Current Issue

30 June 2023, Volume 7 Issue 2

Previous Issue   

Topic: Intellisense Technology

New flexible sensor for the internet of things

Jing WU, Sheng LI, Jing ZHANG, Ming XIN, Ruowen TAO, Zhou ZHOU, Lijia PAN, Yi SHI

2023, 7(2):  1-14.  doi:10.11959/j.issn.2096-3750.2023.00294

Asbtract ( 253 )   HTML ( 64)   PDF (22773KB) ( 146 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

The new flexible sensing technology integrates electronic information, material chemistry, biomedical engineering and nano-processing, to provide intelligent sensing terminals for the internet of things, which will be widely used in health monitoring, smart home, smart manufacturing and other fields.For the development of new materials and research on novel mechanisms, flexible sensors have gained much progress in basic parameters, such as sensitivity, response range, response time, linearity, hysteresis, and stability.The flexible sensing technology is not only limited to materials and devices, but also expands from a single device to system-level integrations.Aimed at the hot research issues in recent years, the development of new flexible sensors was introduced from four aspects: multimode integrated sensors, bionic electronic sensors, wireless signal transmission and energy supply, and intelligent information processing.

An environment adaptive gesture recognition system based on visible light

Zhu WANG, Hualei ZHANG, Qianhong HU, Zhiwen YU

2023, 7(2):  15-25.  doi:10.11959/j.issn.2096-3750.2023.00344

Asbtract ( 185 )   HTML ( 20)   PDF (6496KB) ( 164 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Gesture-based human-machine interaction is becoming more and more important, which can provide users with a better experience in scenarios such as video games and virtual reality.In recent years, researchers have explored different sensing technologies to facilitate gesture recognition, including RF signal, acoustic signal, etc.Compared with these approaches, visible light-based gesture recognition is a more pervasive option.The basic principle is that different gestures will produce unique shadow patterns as they block the visible light, and gesture recognition can be achieved by capturing shadow changes through photoelectric sensors.To address the environment-dependent problem faced by existing solutions, a digit gesture recognition system was designed based on the photoelectric sensor array.In particular, by modeling recordings of the sensor array as images, the temporal and spatial correlation between different sensor recordings was discovered.An environment adaptive gesture recognition method was designed based on CNN-RNN by fusing the spatio-temporal features.To verify the effectiveness of the proposed method, a prototype gesture recognition system was designed, named Vi-Gesture.Experimental results show that the proposed method outperforms baselines by more than 10% in recognition accuracy.

Research progress of human health IoT based on wearable and implantable techniques

Junge LIANG, Yiran SONG, Yangfan SUN, Yingying JI, Lijia PAN, Yi SHI

2023, 7(2):  26-34.  doi:10.11959/j.issn.2096-3750.2023.00343

Asbtract ( 213 )   HTML ( 45)   PDF (4172KB) ( 268 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Monitoring various physiological parameters in healthy people daily life can provide early warning of abnormalities and diseases, as well as reduce the pressure on national public health and medical resources.Hence, it is necessary to develop human health IoT system based on wearable and implantable sensors.The bioinformatics detection based on body fluids and electrical signals as the entry point was used to divide sensing technologies, and the key technologies such as self-powered and near-end communication for human health IoT were introduced for discussion.Finally, the technological progress and industrial application in the field of health data management and disease diagnosis and prevention were explored, and the concept of human health IoT based on wearable and implantable technologies was attempted to be constructed.

Intelligent phase imaging guided by physics models

Zhen LIU, Hao ZHU, You ZHOU, Zhan MA, Xun CAO

2023, 7(2):  35-42.  doi:10.11959/j.issn.2096-3750.2023.00345

Asbtract ( 121 )   HTML ( 18)   PDF (12710KB) ( 69 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Implicit neural representation characterizes the mapping between the signal’s coordinate to its attributes, and has been widely used in the optimization of inverse problems by embedding the physics process into the loss function.However, the implicit neural representation is suffering the low convergence speed and accuracy from the random initialization of the network parameters.The meta-learning algorithm for providing implicit neural representation with a strong prior of network parameters was proposed, thus enhancing the optimization efficiency and accuracy for solving inverse problems.To address the important issue of lens less phase imaging, an intelligent method on phase imaging was proposed based on the snapshot lens less sensing model.By embedding the optical diffraction propagation theory into the design of loss function for implicit neural representation, the dependency of large-scale labelled dataset in traditional deep learning-based methods was eliminated and accurate phase image from a single diffraction measurement image was provided.Furthermore, the meta-learning model was introduced for initializing network to further improve the efficiency and accuracy of network training.Numerical simulation results show that the proposed method can achieve a PSNR improvement of more than 11 dB compared to the conventional method.The experimental results in real data show that the phase image reconstructed by the proposed method is clearer and has fewer artifacts.

