Correlation Properties Research Articles

A Complex Network Important Node Identification Based on the KPDN Method

In complex networks, identifying influential nodes is of great significance for their wide application. The proposed method integrates the correlation properties of local and global, and in terms of global features, the K-shell decomposition method of fusion degree is used to improve the actual discrimination degree of each node. In terms of local characteristics, the Solton index is introduced to effectively show the association relationship between each node and adjacent nodes. Through the analysis and comparison of multiple existing methods, it is found that the proposed method can identify key nodes more accurately so as to help quickly disintegrate the network. The final manual network verification also shows that this method is also suitable for the identification of important nodes of small-world networks and community networks.

Novel technique to cancel MAI for OCDMA system

Abstract Code division multiple access in optical domain has a potential in next-generation long-haul transmission systems. However, multiple user interference is a snag in functioning of system employing OCDMA. A detailed analysis of multiple access interference (MAI) which causes a severe degradation in the system performance is done in this paper. An optimum MAI cancellation scheme has been proposed. This cancellation scheme depends not only on using the excellent correlation property of the double padded modified prime codes (DPMPC) as spreading sequence but also on employing minimum shift keying (MSK). The performance of the proposed cancellation scheme is compared with the existing techniques of FSK-OCDMA and PPM-OCDMA cancellation techniques. MSK-OCDMA cancellation technique achieves the lowest bit error rate of 10−20. It is also efficient in power. It achieves lowest bit error rate at same number of mean photon number when compared to the existing techniques. The investigations reveal that MSK-OCDMA cancellation technique outperforms FSK-OCDMA and PPM-OCDMA cancellation schemes.

Antimicrobial Activities of Mangrove Species in Southeast Asia: A Systematic Review

Antimicrobial resistance in Southeast Asia is a macro-level health dilemma that may cause substantial casualties annually. Fortunately, Southeast Asian mangroves are potent reservoirs of bioactive compounds with antimicrobial properties in correlation to their traditional usage. Hence, this study aimed to systematically review studies concerning antimicrobial activities of mangroves in Southeast Asia as well as provide a technological overview of its prospective use in pharmaceutical industry applications through patents. Through the PRISMA protocol, the search for peer-reviewed studies originated from Southeast Asia and published between 2010 to 2022 were conducted over databases such as CORE, Google Scholar, PubMed (MedLine), Science Direct, Semantic Scholar, Scopus, and Web of Science. Additionally, a patent search was also performed on the Espacenet Patent Search, Google Patents, National Institute of Industrial Property, and United States Patent and Trademark Office. The studies and patents were collated on Mendeley Reference Manager as well as tabulated and assessed on Microsoft Office Excel 2016. After the four-phase screening, 59 studies and one patent regarding antimicrobial activities of Southeast Asian mangroves passed the criteria for the systematic review. It was inferred that Southeast Asia constitutes potentially most species-diverse mangroves with highly varied antimicrobial properties and can form synergism with drugs. The existing studies and patents may provide enlightenment on the future path of studies and inventions which must be continually substantiated in animal and clinical experiments for prospective pharmaceutical industry use. In this way, mangroves can be conserved while antimicrobial resistance and anthropogenic activities can be lowered.

CCA-Net: A Lightweight Network Using Criss-Cross Attention for CSI Feedback

To enhance the performance of a massive multiple-input multiple-output (MIMO) system, the downlink channel state information (CSI) should be compressed and fed back to the base station (BS) effectively in a frequency division duplex (FDD) mode. In contrast to traditional compression methods that rely on the particular prior assumptions, deep learning (DL) based CSI feedback has the potential to learn informatively from the hidden complex models behind the data. However, most of the current DL based methods stay in the early stage that uses DL modules directly from image processing, while the distinctive channel semantic information has been largely ignored. In this letter, considering the spatial-temporal correlation property of CSI matrices, we propose an architecture based on criss-cross attention, called CCA-Net, which extracts the channel features more precisely by leveraging the contextual information along the delay and angle domain. Meanwhile, a real and imaginary part combination module is used to assist in extracting the internal features of CSI. In addition, to reap the computational resources at BSs, a scalable decoder is designed to extend the reconstruction abilities. Experiments demonstrate that our network achieves better normalized mean square error with lower flops than existing state-of-the-art algorithms.

Discrete-time quantum walk dispersion control through long-range correlations.

