Yu Yu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Yu Yu has a total of 37 co-authors. Here are the top co-authors by number of shared publications:
- Wen-Jun Lu
Jiangsu Key Laboratory of Wireless Communications and Internet-of-Things, Nanjing University of Posts and Telecommunications, Nanjing, China
8 shared publications - Hongbo Zhu
Nanjing University of Posts and Telecommunications
4 shared publications - Yang Liu
School of Internet of Things Engineering, Jiangnan University, Wuxi, China
4 shared publications - Yongxu Zhu
National Key Laboratory of Mobile Communications, Southeast University, Nanjing, China
4 shared publications - Jun Sun
Jiangsu Key Laboratory of Wireless Communications and Internet-of-Things, Nanjing University of Posts and Telecommunications, Nanjing, China
3 shared publications - Lihua Yang
Jiangsu Key Laboratory of Wireless Communications and Internet-of-Things, Nanjing University of Posts and Telecommunications, Nanjing, China
3 shared publications
Latest Publications
- 19 May 2026
- Published byĀ Institute of Electrical and Electronics Engineers (IEEE)Ā inĀ IEEE Transactions on Wireless Communications
- Vol.Ā 25 Ā (15361276) , 17102-17115
Auto-Polarization Fluid Antennas (APFAs): Evolution to Future Kinetic-Reconfigurable Wearable Wireless Technology?
An auto-polarization fluid antenna (APFA) is developed for indoor wireless channel sounding and employed to reveal a novel āfluid polarization effectā (FPE) in wireless communications. Unlike conventional fluid antennas (FAs) that are steering their beams/nulls with the aid of external mechanical/electronic actuators, the APFA only relies on the natural swinging of human arms to yield a self-driven polarization switching ability. Compared with the conventional fixed circularly polarized antennas, the wrist-worn, self-driven APFA in indoor wireless channel sounding systems effectively reduces multipath clusters (MPCs), attains smaller path loss exponent (PLE), and consequently yields the FPE. Compared to the fixed circularly polarized case with PLE= 1.62, the measured PLE is reduced by 14% to 1.38, and the system packet error rate (PER) is improved by 76%. It realizes robust anti-multipath fading performance owing to the much-improved FPE. The fluid effect in polarization domain is anticipated to remarkably enhance the anti-multipath fading performance of wearable wireless communication systems. It opens a new horizon to develop self-driven, cost-effective fluid antenna systems (FASs) for universal applications.
- 19 May 2026
- Published byĀ Institute of Electrical and Electronics Engineers (IEEE)Ā inĀ IEEE Transactions on Wireless Communications
- Vol.Ā 25 Ā (15361276) , 17102-17115
Auto-Polarization Fluid Antennas (APFAs): Evolution to Future Kinetic-Reconfigurable Wearable Wireless Technology?
An auto-polarization fluid antenna (APFA) is developed for indoor wireless channel sounding and employed to reveal a novel āfluid polarization effectā (FPE) in wireless communications. Unlike conventional fluid antennas (FAs) that are steering their beams/nulls with the aid of external mechanical/electronic actuators, the APFA only relies on the natural swinging of human arms to yield a self-driven polarization switching ability. Compared with the conventional fixed circularly polarized antennas, the wrist-worn, self-driven APFA in indoor wireless channel sounding systems effectively reduces multipath clusters (MPCs), attains smaller path loss exponent (PLE), and consequently yields the FPE. Compared to the fixed circularly polarized case with PLE= 1.62, the measured PLE is reduced by 14% to 1.38, and the system packet error rate (PER) is improved by 76%. It realizes robust anti-multipath fading performance owing to the much-improved FPE. The fluid effect in polarization domain is anticipated to remarkably enhance the anti-multipath fading performance of wearable wireless communication systems. It opens a new horizon to develop self-driven, cost-effective fluid antenna systems (FASs) for universal applications.
