Low Computational Complexity Research Articles

Enhancement of single-lead dry-electrode ECG through wavelet denoising

Neonatal electrocardiogram (ECG) monitoring is an important diagnostic tool for identifying cardiac issues in infants at birth. Long-term remote neonatal dry-electrode ECG monitoring solutions can be an additional step for preventive healthcare measures. In these solutions, power and computationally efficient embedded signal processing techniques for denoising newborn ECGs can assist in increasing neonatal medical wearable time. Wavelet denoising is an appropriate denoising mechanism with low computational complexity that can be implemented on embedded microcontrollers for long-term remote ECG monitoring. Discrete wavelet transform (DWT) denoising for neonatal dry-electrode ECG using different wavelet families is investigated. The wavelet families and mother wavelets used include Daubechies (db1, db2, db3, db4, and db6), symlets (sym5), and coiflets (coif5). Different levels of added white Gaussian noise (AWGN) were added to 19 newborn ECG signals, and denoising was performed to select the appropriate wavelets for neonatal dry-electrode ECG. The selected wavelets then undergo real noise additions of baseline wander and electrode motion to determine their robustness and accuracy. Signal-to-noise ratio (SNR), mean squared error (MSE), and power spectral density (PSD) are used to examine denoising performance. db1, db2, and db3 wavelets are eliminated from analysis where the 30 dB AWGN led to negative SNR improvement for at least one newborn ECG, removing important ECG information. db4 and sym5 are eliminated from selection due to their different waveform morphology compared to the dry-electrode newborn ECG’s QRS complex. db6 and coif5 are selected due to their highest SNR improvement and lowest MSE of 6.26 × 10−6 and 1.65 × 10−7 compared to other wavelets, respectively. Their wavelet shapes are more like a newborn ECG’s QRS morphology, validating their selection. db6 and coif5 showed similar denoising performance, decreasing electrode motion and baseline wander noisy ECG signals by 10 dB and 14 dB, respectively. Further denoising of inherent dry-electrode noise is observed. DWT with coif5 or db6 wavelets is appropriate for denoising newborn dry-electrode ECGs for long-term neonatal dry-electrode ECG monitoring solutions under different noise types. Their similarity to newborn dry-electrode ECGs yields accurate and robust reconstructed denoised newborn dry-electrode ECG signals.

Kalman Filter Using a Third-Order Tensorial Decomposition of the Impulse Response

For system identification problems associated with long-length impulse responses, the recently developed decomposition-based technique that relies on a third-order tensor (TOT) framework represents a reliable choice. It is based on a combination of three shorter filters, which merge their estimates in tandem with the Kronecker product. In this way, the global impulse response is modeled in a more efficient manner, with a significantly reduced parameter space (i.e., fewer coefficients). In this paper, we further develop a Kalman filter based on the TOT decomposition method. As compared to the recently designed recursive least-squares (RLS) counterpart, the proposed Kalman filter achieves superior performance in terms of the main criteria (e.g., tracking and accuracy). In addition, it significantly outperforms the conventional Kalman filter, while also having a lower computational complexity. Simulation results obtained in the context of echo cancellation support the theoretical framework and the related advantages.

Laminography as a tool for imaging large-size samples with high resolution.

Despite the increased brilliance of the new generation synchrotron sources, there is still a challenge with high-resolution scanning of very thick and absorbing samples, such as a whole mouse brain stained with heavy elements, and, extending further, brains of primates. Samples are typically cut into smaller parts, to ensure a sufficient X-ray transmission, and scanned separately. Compared with the standard tomography setup where the sample would be cut into many pillars, the laminographic geometry operates with slab-shaped sections significantly reducing the number of sample parts to be prepared, the cutting damage and data stitching problems. In this work, a laminography pipeline for imaging large samples (>1 cm) at micrometre resolution is presented. The implementation includes a low-cost instrument setup installed at the 2-BM micro-CT beamline of the Advanced Photon Source. Additionally, sample mounting, scanning techniques, data stitching procedures, a fast reconstruction algorithm with low computational complexity, and accelerated reconstruction on multi-GPU systems for processing large-scale datasets are presented. The applicability of the whole laminography pipeline was demonstrated by imaging four sequential slabs throughout an entire mouse brain sample stained with osmium, in total generating approximately 12 TB of raw data for reconstruction.