We investigate the evolution dynamics of inhomogeneous discrete-time one-dimensional quantum walks displaying long-range correlations in both space and time. The associated quantum coin operators of internal states are built to exhibit random inhomogeneity distribution of long-range correlations embedded in the time evolution protocol through a fractional Brownian motion with spectrum following a power-law behavior, S(k)∼1/k^{ν}. From extensive numerical simulations with averages over a large number of independent realizations of the phases of quantum coins, the power-law correlated disorder encoded in the coin phases is shown to give rise to a wide variety of spreading patterns of the qubit states, from localized to subdiffusive, diffusive, and superdiffusive (including ballistic) behavior, depending on the relative strength of the parameters driving the correlation degree. Dispersion control is possible in one-dimensional discrete-time quantum walks by tuning the long-range correlation properties assigned to the inhomogeneous quantum coin operator.

MIMO waveforms design with good correlation properties under bound constraints

We develop a new fast and efficient algorithm for designing multiple-input multiple-output (MIMO) waveforms with low auto- and cross-correlation sidelobes. Unlike the existing methods that consider unimodular signals, we design both signal amplitudes and phases to minimize the weighted integrated sidelobe level (WISL) of signals. Bound constraints on signal amplitudes are forced into the optimization to balance the transmitter efficiency and the degrees of optimization freedom, which also can provide a feasible set that facilitates efficient optimization. The amplitude-bounded signals design problem is formulated as a quartic optimization problem with bound constraints and equality (constant energy) constraints. We propose a novel algorithm based on the limited-memory Broyden-Fletcher-Goldfarb-Shanno with bound constraints (L-BFGS-B) method to solve the resultant problem. An algorithmic enhancement via nonlinear conjugate gradient (NCG) method is also developed to improve the proposed method. In addition, we derive the gradients of the cost function with respect to (w.r.t.) the signal amplitudes and phases and calculate them with the help of fast Fourier transformation (FFT), which makes the proposed method very computationally efficient. Numerical simulations illustrate that the proposed algorithm outperforms the state-of-the-art methods in both optimization performance and computational efficiency.

Dynamics of non-equilibrium concentration fluctuations during free-diffusion in highly stratified solutions of glycerol and water.

We investigate the non-equilibrium fluctuations occurring during free diffusion between two solutions of glycerol and water with various concentration differences. The non-linearity of the system, determined by the strong stratification of the sample, requires introducing an interpretation model able to characterize the dependence of the correlation properties of the non-equilibrium fluctuations on the local thermophysical variables of the system. The proposed model allows us to characterize the dynamics of non-equilibrium fluctuations in the presence of a wide range of relaxation times determined by the strong stratification of the sample, at variance with the cumulant methods commonly used in dynamic light scattering experiments, which work well in the presence of a moderate dispersion of relaxation times.

From celestial correlators to AdS, and back

We present a general relation between celestial correlation functions in d-dimensions and Witten diagrams in (d + 1)-dimensional Euclidean anti-de Sitter (EAdS) space, to all orders in perturbation theory. Contact diagram processes are proportional to contact Witten diagrams and particle exchanges can be recast as a continuum of particle exchanges in EAdS where the exchanged particles carrying unitary Principal Series representations of SO(d + 1, 1). One can then try to import familiar EAdS techniques to study the properties of celestial correlators. In this work we use this relation to infer the analytic structure of the spectral density in the conformal partial wave expansion of celestial correlators which, at least perturbatively, should be a meromorphic function of the spectral parameter. We also discuss non-perturbative constraints from unitarity in Euclidean Conformal Field Theory, which requires positivity of the spectral density. This extends similar relations recently uncovered between boundary correlation functions in de Sitter space and Witten diagrams in EAdS, suggesting that EAdS could play a central role in efforts towards holography for all lambdas.

Features of application of fractal Gaussian noise type signals in telecommunication systems

Problem statement. At present, the properties of a separate type of pseudorandom signals such as fractal Gaussian noise remain insufficiently investigated. The existing discrete model of fractal Gaussian noise is characterized by the value of the Hurst exponent and the time scaling factor, the influence of which on the properties of fractal Gaussian noise is insufficiently investigated. The issue of synthesis of new types of fractal broadband signals of complex design, which can be used in noise-resistant telecommunication systems, is relevant. Objective. Based on the mathematical model of fractal Gaussian noise, investigate the effect of the time scaling factor on the statistical, correlation and energy properties of fractal Gaussian noise signals. Results. By the method of averaging over realizations the dependences of energy, correlation and statistical properties of signals of fractal Gaussian noise type, which can be used for coding information bits, on the time scaling factor were obtained. Practical significance. According to the results of research on the properties of fractal Gaussian noise, the regularities of the passage of self-similar flows of Internet traffic in the approximation that they are self-similar and are described by a mathematical model of fractal Gaussian noise are established. When the intensity is increased up to requests/hour, the results of modeling the average transit time of traffic through the system for flows with different types of distribution show significant differences in the value of the average transit time of each request.

