Adaptive collaboration for heterogeneous sensor networks in dynamic environments

Collaboration between the low-quality sensor and high-quality sensor can achieve the tradeoff between accuracy and energy efficiency in heterogeneous sensor networks (HSNs). Generally, HSNs are deeply integrated with dynamic physical environments. Dynamics of the monitored target are the most important and common factors of the dynamic environments and have great influence on the system performance. If the state of the monitored target changes, some important parameters (e.g., active opportunity and sampling frequency) fails to adapt to the changes, which undermines the collaboration´s performance. Even the performance of the system is not up to the requirements or a large amount of energy is consumed. To address this problem, we propose an adaptive collaboration method (EasiAC) by the collaboration between magnetic and camera sensors. First, for the dynamics of the monitored target, EasiAC utilizes the magnetic sensors to predict the target´s state via Bayesian filtering. Second, to achieve good performance of collaboration between the above two kinds of sensors, EasiAC adjusts the camera sensors´ sampling frequency and active opportunity dynamically according to the estimated results from the magnetic sensors. Finally, we evaluate EasiAC through simulations and real road environment experiments. The results demonstrate that EasiAC needs less energy consumption than traditional solutions, while maintaining the performance at acceptable level in the presence of dynamics of the monitored target.