Yifan Chen

Accuracy of geometric channel-modeling methods

When constructing a propagation channel model, a substitute is often created by giving an arbitrary shape or form to scatterer distributions based on its intuitive appeal for a certain radio environment. However, such models do not necessarily represent the actual propagation process and may yield inaccurate results. The main objective of this paper is to provide an insight into the underlying relationship between geometric models and the particular physical propagation process they represent. The workhorse is the semi-geometrically based statistical (SGBS) model and the two heuristic rules. The SGBS model defines the distribution of dominant scatterers contributing to the last reradiation of multipath components to the base station. The earlier multiple-reflection process is modeled using the composite Nakagami/log-normal probability density function. Two parameters are then introduced; namely, the effective path length and the normalized space-dependent intensity measure. Using these two metrics, two heuristic rules are subsequently proposed to provide the missing link between the canonical models and the physical channel. The rules are then applied to revisit several widely used geometric models in macro- and microcellular environments. As a working example, the Gaussian scatterer density model is further extended using such an approach. Important channel parameters such as power azimuthal spectrum, power delay spectrum, and azimuthal and delay spreads are then calculated and compared with simulation results.

Dynamic simulation model of indoor wideband directional channels

Culled from existing channel measurements, a dynamic indoor wideband directional-channel simulator is constructed by incorporating the spatial-temporal properties as well as the time-varying nature of propagation environments. A two-state semi-Markov model is used to account for the births and deaths of scattering clusters, which represents an improvement over the conventional Markov model where the duration in each state is forced to follow an exponential distribution. The spatial-temporal variations of clusters within their lifespan are modeled based on the real-life measurement results. The ranges of angle of arrival (AOA), angle of departure (AOD), and time of arrival (TOA) for each cluster are partitioned into finite number of intervals using the deterministic simulation methods. Subsequently, the cluster properties at different intervals can be derived from the empirical findings in previous channel-measurement campaigns. The efficacy of the proposed simulator is demonstrated by applying it to characterize a time-varying multiple-input multiple-output (MIMO) channel.

A generic channel model in multi-cluster environments

We present a new method of defining spatio-temporal clusters using geometric modeling approach, which simulate the phenomenon that has a geometric analog. Various clusters are given different shapes or forms to scatterer distributions based on their logical appeal to certain radio environments. This classification methodology is based on the semi-geometrically-based statistical model developed in this paper. We then derive the analytical expressions of power azimuthal spectrum, power delay spectrum and cluster power using the generic model. Armed with the model preliminaries, we further parameterize some important geometric clusters, which reproduce the significant effects of typical scattering environments. The modeling approach presented in this paper is useful for system-level simulation and performance evaluation.

Ultrawideband source localization using a particle-swarm-optimized Capon estimator

We present a realistic frequency-dependent channel model for ultrawideband (UWB) communications and develop a generalized broadband Capon estimator for localization of the spatially dispersed UWB signals. The proposed estimator is able to address the three key aspects of practical UWB radio wave propagation: presence of local scattering, wideband array signals, and frequency-dependent dispersive effects. The particle swarm optimization, which is a high-performance optimizer based on the intelligence of swarms, is then implemented to perform a multidimensional parameter search to jointly estimate the source central angles, the polynomial regression coefficients for angle spreads, and the frequency dependence of propagation modes.

Diversity Gains of Generalized Distributed Antenna Systems with Cooperative Users

A geometry-based channel model is proposed to describe the topology of a generalized distributed antenna system with cooperative diversity (GDAS-CD). The system architecture comprises a number of largely separated access points (APs) each with multiple antennas within an AP at one side of the link, and several geographically closed user terminals (UTs) each having multiple antennas within a UT at the other side. The UTs are assumed to be cooperative devices. The average cross-correlation of signals received from non-collocated APs and UTs is derived based on the system topology and the empirical models proposed by Sorensen. Subsequently, we investigate the diversity gains obtainable from the GDAS-CD based on the proposed model, which would provide insight into the degree of possible performance improvement when combining multiple copies of the received signal.

