Shuang Zhou
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
Shuang Zhou has a total of 40 co-authors. Here are the top co-authors by number of shared publications:
- Zhaoming Lu
Beijing University of Posts and Telecommunications
10 shared publications - Xiangming Wen
Beijing Key Laboratory of Network System Architecture and Convergence and Beijing Laboratory of Advanced Information Networks, Beijing University of Posts and Telecommunications, Beijing, China
6 shared publications - Zijun Han
Key Laboratory of High Confidence Software Technologies (Ministry of Education), school of Computer Science, Peking University, Beijing, China
4 shared publications - Jingbo Zhao
School of Information and Communication Engineering, Beijing University of Posts and Telecommunications, Beijing, China
3 shared publications - Lingchao Guo
School of Information and Communication Engineering, Beijing University of Posts and Telecommunications, Beijing, China
3 shared publications - Wanming Hao
Zhengzhou University
3 shared publications
Latest Publications
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13  (22) , 55459-55475
AbsTrack: Multiperson Absolute Localization and Trajectory Tracking Using Commodity Wi-Fi
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13  (22) , 55459-55475
AbsTrack: Multiperson Absolute Localization and Trajectory Tracking Using Commodity Wi-Fi
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 26  (15) , 1
Robust Human Sensing through Boundary Control with Commodity WiFi
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 26  (15) , 1
Robust Human Sensing through Boundary Control with Commodity WiFi
- 18 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 25  (15361276) , 3960-3977
Secure Energy Efficiency Optimization for Sub-Connected Active RIS-Assisted mmWave ISAC System
In this paper, we investigate a millimeter-wave secure integrated sensing and communication system assisted by the sub-connected (SC) active reconfigurable intelligent surface (SC-ARIS), where the dual-function radar and communication station (DFBS) employs a hybrid precoding structure. We formulate an optimization problem to jointly design the DFBS hybrid precoding and SC-ARIS beamforming, aiming to maximize the secure energy efficiency while ensuring communication and sensing qualities. To address the above non-convex problem, we utilize alternating optimization technique to decouple it into two subproblems, where DFBS hybrid precoding and SC-ARIS beamforming are respectively optimized. For the first one, we first propose an iterative algorithm to solve the equivalent digital precoding based on constrained concave-convex procedure, Taylor expansion, semidefinite relaxation (SDR) and fractional programming techniques. Then, the hybrid precoding is obtained rely on the manifold optimization alternating minimization technique. For the later one, we propose an iterative algorithm based on the SDR. Considering a more realistic scenario, we extend to the imperfect eavesdropping channel, and propose a robust beamforming design scheme. Finally, simulation results show the effectiveness of the proposed schemes.
- 18 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 25  (15361276) , 3960-3977
Secure Energy Efficiency Optimization for Sub-Connected Active RIS-Assisted mmWave ISAC System
In this paper, we investigate a millimeter-wave secure integrated sensing and communication system assisted by the sub-connected (SC) active reconfigurable intelligent surface (SC-ARIS), where the dual-function radar and communication station (DFBS) employs a hybrid precoding structure. We formulate an optimization problem to jointly design the DFBS hybrid precoding and SC-ARIS beamforming, aiming to maximize the secure energy efficiency while ensuring communication and sensing qualities. To address the above non-convex problem, we utilize alternating optimization technique to decouple it into two subproblems, where DFBS hybrid precoding and SC-ARIS beamforming are respectively optimized. For the first one, we first propose an iterative algorithm to solve the equivalent digital precoding based on constrained concave-convex procedure, Taylor expansion, semidefinite relaxation (SDR) and fractional programming techniques. Then, the hybrid precoding is obtained rely on the manifold optimization alternating minimization technique. For the later one, we propose an iterative algorithm based on the SDR. Considering a more realistic scenario, we extend to the imperfect eavesdropping channel, and propose a robust beamforming design scheme. Finally, simulation results show the effectiveness of the proposed schemes.
