This article presents a study on lung cancer detection based electronic nose technology. The pattern recognition algorithm is extremely crucial for an system, but the customary learning algorithms usually prefer majority class imbalance due to assumption of equal misclassification costs. To address this challenge, in article, we propose novel classification method named weighted discriminative extreme machine (WDELM) diagnosis. First, WDELM assigns different weight each particular sample by...

The long-term sensor drift phenomenon seriously restricts the performance of Electronic Nose (E-nose) systems in their various applications. Due to frequent recalibrations, traditional compensation methods are costly and laborious, limited due nonlinear dynamic properties drift. latest proposed Broad Learning System (BLS) has been confirmed be an efficient effective learning technique for many machine problems. However, BLS with cross-domain capability rarely studied. In this paper, a novel...

This article presents a novel sensor array optimization scheme for multisensor electronic nose detection system. A system architecture with is first proposed to implement the medical detection, including bacterial culture medium and animal wound infection detection. The efficiency evaluated by comparing field asymmetric ion mobility spectrometry (FAIMS) To further improve effect reduce number of sensors system, we then derive two procedures based on factor analysis Hilbert–Schmidt...

This paper considers a novel cooperative design for radar-communication spectral sharing system with multiple-input multiple-output (MIMO) structure based on mutual information optimization. We present new framework aiming at maximizing the radar information. Specifically, transmit sequences including waveform and communication codebook are designed simultaneously Subsequently, an iterative procedure is devised to maximize under constraints power, constant modulus requirement waveform,...

In this work, two electronic nose (E-nose) prototypes for lung cancer detection via breath analysis are built. The instrumental variation between them is a critical problem, as it impacts on the generalization performance of classification model. conventional domain adaption (DA) methods based assumption that data quality both devices high enough. However, not true in our case because concentrations volatile organic compounds (VOCs) biomarkers very low and acquisition process samples prone...

Industrial Internet of Things (IoT) connecting society and industrial systems represents a tremendous promising paradigm shift. With IoT, multimodal heterogeneous data from devices can be easily collected, further analyzed to discover device maintenance health related potential knowledge behind. IoT data-based fault diagnosis for is very helpful the sustainability applicability an ecosystem. But how efficiently use fuse this realize intelligent still challenge. In paper, novel Deep...

Gas sensor array (GSA) data usually has high-dimensional features and a small sample size. When classifier is directly used for GSA classification, it prone to overfitting high time cost. The traditional solution perform feature dimensionality reduction before classification. However, selecting suitable method time-consuming laborious, some useful classification may be lost after reduction, especially the weak response low-concentration gases. In this article, we proposed ensemble-based...

The drift of electronic nose (e-nose) sensors is the most challenging problem. An effective calibration method for e-nose proposed in this paper, which based on Projection to Convex Sets (POCS) and Extreme Learning Machine (ELM) method, including an automatic sampling platform sensor model. It realizes a unified universal both recognition regression application solve problem under long-term working conditions. Isopropanol acetone gases that respond were chosen as training Experiments show...

The intermittency, volatility, and randomness of the power output wind farms lead to serious quality problems at point common coupling (PCC) farms. Many waveform recording signals have high requirements for data transmission storage, which is not conducive disturbance analysis. Therefore, a compression method based on integrated wavelet atomic decomposition proposed improve ratio computational complexity. According characteristics signal, small-scale coherent dictionary was established...

Deep learning models can autonomously learn intricate feature representations directly from raw data, thereby alleviating the burden of engineering and achieving notable performance enhancements across various tasks. However, due to limited sample size gas sensor array data sets, deep are susceptible issues such as overfitting in odor classification This paper proposed a pyramid scattering convolutional network framework (PSCN) based on wavelet transform address above issues. PSCN first...

This study aims to enhance the robustness of electronic nose (e-nose) against noise interference and sensor damage through a biomimetic approach. Inspired by structure biological mechanisms mammalian olfactory system, we have proposed novel model for processing signals from gas-sensitive arrays e-nose. simulates functions epithelium bulb, encoding into spike sequences, incorporates concept wavelet filter banks scattering networks, effectively capturing local variations in signals. Utilizing...

Signal processing in electronic nose (e-nose) odor detection involves a series of complex and time-consuming steps, such as noise reduction, filtering, normalization signal preprocessing, well feature generation, selection, dimensionality reduction. When dealing with different gas sensors, these methods might require meticulous adjustments, each sensor need own specific strategy. Such strategies are often tedious overly reliant on intricate engineering, leading to operations poor...

Aiming at the problem of excessive configuration capacity in multi-energy microgrid (MG) planning, this paper first analyzes system optimization MG, establishes flow chart optimal sizing entire system, then proposes constraints considering load loss probability, and a planning model, selects appropriate method for solving. By using particle swarm (PSO) algorithm to calculate analyze, result can be obtained. Furthermore, simulation is conducted combined with case studies compare selection...

In this paper, we consider and analyze a novel spectrum sharing framework for the joint design of colocated MIMO radar communication system wherein both transceiver transmitter designs are investigated, in which filtered effective interference power is minimized under transmit power, similarity, mutual information (MI) requirement. order to solve original non-convex problem effectiveness, an alternating iterative algorithm exploited derived multi-subproblems. Finally, effectiveness proposed...

Regarding the accuracy problem of low-frequency oscillation parameter identification in current power system, this paper proposes a new method extracting os cillation modal parameters, combining Prony algorithm and Levenberg-Marquardt algorithm. First, is used to calculate initial values parameters oscillation, then iterate find optimal parameters. Through simulation 3-machine 9-node effectiveness proposed verified, it has certain advantages compared with traditional terms accuracy.

In this paper, we investigate the multiobjective optimization problem for array radar and multiple-input multiple-output (MIMO) communication spectrum sharing. The aim is to simultaneously implement rates output signal-to-interference plus noise ratio (SINR) optimization. converted into a singleobjective by utilizing weighted Tchebycheff method. We solve resulting formulation via alternative iteration, whose convergence guaranteed closed-form solutions can be derived. Finally, numerical...