A Path Loss and Shadowing Model for Multilink Vehicle-to-Vehicle Channels in Urban Intersections

Abstract

The non line-of-sight (NLOS) scenario in urban intersections is critical in terms of traffic safety—a scenario where Vehicle-to-Vehicle (V2V) communication really can make a difference by enabling communication and detection of vehicles around building corners. A few NLOS V2V channel models exist in the literature but they all have some form of limitation, and therefore further research is need. In this paper, we present an alternative NLOS path loss model based on analysis from measured V2V communication channels at 5.9 GHz between six vehicles in two urban intersections. We analyze the auto-correlation of the large scale fading process and the influence of the path loss model on this. In cases where a proper model for the path loss and the antenna pattern is included, the de-correlation distance for the auto-correlation is as low as 2–4 m, and the cross-correlation for the large scale fading between different links can be neglected. Otherwise, the de-correlation distance has to be much longer and the cross-correlation between the different communication links needs to be considered separately, causing the computational complexity to be unnecessarily large. With these findings, we stress that vehicular ad-hoc network (VANET) simulations should be based on the current geometry, i.e., a proper path loss model should be applied depending on whether the V2V communication is blocked or not by other vehicles or buildings.

Path Loss Prediction Based on Machine Learning Techniques: Principal Component Analysis, Artificial Neural Network, and Gaussian Process

“…Since radio receivers require a certain minimum power (sensitivity) to be able to successfully decode information, path loss prediction is essential in mobile communications network design and planning such as link budget, coverage analysis, and locating base station. Many existing path loss models [1][2][3][4][5][6][7][8][9][10][11][12][13][14] adopt a linear log-distance model, which is empirically derived by assuming a linear proportionality between the path length and the path loss, and by determining a proportional factor through the adequate linear regression analysis of the measured data. The linear log-distance model is simple and tractable, but it does not guarantee accurate path loss prediction performance for all radio propagation environments.…”

Deterministic 3D Ray-Launching Millimeter Wave Channel Characterization for Vehicular Communications in Urban Environments

“…The next generation of autonomous vehicles that will assist or completely replace the driving system in vehicles will be implemented on top of this technology. As a result, there has been extensive research on ADAS systems in recent years, the variety of on-board sensors (e.g., radars, cameras, and LiDARs) are used to reduce the number of traffic accidents and deaths [ 5 ]. Although ADAS applications are still in the early stages, they will transform the driving experience and simultaneously provide road safety.…”

“…One of the main disadvantages of these models is that they must be employed in environments similar to those used in their estimation, requiring strong modifications for their use in other environments, and in many cases, they are simply ineffective outside of these scenarios [ 25 ]. Some empirical studies regarding sub 6 GHz vehicular propagation channels can be found in the literature like [ 5 , 26 , 27 ]. Non-geometric stochastic models (NGSCM) are used to predict the values regarding the modeled system, taking into account the occurrence of random elements within the system.…”

V2I Propagation Loss Predictions in Simplified Urban Environment: A Two-Way Parabolic Equation Approach

“…Therefore, it is usually necessary to develop new empirical models for each new communication system. Empirical models and measurements are used to study the propagation losses between moving cars in the urban environment [7,8]. Similar research has been conducted for the vehicle-to-infrastructure (V2I) scenario [9].…”