Effect of Antenna Directivity on Angular Power Distribution at Mobile Terminal in Urban Macrocells: A Geometric Channel Modeling Approach

We present a geometric channel model to study the effect of antenna directivity on angular power distribution at the mobile terminal in urban macrocells. The methodology reviewed in this paper integrates the antenna effect into the model geometry, thereby facilitating a system-dependent channel characterization. As each device is limited in terms of measurement sensitivity, the effective scatterer distribution is essentially dependent on the antenna beam pattern. Subsequently, two heuristic rules are proposed to establish the underlying relationship between the model geometry and the corresponding wave-propagation processes. It is shown that the influence of directional antenna is twofold. First, it alters the spatial distribution of scatterers by providing a different sample space for the random field, and secondly, it distributes signal components into the angles-of-departure or collects them from the angles-of-arrival by weighted combination. Important channel parameters measured at the mobile terminal such as the angular power distribution, Doppler spectrum, and multipath shape factors are also investigated to further exemplify the usefulness of the proposed model.

Parabolic distribution of scatterers for street-dominated mobile environments

We propose a parabolic scattering model (PSM) to complement the widely used circular scattering model for street-dominated environments, where street canyons favor propagation in narrow angular ranges centered on the street direction. The proposed model is based on the recently introduced semi-geometrically based statistical (SGBS) model and the geometric classification (GEC). The SGBS model defines the distribution of scatterers contributing to the final reradiation of multipath components to the receiver and the GEC defines spatio-temporal clusters using a geometric approach, which simulates different propagation classes based on their suitability to specific radio environments. Expressions are then provided for the power azimuthal spectrum (PAS), angular constriction, variance of received power, and Doppler spectrum (DS) using the PSM, all of which are obtained as seen from the mobile unit. Results for the PSM are also compared with the experimental data provided for dense urban macrocells, for which it has been shown that the proposed model produces PAS and DS that closely agree with the measurement.

Spatio-diversity gain of a system-dependent wideband directional channel

Estimations of spatio-diversity gain (SDIV) in the wideband directional channel suffer from imprecise modeling of multipath angular spread (AS). Using previously published measurements and a synthetic array and channel parameterization, we show that AS exhibits dependence on both line-source distribution and operating frequency. As a result, SDIV is calculated through the integration of a continuous density function defined on the entire signal space. Using this methodology, we show through numerical examples that the conventional diversity estimate for directional systems may lead to systematic prediction error of the achievable diversity gain.

Ultra-wideband source localization using a particle-swarm-optimized Capon estimator from a frequency-dependent channel modeling viewpoint

We introduce a realistic frequency-dependent channel model for ultra-wideband (UWB) communication systems and develop a generalized broadband Capon spatial spectrum estimator for localization of multiple incoherently distributed scattering clusters. The proposed estimator is able to address the three crucial features of practical UWB impulse propagation: presence of local scattering for multiple incoherently distributed clusters, wideband array signals, and frequency-dependent dispersive effects. The particle-swarm optimization, which is a recently invented high-performance optimizer based on the movement and intelligence of swarms, is then implemented to perform a multidimensional parameter search to jointly estimate the source central angles, the polynomial regression coefficients for angle spreads, and the frequency-dependence of various clusters. Numerical experiments are also carried out to examine the performance of the algorithm under various environments and model mismatches.

Effect of antenna directivity on spatial pattern of scatterers in urban macrocellular environment

When directional antennas are employed in the transmission link, the channel response is often obtained by applying a complex weight (antenna voltage pattern) to the waves distributing into/coming from different directions and is considered as a spatial filtering process. An important question is whether the antenna directivity has a substantial influence on the spatial pattern of scatterer distribution. As a consequence, allowing for a system-independent channel characterization may be implausible if the answer is yes. We present a systematic method to study the effect of antenna patterns on scatterer distribution in the multipath channel. The workhorse is the recently introduced semi-geometrically based statistical model and two heuristic rules, which establish the correspondence between geometric models and physical wave propagation processes. It is shown that the effect of a directional antenna is twofold. First, it alters the spatial distribution of scatterers by providing a different sample space for the random field; second, it distributes the signal components into the angles-of-departure or collects them from the angles-of-arrival by weighted combination.