The increasing popularity of Internet of Things (IoT) devices provides a huge data source for intelligent identification. Images captured by unmanned aerial vehicle (UAV) are often exploited in target detection missions for search, rescue and fire prevention. However, without sufficient training samples, the machine learning method is usually difficult to meet application requirements. To solve the problem, a brain-computer interface (BCI) real-time system is applied to UAV target detection. In this study, a novel rapid serial visual presentation (RSVP) paradigm was formulated to present images, enhancing the ability for wide-area target detection. Since there is spatial correlation in captured images, joint decision for identical target can improve the recognition efficiency. To suppress interfering components and improve event-related potential (ERP) detection efficiency, an enhancing ERP component (EEC) algorithm is proposed. Both decision method and EEC algorithm are based on strictly statistical theory. RSVP task was performed by 12 subjects. The interference components and noise correlation were significantly reduced by EEC algorithm. The target detection rate online was 86.6% while the false alarm rate was less than 5%. Besides, the joint decision strategy raised the area under curve (AUC) value from 0.876 to 0.963. The proposed BCI real-time system realizes the complementarity of human intelligence and IoT, ushering UAV target detection into the era of hybrid intelligence.

Hardware Trojan detection method has been given particular attentions by hardware security researchers since the failure of Syrian radars in 2007. Electromagnetic side-channel analysis is a promising method which is widely used in practice due to its advantage of efficiency, non-touch and high accuracy. In this brief, we propose a novel method using electromagnetic side-channel signal for hardware Trojan detection by transfer learning. Firstly, time-frequency information of electromagnetic signal is extracted by continuous wavelet transform to take full advantage of useful information in time domain and frequency domain. Then, time-frequency information is fed to transfer learning network to get the classification result. In order to further improve the result, we use transfer learning to further extract the key features in time-frequency information. Finally, the key features are classified by the support vector machine to improve the accuracy. The experiment is conducted on a stand FPGA board and advanced encryption standard circuit is used as the benchmark circuit. Experimental results show that our methods can improve the result efficiently.

Goal: Evoked or Event-Related Potential (EP/ERP) detection is a major area of interest within the domain of EEG (electroencephalography) signal processing. While traditional methods of EEG processing have mostly focused on enhancing signal components, few have explored background noise suppression techniques. Optimizing the suppression of background noise can play a critical role in improving the performance of EP/ERP detection. Methods: In this study, a spatio-temporal equalization (STE) method was proposed based on the Multivariate Autoregressive (MVAR) model, which has been shown to suppress the spatio-temporal correlation of background noise and improve the EEG signal detection performance. Results: For practical applications, two optimization schemes based on the spatio-temporal equalization method were designed to solve two common challenges in EEG signal detection: P300 and steady state visual evoked potentials. Our results demonstrated that the STE method effectively improves recognition performance of evoked or event-related potential detection. Additionally, the STE method offers fewer parameters, lower computational complexity, and easier implementation. Conclusion: These attributes allow the STE approach to be extended as a preprocessing method which can be used in combination with existing techniques.

The optimization of coding stimulus is a crucial factor in the study of steady-state visual evoked potential (SSVEP)-based brain-computer interface(BCI).This study proposed an encoding approach named Multi-Symbol Time Division Coding (MSTDC). This approach is based on a protocol of maximizing the distance between neural responses, which aims to encode stimulation systems implementing any number of targets with finite stimulations of different frequencies and phases. Firstly, this study designed an SSVEP-based BCI system containing forty targets with this approach. The stimulation encoding of this system was achieved with four temporal-divided stimuli that adopt the same frequency of 30 Hz and different phases. During the online experiments of twelve subjects, this system achieved an average accuracy of $96.77 \pm 2.47$ % and an average information transfer rate (ITR) of 119.05 ± 6.11 bits/min. This study also devised an SSVEP-based BCI system containing 72 targets and proposed a Template Splicing task-related component analysis (TRCA) algorithm that utilized the dataset of the previous system containing forty targets as the training dataset. The subjects acquired an average accuracy of 86.23 ± 7.75% and an average ITR of 95.68 ± 14.19 bits/min. It can be inferred that MSTDC can encode multiple targets with limited frequencies and phases of stimuli. Meanwhile, this protocol can be effortlessly expanded into other systems and sufficiently reduce the cost of collecting training data. This study provides a feasible technique for obtaining a comfortable SSVEP-based BCI with multiple targets while maintaining high information transfer rate.