- 28 October 2025
- Published byĀ Institute of Electrical and Electronics Engineers (IEEE)Ā inĀ IEEE Transactions on Antennas and Propagation
- Vol.Ā 74 Ā (1) , 859-872
Loosely Coupled Surface Wave Wireless Propagation Channel on Bulky Metallic Pipes: Measurement, Modeling, and Analysis
To explore the effect of coupled distances on surface waves (SW) mode excitation in bulky conductor cases, whose diameter is comparable to or exceeding one-wavelength, we design the wirelessly coupled SW launchers for long, bulky conductors of varying diameters, aiming to establish a loosely coupled SW channel model for Industrial Internet-of-Things (IIoT) environment. A path loss (PL) model is proposed, incorporating both the coupled distance and transceiver distance. Unlike traditional SW wireless channels, which require a tightly coupled distance of less than half-wavelength, our results indicate that increasing the conductor diameter can extend the effective coupled distance to up to 2-wavelength. This loosely coupled configuration reduces the PL by approximately 17 dB compared to free-space propagation at the same transceiver distance, suppresses the first multi-path level by at least 5 dB, and yields a channel capacity gain of 1.5~14 Gbps. Furthermore, under a fixed loosely coupled distance, the SW wireless channel becomes increasingly directional as the conductorās diameter grows. The proposed model is expected to significantly enhance the deployment of SW nodes in IIoT environments that utilize multiple bulky conductors.
- 28 October 2025
- Published byĀ Institute of Electrical and Electronics Engineers (IEEE)Ā inĀ IEEE Transactions on Antennas and Propagation
- Vol.Ā 74 Ā (1) , 859-872
Loosely Coupled Surface Wave Wireless Propagation Channel on Bulky Metallic Pipes: Measurement, Modeling, and Analysis
To explore the effect of coupled distances on surface waves (SW) mode excitation in bulky conductor cases, whose diameter is comparable to or exceeding one-wavelength, we design the wirelessly coupled SW launchers for long, bulky conductors of varying diameters, aiming to establish a loosely coupled SW channel model for Industrial Internet-of-Things (IIoT) environment. A path loss (PL) model is proposed, incorporating both the coupled distance and transceiver distance. Unlike traditional SW wireless channels, which require a tightly coupled distance of less than half-wavelength, our results indicate that increasing the conductor diameter can extend the effective coupled distance to up to 2-wavelength. This loosely coupled configuration reduces the PL by approximately 17 dB compared to free-space propagation at the same transceiver distance, suppresses the first multi-path level by at least 5 dB, and yields a channel capacity gain of 1.5~14 Gbps. Furthermore, under a fixed loosely coupled distance, the SW wireless channel becomes increasingly directional as the conductorās diameter grows. The proposed model is expected to significantly enhance the deployment of SW nodes in IIoT environments that utilize multiple bulky conductors.
- Open Access
- 23 October 2025
- Published byĀ MDPI AGĀ inĀ Mathematics
- Vol.Ā 13 Ā (21) , 3377
An Elevation-Aware Large-Scale Channel Model for UAV Air-to-Ground Links
This paper addresses the issue of existing research that fails adequately capture the spatiotemporal nonstationarity caused by the building of occlusion and flight dynamics in air-to-ground channels from unmanned aerial vehicles (UAVs) in urban scenarios. This study focuses on the angular-altitude correlations of three key metrics: path loss (PL), shadow fading, and the Ricean K-factor. A dynamic path-loss model incorporating the look-down angle is proposed, an exponential decay model for the shadow-fading standard deviation is constructed, and a model for the angle-dependent variation of the Ricean K-factor is established based on line-of-sight probability. Simulations were conducted in two urban-geometry scenarios using WinProp to evaluate the combined effects of flight altitude and elevation angle. The results indicate that path loss decreases and subsequently stabilizes with increasing elevation angle, the shadow-fading standard deviation decreases significantly, and the Ricean K-factor increases with angle and saturates at high angles, in agreement with theoretical predictions. These models are more adaptable to UAV mobility scenarios than traditional fixed exponential models and provide a useful basis for UAV link planning and system optimization in urban environments.