Sub-6 GHz to mmWave for 5G-Advanced and Beyond: Channel Measurements, Characteristics and Impact on System Performance

In the 5G-Advanced and beyond systems, multi-frequency cooperative networking will become an inevitable development trend. However, the channels have not been fully investigated at multi-frequency bands and in multi-scenarios by using the same channel sounder, especially for the sub-6 GHz to millimeter-wave (mmWave) bands. In this paper, we carry out channel measurements at four frequency bands (3.3, 6.5, 15, and 28 GHz) in two scenarios including Urban Micro (UMi) and Outdoor-to-Indoor (O2I) with the same channel sounder. The channel characteristics are extracted and modeled, including path loss (PL), shadow fading, frequency dependence of cluster features, root mean square (RMS) delay spread (DS), Ricean K-factor, and the correlation properties. We mainly focus on the analysis of large-scale parameters and more consideration of the link budget coverage problem. We present the frequency dependence model of the channel characteristics. Among them, in the non-line-of-sight (NLoS) condition, it is found that except for theoretical value brought by higher frequency, additional path loss increment will be generated. Based on these channel characteristics, the impact on performance of the wireless system is analyzed including cell coverage radius, data rate and bit error rate (BER). The results can give an insight into the spectrum selection and optimization in 5G-Advanced and beyond multi-frequency communication systems.

Stepped-Frequency Binary Offset Carrier Modulation for Global Navigation Satellite Systems

In this paper, we propose Stepped-Frequency Binary Offset Carrier (SFBOC) modulation for Global Navigation Satellite Systems (GNSS) that use spread spectrum signals. SFBOC modulation generates a Binary Offset Carrier (BOC) signal of which the subcarrier frequency varies stepwise within each period of the pseudorandom-noise (PRN) code. The generated SFBOC signal can have a sharp autocorrelation function (ACF) output and an almost flat spectrum over a wide frequency band. These characteristics result in correlation properties, multipath mitigation, and anti-interference that are superior to those of conventional GNSS modulations. To testify, we compare the performance of SFBOC in various aspects to conventional GNSS modulations and other time-varying subcarrier frequency (TVSF) modulations. From numerical simulations, we show that the proposed SFBOC has higher ranging accuracy, lower tracking ambiguity, and more robustness to noise, interference, and multipath than the conventional GNSS modulations. In addition, we demonstrate that the proposed SFBOC moulation has two advantages while other modulations with TVSF cannot have both; that SFBOC is robust to code-Doppler and that, when filterbank is employed, SFBOC improves the signal-to-noise ratio (SNR) significantly without any loss of weak early arrival path in the Non-line of sight (NLOS) multipath environments.

Statistical upscaling workflow for warm solvent injection processes – Longitudinal and transverse dispersivity and thermal conductivity

Accurately simulating, upscaling, and optimizing the warm solvent injection process (WSI) is challenging due to its complex heat and mass transfer mechanisms. This manuscript presents a comprehensive two-step flow-based upscaling workflow designed explicitly for WSI simulation in heterogeneous reservoirs. The technique builds upon previous research on particle transport in single-phase flow and extends the upscaling method to incorporate multiple correlated properties relevant to the process physics. Specifically, the upscaling methodology considers the correlation between longitudinal and transverse dispersivities. The influence of scale variation on thermal conductivity is examined. The proposed statistical upscaling technique does not rely on explicit assumptions about the distribution of heterogeneity. The results highlight the directionality in dispersivity. Due to the limited temperature variation observed in the case study, upscaling thermal conductivity has little impact. The method's efficiency can be verified by coarser model simulations that capture similar fine-scale model responses with less computational run time.