Pulse detection sensor with green light transceiver integrated optoelectronic chip

Chenchen ZHANG, Xumin GAO, Ziqi YE, Mingming BO, Zefeng HU, Yongjin WANG

2023, 7(2):  43-49.  doi:10.11959/j.issn.2096-3750.2023.00319

Asbtract ( 136 )   HTML ( 15)   PDF (15375KB) ( 135 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Photo plethysmo graphy (PPG) is widely used to detect heart rate and pulse activity due to its simple design, low cost, and signal cycle that can effectively reflect the heart rhythm.Owing to the partial overlap between the emission and the detection spectra of GaN multiple quantum well (MQW) diode, the MQW receiver can respond to the light emitted from the transmitter sharing identical MQW structures.Thus, the fabricate and characterize monolithic integration of GaN transmitter and receiver was proposed as a single unit onto a conventional GaN-on-sapphire platform.This not only can effectively reduce the size and cost of the device, but also fully improves the sensitivity of the system.The experimental results show that the gallium nitride integrated optoelectronic chip can effectively receive the reflected light modulated by the heart pulse.Combined with circuit processing, the tester’s heart rate and pulse beat frequency can be calculated according to the restored PPG signal, so as to play the role of detection and warning.

Theory and Technology

A survey of federated learning for 6G networks

Guanglei GENG, Bo GAO, Ke XIONG, Pingyi FAN, Yang LU, Yuwei WANG

2023, 7(2):  50-66.  doi:10.11959/j.issn.2096-3750.2023.00323

Asbtract ( 350 )   HTML ( 60)   PDF (6843KB) ( 247 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

It is an important feature of the 6G that how to realize everything interconnection through large-scale complex heterogeneous networks based on native artificial intelligence (AI).Thanks to the distinct machine learning architecture of data processing locally, federated learning (FL) is regarded as one of the promising solutions to incorporate distributed AI in 6G scenarios, and has become a critical research direction of 6G.Therefore, the necessity of introducing distributed AI into the future 6G especially for internet of things (IoT) scenarios was analyzed.And then, the potentials of FL in meeting the 6G requirements were discussed, and the state-of-the-arts of FL related technologies such as architecture design, resource utilization, data transmission, privacy protection, and service provided for 6G were investigated.Finally, several key technical challenges and potential valuable research directions for FL-empowered 6G were put forward.

A joint optimization method of multi backhaul link selection and power allocation in 6G

Qingyang LI, Xueting LI, Xiaorong ZHU

2023, 7(2):  67-75.  doi:10.11959/j.issn.2096-3750.2023.00284

Asbtract ( 79 )   HTML ( 13)   PDF (3571KB) ( 256 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Aiming at the problem of limited single backhaul link capability of base stations in hotspot areas of the 6G system, a joint optimization method was proposed for multi-backhaul link selection and power allocation in an elastic coverage system, so that the data packets can select the appropriate backhaul link and transmission power according to their service characteristics and link conditions.Firstly, the transmission delay of data packets on the small cell sub-queue was analyzed by using queuing theory.Then, the optimization objective was modeled with the maximum delay tolerance elasticity value.Finally, the optimization problem were solved by the Hungarian algorithm and the Lagrangian duality and gradient descent method.The simulation results show that, compared with the traditional algorithm, the algorithm proposed reduces the average delay of URLLC (ultra-reliable and low-latency communication) service data packets and eMBB (enhanced mobile broadband) service data packets by 17% and 14% respectively, and effectively improves the network transmission rate.

