Channel Measurement and Noise Estimation in FBMC/OQAM-Based IoT Networks

Channel measurement plays an important role in the emerging 5G-enabled Internet of Things (IoT) networks, which reflects the channel quality and link reliability. In this paper, we address the channel measurement for link reliability evaluation in filter-bank multicarrier with offset quadrature amplitude modulation- (FBMC/OQAM-) based IoT network, which is considered as a promising technique for future wireless communications. However, resulting from the imaginary interference and the noise correlation among subcarriers in FBMC/OQAM, the existing frequency correlation method cannot be directly applied in the FBMC/OQAM-based IoT network. In this study, the concept of the block repetition is applied in FBMC/OQAM. It is demonstrated that the noises among subcarriers are independent by the block repetition and linear combination, instead of correlated. On this basis, the classical frequency correlation method can be applied to achieve the channel measurement. Then, we also propose an advanced frequency correlation method to improve the accuracy of the channel measurement, by assuming channel frequency responses to be quasi-invariant for several successive subcarriers. Simulations are conducted to validate the proposed schemes.

Effect of snowfall on changes in relative seismic velocity measured by ambient noise correlation

In mountainous, cold temperate and polar sites, the presence of snow cover can affect relative seismic velocity changes (dV/V) derived from ambient noise correlation, but this relation is relatively poorly documented and ambiguous. In this study, we analyzed raw seismic recordings from a snowy flat field site located above Davos (Switzerland), during one entire winter season (from December 2018 to June 2019). We identified three snowfall events with a substantial response of dV/V measurements (drops of several percent between 15 and 25 Hz), suggesting a detectable change in elastic properties of the medium due to the additional fresh snow. To better interpret the measurements, we used a physical model to compute frequency-dependent changes in the Rayleigh wave velocity computed before and after the events. Elastic parameters of the ground subsurface were obtained from a seismic refraction survey, whereas snow cover properties were obtained from the snow cover model SNOWPACK. The decrease in dV/V due to a snowfall was well reproduced, with the same order of magnitude as observed values, confirming the importance of the effect of fresh and dry snow on seismic measurements. We also observed a decrease in dV/V with snowmelt periods, but we were not able to reproduce those changes with our model. Overall, our results highlight the effect of the snow cover on seismic measurements, but more work is needed to accurately model this response, in particular for the presence of liquid water in the snowpack.

Recovering and monitoring the thickness, density and elastic properties of sea ice from seismic noise recorded in Svalbard

Due to global warming, the decline in the Arctic sea ice has been accelerating over the last four decades, with a rate that was not anticipated by climate models. To improve these models, there is the need to rely on comprehensive field data. Seismic methods are known for their potential to estimate sea-ice thickness and mechanical properties with very good accuracy. However, with the hostile environment and logistical difficulties imposed by the polar regions, seismic studies have remained rare. Due to the rapid technological and methodological progress of the last decade, there has been a recent reconsideration of such approaches. This paper introduces a methodological approach for passive monitoring of both sea-ice thickness and mechanical properties. To demonstrate this concept, we use data from a seismic experiment where an array of 247 geophones was deployed on sea ice in a fjord at Svalbard, between March 1 and 24, 2019. From the continuous recording of the ambient seismic field, the empirical Green's function of the seismic waves guided in the ice layer was recovered via the so-called 'noise correlation function'. Using specific array processing, the multi-modal dispersion curves of the ice layer were calculated from the noise correlation function, and then inverted for the thickness and elastic properties of the sea ice via Bayesian inference. The evolution of sea-ice properties was monitored for 24 days, and values are consistent with the literature, as well as with measurements made directly in the field.

Modeling Neural Variability in Deep Networks with Dropout

ABSTRACTConvolutional neural networks (CNNs) have been used to model the biological visual system. Compared to other models, CNNs can better capture neural responses to natural stimuli. However, previous successes are limited to modeling mean responses; while another fundamental aspect of cortical activity, namely response variability, is ignored. How the CNN models capture neural variability properties remains unknown. Previous computational neuroscience studies showed that the response variability can have a functional role, and found that the correlation structure (especially noise correlation) influences the amount of information in the population code. However, CNN models are typically deterministic, so noise (and correlations) in CNN models have not been studied. In this study, we developed a CNN model of visual cortex that includes neural variability. The model includes Monte Carlo dropout, namely a random subset of units is silenced at each presentation of the input image, inducing variability in the model. We found that our model captured a wide-range of neural variability findings in electrophysiology experiments, including that response mean and variance scale together, noise correlations are small but positive on average, both evoked and spontaneous noise correlation are larger for neurons with similar tuning, and the noise covariance is low-dimensional. Further, we found that removing the correlation can boost trial-by-trial decoding performance in the CNN model.