- 27 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74  (00189456) , 1-16
MUID: Multiperson Gait Identification With Commodity Wi-Fi
With the popularization of smart homes and Internet of Things, human gait identification using Wi-Fi channel state information (CSI) has attracted much attention. However, it remains a challenge to accurately recognize multi-person identities when they walk simultaneously in the sensing area. Existing researches separate the walking signals of each person from the multipath superposed CSI and then construct the Doppler spectrum of separated signals to extract human gait features for multi-person identification. But these systems suffer from the unsatisfactory accuracy since the extracted gait features are inaccurate and inadequate from the original Doppler spectrum, which is position-dependent and has low temporal resolution. To this end, we propose MUID, a Wi-Fi-based system capable of accurately recognizing multi-person identities using commercial devices. In MUID, we first leverage an existing model proposed by our previous work to separate each person’s walking signals and propose a spectrum calibration method to remove the position dependency of the Doppler spectrum of separated signals. Next, we novelly extract gait Temporal AutoCorrelation (TAC) information and use it to compensate for the low temporal resolution of the Doppler spectrum. Finally, we design an attention-based two-branch feature fusion network, which can effectively combine the gait features extracted from the calibrated Doppler spectrum and TAC for accurate multi-person identification. Extensive experiments are conducted in three different environments and MUID can respectively achieve 93.56% and 90.59% average accuracy for two-person and three-person identification, superior to the state-of-the-art methods.
- 27 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74  (00189456) , 1-16
MUID: Multiperson Gait Identification With Commodity Wi-Fi
With the popularization of smart homes and Internet of Things, human gait identification using Wi-Fi channel state information (CSI) has attracted much attention. However, it remains a challenge to accurately recognize multi-person identities when they walk simultaneously in the sensing area. Existing researches separate the walking signals of each person from the multipath superposed CSI and then construct the Doppler spectrum of separated signals to extract human gait features for multi-person identification. But these systems suffer from the unsatisfactory accuracy since the extracted gait features are inaccurate and inadequate from the original Doppler spectrum, which is position-dependent and has low temporal resolution. To this end, we propose MUID, a Wi-Fi-based system capable of accurately recognizing multi-person identities using commercial devices. In MUID, we first leverage an existing model proposed by our previous work to separate each person’s walking signals and propose a spectrum calibration method to remove the position dependency of the Doppler spectrum of separated signals. Next, we novelly extract gait Temporal AutoCorrelation (TAC) information and use it to compensate for the low temporal resolution of the Doppler spectrum. Finally, we design an attention-based two-branch feature fusion network, which can effectively combine the gait features extracted from the calibrated Doppler spectrum and TAC for accurate multi-person identification. Extensive experiments are conducted in three different environments and MUID can respectively achieve 93.56% and 90.59% average accuracy for two-person and three-person identification, superior to the state-of-the-art methods.
- 11 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Vehicles
- Vol. 10  (1) , 359-372
Autonomous Driving via Knowledge-Enhanced Safe Reinforcement Learning
Recently, the autonomous driving technology is at a critical phase evolving from typical, closed scenarios to largescale, open driving scenarios, which is challenged by the diversity and complexity of the driving environment. Reinforcement learning has the potential to address this challenge, but its application has been hindered by safety issues. To address the safety concern, we propose a Lagrangian framework called Knowledge-Enhanced Safe Reinforcement Learning (KESRL), which integrates prior knowledge of model-based methods into the model-free safe RL framework. Firstly, a model-based constraint based on high-order control barrier function is derived to deduce the safety criteria for the control system. Secondly, we propose a model-free constraint which estimates the distribution of accumulated safety cost and restricts the worst-case safety cost value. Furthermore, the above constraints are unified into the proposed Lagrangian framework to equip the end-to-end autonomous driving system with safety constraints. Experimental results conducted in the CARLA simulator demonstrate the effectiveness of the proposed framework KESRL in decreasing the collision rate and enhancing the control safety compared with several state-of-the-art RL methods.
