Spatial Wireless Channel Prediction under Location Uncertainty

Abstract

Abstract-Spatial wireless channel prediction is important for future wireless networks, and in particular for proactive resource allocation at different layers of the protocol stack. Various sources of uncertainty must be accounted for during modeling and to provide robust predictions. We investigate two channel prediction frameworks, classical Gaussian processes (cGP) and uncertain Gaussian processes (uGP), and analyze the impact of location uncertainty during learning/training and prediction/testing, for scenarios where measurements uncertainty are dominated by large-scale fading. We observe that cGP generally fails both in terms of learning the channel parameters and in predicting the channel in the presence of location uncertainties. In contrast, uGP explicitly considers the location uncertainty. Using simulated data, we show that uGP is able to learn and predict the wireless channel.Index Terms-Gaussian processes, uncertain inputs, location uncertainty, spatial predictability of wireless channels.

RCAMP: A resilient communication-aware motion planner for mobile robots with autonomous repair of wireless connectivity

“…We use Gaussian Processes for regression (GPR) [24] for modeling the radio signal distribution as demonstrated in [8], [25], [26]. A key difference compared to the previous approaches is that we employ online learning with dynamic training size that adapts to the changes in the environment (e.g.…”

Formation Control of Multi-Agent Systems with Location Uncertainty

“…The Gaussian process (GP) framework of [23], called uncertain GP (uGP), is here adapted for learning and prediction of the wireless channel considering TX and RX location uncertainty. This uGP framework is contrasted to classical GP (cGP), wherein location uncertainty is ignored.…”

“…They are then perturbed by location uncertainty. In accordance with [23], heterogeneous location errors have been considered with error variance i . The error variance follows an exponential distribution parametrised by the average location error standard deviation .…”

“…This was partially addressed in [37], which extended [20] to determine the impact of localisation errors on channel prediction. Location uncertainty at the receiver but not the transmitter side was explicitly accounted for in [23].…”

LAPRA: Location-Aware Proactive Resource Allocation

“…In particular, we evaluate a spatial UIL resource allocation method that utilizes Gaussian processes [12], [13] for channel prediction by exploiting user location information. We formulate a decentralized binary integer optimization problem for the user assignment.…”