The efficiency of drug development defined as a number of successfully launched new pharmaceuticals normalized by financial investments has significantly declined. Nonetheless, recent advances in high-throughput experimental techniques and computational modeling promise reductions in the costs and development times required to bring new drugs to market. The prediction of toxicity of drug candidates is one of the important components of modern drug discovery. In this work, we describe eToxPred, a new approach to reliably estimate the toxicity and synthetic accessibility of small organic compounds. eToxPred employs machine learning algorithms trained on molecular fingerprints to evaluate drug candidates. The performance is assessed against multiple datasets containing known drugs, potentially hazardous chemicals, natural products, and synthetic bioactive compounds. Encouragingly, eToxPred predicts the synthetic accessibility with the mean square error of only 4% and the toxicity with the accuracy of as high as 72%. eToxPred can be incorporated into protocols to construct custom libraries for virtual screening in order to filter out those drug candidates that are potentially toxic or would be difficult to synthesize. It is freely available as a stand-alone software at https://github.com/pulimeng/etoxpred .

The millimeter-wave (mmWave) radar technology has attracted significant attention because it is susceptible to environmental lighting, wall shielding, and privacy concern. This article proposes a novel noninvasive human activity recognition system using a mmWave radar. The proposed framework first transforms mmWave signals into point clouds. Generally speaking, it consists of four major components: denosing, enhanced voxelization, data augmentation, and dual-view machine learning to lead to accurate and efficient human activity recognition. The proposed new methodology considers the spatial–temporal point clouds in physical environments through a modified voxelization approach, enriches the sparse data based on the symmetry property of radar rotations, and learns the activity using a dual-view convolutional neural network. To evaluate the performance of the proposed learning models, a dataset involving seven different activities has been established using a mmWave radar platform. The experimental results have demonstrated that the proposed system can achieve 97.61% and 98% accuracies during the tests of fall detection and activity classification, respectively. In comparison, the proposed scheme greatly outperforms four other conventional machine learning schemes in terms of the overall accuracy.

Specific emitter identification (SEI) is a highly potential technology for physical-layer authentication that is one of the most critical supplements for the upper-layer authentication. SEI is based on radio frequency (RF) features from circuit difference, rather than cryptography. These features are inherent characteristics of hardware circuits, which are difficult to counterfeit. Recently, various deep learning (DL)-based conventional SEI methods have been proposed, and achieved advanced performances. However, these methods are proposed for close-set scenarios with massive RF signal samples for training, and they generally have poor performance under the condition of limited training samples. Thus, we focus on few-shot SEI (FS-SEI) for aircraft identification via automatic dependent surveillance-broadcast (ADS-B) signals, and a novel FS-SEI method is proposed, based on deep metric ensemble learning (DMEL). Specifically, the proposed method consists of feature embedding and classification. The former is based on metric learning with a complex-valued convolutional neural network (CVCNN) for extracting discriminative features with compact intracategory distance and separable intercategory distance, while the latter is realized by an ensemble classifier. Simulation results show that if the number of samples per category is more than 5, the average accuracy of our proposed method is higher than 98%. Moreover, feature visualization demonstrates the advantages of our proposed method in both discriminability and generalization. The code and the dataset can be downloaded from https://github.com/BeechburgPieStar/FS-SEI .

Cognitive radio network (CRN) and non-orthogonal multiple-access (NOMA) play the crucial roles in the 5G (fifth generation) wireless communication systems. However, the physicallayer security still remains quite challenging. In this letter, a downlink cascaded transmitting zero-forcing-beamforming (ZFBF) technique is proposed to secure communications in a two-cell multiple-input multiple-output (MIMO), NOMA based CRN. The proposed technique protects the information from illegitimate users (eavesdroppers) within the same and adjacent cells. The restriction on the number of at the transmitter can also be relaxed by applying the signal alignment for the ZFBF technique. This proposed new method can maximize the total secrecy rate of the MIMO-NOMA based CRNs in the lack of conventional transceiver beamforming techniques. Numerical and simulation evaluations have demonstrated the effectiveness of the proposed method.

Biological systems present remarkable adaptation, reliability, and robustness in various environments, even under hostility. Most of them are controlled by the individuals in a distributed and self-organized way. These biological mechanisms provide useful resources for designing the dynamical and adaptive routing schemes of wireless mobile sensor networks, in which the individual nodes should ideally operate without central control. This paper investigates crucial biologically inspired mechanisms and the associated techniques for resolving routing in wireless sensor networks, including Ant-based and genetic approaches. Furthermore, the principal contributions of this paper are as follows. We present a mathematical theory of the biological computations in the context of sensor networks; we further present a generalized routing framework in sensor networks by diffusing different modes of biological computations using Ant-based and genetic approaches; finally, an overview of several emerging research directions are addressed within the new biologically computational framework.

A novel automatic modulation classification method based on the graph presentation of the cyclic spectrum is proposed. In our proposed scheme, the periodicity and the symmetry of the cyclic spectrum will be exploited to establish a concise feature representation of multiple graphs. The modulated signal is first transformed from the cycle-frequency domain into the graph domain. Consequently, the concise graph-presentation, namely, a set of weighted directed rings, will be formulated as the robust features of the original signal. Those features can be easily expressed by the corresponding adjacency matrices. It can be verified that the adjacency matrices are sparse and the non-zero entries therein can be registered as the efficient feature parameters. Through the Hamming distance measure to enumerate the difference between the feature parameters resulting from the training data and the test data, one can perform the modulation classification. Monte Carlo simulation results demonstrate that our proposed method can achieve much better classification accuracy than the existing technique when the cyclic spectrum is used.