SVM+KF Target Tracking Strategy Using the Signal Strength in Wireless Sensor Networks.

Area of Science:

• Wireless Sensor Networks
• Signal Processing
• Machine Learning

Background:

• Target Tracking (TT) is crucial for wireless sensor networks.
• RSSI-based TT is cost-effective but lacks stability and precision due to environmental factors.
• Existing methods struggle with nonlinear data and dynamic conditions.

Purpose of the Study:

• To develop an innovative and robust Target Tracking algorithm for wireless sensor networks.
• To improve the accuracy and stability of RSSI-based TT.
• To address the limitations of conventional TT methods in complex environments.

Main Methods:

• Proposed a novel SVM+KF algorithm combining Support Vector Machine (SVM) for initial nonlinear estimation and an improved Kalman Filter (KF) for refinement.
• Implemented an enhanced KF with a dynamically adjusted innovation update threshold based on target speed and network parameters.
• Utilized Received Signal Strength Indication (RSSI) for target localization.

Main Results:

• The SVM+KF algorithm demonstrated superior tracking accuracy compared to existing methods.
• The enhanced Kalman filter significantly improved the stability of the tracking results.
• Simulations and real-world experiments validated the algorithm's effectiveness across diverse scenarios.

Conclusions:

• The SVM+KF method offers a significant advancement in RSSI-based Target Tracking for wireless sensor networks.
• The dynamic innovation threshold in the improved KF is key to achieving robust and stable tracking.
• This algorithm provides a more reliable solution for applications requiring precise and stable target localization.