Research on medical small sample data classification based on SMOTE and gcForest

Wenchang LIU, Yun WEI, Haoxuan YUAN, Yue GAO

2023, 7(2):  76-87.  doi:10.11959/j.issn.2096-3750.2023.00337

Asbtract ( 119 )   HTML ( 23)   PDF (1804KB) ( 343 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Aiming at the problem of poor classification performance in traditional machine learning models caused by shallow model structure and complex data characteristics in small medical sample data, an combine multi- grained improved cascade forest (cgicForest) model was proposed.It enhances the representation learning ability of the model by adding random sampling into the multi-grained scanning and optimizing the transformation features.It also enhances the model's classification ability by updating the cascade forest’s hierarchical structure.Considering category imbalance problems in datasets, the safe-borderline-SMOTE (SBS) algorithm was proposed to dynamic interpolate around the few class samples belonging to the safety boundary, which can improve the quality of training data.The cgicForest was applied for training and learning, thus the SBS-cgicForest classification model was obtained which can support imbalanced medical small samples data.The model is used on three medical datasets for classification experiments.The results show that the performance indexes of the cgicForest model in the classification of medical small sample data with complex characteristics have increased by 4.1～5.4 percentage points, compared with the multi-grained cascade forest (gcForest) model.The performance indexes have increase by 6.6～11.2 percentage points after the combination with SBS algorithm, the F₁ score was 2～2.5 percentage points higher than that obtained by traditional sampling methods.It provides a reference for solving the classification problem of small medical sample data, and includes support for internet of things applications in smart medical scenarios.

Time synchronization attack detection for industrial wireless network

Sichao ZHANG, Wei LIANG, Xudong YUAN, Yinlong ZHANG, Meng ZHENG

2023, 7(2):  88-97.  doi:10.11959/j.issn.2096-3750.2023.00334

Asbtract ( 91 )   HTML ( 10)   PDF (2806KB) ( 352 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

High-precision time synchronization is the basis for ensuring the secure and reliable transmission of industrial wireless network (IWN).Delay attacks, as a class of time synchronization attacks which cannot be solved by cryptographic techniques, seriously threaten the secure operation of IWN.Firstly, based on the in-depth analysis on the time synchronization mechanisms of IWN, three-time synchronization attack models were proposed, including the one-way full life cycle delay attack, two-way full life cycle delay attack, and one-way non-full-life cycle delay attack.Stealthier delay attacks could be realized by the attack models under the premise that target nodes were not captured.Secondly, considering the problem that existing detection algorithms are difficult to detect stealthier delay attacks without obvious changes in time features, an attack detection algorithm based on a Bayesian model was proposed that extracts four representative features, including transmission rate, transmission delay, transmission success rate and time synchronization interval.In addition, in order to ensure the accuracy of the attack detection and classification in the presence of noise interference, the noise model of wireless channel was introduced to the Bayesian feature information matrix.Experimental results show that the proposed algorithm can effectively detect three kinds of attacks in the presence of noise.

Interference alignment based secure transmission scheme in multi-user interference networks

Lin HU, Jiabing FAN, Hong WEN, Jie TANG, Qianbin CHEN

2023, 7(2):  98-108.  doi:10.11959/j.issn.2096-3750.2023.00331

Asbtract ( 102 )   HTML ( 14)   PDF (3010KB) ( 118 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Faced with the requirement of information security in internet of things, a scenario of multi-user interference networks with multiple eavesdroppers was considered, and an interference alignment (IA) scheme based physical layer secure transmission was proposed.Traditional IA security algorithm may result in secret signal cancellation.To overcome this threat, a modified alternating minimization (AM) method was proposed.The multi-user interference was eliminated by alternatively optimizing transceiver matrices, and artificial noise (AN) aided max-eigenmode beamforming was employed for secure transmission.To obtain a more accurate analysis of IA feasibility, the IA equation was divided into independent subsets and their combinations.By analyzing each case, a much tighter necessary condition for IA feasibility was established.Finally, the power allocation ratio between the secret signal and the AN signal was optimized to maximize the secrecy outage probability (SOP) constrained secrecy rate.Numerical results confirm that both the quality and security of the secret signal have been enhanced.Therefore, the proposed scheme is more suitable and reliable for security applications in interference networks.

Dual-functional radar-communication waveform trade-off optimal design with peak average power ratio constraint

Yuefeng HOU, Xiao ZHAO, Jianing LIU, Rongkang ZHOU, Wenbo WANG, Feng TIAN

2023, 7(2):  109-117.  doi:10.11959/j.issn.2096-3750.2023.00338

Asbtract ( 116 )   HTML ( 19)   PDF (2538KB) ( 156 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Based on the same hardware platform, spectrum resources, and transmit waveforms, dual-function radar-communication (DFRC) can achieve radar and communication functions at the same time, which provides a strong support for realizing integrated sensing and communication (ISAC).Aiming at the dual-function radar communication system, radar and communication waveform trade off optimal design was studied, and a non-convex optimization problem was constructed with quadratic equality and inequality constraints under the constraints of transmission energy consumption and peak-to-average ratio (PAPR).The nonlinear equality constrained alternative direction method of multipliers was used to decompose the non-convex optimization problem into three sub-problems.The simulation results show that the optimized dual-function radar communication waveform can approach the ideal radar waveform and realize the optimal performance tradeoff between radar and communication.