EEG motor imagery decoding: a framework for comparative analysis with channel attention mechanisms

Objective. The objective of this study is to investigate the application of various channel attention mechanisms within the domain of brain-computer interface (BCI) for motor imagery decoding. Channel attention mechanisms can be seen as a powerful evolution of spatial filters traditionally used for motor imagery decoding. This study systematically compares such mechanisms by integrating them into a lightweight architecture framework to evaluate their impact. Approach. We carefully construct a straightforward and lightweight baseline architecture designed to seamlessly integrate different channel attention mechanisms. This approach is contrary to previous works which only investigate one attention mechanism and usually build a very complex, sometimes nested architecture. Our framework allows us to evaluate and compare the impact of different attention mechanisms under the same circumstances. The easy integration of different channel attention mechanisms as well as the low computational complexity enables us to conduct a wide range of experiments on four datasets to thoroughly assess the effectiveness of the baseline model and the attention mechanisms. Results. Our experiments demonstrate the strength and generalizability of our architecture framework as well as how channel attention mechanisms can improve the performance while maintaining the small memory footprint and low computational complexity of our baseline architecture. Significance. Our architecture emphasizes simplicity, offering easy integration of channel attention mechanisms, while maintaining a high degree of generalizability across datasets, making it a versatile and efficient solution for electroencephalogram motor imagery decoding within BCIs.

Meta‐heuristic optimization techniques for scheduling in multiple‐input multiple‐output multiple‐access channel system

SummaryIn a highly dense multiple‐input multiple‐output (MIMO) communication system, proper resource allocation is exceedingly challenging. The simultaneous selection of antennas at the base‐station (BS) and scheduling of users in an uplink multiple access channel (MAC) system add up to the signal processing load in the MIMO wireless system. The method chosen for combining the user scheduling and antenna selection must have a very low computational complexity. However, the exhaustive search algorithm (ESA), which achieves the best sum‐rate capacity in a MIMO‐MAC system, has a very high level of complexity. In this paper, we use a metaheuristic‐based optimization algorithms to optimize the sum‐rate capacity in a MAC system with significantly less computational complexity. First, we use the beluga whale optimization (BWO) metaheuristic method to solve the optimization problem. The BWO is a swarm‐based algorithm that draws inspiration from beluga whale behavior. An improved BWO (iBWO) is also presented and is utilized for the same optimization challenge. The proposed iBWO scheme improves the search capability of the algorithm and thereby ensures better exploration and exploitation capabilities of the beluga whales. We compare the performance of these algorithms with a standard differential evolution (DE) algorithm. The simulation results demonstrate that the proposed iBWO is a very competitive meta‐heuristic algorithm for solving optimization problems. The proposed iBWO algorithm achieves near optimal throughput in a MIMO‐MAC system. Also, the computational complexity related to the suggested optimization technique in a MIMO‐MAC system is found to be very less compared with the complexity related to ESA.

On-board model predictive control for autonomous lane keeping with fuzzy preview distance: Design and experiment

