Analysis of multipath channel delay estimation using subspace fitting

TL;DR

This paper investigates how unresolved multipath components in dense scattering environments bias delay estimates in radio channels, deriving an analytical expression for the bias and validating it through numerical experiments.

Contribution

It introduces a first-order perturbation analysis of subspace fitting delay estimation, revealing how unresolved MPCs affect bias in multipath channel analysis.

Findings

Derived an analytical bias expression for delay estimation

Validated the bias model with numerical experiments

Showed bias depends on unresolved MPC power and delay

Abstract

The presence of rich scattering in indoor and urban radio propagation scenarios may cause a high arrival density of multipath components (MPCs). Often the MPCs arrive in clusters at the receiver, where MPCs within one cluster have similar angles and delays. The MPCs arriving within a single cluster are typically unresolvable in the delay domain. In this paper, we analyze the effects of unresolved MPCs on the bias of the delay estimation with a multiband subspace fitting algorithm. We treat the unresolved MPCs as a model error that results in perturbed subspace estimation. Starting from the first-order approximation of the perturbations, we derive the bias of the delay estimate of the line-of-sight (LOS) component. We show that it depends on the power and relative delay of the unresolved MPCs in the first cluster compared to the LOS component. Numerical experiments are included to show that the derived expression for the bias well describes the effects of unresolved MPCs on the delay estimation.