts of ASTM E2213–03 to be obsolete in favor of IEEE 802.11 and 802.11p. The United States Federal Communications Commission (FCC) the regulations for DSRC [12, 13], however, still incorporate by reference rules contained in ASTM E2213–03. It was anticipated that FCC regulations would eventually be developed and updated to instead require compliance with IEEE 802.11 and 802.11p.

3.1.3 Frame Structure

Figure 3 shows the physical layer data frame structure.A1-A10 are ten identical short training symbols, each 16 samples long.

Fig. 3.
figure 3

DSRC PHY frame format

Full size image

A subset of these symbols are used for automatic packet detection gain control (AGC) and various diversity combination schemes. The remaining short training symbols are used for the coarse estimate of the frequency offset and the coarse estimate of the symbol timing.

These training symbols are followed by two identical long training symbols, C1-C2, which is used for channel estimation, fine frequency and symbol timing estimation.

C1 and C2 are 64 samples long and the 32 sample long CP1 is the cyclic prefix which protects against intersymbol interference (ISI) from short training symbols. After short and long drive symbols, comes the modulated actual OFDM payload symbols. The first OFDM data symbol is the physical layer header which is BPSK modulated and specifies the modulation scheme used in the following payload OFDM symbols.

Each OFDM symbol consists of 64 samples and a length of 16 CP samples which is pre-affixed for each OFDM symbol to combat ISI [14].

3.1.4 DSRC Frequency Offset Estimation Algorithm

For the DSRC receiver, there are two steps to estimate and correct frequency errors. The detailed steps can be seen in Algorithm 1.

1: The short training sequences (for coarse frequency offset estimation) and the long training sequence (for Frequency offset estimation) are utilized in the PLCP preamble to correct the frequency error, and the integer and non-integer parts of the frequency error can be corrected at the same time;

2: Four pilot subcarriers of every OFDM symbol are used for carrier phase tracking to alleviate the residual frequency error and phase noise [15].

3.2 LTE V2X (Long Term Evolution Vehicle to Everything)

  • Definition

LTE based V2V (Vehicle to Vehicle) WI (Work Item) was approved in December 2015 [16], and LTE based V2X (Vehicle to Everything) WI was approved in September 2016 [17] in 3GPP.

LTE-V2X is relatively considered to be a new technology and is specifically designed to support vehicular communication scenarios.

As already mentioned the first version of LTE-V2X was released by 3GPP in 2016 under the umbrella of LTE-release 14 specification, as an extension of LTE Device-to-Device (D2D) functionality which is standardized in LTE version 12.

LTE-V2X uses a secondary link which describes the physical channels and is basically based on the waveform of the LTE uplink. In LTE-V2X, there are two communication radio interfaces: firstly LTE-PC5 which is also known as LTE side link (PC5 refers to the radio interface name where user equipment (UE) communicates directly with another UE on the direct channel) and secondly LTEUu (UTRAN (Universal Terrestrial Radio Access Network), (The radio interface between the eNodeB and the user Equipment) as shown in Fig. 4 [18].

Fig. 4.
figure 4

LTE-V2X architecture[15].

Full size image

3.2.1 Frame Structure

The following figure shows the frame structure of LTE V2X. In the frame structure, there are 14 TTI (Transmission.

Time intervals), in which four DMRS (Demodulation Reference signals) and a GP (guard period) are included, and the rest are data symbols (Fig. 5).

Fig. 5.
figure 5

Frame structure of LTE V2X [12]

Full size image

For V2V, the data frame structure of D2D defined in 3GPP TS 36.211 and 3GPP TS 36.212 is reused:

There are 14 symbols in a TTI which lasts 1 ms, and the last symbol is used as the on-call period.

In PSSCH/PSCCH/PSDCH of 3GPP Rel 12/13 D2D, there are two DMRS per PRB and the DMRS time interval is 0.5 ms. When the speed of the mobile terminal increases, for example 140 km / h, and the center frequency of the signal is 6.0GHz, the coherence time (about 0.277ms) of the signal will be less than the current DMRS time interval (approximately 0.5 SP). On the other hand, the demodulation performance of the data will rise sharply due to poor channel estimation and a consequent lack of channel information. There is a consensus that the DMRS density over time the domain should be increased to four symbols.

