This study proposes a new component of the composite loss function minimised during training of the Super-Resolution(SR)algorithms—the normalised structural similarity index loss LSSIMN,which has the potential to imp...This study proposes a new component of the composite loss function minimised during training of the Super-Resolution(SR)algorithms—the normalised structural similarity index loss LSSIMN,which has the potential to improve the natural appearance of reconstructed images.Deep learning-based super-resolution(SR)algorithms reconstruct high-resolution images from low-resolution inputs,offering a practical means to enhance image quality without requiring superior imaging hardware,which is particularly important in medical applications where diagnostic accuracy is critical.Although recent SR methods employing convolutional and generative adversarial networks achieve high pixel fidelity,visual artefacts may persist,making the design of the loss function during training essential for ensuring reliable and naturalistic image reconstruction.Our research shows on two models—SR and Invertible Rescaling Neural Network(IRN)—trained on multiple benchmark datasets that the function LSSIMN significantly contributes to the visual quality,preserving the structural fidelity on the reference datasets.The quantitative analysis of results while incorporating LSSIMN shows that including this loss function component has a mean 2.88%impact on the improvement of the final structural similarity of the reconstructed images in the validation set,in comparison to leaving it out and 0.218%in comparison when this component is non-normalised.展开更多
Considering high-order digital modulation schemes, the bottleneck in consumer products is the detector rather than the modulator. The complexity of the optimal a posteriori probability （APP） detector increases expon...Considering high-order digital modulation schemes, the bottleneck in consumer products is the detector rather than the modulator. The complexity of the optimal a posteriori probability （APP） detector increases exponentially with respect to the number of modulated bits per data symbol. Thus, it is necessary to develop low-complexity detection algorithms with an APP-like performance, especially when performing iterative detection, for example in conjunction with bit interleaved coded modulation. We show that a special case of superposition modulation, dubbed Direct Superposition Modulation （DSM）, is particularly suitable for complexity reduction at the receiver side. As opposed to square QAM, DSM achieves capacity without active signal shaping. The main contribution is a low-cost detection algorithm for DSM, which enables iterative detection by taking a priori information into account. This algorithm exploits the approximate piecewise linear behavior of the soft outputs of an APP detector over the entire range of detector input values. A theoretical analysis and simulation results demonstrate that at least max-log APP performance can be reached, while the complexity is significantly reduced compared to classical APP detection.展开更多
To reduce the energy consumption of the LTE-A system,a distributed energy-saving mechanism based on Co MP(Co MPDESM) is proposed to solve the inadequate coverage problem under the dormant cells.First,the network is di...To reduce the energy consumption of the LTE-A system,a distributed energy-saving mechanism based on Co MP(Co MPDESM) is proposed to solve the inadequate coverage problem under the dormant cells.First,the network is divided into clusters based on the equivalent cell principle.Then,we transfer global optimization into a group of subproblems.Second,a joint processing-based cooperative cell selection model is constructed to determine cooperative cells and dormant cells.Third,the compensative cells with a determined threshold are selected to control users' access.Finally,a simulation is implemented in Matlab.Results show that the energy-saving rate can reach 36.4% and that the mechanism meets the network coverage requirement.Thus,joint processing can be effectively applied in an energy saving mechanism and used to improve the network performance of edge users without increasing transmission power.展开更多
Since precise self-position estimation is required for autonomous flight of aerial robots, there has been some studies on self-position estimation of indoor aerial robots. In this study, we tackle the self-position es...Since precise self-position estimation is required for autonomous flight of aerial robots, there has been some studies on self-position estimation of indoor aerial robots. In this study, we tackle the self-position estimation problem by mounting a small downward-facing camera on the chassis of an aerial robot. We obtain robot position by sensing the features on the indoor floor.In this work, we used the vertex points(tile corners) where four tiles on a typical tiled floor connected, as an existing feature of the floor. Furthermore, a small lightweight microcontroller is mounted on the robot to perform image processing for the onboard camera. A lightweight image processing algorithm is developed. So, the real-time image processing could be performed by the microcontroller alone which leads to conduct on-board real time tile corner detection. Furthermore, same microcontroller performs control value calculation for flight commanding. The flight commands are implemented based on the detected tile corner information. The above mentioned all devices are mounted on an actual machine, and the effectiveness of the system was investigated.展开更多
