Enhanced Wireless Channel Estimation Through Parametric Optimization of Hybrid Ray Launching-Collaborative Filtering Technique

Abstract

In this paper, an enhancement of a hybrid simulation technique based on combining collaborative filtering with deterministic 3D ray launching algorithm is proposed. Our approach implements a new methodology of data depuration from low definition simulations to reduce noisy simulation cells. This is achieved by processing the maximum number of permitted reflections, applying memory based collaborative filtering, using a nearest neighbors' approach. The depuration of the low definition ray launching simulation results consists on discarding the estimated values of the cells reached by a number of rays lower than a set value. Discarded cell values are considered noise due to the high error that they provide comparing them to high definition ray launching simulation results. Thus, applying the collaborative filtering technique both to empty and noisy cells, the overall accuracy of the proposed methodology is improved. Specifically, the size of the data collected from the scenarios was reduced by more than 40% after identifying and extracting noisy/erroneous values. In addition, despite the reduced amount of training samples, the new methodology provides an accuracy gain above 8% when applied to the real-world scenario under test, compared with the original approach. Therefore, the proposed methodology provides more precise results from a low definition dataset, increasing accuracy while exhibiting lower complexity in terms of computation and data storage. The enhanced hybrid method enables the analysis of larger complex scenarios with high transceiver density, providing coverage/capacity estimations in the design of heterogeneous IoT network applications. INDEX TERMS Collaborative filtering, 3-D ray launching, pattern recognition, wireless channel.

Empirical and Modeling Approach for Environmental Indoor RF-EMF Assessment in Complex High-Node Density Scenarios: Public Shopping Malls Case Study

“…For that purpose, a combination of electromagnetic theories and equations based on Geometrical Optics (GO) and Geometrical Theory of Diffraction (GTD) [116], have been considered. Moreover, the in-house developed 3D-RL code has been optimized in order to decrease processing time by means of hybrid simulation methodologies, as neural networks, or using collaborative filtering and the diffusion equation, enabling the evaluation of large complex heterogeneous environments [117][118][119]. In this sense, the simulation tool has been widely tested and validated for wireless propagation channel characterization and EMF exposure assessment in indoor and outdoor urban scenarios [120][121][122].…”

5G Spatial Modeling of Personal RF-EMF Assessment Within Aircrafts Cabin Environments

“…Following the same methodology as described in [41][42], the proposed 3D-RL EMF algorithm provides an absolute mean error of 1-3 dB and a standard deviation of 1-6 dB, when compare with real measurements, as presented in [43][44]. In addition, in order to decrease the computational time of the presented algorithm, hybrid modeling approaches have been proposed combining the RL approach with different techniques, such as Neural Networks (NN) [43], Diffusion Equation (DE) [45], Collaborative Filtering (CF) [46] or Machine Learning (ML) techniques [47]. These hybrid methods achieve precise results whilst reducing the computational cost, leading to an Optimized 3D-RL approach, more efficient and robust for complex scenarios.…”

Diffuse-Scattering-Informed Geometric Channel Modeling for THz Wireless Communications Systems

“…The different hybrid methodologies implemented in the software are also presented in Fig. 2 for completeness, although they have not been used here; namely the combination of Ray Launching -Neural Network [25], where a lower number of launched rays in the simulation scenario can be used whereas intermediate points are predicted using neural network; Ray Launching -Diffusion Equation [26], [47], where the 3D-RL algorithm is combined with a diffusion equation method based on the equation of transfer; or Ray Launching -Collaborative Filtering [48], [49], where database learning techniques are applied to lessen the poor-quality results of low-definition simulations. These acceleration techniques manage to obtain accurate results with lower simulation times.…”