Collaborative Peer-to-Peer Training and Target Classification in Wireless Sensor Networks

Xue Wang,Sheng Wang,Daowei Bi,Liang Ding,Junjie Ma

DOI: https://doi.org/10.1109/FGCN.2007.87

2007-01-01

Abstract:30480208

What problem does this paper attempt to address?

Collaborative Target Classification for Image Recognition in Wireless Sensor Networks

Xue Wang,Sheng Wang,Junjie Ma

DOI: https://doi.org/10.1007/978-3-540-73871-8_19

2007-01-01

Abstract:Target classification, especially visual target classification, in complex situations is challenging for image recognition in wireless sensor networks (WSNs). The distributed and online learning for target classification is significant for highly-constrained WSNs. This paper presents a collaborative target classification algorithm for image recognition in WSNs, taking advantages of the collaboration for the data mining between multi-sensor nodes. The proposed algorithm consists of three steps, target detection and feature extraction are based on single-sensor node processing, whereas target classification is implemented by collaboration between multi-sensor nodes using collaborative support vector machines (SVMs). For conquering the disadvantages of inevitable missing rate and false rate in target detection, the proposed collaborative SVM adopts a robust mechanism for adaptive sample selection, which improves the incremental learning of SVM by just fusing the information from a selected set of wireless sensor nodes. Furthermore, a progressive distributed framework for collaborative SVM is also introduced for enhancing the collaboration between multi-sensor nodes. Experimental results demonstrate that the proposed collaborative target classification algorithm for image recognition can accomplish target classification quickly and accurately with little congestion, energy consumption and execution time.
Peer-To-Peer Collaborative Signal Processing for Target Tracking in Wireless Sensor Networks

Xue Wang,Sheng Wang

DOI: https://doi.org/10.1109/gcc.2006.69

2006-01-01

Abstract:A distributed dynamic collaboration approach built in peer-to-peer (P2P) architecture is presented to acquire, transmit and process information in wireless sensor networks (WSNs). Several practically feasible measures of energy consumption and information utility was introduced in collaborative signal processing (CSP) for nodes selection. Then we introduce a target tracking method and present a simple but robust algorithm to integrate the results from multiple nodes. Furthermore, we introduce a progressive fusion tracking strategy based on P2P collaborative signal processing (P2P-CSP) mechanism. Finally, we illustrate an indoor tracking experiment, as well as compare the energy consumption and time delay of P2P-CSP and client/server mechanism. Experimental results verify that the proposed algorithm is a scalable, progressive, energy-efficient and robust way for target tracking.
Dynamic Collaborative Support Vector Machine For Target Classification In Wireless Sensor Networks

Wang Xue,Wang Sheng,Bi Daowei,Ding Liang

IF: 1.019

2008-01-01

Chinese Journal of Electronics

Abstract:Target classification is a significant topic addressed in Wireless sensor networks (WSNs). The distributed learning for target classification is desired for highly-constrained WSN since the processing ability, energy and bandwidth are strictly limited. This paper presents a dynamic data fusion based collaborative Support vector machine (SVM) for target classification in WSN. With the collaboration of multiple sensor nodes, the dynamic collaborative SVM can independently select the optimal set of sensor nodes according to the estimated energy consumption and information contribution. Then the training process of SVM is incrementally carried out in the dynamic data fusion framework with the consideration of tradeoff between energy consumption and information contribution. Because of the purposeful sensor nodes selection strategy, the dynamic collaborative SVM can conquer the inevitable missing rate and false rate of samples in WSN and save the energy for prolonging the lifetime of WSN. Furthermore, the dynamic collaborative SVM can also improve the robustness in target classification, and achieve the tradeoff between the performance and the cost. The experimental results demonstrate that the proposed dynamic collaborative SVM can effectively implement target classification in WSN. It is also verified that the proposed dynamic collaborative SVM has outstanding performance in energy efficiency and time delay.
Collaborative Target Classification with Multiagent System in Wireless Multimedia Sensor Networks

