Traditional analog-to-digital converters (ADCs) often struggle to balance high sampling rates with power efficiency, limiting their effectiveness in advanced radar and communication systems. Neuromorphic ADCs capture ...
详细信息
ISBN:
(数字)9798350329209
ISBN:
(纸本)9798350329216
Traditional analog-to-digital converters (ADCs) often struggle to balance high sampling rates with power efficiency, limiting their effectiveness in advanced radar and communication systems. Neuromorphic ADCs capture samples only when a signal crosses a specific threshold. This asynchronous sampling strategy effectively compresses incoming datastreams, enabling neuromorphic ADCs to achieve higher sampling rates at reduced power consumption compared to conventional ADCs. However, existing algorithms are poorly suited for processing the asynchronous samples. Conventional techniques like matched filters are inapplicable and most established deep learning algorithms expect regularly sampled data. This work introduces a spiking neural network (SNN) architecture specifically designed for processing asynchronous radar samples. Our novel approach is applied to Radar High-Resolution Range Profile (HRRP) based target classification. Remarkably, our experiments demonstrate that combining a neuromorphic ADC with an SNN achieves performance on par with high-sample-rate conventional ADCs paired with Convolutional Neural Networks (CNNs) while reducing the overall sampling rate by more than 95%.
One of the most physically and psychologically damaging neurological conditions that affect people of all ages is an epileptic seizure. The abnormality should be recognized early so that the proper treatment is timely...
详细信息
ISBN:
(数字)9798350389449
ISBN:
(纸本)9798350389456
One of the most physically and psychologically damaging neurological conditions that affect people of all ages is an epileptic seizure. The abnormality should be recognized early so that the proper treatment is timely. It is possible only with advancedsignalprocessing techniques to distinguish and predict epileptic patterns in which substantial effort is invested. Therefore, efficient seizure detection and classification methods are proposed for machine learning and deep learning algorithms. The proposed method uses deep and machine-learning algorithms for seizure detection and classification. The objective is to analyze the performance and efficiency of deep and machine learning classifiers by comparing the various classifiers. This proposed work uses 11,500 EEG data samples from the UCI machine learning repository. To suggest an Improved Fitness Function Genetic Algorithm (IGA) technique for optimal feature selection to improve the detection rate and CNN-RNN algorithm used as the classifier. The analysis proved that the hybrid CNNRNN(LSTM with GRU) classifier with GA-based feGAbased-lection provides better densification accuracy results of 98% when compared with all other classifiers.
Synthetic aperture radar (SAR) is widely used in various military and civilian fields due to its advantages such as high resolution, strong penetration and all-weather working ability. With the development of SAR imag...
详细信息
ISBN:
(数字)9798350355895
ISBN:
(纸本)9798350355901
Synthetic aperture radar (SAR) is widely used in various military and civilian fields due to its advantages such as high resolution, strong penetration and all-weather working ability. With the development of SAR imaging technology, the quantity of SAR image data continues to increase, which puts higher requirements on data processing capabilities. Moreover, the lightweight and miniaturized design of radar systems is the future development trend. Traditional computing models can no longer meet the above development needs, and new solutions are urgently needed. The SAR raw image contains a large amount of coherent speckle noise, which requires preprocessing to ensure the accuracy of subsequent target recognition. In order to solve the above problems, this paper implements the preprocessing of SAR target recognition based on coarse-grained reconfigurable array (CGRA). Experimental results show that compared with the implementation based on the most advanced FPGA, the proposed architecture has greatly improved energy efficiency.
Traditional polysomnography monitoring is complicated and time-consuming, and bracelets and other sleep monitoring devices cannot directly detect respiratory airflow. In this work, we designed a portable sleep monitor...
详细信息
ISBN:
(数字)9798331504205
ISBN:
(纸本)9798331504212
Traditional polysomnography monitoring is complicated and time-consuming, and bracelets and other sleep monitoring devices cannot directly detect respiratory airflow. In this work, we designed a portable sleep monitoring equipment that connects to the forehead and captures data on air flow directly. The sleep monitoring system can gather data from the following modules: blood oxygen, respiration, acceleration sensor, bioelectrical signal collection, and microphone. In addition, this work describes a method for identifying respiratory peak using a dynamic threshold to determine respiratory frequency. Fourteen healthy individuals took part in the investigation, wearing both our portable sleep monitoring equipment and a reference device at night. Due to the obvious phenomenon of baseline drift, the smooth filtering begins after baseline correction. For respiratory events such as oral respiration, apnea, and hypopnea, we set the maximum threshold and minimum threshold update time. The results reveal that the portable sleep monitoring system can accurately identify the breathing peaks.
Quantum computing is rapidly emerging as a revolutionary computing paradigm with significant development potential. Variational quantum algorithms, especially those employing variational quantum circuits, have demonst...