MARC-Net: Terrain Classification in Parallel Network Architectures Containing Multiple Attention Mechanisms and Multi-Scale Residual Cascades

To address the problem that traditional deep learning algorithms cannot fully utilize the correlation properties between spectral sequence information and the feature differences between different spectra, this paper proposes a parallel network architecture land-use classification based on a combined multi-head attention mechanism and multiscale residual cascade called MARC-Net. This parallel framework is firstly implemented by deeply mining the features generated by grouped spectral embedding for information among spectral features by adding a multi-head attention mechanism, which allows semantic features to have expressions from more subspaces while fully considering all spatial location interrelationships. Secondly, a multiscale residual cascade CNN (convolutional neural network) is designed to fully utilize the fused feature information at different scales, thereby improving the network’s ability to represent different levels of information. Lastly, the features obtained by the multi-head attention mechanism are fused with those obtained by the CNN, and the merged resultant features are downgraded through the fully connected layer to obtain the classification results and achieve pixel-level multispectral image classification. The findings show that the algorithm proposed in this paper has an aggregate precision of 97.22%, compared to that of the Vision Transformer (ViT) with 95.08%; its performance on the Sentinel-2 dataset shows a huge improvement. Moreover, this article mainly focuses on the change rate of forest land in the study area. The Forest land area was 125.1143 km2 in 2017, 105.6089 km2 in 2019, and 76.3699 km2 in 2021, with an increase of 15.59%, an decrease of 0.97%, and increase of 14.76% in 2017–2019, 2019–2021 and 2017–2021, respectively.

Construction of Local-Shape-Controlled Quartic Generalized Said-Ball Model

Said-Ball curves and surfaces are extensively applied in the realm of geometric modeling. Their appearance is only decided by the control points, which produces a great deal of inconvenience for the shape design of sophisticated products. To overcome this defect, we construct a novel kind of quartic generalized Said-Ball (QGS-Ball, for short) curves and surfaces, which contain multiple shape parameters, and the global and local shape can be easily modified via shape parameters. The specific research contents are as follows: Firstly, the QGS-Ball basis functions carrying multiple shape parameters are defined, and the correlative properties are proved. Secondly, the QGS-Ball curve is proposed according to the QGS-Ball basis functions, and the effect of shape parameters on the curve is discussed. Thirdly, in view of the constructed QGS-Ball curve, we further propose the combined quartic generalized Said-Ball (CQGS-Ball, for short) curves, and deduce the conditions of first-order and second-order geometric continuity (namely, G1 and G2 continuity). Finally, the QGS-Ball surface is defined by tensor product method, and the influence of shape parameters on the surface is analyzed. The main contribution of this article is to construct the QGS-Ball curve model, and deduce the G1 and G2 geometric joining conditions of QGS-Ball curves. Combined with some modeling examples, it further illustrates that the QGS-Ball curve as a new geometric model provides a powerful supplement for the geometric design of sophisticated form in computer-aided design (CAD) and computer-aided manufacturing (CAM) systems.

Joint Learning of Correlation-Constrained Fuzzy Clustering and Discriminative Non-Negative Representation for Hyperspectral Band Selection

Hyperspectral band selection plays an important role in overcoming the curse of dimensionality. Recently, clustering-based band selection methods have shown promise in the selection of informative and representative bands from hyperspectral images (HSIs). However, most existing clustering-based band selection methods involve the clustering of original HSIs, limiting their performance because of the high dimensionality of hyperspectral bands. To tackle this problem, a novel hyperspectral band selection method termed joint learning of correlation-constrained fuzzy clustering and discriminative non-negative representation for hyperspectral band selection (CFNR) is presented. In CFNR, graph regularized non-negative matrix factorization (GNMF) and constrained fuzzy C-means (FCM) are integrated into a unified model to perform clustering on the learned feature representation of bands rather than on the original high-dimensional data. Specifically, the proposed CFNR aims to learn the discriminative non-negative representation of each band for clustering by introducing GNMF into the model of the constrained FCM and making full use of the intrinsic manifold structure of HSIs. Moreover, based on the band correlation property of HSIs, a correlation constraint, which enforces the similarity of clustering results between neighboring bands, is imposed on the membership matrix of FCM in the CFNR model to obtain clustering results that meet the needs of band selection. The alternating direction multiplier method is adopted to solve the joint optimization model. Compared with existing methods, CFNR can obtain a more informative and representative band subset, thus can improve the reliability of hyperspectral image classifications. Experimental results on five real hyperspectral datasets demonstrate that CFNR can achieve superior performance compared with several state-of-the-art methods.