3.2.2 LTE V2X Frequency Offset Estimation Algorithm

  • 1: Timing detection by searching the peak of channel estimation transformed to the time domain, ! d;

  • 2: Local DMRS sequence is transformed to the time domain, ! P(n);

  • 3: Sequence shift of sequence in Step 2 according to timing in Step 1, ! e P(n) = P(mod(n + d;N));

  • 4: Received DMRS symbol is transformed to the time Domain, ! r(n);

  • 5: Correlation is done for sequence in Step 3 and Step 4;

  • 6: Frequency offset is estimated by comparing the angle Difference offset half and second half of sequence in Step 5.

$$ \begin{array}{*{20}l} {f = \frac{1}{{2\pi \Delta t\,\tan^{ - 1} }}\left\{ {\sum\limits_{n = 0}^{N/2 - 1} {\tilde{P}(n)r(n)} } \right\}} \hfill \\ { \times \left\{ {\sum\limits_{n = 0}^{N/2 - 1} {\tilde{P}(n + \frac{N}{2})r(n + \frac{N}{2})} } \right\}} \hfill \\ \end{array} $$

To conclude this part shows us several points of difference allowing to see the difference between LTE V2X and DSRC namely the architecture of each of the two technologies, physical layer, frame structure the next section will help us to have more clarification on the comparison between these two technohnologies based on a simulation that evaluates the delivery performance of the packages.

4 DSRC vs LTE V2X Comparative Study

In this section a comparative study was carried out between the DSRC protocol and LTE V2X as well a table is presented which summarizes a comparative study between DSRC and LTE V2X based on several criteria allowing differentiating the performance of each protocol.

The DSRC is based on the IEEE 802.11p standard, which is an amended version of the IEEE Std. 802.11a to take advantage of the distributed capability and simplicity of operation of 802.11 networks, such as dynamic spectrum access, rapid deployment, and efficient network access. In a matter of fact, various V2X technologies have been developed to support ubiquitous, large-scale, high-performance communication methods for vehicle users, including both IEEE 802.11 V2X and cellular V2X (C-V2X). There are three major steps in order to improve V2X applications. The first and second stages focus on the areas of ITS telematics and advanced auxiliary driving, respectively. As the era of 5G approaches, V2X technology moves into the third stage which can support a wider range of advanced automotive applications, such as autonomous vehicles, remote and cooperative driving, and environmental perception and control. Real-time ITS [19],With its development, LTE has marked great success around the world, LTE-V2X can greatly benefit from the design, integration and scale of the LTE market. With the versatile communication types of one-to-one to one-to-many transmissions in LTE and the harmonization of the re-use of the application layer standard of SAE, the standardization of LTE-V2X in 3GPP can focus on standard developments for radio and layered network with a spectrally efficient air interface, the performance of LTE V2X can be proven to be superior to that of IEEE 802.11pe, based on simulation results, LTE- V2X can provide better performance and leverage successful deployments and ecosystem [20].

The following table shows the comparison points between DRSC and LTE VX based on several criteria namely Channel width, Frequency band, bite rate, range, capacity, coverage, mobility assistance, Market penetration (Table 1).

Table 1. Table showing a comparison between DSRC and LTE V2X
Full size table

To summarize this part we can see from the comparative table that the LTE VX exceeds the DSRC especially for Channel Width, a capacity which is very high with the LTE V2X as well as mobility support and market penetration.

5 Simulation and Performance Evaluation of DSRC and LTE V2X Protocol

In this section, the article presents the approaches used to compare DSRC and LTE for vehicular communication.

There are several software tools, namely OMNET  ++, and ns-3 widely used to develop V2X simulations, for.

The evaluation of DSRC versus LTE in this article we used the software Ns3, which is a discrete network simulator that uses.

C/C++ programming to create networking scenarios for DSRC and LTE. In order to analyze the results of the simulation, the simulator the pure data output is imported to other software, such as MATLAB, to create a visual representation of the Results.

  • Motivation for using the simulator NS3:

  • NS-3 provides a controlled environment to perform experimental evaluation of protocols when equipment is limited.

  • NS3 contains model libraries to simulate the architecture of the vehicle environment wireless access system (WAVE), it is considered one of the most reliable simulators for testing V2X protocols.

  • NS3 is a discrete event simulator, which means the simulation time update is event based.

  • the NS3 is developed to study the V2V, V2I and V2X communication models in an urban and road environment. In this design, several parameters are taken into account and each of them must be carefully configured in order to avoid complications.

  1. A.

    DSRC setup

In order to test the performance of the DSRC, various tests were created with parameters such as traffic type, maximum latency, congestion and range. In each scenario, the packet delivery success rate is measured against each parameter. The overall maximum latency was one of the major factors in the packet delivery success rate. He determines the time required delivering a message in a sometime and if not, it is considered a failure message.

Congestion and reach measure the number of vehicles on the road and distance between vehicles deliver messages, respectively. The DSRC simulation output is the success rate of delivering packets as a percentage at different ranges for each of the different parameters. For DSRC tests, congestion tests started at 20 vehicles and increased to 160 vehicles in ten increments.