Xue Wang,Xinyao Sun,Daowei Bi

DOI: https://doi.org/10.1109/i2mtc.2012.6229232

2012-01-01

Abstract:Recent technological advances in hardware design and wireless communications, together with the availability of low cost microphones and video camera, have stimulated the emergence of wireless multimedia sensor networks (WMSNs). This paper proposes to explore such novel approaches as inspired by multi-agent system (MAS) researches that conduct systematic investigation of interaction between autonomous entities. Within the MAS framework, a hierarchical WMSN architecture is established and swarm intelligence is introduced to facilitate its formation. Contract net protocol is employed to achieve efficient resource allocation among the sensor nodes for the purpose of intruding target classification. To make the best global decision, the mechanism of committee voting is exploited to combine all the individual decisions. The supervised machine learning techniques of Gaussian process classifier is presented for investigating the performance of the proposed hierarchical WMSN architecture. Simulation experiments with real-world data have been extensively performed to evaluate the proposed architecture, resource allocation and decision fusion mechanisms, and the results show that the proposed MAS approaches are effective and efficient.
Distributed Peer-to-Peer Target Tracking in Wireless Sensor Networks

Xue Wang,Sheng Wang,Dao-Wei Bi,Jun-Jie Ma

DOI: https://doi.org/10.3390/s7061001

2007-01-01

Abstract:Target tracking is usually a challenging application for wireless sensor networks(WSNs) because it is always computation-intensive and requires real-time processing. Thispaper proposes a practical target tracking system based on the auto regressive movingaverage (ARMA) model in a distributed peer-to-peer (P2P) signal processing framework.In the proposed framework, wireless sensor nodes act as peers that perform target detection,feature extraction, classification and tracking, whereas target localization requires thecollaboration between wireless sensor nodes for improving the accuracy and robustness.For carrying out target tracking under the constraints imposed by the limited capabilities ofthe wireless sensor nodes, some practically feasible algorithms, such as the ARMA modeland the 2-D integer lifting wavelet transform, are adopted in single wireless sensor nodesdue to their outstanding performance and light computational burden. Furthermore, aprogressive multi-view localization algorithm is proposed in distributed P2P signalprocessing framework considering the tradeoff between the accuracy and energyconsumption. Finally, a real world target tracking experiment is illustrated. Results fromexperimental implementations have demonstrated that the proposed target tracking systembased on a distributed P2P signal processing framework can make efficient use of scarceenergy and communication resources and achieve target tracking successfully.
Agent Collaborative Target Localization And Classification In Wireless Sensor Networks

Xue Wang,Dao-Wei Bi,Liang Ding,Sheng Wang

DOI: https://doi.org/10.3390/s7081359

IF: 3.9

2007-01-01

Sensors

Abstract:Wireless sensor networks (WSNs) are autonomous networks that have been frequently deployed to collaboratively perform target localization and classification tasks. Their autonomous and collaborative features resemble the characteristics of agents. Such similarities inspire the development of heterogeneous agent architecture for WSN in this paper. The proposed agent architecture views WSN as multi-agent systems and mobile agents are employed to reduce in-network communication. According to the architecture, an energy based acoustic localization algorithm is proposed. In localization, estimate of target location is obtained by steepest descent search. The search algorithm adapts to measurement environments by dynamically adjusting its termination condition. With the agent architecture, target classification is accomplished by distributed support vector machine (SVM). Mobile agents are employed for feature extraction and distributed SVM learning to reduce communication load. Desirable learning performance is guaranteed by combining support vectors and convex hull vectors. Fusion algorithms are designed to merge SVM classification decisions made from various modalities. Real world experiments with MICAz sensor nodes are conducted for vehicle localization and classification. Experimental results show the proposed agent architecture remarkably facilitates WSN designs and algorithm implementation. The localization and classification algorithms also prove to be accurate and energy efficient.
Particle Swarm Optimization Clustering for Target Classification in Wireless Sensor Networks