- 23 October 2025
- Published byĀ MDPI AGĀ inĀ Mathematics
- Vol.Ā 13 Ā (21) , 3377
Open AccessAn Elevation-Aware Large-Scale Channel Model for UAV Air-to-Ground Links
This paper addresses the issue of existing research that fails adequately capture the spatiotemporal nonstationarity caused by the building of occlusion and flight dynamics in air-to-ground channels from unmanned aerial vehicles (UAVs) in urban scenarios. This study focuses on the angular-altitude correlations of three key metrics: path loss (PL), shadow fading, and the Ricean K-factor. A dynamic path-loss model incorporating the look-down angle is proposed, an exponential decay model for the shadow-fading standard deviation is constructed, and a model for the angle-dependent variation of the Ricean K-factor is established based on line-of-sight probability. Simulations were conducted in two urban-geometry scenarios using WinProp to evaluate the combined effects of flight altitude and elevation angle. The results indicate that path loss decreases and subsequently stabilizes with increasing elevation angle, the shadow-fading standard deviation decreases significantly, and the Ricean K-factor increases with angle and saturates at high angles, in agreement with theoretical predictions. These models are more adaptable to UAV mobility scenarios than traditional fixed exponential models and provide a useful basis for UAV link planning and system optimization in urban environments.
- 9 September 2025
- Published byĀ Institute of Electrical and Electronics Engineers (IEEE)Ā inĀ IEEE Transactions on Wireless Communications
- Vol.Ā 25 Ā (15361276) , 3514-3525
Surface Wave Wireless Propagation Channel With Antenna Rotation for Industrial Internet-of-Things: Measurement, Modeling, and Analysis
Investigations on surface waves (SW) launchers wireless coupled to a long, single conductor in arbitrary azimuth angles rotation are carried out to inspire a novel SW wireless channel model in Industrial Internet-of-Things (IIoT) environment. A path loss (PL) model with two degrees in freedom, i.e., the coupled azimuth angle α and the transceiver separation d, is modeled at first. It is revealed that the angular factor governs the excitation degree of the SW propagation mode. Then, a dual-factor root mean square (RMS) delay extension model is developed. Next, the fast-fading distribution is modeled as a nonlinear combination of harmonic functions of the coupled azimuth angle α. It is validated that the PL of the SW propagation mode is reduced by 16~28dB, with the channel impulse response (CIR) principal path level increased by 18dB and the first multi-path suppressed by 7.5dB compared to the free-space propagation mode. Finally, calculations on channel capacity (CC) are performed to demonstrate a CC enhancement of 7~17Gbps. The advanced separation-angle joint channel model is expected to provide useful guidelines in future SW communication nodes deployments in IIoT scenarios.
- 9 September 2025
- Published byĀ Institute of Electrical and Electronics Engineers (IEEE)Ā inĀ IEEE Transactions on Wireless Communications
- Vol.Ā 25 Ā (15361276) , 3514-3525
Surface Wave Wireless Propagation Channel With Antenna Rotation for Industrial Internet-of-Things: Measurement, Modeling, and Analysis
Investigations on surface waves (SW) launchers wireless coupled to a long, single conductor in arbitrary azimuth angles rotation are carried out to inspire a novel SW wireless channel model in Industrial Internet-of-Things (IIoT) environment. A path loss (PL) model with two degrees in freedom, i.e., the coupled azimuth angle α and the transceiver separation d, is modeled at first. It is revealed that the angular factor governs the excitation degree of the SW propagation mode. Then, a dual-factor root mean square (RMS) delay extension model is developed. Next, the fast-fading distribution is modeled as a nonlinear combination of harmonic functions of the coupled azimuth angle α. It is validated that the PL of the SW propagation mode is reduced by 16~28dB, with the channel impulse response (CIR) principal path level increased by 18dB and the first multi-path suppressed by 7.5dB compared to the free-space propagation mode. Finally, calculations on channel capacity (CC) are performed to demonstrate a CC enhancement of 7~17Gbps. The advanced separation-angle joint channel model is expected to provide useful guidelines in future SW communication nodes deployments in IIoT scenarios.