These congestion tests are tested to three max latencies of 10 ms, 50 ms and 100 ms. These latencies were chosen due to standard being within 100ms for message delivery.

  1. B.

    LTE V2X setup

In fact LTE code settings were not as flexible as the parameters of the DSRC code and it also did not provide a user Friendly outing. The parameters tested were latency and congestion using the packet delivery success rate as a test metric.

For output, he gave a list of all messages received by different ports using the LTE communication standard with the success rate of delivering packets attached to the message. For comparison, LTE and DSRC were tested in the same conditions using congestion and latency parameters. The congestion test was carried out at 200 m with the motorway scenario at the same three latencies. Though there are only three different tests, the data from these tests provide a good comparison with the highway part of the DSRC congestion tests.

  1. C.

    Simulation Parameters

(See Table 2).

Table 2. Simulation parameters
Full size table
  1. D.

    Simulation Results

This part presents the result of the LTE vs. DSRC congestion comparison at 10 ms, 20 ms, 50 ms latency based on the effect of the congestion on the packet delivery success rate.

Fig. 6.
figure 6

Congestion comparison at 10 ms latency LTE vs. DSRC

Full size image

It can be seen from Fig. 6 that LTE performs better than DSRC at the lowest latency which is 10ms; this comparison is based on how the packet delivery success rate is affected by the congestion.

Fig. 7. 
figure 7

Congestion comparison at 50 ms latency LTE vs. DSRC

Full size image

We can notice that for a latency of 50ms we can see that the DSRC improves and approaches LTE, so the packet delivery success rate improves when the latency increases.

Fig. 8.
figure 8

Congestion comparison at 100 ms latency LTE vs. DSRC

Full size image

Now we can see from Fig. 8 that for the maximum latency which is 100ms both have the same level of packet delivery rate.

To conclude this part, we see that LE V2X is more efficient than DSRC on the other hand for the simulation for a latency set to the maximum we have the same result for both.

6 Discussion

We notice From the comparative table that based on the channel Width criterion the LTE V2x can reach up to 20 MHz on the other hand,DSRC just 10 MHz so for Frequency Band we notice that there is an equity because both can reach up to 5.9 GHz, for Bit Rate the Lte V2X exceeds Dsrc so for the range the LTE V2x can reach up to 30 km/h see on other simulations we can test it up to 120 km/h and for the capacity the LTE V2X is more efficient than the DSRC the same thing for mobility support and market penetration the LTE V2X takes the higher part on DSRC.

Based on the simulation and performance evaluation of DSRC and LTE V2X protocol, LTE testing is not as thorough as DSRC testing; a comparison can be made based on the effect of congestion on the package delivery success rate. LTE and DSRC data shows that LTE performs very well and better than DSRC for low maximum latency of 10 ms, however higher congestion levels decrease packet delivery success rate as long as maximum latency authorized increases, we can notice that the performances improve in Fig. 7 and 8, we see that the DSRC improves and approaches LTE until the latency is set maximum to 100 ms we see that both are the same level of package delivery success.

7 Conclusion and Future Works

To conclude, Several studies and research have been carried out to compare the effectiveness of direct communication technologies between LTE-V2X PC5 and 802.11p from the point of view of the accident avoided and the reduction of fatal and serious injuries. The study shows that LTE-V2X achieves a high level of accident avoidance and injury reduction. It also indicates that LTE-V2X achieves a high percentage of successful packet delivery and communication range.

From the simulation in our paper, we can notice that LTE V2X exceeds DSRC technology on several levels but there are also equalities that we observed from the simulation on the effect of congestion on the packet delivery success rate on each of the two especially when the maximum latency is set, So lTE V2x has a great capacity and performance, on the other hand the best solution to have an efficient performance of autonomous vehicles is the combination between the two technologies.

In our next work we will propose a coexistence solution: a hybrid approach deploying both DSRC and LTE-V2X would combine the advantages of both technologies to generate more efficient solution promising for vehicular communication. For example, DSRC supports more robust security message delivery than LTE-V2X, while the data transmission rate is provided by LTE V2X.

This solution is based on a selection algorithm that enables a heterogeneous LTE / DSRC solution, where LTE and / or DSRC are selected according to services. Each vehicle is assumed to be equipped with both LTE and DSRC interfaces, this proposed heterogeneous LTE / DSRC approach will be based on available radio access technologies and infrastructure to support future automated driving with high reliability and low latency requirements. The approach will make it possible to provide low latency for messages linked to transmission security by the DSRC, on the other hand high reliability for bandwidth-intensive services by LTE, and integrating these two radio access technologies taking into account the requirements, service and network performance in real time.