Daowei Bi,Xue Wang,Sheng Wang

DOI: https://doi.org/10.1109/ICNC.2008.135

2008-01-01

Abstract:Wireless sensor networks (WSNs) is an emerging technology that enables information retrieval from the environment by densely deployed tiny, low-cost and low-power wireless device called sensor nodes. In this paper, we explore a two-tiered WSN model containing both static and mobile sensor nodes, and focus on vehicular target classification with the small sample kernel classifier of support vector machine (SVM). Clustering is employed to achieve energy efficiency in battery powered WSNs and facilitate collaborative processing that promises to improve classification accuracy. Since clustering is an NP-hard problem, particle swarm optimization (PSO), a stochastic optimization technique emulating the behavior of a flock of birds, is used to search for the optimal cluster formation. In addition, a simple yet effective cluster number estimate technique is put forward, which takes into account the maximum communication range. Collaborative target classification is implemented with a simple voting scheme. Simulation experiments show that PSO clustering is effective and collaborative SVM classification markedly improves target classification accuracy.
Collaborative Probability Based Multimodel Target Identification in Wireless Sensor Networks

SUN Xin-yao,WANG Xue,WANG Sheng

DOI: https://doi.org/10.3969/j.issn.1000-436x.2011.06.019

2011-01-01

Journal of Communications

Abstract:A collaborative probability based multimodel target identification method for the applications in wireless sen-sor networks was proposed.Wavelet package was used to extract the features of the target information sample,and in or-der to decrease the computing complexity and reserve the key information as well,feature compaction,which achieved by principal components analysis,was carried out before learning.Different types of Gaussian classifier were developed for classifier learning.To gain global optimal results,committee decision was employed to select and combine individual decisions with dynamically adjusted weights.The experiment results verified that the method introduced can adapt to the interference factor of the environment and can implement target identification accurately and stably.
Distributed visual-target-surveillance system in wireless sensor networks

Xue Wang,Sheng Wang,Daowei Bi

DOI: https://doi.org/10.1109/TSMCB.2009.2013196

Abstract:A wireless sensor network (WSN) is a powerful unattended distributed measurement system, which is widely used in target surveillance because of its outstanding performance in distributed sensing and signal processing. This paper introduces a multiview visual-target-surveillance system in WSN, which can autonomously implement target classification and tracking with collaborative online learning and localization. The proposed system is a hybrid system of single-node and multinode fusion. It is constructed on a peer-to-peer (P2P)-based computing paradigm and consists of some simple but feasible methods for target detection and feature extraction. Importantly, a support-vector-machine-based semisupervised learning method is used to achieve online classifier learning with only unlabeled samples. To reduce the energy consumption and increase the accuracy, a novel progressive data-fusion paradigm is proposed for online learning and localization, where a feasible routing method is adopted to implement information transmission with the tradeoff between performance and cost. Experiment results verify that the proposed surveillance system is an effective, energy-efficient, and robust system for real-world application. Furthermore, the P2P-based progressive data-fusion paradigm can improve the energy efficiency and robustness of target surveillance.
Distributed sensor activation algorithm for target tracking with binary sensor networks

Jiming Chen,Kejie Cao,Keyong Li,Youxian Sun

DOI: https://doi.org/10.1007/s10586-009-0092-0

2009-01-01

Cluster Computing

Abstract:Target tracking with wireless sensor networks (WSNs) has been a hot research topic recently. Many works have been done to improve the algorithms for localization and prediction of a moving target with smart sensors. However, the results are frequently difficult to implement because of hardware limitations. In this paper, we propose a practical distributed sensor activation algorithm (DSA 2 ) that enables reliable tracking with the simplest binary-detection sensors. In this algorithm, all sensors in the field are activated with a probability to detect targets or sleep to save energy, the schedule of which depends on their neighbor sensors’ behaviors. Extensive simulations are also shown to demonstrate the effectiveness of the proposed algorithm. Great improvement in terms of energy-quality tradeoff and excellent robustness of the algorithm are also emphasized in the simulations.
Collaborative hybrid classifier learning with ant colony optimization in wireless multimedia sensor networks

