Hyperspectral images contain mixed pixels due to low spatial resolution of hyperspectral sensors. Spectral unmixing problem refers to decomposing mixed pixels into a set of endmembers and abundance fractions. Due to nonnegativity constraint on abundance fractions, nonnegative matrix factorization (NMF) methods have been widely used for solving spectral unmixing problem. In this letter we proposed using multilayer NMF (MLNMF) for the purpose of hyperspectral unmixing. In this approach, spectral signature matrix can be modeled as a product of sparse matrices. In fact MLNMF decomposes the observation matrix iteratively in a number of layers. In each layer, we applied sparseness constraint on spectral signature matrix as well as on abundance fractions matrix. In this way signatures matrix can be sparsely decomposed despite the fact that it is not generally a sparse matrix. The proposed algorithm is applied on synthetic and real data sets. Synthetic data is generated based on endmembers from U.S. Geological Survey spectral library. AVIRIS Cuprite data set has been used as a real data set for evaluation of proposed method. Results of experiments are quantified based on SAD and AAD measures. Results in comparison with previously proposed methods show that the multilayer approach can unmix data more effectively.

The integration of more renewable energy resources into distribution networks makes the operation of these systems more challenging compared to the traditional passive networks. This is mainly due to the intermittent behavior of most renewable resources such as solar and wind generation. There are many different solutions being developed to make systems flexible such as energy storage or demand response. In the context of demand response, a key factor is to estimate the amount of load over time properly to better manage the demand side. There are many different forecasting methods, but the most accurate solutions are mainly found for the prediction of aggregated loads at the substation or building levels. However, more effective demand response from the residential side requires prediction of energy consumption at every single household level. The accuracy of forecasting loads at this level is often lower with the existing methods as the volatility of single residential loads is very high. In this paper, we present a hybrid method based on time series image encoding techniques and a convolutional neural network. The results of the forecasting of a real residential customer using different encoding techniques are compared with some other existing forecasting methods including SVM, ANN, and CNN. Without CNN, the lowest mean absolute percentage of error (MAPE) for a 15 min forecast is above 20%, while with existing CNN, directly applied to time series, an MAPE of around 18% could be achieved. We find the best image encoding technique for time series, which could result in higher accuracy of forecasting using CNN, an MAPE of around 12%.

Editorial: Hyperspectral imaging in environmental monitoring and analysis

High penetration of renewable energy resources in distribution systems brings more uncertainty for system control and management due their intermittent behaviour. In this context, besides generation side, demand side should be also controlled and managed. Since demand side has variant flexibility over time, in order to timely facilitate Demand Response (DR), distribution system operators (DSO) should be aware of DR potential in advance to see whether it is sufficient for different services, and how much and when to send DR signals. This indeed requires accurate short-term or medium-term load forecasting. There are many methods for predicting aggregated loads, but more effective DR schemes should involve individual residential households which would require load forecasting of single residential loads. This is much more challenging due to high volatility in load curves of single customers. In this paper, we present a novel method of forecasting individual household power consumption using recurrence plots and deep learning. We use Convolutional Neural Network (CNN) for such a two-dimensional deep learning approach, and compare it with one-dimensional CNN, as well as Support Vector Machine (SVM) and Artificial Neural Network (ANN). Demonstrating some experimental tests on a real case proved that our approach outperforms the other existing solutions.

The problem of received signal strength (RSS) source localization in wireless sensor networks has raised attention in recent years due to its low complexity. However, the performance of the RSS localization algorithms is degraded due to malicious phenomena such as multipath and fading. Hence, machine learning and especially deep neural networks (DNNs) are applied to solve this problem. So far, convolutional the neural network (CNN), the recurrent neural network (RNN), and multilayer perceptron (MLP) are used to this end. In this letter, neural networks such as MLP, CNN, and RNN are suggested to be combined pairwise. In this regard, we propose to use a fusion of two general neural network localization estimators. To design the combination coefficients, a heuristic approach based on the errors of the training step of two neural networks are suggested. Moreover, two optimized approaches for selecting the combination coefficients are provided based on mean and variances and also based on the mean square error of the combined estimator. The simulation results show improvement of the localization accuracy by fusion of two neural networks.