Search

A wide variety of adversarial attacks have been proposed and explored using image and audio data. These attacks are notoriously easy to generate digitally when the attacker can directly manipulate the input to a model, but are much more difficult to implement in the real world. In this paper we present a universal, time invariant attack for general time series data such that the attack has a frequency spectrum primarily composed of the frequencies present in the original data. The universality of the attack makes it fast and easy to implement as no computation is required to add it to an input, while time invariance is useful for real world deployment. Additionally, the frequency constraint ensures the attack can withstand filtering defenses. We demonstrate the effectiveness of the attack on two different classification tasks through both digital and real world experiments, and show that the attack is robust against common transform-and-compare defense pipelines.

Although the field of low-voltage power-line communications (PLC) has been maturing over the past few decades, the effective, economic and safe coupling of a modem to the live power grid remains a challenging topic. In literature, PLC coupling is covered sparsely, compared to other relevant PLC topics, and limited information is available to PLC researchers and designers. Therefore, this paper aims to achieve the following goals: (i) to guide the reader to a practical understanding of PLC coupling concepts; (ii) to expose the reader to a wide range of low-voltage PLC couplers and their operating principles; (iii) to brief the reader on typical PLC coupler design steps for the various classes of couplers; (iv) to establish good safety principles for these couplers; and (v) to assist researchers and designers with PLC coupler selection, through a qualitative comparison of various coupler pros and cons. This comparison illustrates why the transformer-capacitor coupler remains a coupler of choice, and that the potential of inductor-capacitor (L-C) couplers is perhaps overlooked.

Accurate short-term load forecasting is essential to modern power systems and smart grids. The utility can better implement demand-side management and operate power system stably with a reliable load forecasting system. The load demand contains a variety of different load components, and different loads operate with different frequencies. The conventional load forecasting methods, e.g., linear regression (LR), auto-regressive integrated moving average (ARIMA), deep neural network, ignore the frequency domain and can only use time-domain load demand as inputs. To make full use of both time-domain and frequency-domain features of the load demand, a load forecasting method based on hybrid empirical wavelet transform (EWT) and deep neural network is proposed in this paper. The proposed method first filters noises via wavelet-based denoising technique, and then decomposes the original load demand into several sub-layers to show the frequency features while the time-domain information is preserved as well. Then, a bidirectional long short-term memory (LSTM) method is trained for each sub-layer independently. In order to better tune the hyper-parameters, a Bayesian hyperparameter optimization (BHO) algorithm is adopted in this paper. Three case studies are de-signed to evaluate the performance of the proposed method. From the results, it is found that the proposed method improves the prediction accuracy compared with other load forecasting method.

The Square Kilometre Array (SKA) project is an international effort to build the world's largest radio telescope, enabling science with unprecedented detail and survey speed.The project spans over a decade and is now at a mature stage, ready to enter the construction and integration phase. In the fully deployed state, the MID-Telescope consists of a 150-km diameter array of offset Gregorian antennas installed in the radio quiet zone of the Karoo desert (South Africa). Each antenna is equipped with three feed packages, that are precision positioned in the sub-reflector focus by a feed indexer platform. The total observational bandwidth (0.35-15.4GHz) is segmented into seven bands. Band 1 (0.35 - 1.05 GHz) and Band 2 (0.95 - 1.76 GHz) are implemented as individual feed packages. The remaining five bands (Bands 3, 4, 5a, 5b, and 6) are combined in a single feed package. Initially only Band 5a (4.6 - 8.5 GHz) and Band 5b (8.3 - 15.4 GHz) will be installed. This paper provides an overview of recent progress on design, test and integration of each feed package as well as project and science goals, timeline and path to construction.

Monitoring large bodies of water, such as the Laurentian Great Lakes in North America, can be challenging and costly. The bathymetry, the diffuse attenuation coefficient for downwelling irradiance ($K_{d}$), and the particulate backscattering coefficient ($b_{\text {bp}}$) are important metrics in monitoring water quality in lakes and have typically been measured in two ways: 1) via in situ sampling campaigns, which are expensive, time-consuming, and have a low spatial resolution; and 2) via passive optical imagery, which can have errors in excess of 50%. Recently, Ice, Cloud, and land Elevation Satellite-2 (ICESAT-2), an active light detection and ranging (LiDAR)-based satellite, has proven effective in deriving the bathymetry, $K_{d}$ , and $b_{\text {bp}}$ in the global oceans. However, validation of such metrics has never been done on satellite flyovers taken on the same day as in situ measurements. Likewise, studies on freshwater environments have been limited. Here, we compare in situ data sampled from Lake Michigan and Big Glen Lake between August 13th and 14th, 2021, and results derived from an ICESat-2 flyover in the same region on August 14th, 2021. We find excellent agreement between the in situ values and the satellite-derived values for all three metrics. This suggests that ICESat-2 and other future LiDAR-based satellites will be powerful tools for monitoring large freshwater lakes. [ABSTRACT FROM AUTHOR]