- Open Access
Image Fusion and Target Detection Based on Dual ResNet for Power Sensing Equipment
Target detection helps to identify, locate, and monitor key components and potential issues in power sensing networks. The fusion of infrared and visible light images can effectively integrate the target the indication characteristics of infrared images and the rich scene detail information of visible light images, thereby enhancing the ability for target detection in power equipment in complex environments. In order to improve the registration accuracy and feature extraction stability of traditional registration algorithms for infrared and visible light images, an image registration method based on an improved SIFT algorithm is proposed. The image is preprocessed to a certain extent, using edge detection algorithms and corner detection algorithms to extract relatively stable feature points, and the feature vectors with excessive gradient values in the normalized visible light image are truncated and normalized again to eliminate the influence of nonlinear lighting. To address the issue of insufficient deep information extraction during image fusion using a single deep learning network, a dual ResNet network is designed to extract deep level feature information from infrared and visible light images, improving the similarity of the fused images. The image fusion technology based on the dual ResNet network was applied to the target detection of sensing insulators in the power sensing network, improving the average accuracy of target detection. The experimental results show that the improved registration algorithm has increased the registration accuracy of each group of images by more than 1%, the structural similarity of image fusion in the dual ResNet network has been improved by about 0.2% compared to in the single ResNet network, and the mean Average Precision (mAP) of the fusion image obtained via the dual ResNet network has been improved by 3% and 6% compared to the infrared and visible light images, respectively.
Open AccessImage Fusion and Target Detection Based on Dual ResNet for Power Sensing Equipment
Target detection helps to identify, locate, and monitor key components and potential issues in power sensing networks. The fusion of infrared and visible light images can effectively integrate the target the indication characteristics of infrared images and the rich scene detail information of visible light images, thereby enhancing the ability for target detection in power equipment in complex environments. In order to improve the registration accuracy and feature extraction stability of traditional registration algorithms for infrared and visible light images, an image registration method based on an improved SIFT algorithm is proposed. The image is preprocessed to a certain extent, using edge detection algorithms and corner detection algorithms to extract relatively stable feature points, and the feature vectors with excessive gradient values in the normalized visible light image are truncated and normalized again to eliminate the influence of nonlinear lighting. To address the issue of insufficient deep information extraction during image fusion using a single deep learning network, a dual ResNet network is designed to extract deep level feature information from infrared and visible light images, improving the similarity of the fused images. The image fusion technology based on the dual ResNet network was applied to the target detection of sensing insulators in the power sensing network, improving the average accuracy of target detection. The experimental results show that the improved registration algorithm has increased the registration accuracy of each group of images by more than 1%, the structural similarity of image fusion in the dual ResNet network has been improved by about 0.2% compared to in the single ResNet network, and the mean Average Precision (mAP) of the fusion image obtained via the dual ResNet network has been improved by 3% and 6% compared to the infrared and visible light images, respectively.
- 18 April 2025
- Published byĀ Institute of Electrical and Electronics Engineers (IEEE)Ā inĀ IEEE Wireless Communications Letters
- Vol.Ā 14 Ā (7) , 2059-2063
Surface Wave Wireless Channel on Right-Angle Bent Metal Pipes in Industrial Environment: Modeling and Analysis
This study investigates surface waves (SW) launchers wirelessly coupled to a 90ā-cornered metal pipe to develop a wireless SW channel model in Industrial Internet-of-Things (IIoT) environment. A pair of orthogonal SW modes is excited on a right-angle bent metal pipe, in which demonstrating a path loss can be reduced of 3-10 dB compared to free-space propagation. Furthermore, the multi-path characteristic of the right-angle SW channel can be mitigated by suppressing the first multi-path level of about 7.5dB compared to the free-space propagation scenario. Simulations of channel capacity (CC) are performed to demonstrate a CC enhancement of 3.5Gbps, when BW is 2GHz and SNR is 3.7 dB. The advanced wireless channel models for right-angle metal pipe are expected to provide beneficial strategy for optimizing SW-node deployments in complex IIoT scenarios.
