On Parameter Estimation for Ultra-Wideband Channels with Clustering Phenomenon

IEEE Open J. Antennas Propag.

2020

Estimating parameters of stochastic radio channel models based on new measurement data is an arduous task usually involving multiple steps such as multipath extraction and clustering. We propose two different machine learning methods, one based on approximate Bayesian computation (ABC) and the other on deep learning, for fitting stochastic channel models to data directly. The proposed methods make use of easy-to-compute summary statistics of measured data instead of relying on extracted multipath components. Moreover, the need for post-processing of the extracted multipath components is omitted. Taking the polarimetric propagation graph model as an example stochastic model, we present relevant summaries and evaluate the performance of the proposed methods on simulated and measured data. We find that the methods are able to learn the parameters of the model accurately in simulations. Applying the methods on 60 GHz indoor measurement data yields parameter estimates that generate averaged power delay profile from the model that fits the data.

Section: Introductionmentioning

confidence: 99%

Learning Parameters of Stochastic Radio Channel Models From Summaries

Bharti

Adeogun

IEEE Open J. Antennas Propag.

2020

“…They have been used to calibrate the Turin model [1], the Saleh-Valenzuela (S-V) model [2] and the polarized propagation graph (PG) model [15]. These calibration methods rely either on a Monte Carlo approximation of the likelihood [16], [17], the method of moments [18], [19], or a summarybased likelihood-free inference framework [20]- [23] such as approximate Bayesian computation (ABC). First developed in the field of population genetics in 1997, ABC has since become a popular method for calibrating models with intractable likelihoods in various fields, see [24] for an overview.…”

Section: Introductionmentioning

confidence: 99%

A General Method for Calibrating Stochastic Radio Channel Models With Kernels

Bharti¹,

Briol

IEEE Trans. Antennas Propagat.

2022

Calibrating stochastic radio channel models to new measurement data is challenging when the likelihood function is intractable. The standard approach to this problem involves sophisticated algorithms for extraction and clustering of multipath components, following which, point estimates of the model parameters can be obtained using specialized estimators. We propose a likelihood-free calibration method using approximate Bayesian computation. The method is based on the maximum mean discrepancy, which is a notion of distance between probability distributions. Our method not only by-passes the need to implement any high-resolution or clustering algorithm, but is also automatic in that it does not require any additional input or manual pre-processing from the user. It also has the advantage of returning an entire posterior distribution on the value of the parameters, rather than a simple point estimate. We evaluate the performance of the proposed method by fitting two different stochastic channel models, namely the Saleh-Valenzuela model and the propagation graph model, to both simulated and measured data. The proposed method is able to estimate the parameters of both the models accurately in simulations, as well as when applied to 60 GHz indoor measurement data.

“…Recently, calibration methods which circumvent the need for resolving the multipath components have been introduced [6]- [10]. In [6], [7], a method of moments approach is used for calibration.…”

Section: Introductionmentioning

confidence: 99%

Auto-Generated Summaries for Stochastic Radio Channel Models

Bharti

Adeogun

2021 15th European Conference on Antennas and Propagation (EuCAP)

2021

Recently, a calibration method has been proposed for estimating the parameters of stochastic radio channel models using summaries of channel impulse response measurements without multipath extraction. In this paper, we attempt to automatically generate summaries using an autoencoder for calibration of channel models. This approach avoids the need for explicitly designing informative summaries about the model parameters, which can be tedious. We test the method by calibrating the stochastic polarized propagation graph model on simulated as well as measured data. The autoencoder is found to generate summaries that give reasonably accurate results while calibrating the considered model.