This paper concerns with the development of computationally efficient lane keeping control method for the autonomous vehicles. To obtain autonomous lane keeping, model predictive control (MPC) scheme was intensively investigated in the previous studies owing to its inherent advantage of dealing with constrained multivariable systems. However, the tradeoff problem between the good tracking performance and the low computational complexity is inevitably raised in developing of MPC-based lane keeping technology. To alleviate the conflict between performance and cost, an on-board MPC with fuzzy preview distance is designed in this study. A linear system dynamics model is used in the MPC design to reduce the computational cost, and a fuzzy logic algorithm is developed to select an appropriate preview distance for enhancing the MPC performance. Further, hardware-in-the-loop test is adopted to explore the effectiveness and efficiency of the proposed control method. In comparison to the proportional-integral controller, the experimental results show that the MPC is more sensitive to the selective value of fixed preview distance. Since the significant impact of preview distance selection on the MPC-based lane keeping performance, the fuzzy logic algorithm is of the essence in terms of selecting the appropriate preview distance for MPC enhancement under different vehicle speed and road curvature. Eventually, experimental results validate that the proposed fuzzy preview distance algorithm can effectively improve the MPC-based lane keeping performance for autonomous vehicles subject to limited computational resource.

Multiplicative Basis Function Based Adaptive DOA Estimation in Optimal MIMO Sparse Antenna Reconfiguration Model

Direction of arrival (DOA) estimation using sparse signal representation has gained significant attention in recent decades. The spatial signals utilized for DOA estimation are reconstructed through a novel compressive sensing (CS) approach. In this study, we used a CS approach to ascertain the adaptive underdetermined DOA for multiple inputs and multiple output (MIMO) sparse media. Accurate DOA estimation through the CS approach contributes to enhanced signal investigation and the suppression of undesired noise in a sparse channel. The method proposed herein encompasses the development and execution of a novel multiplicative basis function known as the multi-kernel supported non-negative sparse Bayesian learning (NNSBL) algorithm, which is implemented on predefined grid values. Concurrently, stochastic grey wolf optimization (GWO) is virtually applied to increase the degrees of freedom (DOF) on a minimum redundancy array (MRA) while considering an optimal antenna reconfiguration model. The advantages of this proposed approach include the generation of a distinct manifold matrix for precise beam steering, resulting in improved DOA estimation accuracy. Additionally, an augmentation of DOF using GWO was observed, with low computational complexity. The simulated result of proposed algorithm leads to the attainment of optimal minimum root mean square error (RMSE) across various optimal wavelengths of the stochastic sources. Finally, a comparative analysis of RMSE and convergence plots confirms the superior performance and high potentiality of the proposed method in comparison to existing techniques across different SNR values. Notably, there is significant reduction in RMSE which is suitable for precise DOA estimation in signal processing applications.

Sla-former: conformer using shifted linear attention for audio-visual speech recognition

Conformer-based models have proven highly effective in Audio-visual Speech Recognition, integrating auditory and visual inputs to significantly enhance speech recognition accuracy. However, the widely utilized softmax attention mechanism within conformer models encounters scalability issues, with its spatial and temporal complexity escalating quadratically with sequence length. To address these challenges, this paper introduces the Shifted Linear Attention Conformer, an evolved iteration of the conformer architecture. Shifted Linear Attention Conformer adopts shifted linear attention as a scalable alternative to softmax attention. We conducted a thorough analysis of the factors constraining the efficiency of linear attention. To mitigate these issues, we propose the utilization of a straightforward yet potent mapping function and an efficient rank restoration module, enhancing the effectiveness of self-attention while maintaining low computational complexity. Furthermore, we integrate an advanced attention-shifting technique facilitating token manipulation within attentional mechanisms, thereby enhancing information flow across various groups. This three-part approach enhances cognitive computations, particularly beneficial for processing longer sequences. Our model achieves exceptional Word Error Rates of 1.9% and 1.5% on the Lip Reading Sentences 2 and Lip Reading Sentences 3 datasets, respectively, showcasing its state-of-the-art performance in audio-visual speech recognition tasks.