- 11 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Vehicles
- Vol. 10  (1) , 359-372
Autonomous Driving via Knowledge-Enhanced Safe Reinforcement Learning
Recently, the autonomous driving technology is at a critical phase evolving from typical, closed scenarios to largescale, open driving scenarios, which is challenged by the diversity and complexity of the driving environment. Reinforcement learning has the potential to address this challenge, but its application has been hindered by safety issues. To address the safety concern, we propose a Lagrangian framework called Knowledge-Enhanced Safe Reinforcement Learning (KESRL), which integrates prior knowledge of model-based methods into the model-free safe RL framework. Firstly, a model-based constraint based on high-order control barrier function is derived to deduce the safety criteria for the control system. Secondly, we propose a model-free constraint which estimates the distribution of accumulated safety cost and restricts the worst-case safety cost value. Furthermore, the above constraints are unified into the proposed Lagrangian framework to equip the end-to-end autonomous driving system with safety constraints. Experimental results conducted in the CARLA simulator demonstrate the effectiveness of the proposed framework KESRL in decreasing the collision rate and enhancing the control safety compared with several state-of-the-art RL methods.
- 17 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 11  (18) , 29455-29469
Joint Beamforming Design for Hybrid RIS-Assisted mmWave ISAC System Relying on Hybrid Precoding Structure
In this article, we investigate a millimeter wave integrated sensing and communication system with aid of the hybrid reconfigurable intelligent surface (HRIS), where the dual-function radar and communication station (DFBS) applies the hybrid precoding (HP) structure. On this basis, we consider the sensing and communication performance, respectively, and formulate two optimization problems. One is to maximize the worst-case illumination power while ensuring the communication quality, and another is to maximize the total achievable rate while satisfying the sensing performance. To solve them, we first decouple each nonconvex problem into three subproblems via the alternative optimization technique. For the former one, we transform DFBS and HRIS beamforming optimization subproblems into the convex ones by the quadratic constrained quadratic programming (QCQP) and semidefinite program relaxation (SDR) techniques, and obtain the solutions by standard convex optimization technique. For the later one, fractional programming is applied to decouple the objective function, and then we transform DFBS and HRIS beamforming design subproblems into the convex ones by QCQP and Taylor expansion techniques, and obtain the solutions by the alternating direction method of multipliers (ADMMs). For the HP design subproblems of DFBS in both problems, a manifold optimization-alternating minimization (MO-AltMin) algorithm based on minimizing the Euclidean distance is used to obtain the solutions. Simulation results show the effectiveness of the proposed schemes.
- 17 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 11  (18) , 29455-29469
Joint Beamforming Design for Hybrid RIS-Assisted mmWave ISAC System Relying on Hybrid Precoding Structure
In this article, we investigate a millimeter wave integrated sensing and communication system with aid of the hybrid reconfigurable intelligent surface (HRIS), where the dual-function radar and communication station (DFBS) applies the hybrid precoding (HP) structure. On this basis, we consider the sensing and communication performance, respectively, and formulate two optimization problems. One is to maximize the worst-case illumination power while ensuring the communication quality, and another is to maximize the total achievable rate while satisfying the sensing performance. To solve them, we first decouple each nonconvex problem into three subproblems via the alternative optimization technique. For the former one, we transform DFBS and HRIS beamforming optimization subproblems into the convex ones by the quadratic constrained quadratic programming (QCQP) and semidefinite program relaxation (SDR) techniques, and obtain the solutions by standard convex optimization technique. For the later one, fractional programming is applied to decouple the objective function, and then we transform DFBS and HRIS beamforming design subproblems into the convex ones by QCQP and Taylor expansion techniques, and obtain the solutions by the alternating direction method of multipliers (ADMMs). For the HP design subproblems of DFBS in both problems, a manifold optimization-alternating minimization (MO-AltMin) algorithm based on minimizing the Euclidean distance is used to obtain the solutions. Simulation results show the effectiveness of the proposed schemes.