Scenario-adaptive wireless fall detection system based on few-shot learning

Yuting ZENG, Suzhi BI, Lili ZHENG, Xiaohui LIN, Hui WANG

2023, 7(2):  118-132.  doi:10.11959/j.issn.2096-3750.2023.00339

Asbtract ( 187 )   HTML ( 27)   PDF (10035KB) ( 161 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

A scenario robust fall detection system based on few-shot learning (FDFL) in wireless environment was designed.The performance of existing fall detection methods based on Wi-Fi channel state information (CSI) degrades significantly across scenarios, which requires collecting and marking a large number of CSI samples in each application scenario, resulting in high cost for large-scale deployment.Therefore, the method of few-shot learning was introduced, which can maintain the performance of fall detection with high accuracy when the number of annotated samples in unfa-miliar scenes is insufficient.The proposed FDFL was mainly divided into two stages, source domain meta-training and target domain meta-learning.The meta training stage of the source domain consists of two parts: data preprocessing and classification training.In the data preprocessing stage, the collected original CSI amplitude and phase data were denoised and segmented.In the classification training stage, a large number of processed source domain data samples were used to train a CSI feature extractor based on convolutional neural network.In the meta-learning stage of the target domain, the limited labeled data sampled in the target domain was effectively extracted based on the feature extractor trained in the meta-training module, and then a lightweight machine learning classifier was trained to detect the fall behavior under the cross-scene.Through several experiments in different scenarios, FDFL can achieve an average accuracy of 95.52% for the four classification tasks of falling, sitting, walking and sit down with only a small number of samples in the target domain, and maintain robust detection accuracy for changes in test environment, personnel target and equipment location.

Human activity recognition system based on low-cost IoT chip ESP32

Chao HU, Bangyan LU, Yanbing YANG, Zhe CHEN, Lei ZHANG, Liangyin CHEN

2023, 7(2):  133-142.  doi:10.11959/j.issn.2096-3750.2023.00330

Asbtract ( 228 )   HTML ( 46)   PDF (7280KB) ( 146 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Human activity recognition widely exists in applications such as sports management and activity classification.The current human activity recognition applications are mainly divided into three types: camera-based, wearable device-based, and Wi-Fi awareness-based.Among them, the camera-based human activity recognition application has the risk of privacy leakage, and the wearable device-based human activity recognition application has problems such as short battery life and poor accuracy.Human activity recognition based on Wi-Fi sensing generally uses Wi-Fi network cards or software-defined radio devices to identify the rules of channel state information changes, so as to infer user activity.It does not have the problems of privacy leakage and short battery life.But Wi-Fi network cards need to rely on computers and software-defined radio platforms are expensive, which greatly limit the application scenarios of Wi-Fi sensing.Aiming at the above problems, a human activity recognition system based on the low-cost IoT chip ESP32 was proposed.Specifically, the Hampel filter and Gaussian filter were used to preprocess the channel state information obtained by ESP32.Then, the principal component analysis and discrete wavelet transform were utilized to reduce the dimension of the data.Finally, the K-nearest neighbor (KNN) algorithm was applied to classify data.The experimental results show that the system can achieve a recognition accuracy which close to the current mainstream Wi-Fi perception system (Intel 5300 network card) when only two ESP32 nodes are deployed, and the average accuracy rate for the six activities is 98.6%.

Wi-freshness: research on CSI-based pork freshness detecting system

Chao NIU, Weidong YANG, Pengming HU, Xiangshang GAO, Erbo SHEN

2023, 7(2):  143-152.  doi:10.11959/j.issn.2096-3750.2023.00332

Asbtract ( 79 )   HTML ( 10)   PDF (6427KB) ( 303 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Effective and rapid detection of pork freshness is important for pork quality.However, traditional sensory evaluation methods are too subjective, physical and chemical analysis methods are time-consuming and destructive.Recently, radio frequency is widely used in the field of location, material identification and human body monitoring, while, the meat freshness detection is ignored.A real-time, non-destructive and low-cost system for pork freshness detecting based on channel state information (CSI) was proposed.It is a new application of internet of things in the field of fresh agricultural products based on ubiquitous network (commercial Wi-Fi).The proposed Wi-freshness consists of four modules: CSI data sensing, data pre-processing, detection modelling and freshness detection.Considering the need for Wi-freshness data characteristic value processing is not much, and high demand for real-time prediction characteristics, a detection model based broad learning system (BLS) was proposed.Experiment shows that Wi-freshness system can achieve more than 93% detection accuracy.