- 18 April 2025
- Published byĀ Institute of Electrical and Electronics Engineers (IEEE)Ā inĀ IEEE Wireless Communications Letters
- Vol.Ā 14 Ā (7) , 2059-2063
Surface Wave Wireless Channel on Right-Angle Bent Metal Pipes in Industrial Environment: Modeling and Analysis
This study investigates surface waves (SW) launchers wirelessly coupled to a 90ā-cornered metal pipe to develop a wireless SW channel model in Industrial Internet-of-Things (IIoT) environment. A pair of orthogonal SW modes is excited on a right-angle bent metal pipe, in which demonstrating a path loss can be reduced of 3-10 dB compared to free-space propagation. Furthermore, the multi-path characteristic of the right-angle SW channel can be mitigated by suppressing the first multi-path level of about 7.5dB compared to the free-space propagation scenario. Simulations of channel capacity (CC) are performed to demonstrate a CC enhancement of 3.5Gbps, when BW is 2GHz and SNR is 3.7 dB. The advanced wireless channel models for right-angle metal pipe are expected to provide beneficial strategy for optimizing SW-node deployments in complex IIoT scenarios.
- 27 November 2023
- Published byĀ Institute of Electrical and Electronics Engineers (IEEE)Ā inĀ IEEE Antennas and Wireless Propagation Letters
- Vol.Ā 23 Ā (2) , 838-842
Millimeter-Wave Wideband Circularly Polarized SIW Cavity-Backed Stacked Dipole Array Antenna
A wideband circularly polarized (CP) substrate integrated waveguide (SIW) cavity-backed stacked dipoles array antenna is proposed for millimeter-wave (mm-wave) applications. Firstly, double-layer SIW cavities are utilized to construct the CP stacked dipole element, which consists of two crossed stacked linearly polarized (LP) electric dipoles and four metallic probes in driven, short-circuited, and suspended states. These probes can achieve orthogonal phase and energy coupling through a magnetic field without additional baluns. Next, wideband CP working principle and parameter analysis of the element are addressed. A full corporate feeding scheme with 4-way SIW power divider and 4 SIW cavities is adopted for array verification. Finally, a 4 Ć 4 array is designed, fabricated, and measured, which demonstrates wide impedance bandwidth of 28.1%, wide axial ratio (AR) bandwidth of 23%, and a peak gain of 19.35 dBic, respectively. These results demonstrate that the CP stacked dipole is a promising candidate for mm-wave wireless systems.
- 27 November 2023
- Published byĀ Institute of Electrical and Electronics Engineers (IEEE)Ā inĀ IEEE Antennas and Wireless Propagation Letters
- Vol.Ā 23 Ā (2) , 838-842
Millimeter-Wave Wideband Circularly Polarized SIW Cavity-Backed Stacked Dipole Array Antenna
A wideband circularly polarized (CP) substrate integrated waveguide (SIW) cavity-backed stacked dipoles array antenna is proposed for millimeter-wave (mm-wave) applications. Firstly, double-layer SIW cavities are utilized to construct the CP stacked dipole element, which consists of two crossed stacked linearly polarized (LP) electric dipoles and four metallic probes in driven, short-circuited, and suspended states. These probes can achieve orthogonal phase and energy coupling through a magnetic field without additional baluns. Next, wideband CP working principle and parameter analysis of the element are addressed. A full corporate feeding scheme with 4-way SIW power divider and 4 SIW cavities is adopted for array verification. Finally, a 4 Ć 4 array is designed, fabricated, and measured, which demonstrates wide impedance bandwidth of 28.1%, wide axial ratio (AR) bandwidth of 23%, and a peak gain of 19.35 dBic, respectively. These results demonstrate that the CP stacked dipole is a promising candidate for mm-wave wireless systems.
