Breast cancer is a leading cause of cancer-related deaths among women worldwide. Early and accurate detection is crucial for improving patient outcomes. Our study utilizes Visible and Near-Infrared Hyperspectral Imaging (VIS-NIR HSI), a promising non-invasive technique, to detect cancerous regions in ex-vivo breast specimens based on their hyperspectral response. In this paper, we present a novel HSI platform integrated with fuzzy c-means clustering for automated breast cancer detection. We acquire hyperspectral data from breast tissue samples, and preprocess it to reduce noise and enhance hyperspectral features. Fuzzy c-means clustering is then applied to segment cancerous regions based on their spectral characteristics. Our approach demonstrates promising results. We evaluated the quality of the clustering using metrics like Silhouette Index (SI), Davies-Bouldin Index (DBI), and Calinski-Harabasz Index (CHI). The clustering metrics results revealed an optimal number of 6 clusters for breast tissue classification, and the SI values ranged from 0.68 to 0.72, indicating well-separated clusters. Moreover, the CHI values showed that the clusters were well-defined, and the DBI values demonstrated low cluster dispersion. Additionally, the sensitivity, specificity, and accuracy of our system were evaluated on a dataset of breast tissue samples. We achieved an average sensitivity of 96.83%, specificity of 93.39%, and accuracy of 95.12%. These results indicate the effectiveness of our HSI-based approach in distinguishing cancerous and non-cancerous regions. The paper introduces a robust hyperspectral imaging platform coupled with fuzzy c-means clustering for automated breast cancer detection. The clustering metrics results support the reliability of our approach in effectively segmenting breast tissue samples. In addition, the system shows high sensitivity and specificity, making it a valuable tool for early-stage breast cancer diagnosis. This innovative approach holds great potential for improving breast cancer screening and, thereby, enhancing our understanding of the disease and its detection patterns.
This paper introduces a novel approach to incorporate the Time Compression OverLap-Add (TC-OLA) technique used in communication systems into Linear Frequency Modulation Pulse Compression (LFM-PC) radar systems. This technique significantly boosts the signal to noise ratio (SNR) and provides a robust processing gain compared to the traditional radar LFMPC systems. In addition, TC-OLA provides a better immunity against powerful jamming techniques. At the transmitter side, we divide a digitized LFM chirp signal into a controlled number of overlapping segments. We then speed up each segment by increasing the sampling rate to account for the segment overlap. At the receiver side, we apply overlap-add to reconstruct the signal with a much higher gain. To simulate and evaluate the performance of the new system, we extend the conventional LFM-PC radar model, which includes Matched Filter (MF) processor, Moving Target Detector (MTD), and two common Constant False Alarm Rate (CFAR) algorithms, by suitably adding Time Compression (TC) and Overlap-add (OLA) blocks at the transmitter and receiver, respectively. Using the TC-OLAbased LFM radar system, we have control over the SNR level and the spectrum spread while preserving the same Doppler shift and target time delay as the conventional LFM radar system. Furthermore, we transform LFM chirp signal into a novel TC signal that inherits LFM properties while possessing better immunity to jamming. Moreover, the proposed radar model relies on high sample rates only when needed and, therefore, does not require changing MF, MTD and CFAR as is the case for a wideband LFM radar with the same processing gain. Detailed comparisons between the conventional LFM and the wideband LFM radars against the proposed model are also presented.
Linear frequency modulation (LFM) signals are pivotal in radar systems, enabling high-resolution measurements and target detection. However, these signals are often degraded by noise, significantly impacting their processing and interpretation. Traditional denoising methods, including wavelet-based techniques, have been extensively used to address this issue, yet they often fall short in terms of optimizing performance due to fixed parameter settings. This paper introduces an innovative approach by combining wavelet denoising with long short-term memory (LSTM) networks specifically tailored for LFM signals in radar systems. By generating a dataset of LFM signals at various signal-to-noise Ratios (SNR) to ensure diversity, we systematically identified the optimal wavelet parameters for each noisy instance. These parameters served as training labels for the proposed LSTM-based architecture, which learned to predict the most effective denoising parameters for a given noisy LFM signal. Our findings reveal a significant enhancement in denoising performance, attributed to the optimized wavelet parameters derived from the LSTM predictions. This advancement not only demonstrates a superior denoising capability but also suggests a substantial improvement in radar signal processing, potentially leading to more accurate and reliable radar detections and measurements. The implications of this paper extend beyond modern radar applications, offering a framework for integrating deep learning techniques with traditional signal processing methods to optimize performance across various noise-dominated domains.