- 21 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 11  (8) , 15000-15021
MuSense: Multiperson Continuous Activity Sensing Using Commodity Wi-Fi
Wi-Fi-based continuous activity sensing is of great importance to personal healthcare, security monitoring, and healthy lifestyle assessment. However, it remains challenging to understand multiperson continuous activities as the Wi-Fi signals reflected by each person are mixed up in the Wi-Fi channel state information (CSI). To this end, we present MuSense, the first Wi-Fi-based system that enables multiperson continuous activity segmentation and recognition using commodity devices. In MuSense, we design a Wi-Fi network interface card (NIC) combination and calibration (NICCC) algorithm to construct a high-resolution receiving array and calibrate the CSI measurement noises of this array. On this basis, we propose a multiperson reflection signal separation (MPRSS) algorithm to completely and practically separate each person’s Wi-Fi reflection signals, obtaining the amplitude attenuation and phase shift corresponding to each person’s activities on the subcarrier dimension. Finally, we design an unsupervised adversarial continuous activity sensing network (UACAS-Net), with a two-stage adversarial training method to achieve generalized multiperson continuous activity sensing. Through two-stage adversarial training, UACAS-Net can capture distinguishing features of continuous activities from separated reflection signal streams in the source and target domains while minimizing the feature domain discrepancies to segment and recognize each person’s continuous activities in different domains. Intensive experiments are conducted under three different scenarios and the results demonstrate the effectiveness and practicality of MuSense for multiperson continuous activity sensing.
- 21 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 11  (8) , 15000-15021
MuSense: Multiperson Continuous Activity Sensing Using Commodity Wi-Fi
Wi-Fi-based continuous activity sensing is of great importance to personal healthcare, security monitoring, and healthy lifestyle assessment. However, it remains challenging to understand multiperson continuous activities as the Wi-Fi signals reflected by each person are mixed up in the Wi-Fi channel state information (CSI). To this end, we present MuSense, the first Wi-Fi-based system that enables multiperson continuous activity segmentation and recognition using commodity devices. In MuSense, we design a Wi-Fi network interface card (NIC) combination and calibration (NICCC) algorithm to construct a high-resolution receiving array and calibrate the CSI measurement noises of this array. On this basis, we propose a multiperson reflection signal separation (MPRSS) algorithm to completely and practically separate each person’s Wi-Fi reflection signals, obtaining the amplitude attenuation and phase shift corresponding to each person’s activities on the subcarrier dimension. Finally, we design an unsupervised adversarial continuous activity sensing network (UACAS-Net), with a two-stage adversarial training method to achieve generalized multiperson continuous activity sensing. Through two-stage adversarial training, UACAS-Net can capture distinguishing features of continuous activities from separated reflection signal streams in the source and target domains while minimizing the feature domain discrepancies to segment and recognize each person’s continuous activities in different domains. Intensive experiments are conducted under three different scenarios and the results demonstrate the effectiveness and practicality of MuSense for multiperson continuous activity sensing.
- 3 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 73  (1) , 1382-1387
Multiple-Target Doppler Frequency Estimation in ISAC With Clock Asynchronism
Doppler frequency estimation is a challenging problem in integrated sensing and communication systems with clock asynchronism between transmitter and sensing receiver. Existing techniques have various limitations and are generally ineffective in multi-target sensing. In this correspondence, we propose an array-based scheme for multiple-target Doppler frequency estimation. We first develop a novel algorithm for angle-of-arrival (AoA) estimation, and then use the estimated AoA to construct orthogonal projections to separate static paths and each dynamic path for clock asynchronism compensation and Doppler frequency estimation. Simulation and experiment results are provided and demonstrate that the proposed scheme is effective in multiple-target scenarios, and its accuracy outperforms some state-of-the-art methods in single-target scenarios.
