Chen Huayan, Zhang Senlin, Liu Meiqin, Zhang Qunfei
State Key Laboratory of Industrial Control Technology, Hangzhou 310027, China.
College of Electrical Engineering, Zhejiang University, Hangzhou 310027, China.
Sensors (Basel). 2017 Apr 27;17(5):971. doi: 10.3390/s17050971.
We study the problem of energy-efficient target tracking in underwater wireless sensor networks (UWSNs). Since sensors of UWSNs are battery-powered, it is impracticable to replace the batteries when exhausted. This means that the battery life affects the lifetime of the whole network. In order to extend the network lifetime, it is worth reducing the energy consumption on the premise of sufficient tracking accuracy. This paper proposes an energy-efficient filter that implements the tradeoff between communication cost and tracking accuracy. Under the distributed fusion framework, local sensors should not send their weak information to the fusion center if their measurement residuals are smaller than the pre-given threshold. In order to guarantee the target tracking accuracy, artificial measurements are generated to compensate for those unsent real measurements. Then, an adaptive scheme is derived to take full advantages of the artificial measurements-based filter in terms of energy-efficiency. Furthermore, a computationally efficient optimal sensor selection scheme is proposed to improve tracking accuracy on the premise of employing the same number of sensors. Simulation demonstrates that our scheme has superior advantages in the tradeoff between communication cost and tracking accuracy. It saves much energy while loosing little tracking accuracy or improves tracking performance with less additional energy cost.
我们研究水下无线传感器网络（UWSN）中的节能目标跟踪问题。由于UWSN的传感器由电池供电，电池耗尽时更换电池是不切实际的。这意味着电池寿命会影响整个网络的寿命。为了延长网络寿命，在保证足够跟踪精度的前提下降低能耗是值得的。本文提出了一种节能滤波器，实现了通信成本与跟踪精度之间的权衡。在分布式融合框架下，如果本地传感器的测量残差小于预先设定的阈值，则不应将其微弱信息发送到融合中心。为了保证目标跟踪精度，生成人工测量值以补偿那些未发送的真实测量值。然后，推导了一种自适应方案，以充分利用基于人工测量的滤波器在能量效率方面的优势。此外，还提出了一种计算效率高的最优传感器选择方案，以在使用相同数量传感器的前提下提高跟踪精度。仿真表明，我们的方案在通信成本与跟踪精度的权衡方面具有显著优势。它在几乎不损失跟踪精度的情况下节省了大量能量，或者以较少的额外能量成本提高了跟踪性能。