Sheng Wang,Xue Wang,Liang Ding,Daowei Bi,Zheng You

DOI: https://doi.org/10.1109/WCICA.2008.4594495

2008-01-01

Abstract:Wireless multimedia sensor network (WMSN) has powerful multimedia signal acquisition and processing abilities. This paper proposes a collaborative hybrid classifier learning algorithm to achieve online support vector machine (SVM) learning for robust target classification in WMSN. The proposed algorithm is carried out in a hybrid computing paradigm, which combines the advantages of progressive computing paradigm and P2P computing paradigm. Importantly, the participant sensor nodes are purposefully selected by evaluating the specific effectiveness. With the sensor nodes selection strategy, the energy consumption and the impact of inevitable missing detection and false detection can be reduced. Besides, ant colony optimization is also used for decreasing the energy consumption in routing. Experimental results demonstrate that the collaborative hybrid classifier learning algorithm can effectively implement target classification in WMSN, and the ant colony optimization based routing and clustering method can largely decrease the energy consumption and time cost.
Bidding and Voting Strategy for Energy Efficient Collaborative Target Classification in Wireless Sensor Networks

Xue Wang,Daowei Bi,Liang Ding,Sheng Wang

DOI: https://doi.org/10.1109/CISP.2008.8

2008-01-01

Abstract:Recent advances in wireless communications and micro-electro-mechanical systems have fostered the development of wireless sensor networks (WSNs) which consist of tiny, low-cost and low-power sensor nodes. Individual sensor nodes only have local views of the environment, but global views can be obtained by collaborative processing between densely deployed sensor nodes. Energy efficiency is critical for battery powered WSNs to prolong network lifetime. It is very challenging to concurrently achieve energy efficiency and collaborative processing in WSNs. Inspired by some commercial practices, we proposed the bidding and voting strategy (BVS), and apply it to vehicular target classification in WSNs. In BVS, bidding efficiently selects sensor nodes for collaborative classification while voting improves classification decision accuracy by combining local decisions. Vehicular targets are classified by means of support vector machine (SVM). Simulation experiments with real world data are conducted and the results show the proposed strategy can guarantee target classification accuracy while achieve energy efficiency.
Hierarchical Wireless Multimedia Sensor Networks for Collaborative Hybrid Semi-Supervised Classifier Learning

Xue Wang,Sheng Wang,Daowei Bi,Liang Ding

DOI: https://doi.org/10.3390/s7112693

IF: 3.9

2007-01-01

Sensors

Abstract:Wireless multimedia sensor networks (WMSN) have recently emerged as one of the most important technologies, driven by the powerful multimedia signal acquisition and processing abilities. Target classification is an important research issue addressed in WMSN, which has strict requirement in robustness, quickness and accuracy. This paper proposes a collaborative semi- supervised classifier learning algorithm to achieve durative online learning for support vector machine (SVM) based robust target classification. The proposed algorithm incrementally carries out the semi- supervised classifier learning process in hierarchical WMSN, with the collaboration of multiple sensor nodes in a hybrid computing paradigm. For decreasing the energy consumption and improving the performance, some metrics are introduced to evaluate the effectiveness of the samples in specific sensor nodes, and a sensor node selection strategy is also proposed to reduce the impact of inevitable missing detection and false detection. With the ant optimization routing, the learning process is implemented with the selected sensor nodes, which can decrease the energy consumption. Experimental results demonstrate that the collaborative hybrid semisupervised classifier learning algorithm can effectively implement target classification in hierarchical WMSN. It has outstanding performance in terms of energy efficiency and time cost, which verifies the effectiveness of the sensor nodes selection and ant optimization routing.
Multi-agent Negotiation Mechanisms for Statistical Target Classification in Wireless Multimedia Sensor Networks.