With the existing anonymous authentication schemes based on biometrics, the user and the server can create the same session key after mutual authentication. If the anonymous authentication scheme is applied in the ele...With the existing anonymous authentication schemes based on biometrics, the user and the server can create the same session key after mutual authentication. If the anonymous authentication scheme is applied in the electronic medical environment, it is also necessary to consider that the patient may access multiple hospital servers. Based on three factors of smart card, random number and biometrics, an anonymous authentication scheme in the electronic medical environment is proposed. In order to reduce the burden of the medical registration and certification center(HC), in the proposed anonymous authentication scheme, the patient only needs to register with HC once, then he/she can apply for visiting each hospital that has joined the medical servers. Security analysis shows that the proposed scheme has anonymity and dual authentication, and can resist various types of attacks, such as insider attack, modification attack, replay attack and smart card loss attack. Efficiency analysis shows that the calculation cost of the proposed scheme in the registration and login phase is lower, and it is slightly higher than Lei's scheme and Khan et al's scheme in the authentication phase. The proposed scheme can not only resist various types of attacks, but also support dual authentication and multi-server environment. With a little modification, the proposed scheme can also be used to other application scenarios requiring anonymous authentication.展开更多
In order to protect the user's privacy identity,authentication requires anonymous authentication.Anonymous authentication is divided into unconditional anonymous authentication and traceable anonymous authenticati...In order to protect the user's privacy identity,authentication requires anonymous authentication.Anonymous authentication is divided into unconditional anonymous authentication and traceable anonymous authentication.Unconditional anonymous authentication can verify that the user belongs to an anonymous set,but the user's true identity cannot be obtained.However,in some applications,it is necessary to trace the true identity of the user.Therefore,a traceable anonymous authentication scheme is proposed.In order to prevent random tracing,the proposed scheme uses threshold joint tracing.When the identity of the authenticator needs to be traced,the threshold number of members can jointly trace the identity of the authenticator.In some special network applications such as anonymous electronic voting,in order to prevent repeated authentications and repeated elections,it is necessary to verify whether the two authentication signatures are signed by the same user without revealing the true identity of the user.Therefore,the proposed anonymous authentication scheme should have selective linkability.In order to achieve linkable authentication,the linkable tag is embedded by linkable ring signature.Compared with similar schemes through the simulation experiments,the implementation time of the proposed scheme is slightly better than other schemes.展开更多
This paper presents a new algorithm for clustering a large amount of data.We improved the ant colony clustering algorithm that uses an ant’s swarm intelligence,and tried to overcome the weakness of the classical clus...This paper presents a new algorithm for clustering a large amount of data.We improved the ant colony clustering algorithm that uses an ant’s swarm intelligence,and tried to overcome the weakness of the classical cluster analysis methods.In our proposed algorithm,improvements in the efficiency of an agent operation were achieved,and a new function "cluster condensation" was added.Our proposed algorithm is a processing method by which a cluster size is reduced by uniting similar objects and incorporating them into the cluster condensation.Compared with classical cluster analysis methods,the number of steps required to complete the clustering can be suppressed to 1% or less by performing this procedure,and the dispersion of the result can also be reduced.Moreover,our clustering algorithm has the advantage of being possible even in a small-field cluster condensation.In addition,the number of objects that exist in the field decreases because the cluster condenses;therefore,it becomes possible to add an object to a space that has become empty.In other words,first,the majority of data is put on standby.They are then clustered,gradually adding parts of the standby data to the clustering data.The method can be adopted for a large amount of data.Numerical experiments confirmed that our proposed algorithm can theoretically applied to an unrestricted volume of data.展开更多