- 3 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 73  (1) , 1382-1387
Multiple-Target Doppler Frequency Estimation in ISAC With Clock Asynchronism
Doppler frequency estimation is a challenging problem in integrated sensing and communication systems with clock asynchronism between transmitter and sensing receiver. Existing techniques have various limitations and are generally ineffective in multi-target sensing. In this correspondence, we propose an array-based scheme for multiple-target Doppler frequency estimation. We first develop a novel algorithm for angle-of-arrival (AoA) estimation, and then use the estimated AoA to construct orthogonal projections to separate static paths and each dynamic path for clock asynchronism compensation and Doppler frequency estimation. Simulation and experiment results are provided and demonstrate that the proposed scheme is effective in multiple-target scenarios, and its accuracy outperforms some state-of-the-art methods in single-target scenarios.
- 28 September 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 10  (2) , 1588-1604
Wi-Monitor: Daily Activity Monitoring Using Commodity Wi-Fi
Daily activity monitoring is essential to healthy lifestyle assessment and personal healthcare, among which Wi-Fi-based solutions have attracted increasing attention due to their no-intrusive and privacy-protected characters. However, related researches are based on the assumption that there is an interval between two activities, during which the target is thought to be static. This assumption falls short of reality as human activities are performed continuously in daily life. Therefore, this article aims to design a nonintrusive and privacy-protected system, namely, Wi-Monitor, to monitor human activities in daily life. In Wi-Monitor, we first fragmentize Wi-Fi channel state information (CSI) streams into CSI bins and design a feature extraction network to extract activity fragmentation features (AFFs) from these CSI bins. From the extracted AFFs, a temporal convolutional network (TCN) is further used to capture activity continuity features (ACFs), which are used as distinguishing characteristics of continuous activities. Finally, Wi-Monitor utilizes these distinguishing characteristics to segment and recognize human activities in daily life simultaneously to achieve daily activity monitoring. In addition, we design an over-segmentation suppression mechanism with two training stages in Wi-Monitor to overcome the over-segmentation issue and enhance the activity monitoring accuracy. Intensive experiments are conducted in three different scenarios and the results demonstrate the effectiveness and practicality of Wi-Monitor for daily activity monitoring.
- 28 September 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 10  (2) , 1588-1604
Wi-Monitor: Daily Activity Monitoring Using Commodity Wi-Fi
Daily activity monitoring is essential to healthy lifestyle assessment and personal healthcare, among which Wi-Fi-based solutions have attracted increasing attention due to their no-intrusive and privacy-protected characters. However, related researches are based on the assumption that there is an interval between two activities, during which the target is thought to be static. This assumption falls short of reality as human activities are performed continuously in daily life. Therefore, this article aims to design a nonintrusive and privacy-protected system, namely, Wi-Monitor, to monitor human activities in daily life. In Wi-Monitor, we first fragmentize Wi-Fi channel state information (CSI) streams into CSI bins and design a feature extraction network to extract activity fragmentation features (AFFs) from these CSI bins. From the extracted AFFs, a temporal convolutional network (TCN) is further used to capture activity continuity features (ACFs), which are used as distinguishing characteristics of continuous activities. Finally, Wi-Monitor utilizes these distinguishing characteristics to segment and recognize human activities in daily life simultaneously to achieve daily activity monitoring. In addition, we design an over-segmentation suppression mechanism with two training stages in Wi-Monitor to overcome the over-segmentation issue and enhance the activity monitoring accuracy. Intensive experiments are conducted in three different scenarios and the results demonstrate the effectiveness and practicality of Wi-Monitor for daily activity monitoring.
