Ray Tracing propagation modelling: Future prospects

Abstract

This contribution presents the future prospects of radio propagation modelling, with emphasis on deterministic propagation modelling, which is becoming of notable interest in view of future wireless systems adopting mm-wave frequencies and/or advanced MIMO transmission schemes. It is often said that there is not much to further investigate on radio propagation. These rumours often come from the same sources wherefrom criticism is raised about the reliability of propagation models, deterministic ones in particular. This view is actually contradictory: if propagation modelling problems are difficult to solve, and often actually unsolved, then there is place - and need - for further work. The fundamental question is: can a better knowledge of the multi-dimensional characteristics of the radio channel help in the design, deployment and optimization of future wireless systems? The answer is yes, and some motivations for this answer are given here. Quality of Service (QoS) is now a widely used term, and this fact suggests that, in a relatively mature application field such as the mobile radio one, QoS is becoming a strategic performance metric to assess the competitive advantage of different technical solutions or services. Still, not enough attention is being devoted to the actual propagation characteristics, which directly and dramatically impact on the QoS. The same is true for MIMO, Beamforming and UWB transmission techniques where the knowledge of the radio channel and of the real-time Channel State Information (CSI) is of fundamental importance. An increasing number of short-range, high frequency applications is envisioned for the next future, including gigabit wireless applications for indoor connectivity [1], mm-wave back-hauling for urban mobile radio networks [2] [3], etc. Besides the greater spectrum availability, one of the advantages of mm-wave frequencies is the small wavelength, which allows to implement compact, high-order MIMO antenna arrays and therefore to put into action beamforming techniques with narrow beams yielding optimum spatial-spectrum use and high signal-to-interference ratios. In such applications the propagation process takes place in a more limited and well-defined environment with respect to public mobile radio systems making use of large cells and lower frequencies. This represents the ideal field of application of deterministic propagation models such as Ray Tracing (RT). Being based on a sound, albeit approximate, theory such as Ray Optics, ray-based models can be considered very reliable when the wavelength is small compared to the size of the obstacles and when the propagation environment is limited, so that a detailed environment description is possible. RT models have been studied, developed and used for over two decades now. In particular the capability of RT models to predict the spatial and temporal dispersion characteristic of the radio channel, which is very valuable for the design of MIMO and beamforming systems, has been object of several studies [4] [5]. Recent investigations have shown that proper Diffused (or Distributed) Scattering models embedded into the RT engine can increase the prediction performance, especially in terms of the mentioned multi-dimensional dispersion parameters [6] [7]. The increasing availability of low-cost computation power and accurate 3D digital building databases will probably encourage the widespread use of deterministic propagation models in the future. Moreover, since future systems will be empowered with accurate positioning capabilities, deterministic propagation models such as ray tracing will be used not only “off-line” to assist in the system design and deployment phase, but also embedded into the system for on-line, real-time channel prediction to help estimate the CSI, thus reducing the need of costly and time-consuming channel sounding techniques. These and other future developments in the field of deterministic radio propagation modelling will be presented and discussed in this contribution.