- 21 July 2023
- Published byĀ Institute of Electrical and Electronics Engineers (IEEE)Ā inĀ IEEE Internet of Things Journal
- Vol.Ā 11 Ā (2) , 3362-3376
A General 3-D Geometry-Based Stochastic Channel Model for B5G mmWave IIoT
The Industrial Internet of things (IIoT) is one of the typical application scenarios in the beyond fifth generation (B5G) wireless communication systems. Due to numerous metal obstacles and machines, the industrial channel, especially at the millimeter-wave (mmWave) bands, exhibits complex characteristics that havenāt been considered in existing literature. This paper proposes an innovative three-dimensional (3D) non-stationary geometry-based stochastic model (GBSM) for IIoT scenarios at mmWave bands. In the proposed model, device reflections (DR) caused by massive metal machines are modeled based on geometrical optics. Furthermore, the generalized extreme value (GEV) distribution and generalized Pareto (GP) distribution are used to parameterize the number of clusters and rays within a cluster, respectively. Further, the Doppler shift is modeled and analyzed using the Gaussian distribution. Some channel statistical characteristics are captured by the proposed model, such as the power delay profile, root-mean-square delay spread, root-mean-square angle spread, inter-cluster delay, and space-time-frequency correlation function. Then, these channel statistical characteristics are well fitted to the ray-tracing simulations and the channel measurements. The excellent fitting results demonstrate the high accuracy of the proposed model, which is crucial for future IIoT communication system design. Whatās more, this paper shows the antenna height and propagation scenarios can significantly affect the DR ratio, which should adapt to various IIoT communication scenarios.
- 21 July 2023
- Published byĀ Institute of Electrical and Electronics Engineers (IEEE)Ā inĀ IEEE Internet of Things Journal
- Vol.Ā 11 Ā (2) , 3362-3376
A General 3-D Geometry-Based Stochastic Channel Model for B5G mmWave IIoT
The Industrial Internet of things (IIoT) is one of the typical application scenarios in the beyond fifth generation (B5G) wireless communication systems. Due to numerous metal obstacles and machines, the industrial channel, especially at the millimeter-wave (mmWave) bands, exhibits complex characteristics that havenāt been considered in existing literature. This paper proposes an innovative three-dimensional (3D) non-stationary geometry-based stochastic model (GBSM) for IIoT scenarios at mmWave bands. In the proposed model, device reflections (DR) caused by massive metal machines are modeled based on geometrical optics. Furthermore, the generalized extreme value (GEV) distribution and generalized Pareto (GP) distribution are used to parameterize the number of clusters and rays within a cluster, respectively. Further, the Doppler shift is modeled and analyzed using the Gaussian distribution. Some channel statistical characteristics are captured by the proposed model, such as the power delay profile, root-mean-square delay spread, root-mean-square angle spread, inter-cluster delay, and space-time-frequency correlation function. Then, these channel statistical characteristics are well fitted to the ray-tracing simulations and the channel measurements. The excellent fitting results demonstrate the high accuracy of the proposed model, which is crucial for future IIoT communication system design. Whatās more, this paper shows the antenna height and propagation scenarios can significantly affect the DR ratio, which should adapt to various IIoT communication scenarios.
- 8 November 2021
- Published byĀ Institute of Electrical and Electronics Engineers (IEEE)Ā inĀ IEEE Internet of Things Journal
- Vol.Ā 9 Ā (13) , 11190-11202
Person Density Dependency on Path Loss and Root Mean Square Delay Spread for Smart Office Scenarios
Novel empirical path-loss and root mean square delay spread (RDS) models for smart office scenarios are proposed. The effects of person density on the path loss and RDS are investigated based on the extensive measurements at 2.3ā2.5 GHz. First, both of the measured path loss and RDS data are modeled as the dual log-distance functions. It is caused by the regular structure and furniture in the office environment. Second, in the proposed path-loss model, the path-loss exponents and the additional attenuation factor are modeled as quadratic functions of the person density. Meanwhile, the RDS is found to be uncorrelated with the person density. These phenomena reveal that the persons in the environments can be regarded as absorbers rather than scatters. Then, the accuracy of the proposed models is validated by the measured data and compared with two traditional models. Finally, the effect of the personsā movements on the path loss and RDS is investigated, and the proposed models are extended to millimeter wave bands by a ray tracing technology. The proposed models and results can provide necessary information for link budget and algorithm design for the Internet of Things smart office scenarios.