Multi-objective design of parametric equivalent-input-disturbance approaches with application to vehicle suspension systems

A multi-objective parametric design method is developed for an equivalent-input-disturbance-based control system with mismatched disturbances. Different from the existing design methods that calculated control parameters by stability conditions, multiple performance objectives including unmeasurable output estimation, disturbance rejection, noise suppression, input refinement, and robustness, are simultaneously optimized in the developed design. First, for the sake of low computation complexity, the closed-loop system is represented. Then, the parameterization of the design conditions for multiple performance objectives is given. Further, a nonlinear optimization method is used to perform the performance-oriented system design. Lastly, the efficacy and benefits of the developed methods are demonstrated through simulations of a quarter-vehicle suspension system and comparisons with other alternative methods.

Interpretable online network dictionary learning for inferring long-range chromatin interactions.

Dictionary learning (DL), implemented via matrix factorization (MF), is commonly used in computational biology to tackle ubiquitous clustering problems. The method is favored due to its conceptual simplicity and relatively low computational complexity. However, DL algorithms produce results that lack interpretability in terms of real biological data. Additionally, they are not optimized for graph-structured data and hence often fail to handle them in a scalable manner. In order to address these limitations, we propose a novel DL algorithm called online convex network dictionary learning (online cvxNDL). Unlike classical DL algorithms, online cvxNDL is implemented via MF and designed to handle extremely large datasets by virtue of its online nature. Importantly, it enables the interpretation of dictionary elements, which serve as cluster representatives, through convex combinations of real measurements. Moreover, the algorithm can be applied to data with a network structure by incorporating specialized subnetwork sampling techniques. To demonstrate the utility of our approach, we apply cvxNDL on 3D-genome RNAPII ChIA-Drop data with the goal of identifying important long-range interaction patterns (long-range dictionary elements). ChIA-Drop probes higher-order interactions, and produces data in the form of hypergraphs whose nodes represent genomic fragments. The hyperedges represent observed physical contacts. Our hypergraph model analysis has the objective of creating an interpretable dictionary of long-range interaction patterns that accurately represent global chromatin physical contact maps. Through the use of dictionary information, one can also associate the contact maps with RNA transcripts and infer cellular functions. To accomplish the task at hand, we focus on RNAPII-enriched ChIA-Drop data from Drosophila Melanogaster S2 cell lines. Our results offer two key insights. First, we demonstrate that online cvxNDL retains the accuracy of classical DL (MF) methods while simultaneously ensuring unique interpretability and scalability. Second, we identify distinct collections of proximal and distal interaction patterns involving chromatin elements shared by related processes across different chromosomes, as well as patterns unique to specific chromosomes. To associate the dictionary elements with biological properties of the corresponding chromatin regions, we employ Gene Ontology (GO) enrichment analysis and perform multiple RNA coexpression studies.

Attack on optical cryptosystems by skip connection networks

Optical encryption methods, due to their efficient operation speed and parallel processing capabilities, hold significant importance in securing multidimensional and large-volume data. Enhancing the security of optical cryptosystems from the perspective of cryptanalysis holds significant importance currently. Presently, attack methods against optical encryption are complex, and the effectiveness of these attacks is insufficient. Security analysis solutions face limitations in both breadth and depth. Therefore, this paper proposes an attack on optical cryptosystems based on a skip connection network, demonstrating the susceptibility of optical cryptosystems to attacks based on neural network algorithms. The network model is trained on plaintext-ciphertext pairs, fitting equivalent keys without various additional conditions. It approximates plaintext information in high-dimensional space, directly obtaining corresponding plaintext through ciphertext information, expanding the applicability and enhancing the effectiveness of the attack scheme. Finally, the feasibility and effectiveness of the attack scheme were verified through computer simulations. The experiments indicate that the method proposed in this paper has low computational complexity, wide applicability, produces high-quality decrypted images, and high decipherment accuracy. This provides a universal approach for analyzing the security of various optical cryptosystems from the perspective of chosen plaintext attacks.

