**5. Channel modeling for V2X applications**

Traditionally, deterministic channel modeling tools are often used in V2X studies to evaluate the performance of antennas designed to operate in free-space when installed on a mast or a vehicle, Advanced Driver-Assistance Systems (ADAS), and interference among different vehicles [21–24]. Specific models have been recently developed for these particular applications utilizing various geometries to assess V2X communication systems in dynamic scenarios, the review of which is provided in [25, 26]. The IEEE 802.11p and LTE-V standards are widely used for V2X communications. However, nowadays, 5G technologies are also being utilized for vehicle communications after the massive development in 5G networks and their capabilities in delivering high speed and reliable links between devices and equipment. In this section, we compare the capabilities and limitations of SCM and RT tools in assessing the performance of V2X communication systems in the physical (PHY) layer.

It should be noted that the RT approach is the method of choice for static indoor and outdoor environments in particular for the initial design and deployment as well as the optimal number of transmitters to ensure coverage for the desired coverage area. Once the network is deployed and operational, the performance will obviously degrade in certain spots due to relative motion between different objects

in the network which can be resolved by increasing the transmitted power, adjusting the tilt of the antenna array or installing additional transmitters as needed. In a realistic simulation involving multiple vehicles moving at different speeds on a highway, multiple trajectories have to be defined for several vehicles involved in the scenario. Each vehicle should be represented with the proper geometry to represent the vehicle type such as, sedans, trucks, motorcycles and SUVs with their respective material properties and included in the RT solution at different positions, the number of which depends on the speed of the fastest vehicle and the data rate of the V2X system in order to relate the sampling in time to the channel coherence time. Obviously, this leads to extremely large computational demands that can only be achieved on highly dedicated cluster computers.

Instead, the approach we followed to model simple V2X scenarios involves several simplifying assumptions in order to make the computational demands tractable. The vehicle structure has not been included in the RT approach since a dense outdoor environment is involved. We selected a path as shown in **Figure 14** for which the capacity has been evaluated along 410 m of discrete path points each separated by 1 m.

The same outdoor urban channel model in the previous section is utilized in studying the channel capacity available to vehicles traversing a path of 410 m at a given speed in a city surrounded by concrete buildings while transmitting and receiving signals from a gNB located close to the pathway in a 2 2 closed-loop MIMO system with no interference as shown in **Figure 14**. The channel parameters obtained from this model can be used in post processing to test different V2X scenarios. The vehicle has a two-element MIMO antenna array operating in the 5G N77/N78 bands at 3.7 GHz. The radiation patterns of the array shown in

**Figure 14.** *V2X route MIMO channel capacity.*

*Stochastic versus Ray Tracing Wireless Channel Modeling for 5G and V2X Applications… DOI: http://dx.doi.org/10.5772/intechopen.101625*

**Figures 5** and **6** are used in the simulation. The antenna is placed on the vehicle rooftop, 1.5 m above ground level.

The 2 2 MIMO system is simulated with 1 W total transmission power from the gNB. The maximum number of ray reflection, transmission and diffraction per path is seven, one and two, respectively. The spacing between the transmitted rays is 0.25° . The model is simulated as a closed-loop MIMO system with beamforming. The orthogonal radiation patterns of the antenna arrays is optimal for beamforming applications as they provide a narrow beam pointing at the vehicle while it moves, with Maximum Ratio Transmission (MRT) as the precoding scheme. The average MIMO capacity perceived at the vehicle's antenna array is 14.05 bps/Hz at an average SNR of 40 dB. **Figure 14** shows the MIMO channel capacity at each vehicle point along the traversed path in Mbit/sec with 20 MHz bandwidth.

In the absence of a vehicular model in the current version of MIMObit, we utilized the 3GPP 3D Urban UMa channel model as the stochastic channel where a gNB is placed at the coordinates (150, 0, 50) as illustrated in **Figure 15** with the same MIMO antenna array used in the RT software. Both LOS and NLOS components are considered. To represent a vehicle movement in the SCM, we developed a new approach in which 20 independent MT antenna arrays are placed at different locations along the path. In this approach, each array is assigned a temporal behavior where it turns on momentarily at the time the vehicle reaches that point. For example, assuming the vehicle is moving at 50 km/h, Rx1 turns on at time, t = 0 s, then turns off, Rx2 turns on at t = 1.8 s then turns off, Rx3 turns on at t = 3.6 s then turns off, and so on. The model is simulated as a closed-loop 2x2 MIMO system with beamforming and the average achieved MIMO channel capacity is 15.3 bps/Hz at an average SNR equal to 40.

A 4.32% difference in the closed-loop beamforming MIMO channel capacity between the two modeling approaches is observed. This is due to that the RT study involves a static environment with no object mobility due to the limitations of the available computational resources and hence there is no time-varying signal distortion caused by mobility. However, the cluster birth-death process in different channel realizations accommodates for the non-static channel behavior of the SCM.

**Figure 15.** *SCM V2X model.*

Additionally, the temporal characteristics of the Rx antenna defines the vehicle movement in the SCM. Nevertheless, incorporating a large number of vehicles in an RT tool moving in an urban environment with different speeds and trajectories places severe limitations in terms of computational time and resources.