Xue Wang,Dao-Wei Bi,Liang Ding,Sheng Wang

DOI: https://doi.org/10.3390/s7102201

2007-01-01

Abstract:The recent availability of low cost and miniaturized hardware has allowedwireless sensor networks (WSNs) to retrieve audio and video data in real worldapplications, which has fostered the development of wireless multimedia sensor networks(WMSNs). Resource constraints and challenging multimedia data volume makedevelopment of efficient algorithms to perform in-network processing of multimediacontents imperative. This paper proposes solving problems in the domain of WMSNs fromthe perspective of multi-agent systems. The multi-agent framework enables flexible networkconfiguration and efficient collaborative in-network processing. The focus is placed ontarget classification in WMSNs where audio information is retrieved by microphones. Todeal with the uncertainties related to audio information retrieval, the statistical approachesof power spectral density estimates, principal component analysis and Gaussian processclassification are employed. A multi-agent negotiation mechanism is specially developed toefficiently utilize limited resources and simultaneously enhance classification accuracy andreliability. The negotiation is composed of two phases, where an auction based approach isfirst exploited to allocate the classification task among the agents and then individual agentdecisions are combined by the committee decision mechanism. Simulation experiments withreal world data are conducted and the results show that the proposed statistical approachesand negotiation mechanism not only reduce memory and computation requi.
Study on signal processing in wireless sensor networks

Chen Jiming,Lin Ruizhong,Sun Youxian

DOI: https://doi.org/10.3321/j.issn:0254-3087.2006.09.030

2006-01-01

Abstract:Wireless sensor network is data-centric signal processing system. In this paper, the concept of Collaborative Signal and Information Processing (CSIP) is proposed for the requirements of signal processing in target tracking. The experiments for evaluating CSIP based on MICA2 and PDA platform were designed, and (m, k)-firm based scheduling scheme was adopted. Experimental results reveal that the (m, k)-firm based scheme is more effective in decreasing the information loss-rate and reducing the signal distortion than EDF.
Collaborative Statistical Learning with Rough Feature Reduction for Visual Target Classification

Sheng Wang,Xue Wang,Daowei Bi,Liang Ding,Zheng You

DOI: https://doi.org/10.1109/ijcnn.2008.4633944

2008-01-01

Abstract:To implement visual target classification, this paper proposes a collaborative statistical learning algorithm for online support vector machine(SVM) classifier learning in wireless multimedia sensor network (WMSN). For achieving robust target classification, classifier learning should be carried out iteratively for updating classifiers according to various situations. Because only unlabeled samples can be acquired, semi-supervised learning is desired to make full use of unlabeled samples. According to the restrict limitation in energy and bandwidth, the proposed algorithm incrementally implement classifier learning with the selected features from multiple sensor nodes, where rough set based feature reduction is used for retaining most of the intrinsic information. Furthermore, some metrics are introduced to evaluate the effectiveness of the samples in specific sensor nodes, and a sensor node selection strategy is also proposed to reduce the impact of inevitable missing detection and false detection. Experimental results demonstrate that the collaborative statistical learning algorithm can effectively implement target classification in WMSN. With the rough set based feature reduction, the proposed algorithm has outstanding performance in energy efficiency and time cost.
Collaborative Signal Processing for Target Tracking in Distributed Wireless Sensor Networks

Xue Wang,Sheng Wang

DOI: https://doi.org/10.1016/j.jpdc.2007.02.001

IF: 4.542

2007-01-01

Journal of Parallel and Distributed Computing

Abstract:Target tracking, especially visual target tracking, in complex situations is challenging, which is always performed in single-view system. Because of the conflict between resolution and tracking range, however, single-view tracking is not robust and accurate. This paper presents a distributed multi-view tracking system using collaborative signal processing (CSP) in distributed wireless sensor networks (DWSNs). In the proposed tracking system, target detection and classification algorithms are based on single-node processing and target tracking is performed in sink node, whereas target localization algorithm is carried out by CSP between multisensor. For conquering the disadvantages of client/server based centralized data fusion, a progressive distributed data fusion are proposed. Finally, an indoor target tracking experiment is illustrated, and then tracking performance, execution time and energy consumption of progressive distributed data fusion are compared with client/server based centralized data fusion. Experimental results demonstrate that the CSP based distributed multi-view tracking system in DWSNs can accomplish multi-target extraction, classification, localization, tracking and association quickly and accurately with little congestion, energy consumption and execution time.
Networked Ultrasonic Sensors for Target Tracking: An Experimental Study

Hongbin Li,Di Miao,Jiming Chen,Youxian Sun,Xuemin Shen

DOI: https://doi.org/10.1109/GLOCOM.2009.5425343

2009-01-01

Abstract:As an emerging application, tracking mobile targets using distributed and networked sensors is gaining noticeable attentions in recent years. In this paper, we utilize ultrasonic sensors, which are traditionally viewed as basic ranging components, to build a target detection and tracking system. An integrated sensor node, with one communication/computation module, one Passive InfraRed (PIR) module and three ultrasonic modules, is designed to support local detection/ranging, data filtering and result reporting. Comprised of a set of the designed sensor nodes, our system is able to track the position of a dynamic target, using Extended Kalman Filter. Simulations and experimental results are given to evaluate performance bound and demonstrate the efficiency of our system.
Adaptive Sensor Activation for Target Tracking in Wireless Sensor Networks.