Solid-liquid phase equilibrium of 2, 4-epibrassinolide in binary mixed solvents: Solubility determination, model correlation, preferential solvation and thermodynamic properties

The solubility of 2, 4-epibrassinolide in four binary mixed solvents including (ethanol + water, isopropanol + water, acetone + water, and acetonitrile + water) with numerous compositions was determined at temperature of 283.15 K to 323.15 K by a laser dynamic method. The solubility of 2, 4-epibrassinolide in all four binary solvent mixtures increases with increasing temperature. At a given temperature, it shows a trend of increasing and then decreasing with increasing mole fraction of water. In the same solvent composition, the order of solubility is isopropanol + water > ethanol + water > acetone + water > acetonitrile + water. The experimental solubility data of 2, 4-epibrassinolide was correlated with the modified Apelblat, Van’t Hoff, NRTL, Wilson, and three-suffix Margules model. The preferential solvation mechanism of 2, 4-epibrassinolide in four aqueous solutions was also discussed by IKBI method. Furthermore, the Wilson model was worked to analyze the thermodynamic mixing properties of the solutions, including ΔmixG, ΔmixH and ΔmixS. All ΔmixG values are negative indicating that the mixing process of 2, 4-epibrassinolide in these solvents is spontaneous.

Comportamiento del mercado laboral profesional en la ciudad de Quito

This publication is the result of an investigation carried out from the application of a survey to a sample of 79 companies, from different economic sectors, in the city of Quito, in order to determine the profile of professionals demanded by the organizations consulted. From the application of a descriptive and correlational approach, properties and characteristics required of human talent could be determined. As a conclusion, from the present research work, it was established that the university has to train professionals trained both in the use of disruptive technological tools and in the programming of managerial information that supports decision making, data administration, the generation of reports of an executive nature, that present a strong conviction towards aspects related to professional ethics, human, social and corporate values and that adapt to the growth needs of the national productive apparatus.

Mechanical Strength and Microstructure of GGBS-SCBA based Geopolymer Concrete

This paper deals with the attempt to develop and study the performance of ground granulated blast furnace slag (GGBS) and sugarcane bagasse ash (SCBA) based sustainable geopolymer concrete. NaOH (8M, 10M, and 12M) and Na2SiO3 were used as alkaline activators with a ratio of 2.5. SCBA mainly acted as amorphous silica and has been utilized as a substitute material for GGBS. The effect of SCBA contents (0%, 5%, 10%, 15% & 20% by the mass of binder) in terms of fresh, hardened, microstructural, and correlation properties of geopolymer concrete developed have been evaluated. Different tests such as the slump cone test, compression test, split tensile test, flexure test, and ultrasonic pulse velocity test were conducted. Scanning electron microscopy, Energy dispersive analysis, and X-ray diffraction analysis were investigated for understanding the microstructural properties. The research findings have shown that with an increase in molarity from 8M to 12M there is an increase in the strength properties of geopolymer concrete. The results in this current study show that 28 days compressive strength was found to increase by 4–15% when the NaOH molarity was increased from 8M to 10M and 8–21% when the NaOH molarity was increased from 8M to 12M. The geopolymer concrete developed with 20% SCBA and 80% GGBS with 8M NaOH solution and SS/SH ratio of 2.5 can be used for a target strength of 30–35 MPa. Scanning electron microscope images show a packed and dense matrix, which clearly outlines the reason behind the attainment of higher strength in higher molarity of GGBS-SCBA based geopolymer concrete samples and the presence of CASH gel confirmed this in the geopolymer matrix. Furthermore, there is a strong correlation between the experimental findings and the model equations proposed. These presented models will be useful in improving the strength of geopolymer concrete incorporating agricultural and industrial wastes.

PO-05-133 PROLONGED NEUROMODULATION EFFECTS AFTER PULMONARY VEIN ISOLATION IN PATIENTS WITH PAROXYSMAL ATRIAL FIBRILLATION: INSIGHT FROM THE NONLINEAR HEART RATE DYNAMICS