A challenging task when applying high-order digital modulation schemes is the complexity of the detector. Particularly, the complexity of the optimal a posteriori probability （APP） detector increases exponentially w...A challenging task when applying high-order digital modulation schemes is the complexity of the detector. Particularly, the complexity of the optimal a posteriori probability （APP） detector increases exponentially with respect to the number of bits per data symbol. This statement is also true for the Max-Log-APP detector, which is a common simplification of the APP detector. Thus it is important to design new detection algorithms which combine a sufficient performance with low complexity. In this contribution, a detection algorithm for two- dimensional digital modulation schemes which cannot be split-up into real and imaginary parts （like phase shift keying and phase-shifted snperposition modulation （PSM）） is proposed with emphasis on PSM with equal power allocation. This algorithm exploits the relationship between Max-Log-APP detection and a Voronoi diagram to determine planar surfaces of the soft outputs over the entire range of detector input values. As opposed to state-of-the-art detectors based on Voronoi surfaces, a priori information is taken into account, enabling iterative processing. Since the algorithm achieves Max-Log-APP performance, even in the presence of a priori information, this implies a great potential for complexity reduction compared to the classical APP detection.展开更多
The strong La line 6520.644 ? is present in a Fourier Transform Spectrum (signal to noise ratio of 240), but its wavelength in commonly used tables (e.g. [1]) is given as 6520.770 ?, while in [2] the wavelength is giv...The strong La line 6520.644 ? is present in a Fourier Transform Spectrum (signal to noise ratio of 240), but its wavelength in commonly used tables (e.g. [1]) is given as 6520.770 ?, while in [2] the wavelength is given as 6520.74 ?, unclassified, with the remark “h” (hazy). The line could not be classified using known energy levels of the La atom (La I) and its first ion (La II). It appears as a single broadened peak. By a combination of laser optogalvanic spectroscopy, laser-induced fluorescence and Doppler-reduced saturation spectroscopy we could introduce a new even parity La I energy level, 35449.041 cm -1, J = 13/2, with hyperfine (hf) constants A = ?8.0(5) MHz, B = 10(10) MHz. For a second, up to now unknown neighbouring La I line (wavelength 6519.869 ?) we introduced another even parity energy level involved in the formation of the line, 41207.994 cm?-1, J = 13/2, A = 91.6(5) MHz, B = 170(50) MHz. We tried also to explain why in old tables the wavelength given was so different.展开更多
基金support from the following institutional grant.Internal Grant Agency of the Faculty of Economics and Management,Czech University of Life Sciences Prague,grant no.2023A0004(https://iga.pef.czu.cz/,accessed on 6 June 2025).
文摘This study proposes a new component of the composite loss function minimised during training of the Super-Resolution(SR)algorithms—the normalised structural similarity index loss LSSIMN,which has the potential to improve the natural appearance of reconstructed images.Deep learning-based super-resolution(SR)algorithms reconstruct high-resolution images from low-resolution inputs,offering a practical means to enhance image quality without requiring superior imaging hardware,which is particularly important in medical applications where diagnostic accuracy is critical.Although recent SR methods employing convolutional and generative adversarial networks achieve high pixel fidelity,visual artefacts may persist,making the design of the loss function during training essential for ensuring reliable and naturalistic image reconstruction.Our research shows on two models—SR and Invertible Rescaling Neural Network(IRN)—trained on multiple benchmark datasets that the function LSSIMN significantly contributes to the visual quality,preserving the structural fidelity on the reference datasets.The quantitative analysis of results while incorporating LSSIMN shows that including this loss function component has a mean 2.88%impact on the improvement of the final structural similarity of the reconstructed images in the validation set,in comparison to leaving it out and 0.218%in comparison when this component is non-normalised.