Stacking and ridge regression-based spectral ensemble preprocessing method and its application in near-infrared spectral analysis

Spectral preprocessing techniques can, to a certain extent, eliminate irrelevant information, such as current noise and stray light from spectral data, thereby enhancing the performance of prediction models. However, current preprocessing techniques mostly attempt to find the best single preprocessing method or their combination, overlooking the complementary information among different preprocessing methods. These preprocessing techniques fail to maximize the utilization of useful information in spectral data and restrict the performance of prediction models. This study proposed a spectral ensemble preprocessing method based on the rapidly developing ensemble learning methods in recent years and the ridge regression (RR) model, named stacking preprocessing ridge regression (SPRR), to address the aforementioned issues. Different from conventional ensemble learning methods, the proposed SPRR method applied multiple different preprocessing techniques to the original spectral data, generating multiple preprocessed datasets. These datasets were then individually inputted into RR base models for training. Ultimately, RR still served as the meta-model, integrating the output results of each RR base model through stacking. This approach not only produced diversity in base models but also achieved higher accuracy and lower computational complexity by using a single type of base model. On the apple spectral dataset collected by our team, correlation analysis showed significant complementary information among the data produced by different preprocessing techniques. This provided robust theoretical support for the proposed SPRR method. By introducing the currently popular averaging ensemble preprocessing method in a comparative experiment, the results of applying the proposed SPRR method to six datasets (apple, meat, wheat, olive oil, tablet, and corn) demonstrated that compared to the single preprocessing method and averaging ensemble preprocessing method, SPRR yielded the best accuracy and reliability for all six datasets. Furthermore, under the same conditions of the training and test datasets, the proposed SPRR method demonstrated better performance than the four commonly used ensemble preprocessing methods.

Prediction of 3D RNA Structures from Sequence Using Energy Landscapes of RNA Dimers: Application to RNA Tetraloops.

Access to the three-dimensional structure of RNA enables an ability to gain a more profound understanding of its biological mechanisms, as well as the ability to design RNA-targeting drugs, which can take advantage of the unique chemical environment imposed by a folded RNA structure. Due to the dynamic and structurally complex properties of RNA, both experimental and traditional computational methods have difficulty in determining RNA's 3D structure. Herein, we introduce TAPERSS (Theoretical Analyses, Prediction, and Evaluation of RNA Structures from Sequence), a physics-based fragment assembly method for predicting 3D RNA structures from sequence. Using a fragment library created using discrete path sampling calculations of RNA dinucleoside monophosphates, TAPERSS can sample the physics-based energy landscapes of any RNA sequence with relatively low computational complexity. We have benchmarked TAPERSS on 21 RNA tetraloops, using a combinatorial algorithm as a proof-of-concept. We show that TAPERSS was successfully able to predict the apo-state structures of all 21 RNA hairpins, with 16 of those structures also having low predicted energies as well. We demonstrate that TAPERSS performs most accurately on GNRA-like tetraloops with mostly stacked loop-nucleotides, while having limited success with more dynamic UNCG and CUYG tetraloops, most likely due to the influence of the RNA force field used to create the fragment library. Moreover, we show that TAPERSS can successfully predict the majority of the experimental non-apo states, highlighting its potential in anticipating biologically significant yet unobserved states. This holds great promise for future applications in drug design and related studies. With discussed improvements and implementation of more efficient sampling algorithms, we believe TAPERSS may serve as a useful tool for a physics-based conformational sampling of large RNA structures.

Novel Illumination-invariant Face Recognition Approach via Reflectance-luminance and Local Matching Model with Weighted Voting System

