Date（日期）: January 12th

Time(时间): 10:00 AM – 11:00 AM

Location (地点): Room D308, School of Internet of Things Engineering

Abstract （报告简介）：This short talk is devoted to discussing advanced techniques in Differential Received Signal Strength (DRSS) based localization for Wireless Sensor Networks (WSNs). Source localization using Received Signal Strength (RSS) is an important practical problem due to its cost-effectiveness and simplicity, as it does not require complex timing synchronization or antenna arrays. Specifically, DRSS approaches eliminate the dependency on knowing the target's transmit power, making them highly suitable for non-cooperative target tracking.

The objective of this talk is to address two fundamental challenges in DRSS localization: estimation accuracy and sensor deployment strategies. First, we examine the significant bias problem found in closed-form location estimators derived from linearized propagation path loss models. This bias arises from noise injection during linearization, particularly in high-noise environments. We propose a robust solution using the Weighted Instrumental Variable (WIV) method combined with Selective Power Measurement to maintain the correlation between instruments and data, thereby effectively reducing estimation bias.

Secondly, we investigate the impact of sensor-target geometry on localization performance. Utilizing the D-optimality criterion and the Fisher Information Matrix (FIM), we analyze the optimal angular separation of sensors. We demonstrate that while equiangular separation constitutes the best geometry for equal target-sensor ranges, optimal geometries require specific analytical adjustments when the distances between the target and sensors vary.

Introduction of lecture (报告人简介)：Jun Li earned his Bachelor of Science (Honours) in Electronics and Communications from the University of South Australia in 2015, followed by his Doctor of Philosophy in Telecommunications Engineering from the Adelaide University（UniSA） in 2021. In the same year, he also obtained a Graduate Certificate in Data Analysis from the Melbourne Institute of Technology.

Beyond his academic achievements, Dr. Li has collaborated with NASA on lunar exploration initiatives, contributing his expertise to search and rescue operations, orbit design, and signal localization for lunar missions.

Since 2025, he has been affiliated with QMF, where he currently serves as an IT and Business Analyst focusing on system integrity and automated workflow development. From 2023 to 2025, he worked as a Data Engineer and Data Analyst at Mott MacDonald, specializing in hydrological sensor integration and data visualization optimization to support data-driven decision-making. Dr. Li’s professional interests span telecommunications engineering, data science, statistical modelling, and business process automation leveraging machine learning algorithms.