Jiming Chen,Kejie Cao,Youxian Sun,Xuemin (Sherman) Shen

DOI: https://doi.org/10.1109/icc.2009.5198708

2009-01-01

Abstract:This paper presents an adaptive sensor activation for target tracking in wireless sensor networks by dynamically adjusting the range of sensor selective activation instead of fixed one. A closed-loop control algorithm for the range of adaptive sensor activation is designed according to the online feedback of the tracking quality. The failed tracking case can also be handled by the proposed algorithm. Extensive simulation results show that the adaptive sensor activation achieves higher performance in terms of tracking effect and energy efficiency.
Collaborative Target Tracking Strategy in Distributed Wireless Sensor Networks

WANG Xue,WANG Sheng,MA Jun-jie

DOI: https://doi.org/10.3321/j.issn:0372-2112.2007.05.026

2007-01-01

Abstract:Target tracking in complex situations is challenging,especially in visual tracking,which is always performed in single-view system.Because of the conflict between resolution and tracking range,however,single-view tracking is not robust and accurate.This paper presents a distributed multi-view tracking system using collaborative signal processing(CSP) in distributed wireless sensor networks(DWSNs),and introduces a progressive distributed data fusion mechanism.This mechanism dynamically chooses the optimal sensor nodes and schedules the fusion process with several measures of energy consumption and information utility.Finally,an indoor target tracking experiment is illustrated,and then tracking performance,execution time and energy consumption of progressive distributed data fusion based CSP are compared with centralized data fusion based CSP.Experimental results demonstrate that the distributed multi-view tracking system using progressive distributed data fusion based CSP in DWSNs can track the 3-D positions of targets quickly and accurately,and decrease the congestion,energy consumption and time latency in communication.

Collaborative Peer-to-Peer Training and Target Classification in Wireless Sensor Networks

Collaborative Target Classification for Image Recognition in Wireless Sensor Networks

Peer-To-Peer Collaborative Signal Processing for Target Tracking in Wireless Sensor Networks

Dynamic Collaborative Support Vector Machine For Target Classification In Wireless Sensor Networks

Collaborative Target Classification with Multiagent System in Wireless Multimedia Sensor Networks

Distributed Peer-to-Peer Target Tracking in Wireless Sensor Networks

Agent Collaborative Target Localization And Classification In Wireless Sensor Networks

Particle Swarm Optimization Clustering for Target Classification in Wireless Sensor Networks

Collaborative Probability Based Multimodel Target Identification in Wireless Sensor Networks

Distributed visual-target-surveillance system in wireless sensor networks

Distributed sensor activation algorithm for target tracking with binary sensor networks

Collaborative hybrid classifier learning with ant colony optimization in wireless multimedia sensor networks

Bidding and Voting Strategy for Energy Efficient Collaborative Target Classification in Wireless Sensor Networks

Hierarchical Wireless Multimedia Sensor Networks for Collaborative Hybrid Semi-Supervised Classifier Learning

Multi-agent Negotiation Mechanisms for Statistical Target Classification in Wireless Multimedia Sensor Networks.

Study on signal processing in wireless sensor networks

Collaborative Statistical Learning with Rough Feature Reduction for Visual Target Classification

Collaborative Signal Processing for Target Tracking in Distributed Wireless Sensor Networks

Networked Ultrasonic Sensors for Target Tracking: An Experimental Study

Adaptive Sensor Activation for Target Tracking in Wireless Sensor Networks.

Collaborative Target Tracking Strategy in Distributed Wireless Sensor Networks