In this study, a novel approach has been introduced for face recognition that is unaffected by changes in illumination. This method is based on the reflectance-luminance model and incorporates local matching using a weighted voting technique to eliminate any artifacts present in the retina images. A total of 37 different linear and nonlinear filters were tested, including high pass and low pass filters, to achieve the reflectance component of the image, which remains invariant to changes in illumination. Among these filters, the maximum filter, which is a simple filter with low computational complexity, yielded the best results in extracting the illumination invariants. It was observed that the illumination invariants obtained through this method outperformed other methods such as QI, SQI, and image enhancement techniques in terms of recognition accuracy. Importantly, the proposed method does not require any prior knowledge about the facial shape or illumination conditions and can be applied to individual images independently. Unlike many existing methods, this approach does not rely on multiple images during the training stage and does not require any parameter selection to generate the illumination invariants. To further enhance the robustness of illumination, a weighted voting system was introduced. Certain regions of the image, which may adversely affect the recognition outcome due to poor illumination, occlusion, noise, or lack of distinctive information, were identified using predefined factors such as grayscale mean, image entropy, and mutual information. The proposed method was also compared to other face recognition methods in the presence of occlusions, and it demonstrated promising results outperforming existing methods. The Python algorithm successfully detects obstructed faces, genders, and ages in videos with a face matching accuracy between 80.9% to 96.9% based on proximity.

STF-Net: sparsification transformer coding guided network for subcortical brain structure segmentation.

Subcortical brain structure segmentation plays an important role in the diagnosis of neuroimaging and has become the basis of computer-aided diagnosis. Due to the blurred boundaries and complex shapes of subcortical brain structures, labeling these structures by hand becomes a time-consuming and subjective task, greatly limiting their potential for clinical applications. Thus, this paper proposes the sparsification transformer (STF) module for accurate brain structure segmentation. The self-attention mechanism is used to establish global dependencies to efficiently extract the global information of the feature map with low computational complexity. Also, the shallow network is used to compensate for low-level detail information through the localization of convolutional operations to promote the representation capability of the network. In addition, a hybrid residual dilated convolution (HRDC) module is introduced at the bottom layer of the network to extend the receptive field and extract multi-scale contextual information. Meanwhile, the octave convolution edge feature extraction (OCT) module is applied at the skip connections of the network to pay more attention to the edge features of brain structures. The proposed network is trained with ahybrid loss function. The experimental evaluation on two public datasets: IBSR and MALC, shows outstanding performance in terms of objective and subjective quality.

Energy-efficient resource allocation for bidirectional wireless power and information transfer over interference channels

Energy efficiency is an important performance metric for the sustainable operation of energy-limited communication networks. This study investigates an energy-efficient resource allocation strategy to maximize the energy efficiency of wireless-powered bidirectional communication over interference channels. The receivers in the system harvest energy and decode information simultaneously using a time switching or power splitting policy from data signals sent by transmitters in the forward link. The harvested energy is subsequently used to send response signals back to the transmitters in the backward link. Our objective is to jointly optimize the transmit power of the transmitters and the energy harvesting ratio of the receivers to maximize the energy efficiency of the entire system while maintaining the minimum required data rates in both the forward and backward links. Considering the non-convexity and NP-hardness of this optimization problem, we first apply nonlinear fractional programming and a dual method to derive an analytical expression for each control parameter and then propose an iterative algorithm to find suboptimal solutions with low computational complexity. The simulation results show that the proposed algorithm adjusts both the transmit power and energy harvesting ratio adaptively considering the co-channel interference and thus achieves near-optimal energy efficiency performance within a short computation time.

Volatility-based diversity awareness for distributed data storage of Mobile Crowd Sensing

With the widespread utilization of sensors and the rapid development of Mobile Crowd Sensing (MCS), the framework of distributed data storage, which makes use of the limited storage capacity of sensing devices, is providing an alternative to cloud-based data storage. Since the construction of the distributed storage framework for MCS has to be completed by Compressed Sensing (CS) theory, a reasonable measurement allocation becomes a no-brainer if we want to obtain more recovery precision of data without wasting measurement resources. To more efficiently address the above problems, we present a novel strategy of Volatility-Based Diversity Awareness for CS measurement allocation called VBDA, which collectively achieves lower computational complexity, adaptive sampling rate allocation for non-visual data, and high-quality reconstruction. We first process the target monitoring region in blocks. Then, we calculate the degree of data diversity of each area using the volatility, which is used to assess the importance of the different areas. Worthy note, based on the volatility calculation, we can obtain the magnitude of the variation in diversity that is only present in the real data, even when using very ambiguous recovered data. This is primarily due to the volatility unique design concept, which attempts to offset partially identical errors in adjacent recovered data using volatility calculations. Finally, to rationally allocate measurement resources, a volatility-based sampling rate allocation scheme is proposed. We further provide an implementation for practical deployments of VBDA in various related contexts. Numerous experimental show that the VBDA performs excellently. In comparison to the current state-of-the-art strategy, it improves data recovery accuracy by 12.2% without the use of any prior knowledge, while being adaptable to many different distribution types of data.