1 July 2021 in IEEE Transactions on Signal Processing
This paper presents a new radar system that combines any number and any kind of radar signals and transmits them simultaneously, thereby obtaining all the advantages of each of the radar waveforms used. This new system is based on a low-rate coded time compression overlap-add (TC-OLA) technique, and allows radar designers to address the challenge of having many features, such as high-time resolution, high-Doppler tolerance, low-sidelobe level, and low probability of intercept (LPI), in a single waveform. Since different types of radar waveforms usually offer different features, the new system allows the designers to obtain optimality via pluralism. The performance evaluation of the new radar system shows its superiority when examined under noise and heavy-tailed clutter background over the classical radar system and other methods of combination by operating on $N$ -radar signals simultaneously, making use of features from each waveform, canceling the Doppler from a Doppler-intolerant signal, and adding an extra processing gain to the proposed system.
This paper introduces a novel smeared synthesized LFM (SSLFM) time compression overlap-add (TC-OLA) radar system. The new system allows us to control the signal to noise ratio level, and, therefore, obtain a higher processing gain compared to the traditional LFM-PC radar systems. In addition, it allows us to control the signal spectrum spreading, making it more immune to noise jamming. The new SSLFM signal is obtained by either multiplying the LFM waveform with a complex unit signal with the random phase or by encoding the time compression signal with the random phase at the transmitter. A denoising processor, placed either before or after the OLA processor, is used to remove the random phase from the SSLFM and forward the resulted LFM signal to the rest of the conventional LFM-PC radar receiver system. The new SSLFM TC-OLA radar system enjoys a better low probability of intercept feature while maintaining the LFM time sidelobe and Doppler tolerance properties. Moreover, the additional modules in the new radar system do not require changing the core LFM radar components. Using TC-OLA and denoising requires a synchronization system (SS) to properly recover the LFM signal. We, therefore, offer three SSs. The performance evaluation of the new radar system shows its superiority over the traditional LFM, the wideband LFM, and the TC-OLA-based LFM radars, especially under powerful noise jamming. The synchronization system is implemented and tested experimentally using software-defined radar.
Radar-based hand motion recognition (HMR) presents several challenges, including sensor interference, clutter, and the limitations of small datasets, which collectively hinder the performance and real-time deployment of deep learning (DL) models. To address these issues, this paper introduces a novel real-time HMR framework that integrates advanced signal pre-processing, a hybrid convolutional neural network–support vector machine (CNN–SVM) architecture, and efficient hardware deployment. The pre-processing pipeline applies filtration, squared absolute value computation, and normalization to enhance radar data quality. To improve the robustness of DL models against noise and clutter, time-series radar signals are transformed into binarized images, providing a compact and discriminative representation for learning. A hybrid CNN-SVM model is then utilized for hand motion classification. The proposed model achieves a high classification accuracy of 98.91%, validating the quality of the extracted features and the efficiency of the proposed design. Additionally, it reduces the number of model parameters by approximately 66% relative to the most accurate recurrent baseline (CNN–GRU–SVM) and by up to 86% relative to CNN–BiLSTM–SVM, while achieving the highest SVM test accuracy of 92.79% across all CNN–RNN variants that use the same binarized radar images. For deployment, the model is quantized and implemented on two System-on-Chip (SoC) FPGA platforms—the Xilinx Zynq ZCU102 Evaluation Kit and the Xilinx Kria KR260 Robotics Starter Kit—using the Vitis AI toolchain. The system achieves end-to-end accuracies of 96.13% (ZCU102) and 95.42% (KR260). On the ZCU102, the system achieved a 70% reduction in execution time and a 74% improvement in throughput compared to the PC-based implementation. On the KR260, it achieved a 52% reduction in execution time and a 10% improvement in throughput relative to the same PC baseline. Both implementations exhibited minimal accuracy degradation relative to a PC-based setup—approximately 1% on ZCU102 and 2% on KR260. These results confirm the framework’s suitability for real-time, accurate, and resource-efficient radar-based hand motion recognition across diverse embedded environments.