**4. Case study 3: outdoor channel modeling**

To further investigate the strengths and limitations of RT and SCM tools, the third case study considers the MIMO channel capacity in a realistic outdoor environment. The same software, MIMObit and Wireless InSite are used to model the stochastic and deterministic outdoor channels, respectively. A 2x2 MIMO antenna array operating in the 5G NR N77/N78 bands is used in the gNB. The 3GPP 3D Urban Macro cell (UMa) channel model is used as the SCM of the urban model.

The deterministic urban channel model is imported from Remcom's example library and is shown in **Figure 12**. It contains 39 buildings with different structures and heights. The gNB MIMO antenna used in this study is the same two-element array from the indoor study operating at 3.7 GHz and is placed at the edge of the rooftop on a building (the green box in **Figure 12**) at a height of 126.57 m with 180° tilt about the x-axis so that the antenna pattern's main-lobes are pointing towards the ground. The MT antennas are two-element dipole arrays operating at 3.7 GHz. 500 MT antennas are placed randomly across the city at different heights ranging from 1.5 to 50 m.

The antenna array at the gNB is fed with a total of 1 W power. 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 MIMO open-loop diversity method is chosen for the study with no precoding or beamforming with 20 MHz allocated signal bandwidth. Equal Gain Combining is used as the combining method operated at the receivers. The average MIMO channel capacity perceived at the MTs is displayed in **Figure 13** under different SNR values ranging from 5 to 30 dB.

The 3GPP 3D UMa SCM is used to represent the stochastic urban environment in a squarish geometry with a length of 500 m. The 3GPP UMa model is used for cities with gNBs located above roof tops of building [20]. The number of clusters and rays per cluster vary with the model and could reaching up to 23 clusters and 20 rays per cluster [20]. Both LOS and NLOS propagations are considered in this

**Figure 12.** *RT outdoor channel model.*

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

**Figure 13.** *Outdoor MIMO channel capacity.*

model. The same gNB antenna operating at 3.7 GHz is placed at a height of 126.57 m and a location/orientation similar to that from the RT software. The receiving MT antenna is a two-element half-wavelength dipole array operating at 3.7 GHz and is placed at the center of the model at a height of 1.5 m. The model is simulated under 1000 instantiations and the average MIMO channel capacity for SNRs ranging from 5 to 30 dB is shown in **Figure 13**.

Similar to the results obtained in the indoor environment case study, the MIMO channel capacity in the SCM is close to the one obtained from RT. The difference in the capacity between the two channel modeling tools is 12.3% at SNR = 5 dB where the SCM achieved 2.41 bps/Hz and the deterministic model achieved 2.13 bps/Hz, 14.9% at SNR = 15 dB where 5.97 bps/Hz and 5.14 bps/Hz capacities are achieved in the stochastic and deterministic models, respectively, and 31.9% at SNR = 30 dB where 13.8 bps/Hz and 10 bps/Hz capacities are achieved in the stochastic and deterministic models, respectively.