- 3 September 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Signal Processing
- Vol. 15  (6) , 1423-1438
Emergency Semantic Feature Vector Extraction From WiFi Signals for In-Home Monitoring of Elderly
The elderly population worldwide is growing drastically. Most elders prefer to live independently in their own homes, which makes them more vulnerable to emergencies and leads to higher death rates. Timely detection and notification of emergencies can minimize the impact of the adversity. In this paper, we try to extract the Emergency Semantic Feature Vector ($ESFV$ ) from the ubiquitous WiFi signals in indoor environments for non-intrusive, cost-effective monitoring of the elderly and effective emergency notification. Common emergencies of the elderly are analyzed and an$ESFV$ is defined as$ESFV = ({\lbrace \pi \rbrace,\lbrace \beta \rbrace, \lbrace \rho \rbrace })$ , which is a combination of the P osition, (abnormal) B ehavior and (abnormal) R espiration of the elderly. To extract the$ESFV$ from WiFi signals, we design a WiFi signal structuring approach, which converts WiFi signals into two structured features, i.e., keypoint maps and respiration graphs, using WiFi Channel State Information (CSI). For keypoint map construction, we extract human-related WiFi signals, based on which we generate CSI maps. Meanwhile, a synchronized camera is used to extract human keypoints and annotate WiFi signals. A neural network is designed to transform the CSI maps into keypoint maps. For respiration graph extraction, we design a respiration-related WiFi signal extraction method and a subcarrier selection algorithm to obtain the respiration graphs. Then we extract the$\lbrace \pi \rbrace$ ,$\lbrace \beta \rbrace$ ,$\lbrace \rho \rbrace$ from the two structured features:$\lbrace \pi \rbrace$ and$\lbrace \beta \rbrace$ are extracted from the keypoint maps while$\lbrace \rho \rbrace$ is extracted from the respiration graphs. Finally, the feature vector$(\lbrace \pi \rbrace,\lbrace \beta \rbrace, \lbrace \rho \rbrace)$ of an emergency semantic is formed for notification. We build a prototype system and conduct extensive experiments. Experiment results show that our approaches perform well in both signal structuring (keypoint map construction and respiration graph extraction) and$ESFV$ extraction. These demonstrate the effectiveness of our approaches for in-home monitoring and emergency notification.- 3 September 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Signal Processing
- Vol. 15  (6) , 1423-1438
Emergency Semantic Feature Vector Extraction From WiFi Signals for In-Home Monitoring of Elderly
The elderly population worldwide is growing drastically. Most elders prefer to live independently in their own homes, which makes them more vulnerable to emergencies and leads to higher death rates. Timely detection and notification of emergencies can minimize the impact of the adversity. In this paper, we try to extract the Emergency Semantic Feature Vector ($ESFV$ ) from the ubiquitous WiFi signals in indoor environments for non-intrusive, cost-effective monitoring of the elderly and effective emergency notification. Common emergencies of the elderly are analyzed and an$ESFV$ is defined as$ESFV = ({\lbrace \pi \rbrace,\lbrace \beta \rbrace, \lbrace \rho \rbrace })$ , which is a combination of the P osition, (abnormal) B ehavior and (abnormal) R espiration of the elderly. To extract the$ESFV$ from WiFi signals, we design a WiFi signal structuring approach, which converts WiFi signals into two structured features, i.e., keypoint maps and respiration graphs, using WiFi Channel State Information (CSI). For keypoint map construction, we extract human-related WiFi signals, based on which we generate CSI maps. Meanwhile, a synchronized camera is used to extract human keypoints and annotate WiFi signals. A neural network is designed to transform the CSI maps into keypoint maps. For respiration graph extraction, we design a respiration-related WiFi signal extraction method and a subcarrier selection algorithm to obtain the respiration graphs. Then we extract the$\lbrace \pi \rbrace$ ,$\lbrace \beta \rbrace$ ,$\lbrace \rho \rbrace$ from the two structured features:$\lbrace \pi \rbrace$ and$\lbrace \beta \rbrace$ are extracted from the keypoint maps while$\lbrace \rho \rbrace$ is extracted from the respiration graphs. Finally, the feature vector$(\lbrace \pi \rbrace,\lbrace \beta \rbrace, \lbrace \rho \rbrace)$ of an emergency semantic is formed for notification. We build a prototype system and conduct extensive experiments. Experiment results show that our approaches perform well in both signal structuring (keypoint map construction and respiration graph extraction) and$ESFV$ extraction. These demonstrate the effectiveness of our approaches for in-home monitoring and emergency notification.