- 8 November 2021
- Published byĀ Institute of Electrical and Electronics Engineers (IEEE)Ā inĀ IEEE Internet of Things Journal
- Vol.Ā 9 Ā (13) , 11190-11202
Person Density Dependency on Path Loss and Root Mean Square Delay Spread for Smart Office Scenarios
Novel empirical path-loss and root mean square delay spread (RDS) models for smart office scenarios are proposed. The effects of person density on the path loss and RDS are investigated based on the extensive measurements at 2.3ā2.5 GHz. First, both of the measured path loss and RDS data are modeled as the dual log-distance functions. It is caused by the regular structure and furniture in the office environment. Second, in the proposed path-loss model, the path-loss exponents and the additional attenuation factor are modeled as quadratic functions of the person density. Meanwhile, the RDS is found to be uncorrelated with the person density. These phenomena reveal that the persons in the environments can be regarded as absorbers rather than scatters. Then, the accuracy of the proposed models is validated by the measured data and compared with two traditional models. Finally, the effect of the personsā movements on the path loss and RDS is investigated, and the proposed models are extended to millimeter wave bands by a ray tracing technology. The proposed models and results can provide necessary information for link budget and algorithm design for the Internet of Things smart office scenarios.
- 10 March 2020
- Published byĀ Institute of Electrical and Electronics Engineers (IEEE)Ā inĀ IEEE Internet of Things Journal
- Vol.Ā 7 Ā (6) , 5580-5589
Neural-Network-Based Root Mean Delay Spread Model for Ubiquitous Indoor Internet-of-Things Scenarios
Massive robust communication demands among machines and humans are required in ubiquitous Internet of Things (IoTs) applications. To design appropriate communication system, the knowledge of the propagation characteristics for various IoTs scenarios is necessary. In this paper, a measurement-based neural network based root mean square (RMS) delay spread model for ubiquitous indoor IoTs scenarios is presented. The proposed model is a two-layer feed-forward neural network plus a random variable, characterizing the average RMS delay spread and uncertain shadowing effect respectively. The neural network consists of five inputs, including transmitting/receiving antennas (Tx/Rx) separation, frequency, antenna height, environment, and line-of-sight/non-line-of-sight (LOS/NLOS) propagation condition, seven hidden layer neurons and one output layer neuron. Compared with different configurations of the neural network, the hyperbolic tangent sigmoid functions and the Levenberg-Marquardt back propagation algorithm are selected as neuronsā activation functions and training method, respectively. Additionally, the random variable is found to follow the normal distribution using the maximum likelihood estimation. Finally, the novel model is experimentally validated to be accurate, general and extensible compared with the conventional normally distributed RMS delay spread model. This model is well applicable to the design and planning of the ubiquitous communication links for future IoTs scenarios.
- 10 March 2020
- Published byĀ Institute of Electrical and Electronics Engineers (IEEE)Ā inĀ IEEE Internet of Things Journal
- Vol.Ā 7 Ā (6) , 5580-5589
Neural-Network-Based Root Mean Delay Spread Model for Ubiquitous Indoor Internet-of-Things Scenarios
Massive robust communication demands among machines and humans are required in ubiquitous Internet of Things (IoTs) applications. To design appropriate communication system, the knowledge of the propagation characteristics for various IoTs scenarios is necessary. In this paper, a measurement-based neural network based root mean square (RMS) delay spread model for ubiquitous indoor IoTs scenarios is presented. The proposed model is a two-layer feed-forward neural network plus a random variable, characterizing the average RMS delay spread and uncertain shadowing effect respectively. The neural network consists of five inputs, including transmitting/receiving antennas (Tx/Rx) separation, frequency, antenna height, environment, and line-of-sight/non-line-of-sight (LOS/NLOS) propagation condition, seven hidden layer neurons and one output layer neuron. Compared with different configurations of the neural network, the hyperbolic tangent sigmoid functions and the Levenberg-Marquardt back propagation algorithm are selected as neuronsā activation functions and training method, respectively. Additionally, the random variable is found to follow the normal distribution using the maximum likelihood estimation. Finally, the novel model is experimentally validated to be accurate, general and extensible compared with the conventional normally distributed RMS delay spread model. This model is well applicable to the design and planning of the ubiquitous communication links for future IoTs scenarios.