详细信息
ISBN:
(数字)9798350390643
ISBN:
(纸本)9798350390650
Quantum computing is rapidly emerging as a revolutionary computing paradigm with significant development potential. Variational quantum algorithms, especially those employing variational quantum circuits, have demonstrated notable advantages in domains including finance, chemistry, and machine learning. However, the practical deployment of these algorithms on Noisy Intermediate-Scale Quantum (NISQ) devices faces challenges due to quantum noise, state collapse, and limited qubits numbers. The real-world performance of these algorithms remains largely unexplored. This paper addresses this gap by proposing and implementing a novel method for training quantum self-attention model for text classification directly on quantum computing chips. This initiative marks the first instance of employing such models on actual quantum hardware. Our approach optimizes quantum circuit designs to mitigate the impact of quantum noise and introduces parallel training strategies to enhance efficiency. Our experiments, conducted on the “Wukong” superconducting 72-qubit quantum computer, demonstrate that our models can be effectively trained on real quantum hardware. The results from the actual quantum chips slightly outperformed those from simulators, confirming the practical feasibility of applying quantum self-attention model to natural language processing tasks. These results validate the effectiveness of our approach and provide valuable insights for the future deployment of quantum machine learning models.
The escalating challenge of traffic congestion in urban areas necessitates innovative solutions at the intersection of computer science and artificial intelligence. The main problem is in the outdated nature of conven...
详细信息
Unmanned Aerial Vehicles (UAVs) are widely used in civil and military applications, increasing the need for reliable detection to protect critical infrastructure. Radar is an effective method for detecting UAVs at lon...
详细信息
ISBN:
(数字)9798331531836
ISBN:
(纸本)9798331531843
Unmanned Aerial Vehicles (UAVs) are widely used in civil and military applications, increasing the need for reliable detection to protect critical infrastructure. Radar is an effective method for detecting UAVs at long distances and in all weather conditions. However, the similarity in size between UAVs and birds often leads to false positives, posing a problem in distinguishing between them. This study addresses this problem by proposing a classifier based on the micro-Doppler effect that discriminates between radar signals from UAVs and birds. The classifier uses a neural network approach, achieving an accuracy of more than 90%. The Engee mathematical computing and dynamic modeling environment was used for the computations. This work highlights the potential of advancedsignalprocessing and machine learning to improve UAV detection systems, reduce false alarms, and improve the safety of critical facilities.
Continual learning and multi-task learning are commonly used machine learning techniques for learning from multiple tasks. However, existing literature assumes multi-task learning as a reasonable performance upper bou...
详细信息
Continual learning and multi-task learning are commonly used machine learning techniques for learning from multiple tasks. However, existing literature assumes multi-task learning as a reasonable performance upper bound for various continual learning algorithms, without rigorous justification. Additionally, in a multi-task setting, a small subset of tasks may behave as adversarial tasks, negatively impacting overall learning performance. On the other hand, continual learning approaches can avoid the negative impact of adversarial tasks and maintain performance on the remaining tasks, resulting in better performance than multi-task learning. This paper introduces a novel continual self-supervised learning approach, where each task involves learning an invariant representation for a specific class of data augmentations. We demonstrate that this approach results in naturally contradicting tasks and that, in this setting, continual learning often outperforms multi-task learning on benchmark datasets, including MNIST, CIFAR-10, and CIFAR-100.
During clinical decision-making, the segmentation process Is considered an important task because it provides valuable information about the location, size, and characterization of the brain tumor. However, manual hum...
详细信息
ISBN:
(数字)9798350351484
ISBN:
(纸本)9798350351491
During clinical decision-making, the segmentation process Is considered an important task because it provides valuable information about the location, size, and characterization of the brain tumor. However, manual human segmentation is error-prone, time-consuming, and requires skilled *** resonance imaging (MRI) could give extremely detailed images for the investigation and diagnosis of glioblastoma brain *** compared evaluated approaches on the BRATS 2021 and BRATS 2022 datasets and found that they outperformed and could compete with state-of-the-art algorithms in comparable *** our research, we focused on two crucial tasks: segmentation and MGMT classification. This study also addresses asn objective evaluation through performance evaluation of cutting-edge DL-based techniques for MR image analysis (Brats 2021-Brats 2022). Based on the findings of the contrasted methods, we can confirm that using a combination of DL techniques will produce more accurate segmentation results than depending on a single, unique methodology. For the second task, five distinct deep learning-based methods were evaluated to predict the methylation state of the MGMT promoter.
This paper presents an investigation into the application of Brain-Computer Interface (BCI) technology for emoji selection by analyzing electroencephalographic (EEG) signals. EEG data was captured to interpret neural ...
详细信息
ISBN:
(数字)9798350352931
ISBN:
(纸本)9798350352948
This paper presents an investigation into the application of Brain-Computer Interface (BCI) technology for emoji selection by analyzing electroencephalographic (EEG) signals. EEG data was captured to interpret neural activity, with a Support Vector Machine (SVM) classifier employed for accurate emoji recognition. The system demonstrated a classification accuracy of 93.28% while implementing advancedsignalprocessing techniques that reduced noise and improved real-time performance. This research highlights the viability of thought-driven emoji selection and its implications for advancing accessible communication technologies and immersive virtual environments.
暂无评论