**1. Introduction**

MIMO communication systems have received significant research activities both in industry and academia since the emergence of 3G systems and are currently attracting many developers of 5G and 6G systems [1–3]. Services such as eMBB and URLLC have played an important role in the development of the 5G NR and Intelligent Transport System (ITS) and their network performance in V2X communications which incorporates Vehicle-to-Vehicle (V2V), Vehicle-to-Infrastructure (V2I) and Vehicle-to-Pedestrians (V2P) communication modes. Wireless channel modeling plays a significant role in designing, assessing, and optimizing the

performance of the systems components including the physical layer, networking protocols, and the antenna arrays at the Transmitter (Tx) and Receiver (Rx) using either RT or SCM tools [4–7]. However, very little research has been reported on the differences between these two channel modeling approaches including their strength, limitations and how they consequently affect the evaluation of the MIMO channel capacity for realistic scenarios [8–14].

Comparisons between deterministic and SCM are reported in [8–14]. However, none of them provided sufficient comparison based on a rigorous representation of the antenna arrays in terms of circuit parameters and 3D far-field patterns for both the co- and cross-polarized vector field components. In [8], the difference in the MIMO channel capacity between the SCM COST 259 and a deterministic urban city model simulated using RT for 3G cellular system is reported. However, only omnidirectional antennas with no consideration for the mutual coupling are used and the results are only simulated in outdoor scenarios under one SNR. In addition, the testing scenarios are not identical, as for the RT model, they used varying heights for the base station towers, but for the SCM, fixed heights are used.

A comparison of the angles of arrival between 3GPP 3D statistical channel model and a deterministic urban channel model is presented in [9]. However, the MIMO channel capacity and throughput results are not evaluated. The authors in [10] considered a large-scale massive MIMO system to compare the downlink throughput between an outdoor urban deterministic model and the statistical i.i.d. Rayleigh model. However, the i.i.d. Rayleigh model is not suitable to represent urban channel models. In addition, the i.i.d. Rayleigh model uses data generated from the RT software.

An evaluation of the MIMO channel capacity is presented in [11] using deterministic and stochastic indoor channel models. TGn C, D and E are used as indoor office SCM and a close representation is created and used in an RT scenario as the deterministic model. The RT channel capacity show close comparison to the results from the TGn stochastic model E. However, in the deterministic channel model, the Mobile Terminal (MT) antennas are placed in only two rooms while ignoring other locations in the building which would subsequently affect the accuracy of the calculated results. Considering the asymmetric distribution of rooms and the small size of the model, the results should be studied for different locations of Tx and a complete distribution of Rx antennas in the entire model's area.

In [12], the difference between RT and SCM for network connectivity is reported. Neither capacity nor throughput results are presented, and all the simulations are for Single-Input Single-Output (SISO) scenarios. A massive MIMO study is presented in [13] comparing RT generated MIMO channel capacity with results from i.i.d. Rayleigh statistical channel model. Similar to [10], the i.i.d. Rayleigh is not a realistic representation of indoor nor urban channel models. The results were also calculated for only one SNR. Lastly, in [14], the authors presented a survey about different channel modeling approaches and the challenges that accompany them in 5G networks. However, the paper did not present comparison data nor case studies.

The main objective of this chapter to provide a fair comparison between the two channel modeling techniques in terms of the MIMO channel capacity. To this end, several important parameters have to be taken into consideration when evaluating the channel capacity, such as the 3D radiation patterns of transmitting and receiving antennas for both the co- and cross-polarized components, the scattering parameters of the antenna arrays under consideration to invoke direct and mutual coupling between the antenna elements, the distribution of the Tx and Rx components using different channel environments (indoor and outdoor) for the 5G NR and the ITS V2X systems. We utilize state-of-art commercial channel modeling tools for our case studies. MIMObit [15] is used for SCM and Wireless InSite [16] is used to

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

represent deterministic models. MIMObit is an electromagnetic propagation simulator that utilizes standardized stochastic spatiotemporal channel models and uses rigorous analytical electromagnetic formulation to produce precise antenna-toantenna channels and evaluate the performance of MIMO systems in different propagation models [15]. Wireless InSite is a 3D RT wireless electromagnetic solver that considers the physical characteristics of the materials in an environment and uses shooting and bouncing rays as electromagnetic waves to track their reflection, diffraction, transmission and scattering through objects and evaluate the received power, capacity and throughput at each point in a study area [16].

The new contributions of this chapter can be summarized as follows. A comprehensive research of the capabilities and limitations of stochastic and deterministic channel modeling tools is presented for the first time in different indoor and outdoor channel environments. The effects of the antenna's 3D radiation patterns and scattering parameters on the MIMO channel capacity for 5G and V2X applications are considered using cutting-edge SCM [15] and RT [16] tools. It should be noted that all the simulations included in this chapter are intended for the downlink transmission utilizing the 2.45 GHz ISM band and the 3.7 GHz 5G NR FR1 N77/N78 bands.

The rest of the chapter is organized as follows. In Section 2, we present a case study involving the evaluation of the SISO and MIMO channel capacity for the Two-Ray model using Wireless InSite and MIMObit in order to validate the results against analytical formulation. In Section 3, we study the MIMO channel capacity in an indoor office environment using RT and SCM. Section 4 presents an evaluation of the MIMO channel capacity in an outdoor scenario using RT and SCM. In Section 5, the performance of the two channel modeling tools is presented for a V2X scenario involving a fixed gNodeB base station (gNB) and a moving vehicle. Finally, the chapter is concluded in Section 6.