Heart rate variability (HRV) reveals information undetected by direct electrocardiogram morphology evaluation. Prior research indicated that patients with paroxysmal atrial fibrillation (PAF) had altered a short-term linear HRV after pulmonary vein isolation (PVI). However, the long-term changes in the nonlinear heart rate dynamics before and after PVI remain unclear. The objective was to investigate the long-term nonlinear HRV changes before and after PVI in PAF patients. We enrolled 42 drug-refractory symptomatic PAF patients who underwent PVI. Holter monitoring were done less than 1 year before, 1∼3, and 6∼12 months after PVI, respectively. The nonlinear HRV variables, including sample entropy (SampEn), multi-scale entropy (MSE), detrended fluctuation analysis (DFA), and Poincaré Plot analysis, were analyzed. Significant reductions in SD1, SD1/SD2, MSE 1∼5, MSE 6∼20, and SampEn occurred 6∼12 months after when compared to those before PVI, respectively. SD2, MSE slope 5, DFA slope 1 decreased 1∼3 months after PVI when compared to those before PVI, respectively, whereas increased significantly 6∼12 months after PVI when compared to those 1∼3 months after PVI, respectively. DFA slope2 increased significantly after PVI when compared to that before PVI. Detailed parameters are shown in the Table 1. As measured by DFA slope2, fractal correlation properties increased significantly following PVI. Heart rate complexity (MSE 1∼5, MSE 6∼20, and SampEn) and unpredictability of an RR time series (SD1/SD2) decreased significantly after PVI, whereas long-term RR interval (SD2) and short-term complexity (MSE slope 5) decreased in the 1∼3 months after PVI and then increased in 6∼12 months after PVI. These results indicated that the neuromodulation effects from PVI was far more prolonged and extensively than expected.Tabled 1Table 1. Serial changes in the HRV parametersPatients with PAF who underwent PVI (n = 42)< 1 year pre-PVI1∼3 months post-PVI6∼12 months post-PVISD121.16 ± 8.416.31 ± 11.08∗14.19 ± 8.63∗SD2177.58 ± 44.38134.92 ± 37∗146.46 ± 37.52∗+SD1/SD20.12 ± 0.060.12 ± 0.090.1 ± 0.06∗+DFA slope11.11 ± 0.240.8 ± 0.31∗0.94 ± 0.31∗+DFA slope21.07 ± 0.11.16 ± 0.13∗1.15 ± 0.11∗MSE 1∼55.62 ± 1.824.76 ± 1.74∗4.72 ± 1.59∗MSE 6∼2019.81 ± 5.6716.01 ± 5.6∗17.3 ± 4.75∗MSE slope50.06 ± 0.07-0.01 ± 0.09∗0.03 ± 0.07+SampEn1 ± 0.321.01 ± 0.290.9 ± 0.26∗+Data are expressed as (mean ± SD); DFA slope1: Detrended fluctuation analysis slope 1; DFA slope2: Detrended fluctuation analysis slope 2; MSE 1∼5, the summation of entropy values of scales 1 to 5; MSE 6∼20, the summation of entropy values of scales 6 to 20; MSE slope 5, the linear-fitted slope of scales 1 to 5; SD1, Poincaré plot standard deviation perpendicular the line of identity; SD2, Poincaré plot standard deviation along the line of identity; SD1/SD2, SD1-to-SD2 ratio; SampEn, sample entropy; *: p < 0.05 versus < 1 year pre-PVI; +: p < 0.05 versus 1∼3 months post-PVI Open table in a new tab

A lightweight neural network compatible with high-correlation broadband optical filters in computational spectroscopy

Computational spectrometer is powerful for portable in situ applications. To achieve better signal to noise ratio (SNR), broadband optical filters are more commonly introduced. As for conventional sparse signals reconstruction algorithms like gradient projection, its performance relies on the low correlation property of optical filters, which poses severe challenge for filters design and fabrication. In this study, we propose a lightweight neural network(NN) named HBOF-Net that can be compatible with high-correlation broadband optical filters to renovate the situation. The spectra reconstruction performs very well under a group of high correlated optical filters and multiscale training datasets. Firstly, we propose a group of various structure design of photonic crystal(PC) slabs and select every 16 to construct a spectrometer megapixel randomly. Then we extract different percentage of spectra from CAVE and ICVL dataset to build up multiscale training dataset groups. Afterwards, the training of the network has been carried out and the testing curve indicate there is no over fitting. A total of 16 trainings have been carried out and the results have been compared and discussed. Lastly, the experiment demonstrates that the mean square error(MSE) of the reconstructed spectra by HBOF-Net can reach 1E-6 by maximum correlation coefficient of 0.93 under multi-scale datasets which is 90%, 10%, 5%, and 3% of the entire dataset. This study successfully prove that NN can be compatible with high correlated optical filters, which is a significant leap in the development of computational spectrometers. It provides a good reference for computational spectrometers based on broadband and high-correlation optical filters.