文摘Considering high-order digital modulation schemes, the bottleneck in consumer products is the detector rather than the modulator. The complexity of the optimal a posteriori probability （APP） detector increases exponentially with respect to the number of modulated bits per data symbol. Thus, it is necessary to develop low-complexity detection algorithms with an APP-like performance, especially when performing iterative detection, for example in conjunction with bit interleaved coded modulation. We show that a special case of superposition modulation, dubbed Direct Superposition Modulation （DSM）, is particularly suitable for complexity reduction at the receiver side. As opposed to square QAM, DSM achieves capacity without active signal shaping. The main contribution is a low-cost detection algorithm for DSM, which enables iterative detection by taking a priori information into account. This algorithm exploits the approximate piecewise linear behavior of the soft outputs of an APP detector over the entire range of detector input values. A theoretical analysis and simulation results demonstrate that at least max-log APP performance can be reached, while the complexity is significantly reduced compared to classical APP detection.
基金partially supported by the National Natural Science Foundation of China(61271187)the "863" Project (2014AA01A701)+2 种基金the Provincial Natural Science Foundation of China(20122BAB211039)the PAPD fundthe CICAEET fund
文摘To reduce the energy consumption of the LTE-A system,a distributed energy-saving mechanism based on Co MP(Co MPDESM) is proposed to solve the inadequate coverage problem under the dormant cells.First,the network is divided into clusters based on the equivalent cell principle.Then,we transfer global optimization into a group of subproblems.Second,a joint processing-based cooperative cell selection model is constructed to determine cooperative cells and dormant cells.Third,the compensative cells with a determined threshold are selected to control users' access.Finally,a simulation is implemented in Matlab.Results show that the energy-saving rate can reach 36.4% and that the mechanism meets the network coverage requirement.Thus,joint processing can be effectively applied in an energy saving mechanism and used to improve the network performance of edge users without increasing transmission power.
基金supported by Branding Research Fund by Shibaura Institute of Technology(SIT)。
文摘Since precise self-position estimation is required for autonomous flight of aerial robots, there has been some studies on self-position estimation of indoor aerial robots. In this study, we tackle the self-position estimation problem by mounting a small downward-facing camera on the chassis of an aerial robot. We obtain robot position by sensing the features on the indoor floor.In this work, we used the vertex points(tile corners) where four tiles on a typical tiled floor connected, as an existing feature of the floor. Furthermore, a small lightweight microcontroller is mounted on the robot to perform image processing for the onboard camera. A lightweight image processing algorithm is developed. So, the real-time image processing could be performed by the microcontroller alone which leads to conduct on-board real time tile corner detection. Furthermore, same microcontroller performs control value calculation for flight commanding. The flight commands are implemented based on the detected tile corner information. The above mentioned all devices are mounted on an actual machine, and the effectiveness of the system was investigated.
基金Supported by the Key Natural Science Foundation of Anhui Higher Education Institutions (KJ2017A857, KJ2019A0727)。
文摘With the existing anonymous authentication schemes based on biometrics, the user and the server can create the same session key after mutual authentication. If the anonymous authentication scheme is applied in the electronic medical environment, it is also necessary to consider that the patient may access multiple hospital servers. Based on three factors of smart card, random number and biometrics, an anonymous authentication scheme in the electronic medical environment is proposed. In order to reduce the burden of the medical registration and certification center(HC), in the proposed anonymous authentication scheme, the patient only needs to register with HC once, then he/she can apply for visiting each hospital that has joined the medical servers. Security analysis shows that the proposed scheme has anonymity and dual authentication, and can resist various types of attacks, such as insider attack, modification attack, replay attack and smart card loss attack. Efficiency analysis shows that the calculation cost of the proposed scheme in the registration and login phase is lower, and it is slightly higher than Lei's scheme and Khan et al's scheme in the authentication phase. The proposed scheme can not only resist various types of attacks, but also support dual authentication and multi-server environment. With a little modification, the proposed scheme can also be used to other application scenarios requiring anonymous authentication.