ED-SS based Cognitive Radio (CR) over Various Fading Channels for Modern Wireless Communications

Cognitive Radio (CR) has already been proved to provide the better solution for the problem of spectrum insufficiency in which every wireless system occupies the limited bandwidth spectrum (LBS) allotting to the licensed users. Sometimes, this LBS is empty in a certain place at a certain time that is called as spectrum holes estimated by spectrum sensing (SS). The spectrum sensing is the main technique used in cognitive radio and various spectrum sensing techniques are available in the literature such as: energy detection (ED), matched filter detection (MFD) and cyclo-stationary feature detection (CFD). In the present work, energy detection-spectrum sensing (ED-SS) based cognitive radio system has been implemented which can analyze the energy of the spectrum. The various fading channels such as additive white gaussian noise (AWGN), Rayleigh, and Nakagami have been incorporated in the present work over which the performance of cognitive radio has been analyzed and compared using the ED-SS technique. The experimental results show thatt the detection of the presence of primary user signal using proposed ED-SS technique is easier with low computational complexities in Cognitive Radio (CR) networks. The present work also demonstrates when the performance of proposed ED-SS technique over aforesaid three fading channels are compared, Nakagami-m fading channels exhibits superior improvement in Pd compared to Rayleigh and AWGN fading channels

YOLOv8-LMG: An Improved Bearing Defect Detection Algorithm Based on YOLOv8

In industrial manufacturing, bearings are crucial for machinery stability and safety. Undetected wear or cracks can lead to severe operational and financial setbacks. Thus, accurately identifying bearing defects is essential for maintaining production safety and equipment reliability. This research introduces an improved bearing defect detection model, YOLOv8-LMG, which is based on the YOLOv8n framework and incorporates four innovative technologies: the VanillaNet backbone network, the Lion optimizer, the CFP-EVC module, and the Shape-IoU loss function. These enhancements significantly increase detection efficiency and accuracy. YOLOv8-LMG achieves a mAP@0.5 of 86.5% and a mAP@0.5–0.95 of 57.0% on the test dataset, surpassing the original YOLOv8n model while maintaining low computational complexity. Experimental results reveal that the YOLOv8-LMG model boosts accuracy and efficiency in bearing defect detection, showcasing its significant potential and practical value in advancing industrial inspection technologies.

An efficient position optimization method based on improved genetic algorithm and machine learning for sparse array

Sparse array antennas are significant in communication and radar systems due to the advantages of high resolution, low complexity, and immunity to interference. In this letter, a novel array synthesis method based on an improved genetic algorithm (IGA) is proposed to optimize the antenna position in the sparse array. Different from the traditional generic methods with high computational complexity, a method combined with Gaussian process regression (GPR) prediction is proposed to optimize the antenna position with low computational complexity. The proposed method can minimize the peak sidelobe level (PSLL) during the beam scanning procedures. Moreover, the minimum antenna spacing constraint is also considered to ensure a physically achievable solution. Simulation results show that the proposed method can achieve a lower PSLL with fewer iterations and provide a global optimization solution for large array synthesis compared with existing methods. Furthermore, a prototype of the sparse array principle optimized by IGA is designed. Measurement results indicate that the optimized sparse array has good synthesis performance in high frequency.