基金Supported by the Key Natural Science Foundation of Anhui Higher Education Institutions(2022AH052536)。
文摘In order to protect the user's privacy identity,authentication requires anonymous authentication.Anonymous authentication is divided into unconditional anonymous authentication and traceable anonymous authentication.Unconditional anonymous authentication can verify that the user belongs to an anonymous set,but the user's true identity cannot be obtained.However,in some applications,it is necessary to trace the true identity of the user.Therefore,a traceable anonymous authentication scheme is proposed.In order to prevent random tracing,the proposed scheme uses threshold joint tracing.When the identity of the authenticator needs to be traced,the threshold number of members can jointly trace the identity of the authenticator.In some special network applications such as anonymous electronic voting,in order to prevent repeated authentications and repeated elections,it is necessary to verify whether the two authentication signatures are signed by the same user without revealing the true identity of the user.Therefore,the proposed anonymous authentication scheme should have selective linkability.In order to achieve linkable authentication,the linkable tag is embedded by linkable ring signature.Compared with similar schemes through the simulation experiments,the implementation time of the proposed scheme is slightly better than other schemes.
基金Project (No.18510132) supported by the Japan Society for the Promotion of Science Grants-in-Aid for Scientific Research
文摘This paper presents a new algorithm for clustering a large amount of data.We improved the ant colony clustering algorithm that uses an ant’s swarm intelligence,and tried to overcome the weakness of the classical cluster analysis methods.In our proposed algorithm,improvements in the efficiency of an agent operation were achieved,and a new function "cluster condensation" was added.Our proposed algorithm is a processing method by which a cluster size is reduced by uniting similar objects and incorporating them into the cluster condensation.Compared with classical cluster analysis methods,the number of steps required to complete the clustering can be suppressed to 1% or less by performing this procedure,and the dispersion of the result can also be reduced.Moreover,our clustering algorithm has the advantage of being possible even in a small-field cluster condensation.In addition,the number of objects that exist in the field decreases because the cluster condenses;therefore,it becomes possible to add an object to a space that has become empty.In other words,first,the majority of data is put on standby.They are then clustered,gradually adding parts of the standby data to the clustering data.The method can be adopted for a large amount of data.Numerical experiments confirmed that our proposed algorithm can theoretically applied to an unrestricted volume of data.
文摘A challenging task when applying high-order digital modulation schemes is the complexity of the detector. Particularly, the complexity of the optimal a posteriori probability （APP） detector increases exponentially with respect to the number of bits per data symbol. This statement is also true for the Max-Log-APP detector, which is a common simplification of the APP detector. Thus it is important to design new detection algorithms which combine a sufficient performance with low complexity. In this contribution, a detection algorithm for two- dimensional digital modulation schemes which cannot be split-up into real and imaginary parts （like phase shift keying and phase-shifted snperposition modulation （PSM）） is proposed with emphasis on PSM with equal power allocation. This algorithm exploits the relationship between Max-Log-APP detection and a Voronoi diagram to determine planar surfaces of the soft outputs over the entire range of detector input values. As opposed to state-of-the-art detectors based on Voronoi surfaces, a priori information is taken into account, enabling iterative processing. Since the algorithm achieves Max-Log-APP performance, even in the presence of a priori information, this implies a great potential for complexity reduction compared to the classical APP detection.
基金supported by The National Centre for Science Poland under the Project No.N519 650740.
文摘The strong La line 6520.644 ? is present in a Fourier Transform Spectrum (signal to noise ratio of 240), but its wavelength in commonly used tables (e.g. [1]) is given as 6520.770 ?, while in [2] the wavelength is given as 6520.74 ?, unclassified, with the remark “h” (hazy). The line could not be classified using known energy levels of the La atom (La I) and its first ion (La II). It appears as a single broadened peak. By a combination of laser optogalvanic spectroscopy, laser-induced fluorescence and Doppler-reduced saturation spectroscopy we could introduce a new even parity La I energy level, 35449.041 cm -1, J = 13/2, with hyperfine (hf) constants A = ?8.0(5) MHz, B = 10(10) MHz. For a second, up to now unknown neighbouring La I line (wavelength 6519.869 ?) we introduced another even parity energy level involved in the formation of the line, 41207.994 cm?-1, J = 13/2, A = 91.6(5) MHz, B = 170(50) MHz. We tried also to explain why in old tables the wavelength given was so different.
