基于压缩感知的土壤呼吸监测传感网动态采样调度策略

Dynamic sampling scheduling policy for soil respiration monitoring sensor networks based on compressive sensing

土壤呼吸监测传感网通常部署在野外,无法直接访问电力,并且土壤呼吸测量过程比较复杂、能耗较高,因此在满足重建准确度要求的前提下,希望以尽量少的采样次数来进行测量.利用土壤呼吸真实物理过程中时间序列的相关性,可以采用压缩感知理论来实现采样调度.本文提出了一种基于压缩感知的分段动态采样调度策略:利用前期测量数据进行分析得到的先验知识,对测量时间区间的数据序列进行分段线性拟合,依据分段数据子序列的线性程度度量指标,动态确定各段的采样率.土壤呼吸测量仪在进行监测时,按照分段动态变化的采样率构造压缩感知采样和重建所需的测量矩阵.实验结果表明,相比平均采样率相同的固定采样策略,本文提出的分段动态采样策略能够得到更好的重建质量,即如果以确定的重建误差阈值作为需求,则本文的动态采样策略具有更小的采样率.虽然计算动态采样率会带来一定的计算开销,但减少的采样次数可以节省更多的能量.本文提出的基于压缩感知的分段动态采样策略,虽然在土壤呼吸监测传感网的应用场景中进行了实验和分析,但其思想对于其他类似应用的采样调度和节能问题也具有借鉴价值和潜在的可应用性.

Sensor networks for soil respiration monitoring are usually deployed in the fields where electric or manual intervention cannot be accessed directly, and the measurement of soil respiration is complex and highly energy-consuming. Therefore, we hope to minimize the number of soil respiration measurements on the premise of reconstruction accuracy. Sampling scheduling can be realized using compressive sensing theory on the basis of temporal correlation of the physical process of soil respiration. Here we propose a segmental dynamic sampling scheduling policy based on compressive sensing. Using a prior knowledge obtained by means of analysis on the earlier measurement data, the data serial in measurement period is partitioned and linear fitted. Then the dynamic sampling rate of each segment is determined according to the linear degree of data in the segment, based on which the measurement matrix is constructed for the sampling and reconstructed for compressive sensing process using the soil respiration measuring instrument. The experimental result shows that the proposed segmental dynamic sampling policy can lead to better reconstructive quality than static sampling policy of the same average sampling rate. That is to say, the proposed dynamic sampling policy needs smaller sampling rate if the reconstructive error threshold is given. The reduction of sampling rate can save more power although the calculation of dynamic sampling rate may consume some power. The proposed segmental dynamic sampling policy based on compressive sensing can also be referenced and potentially used by similar applications for the sampling scheduling and powersaving issues, although it is experimented and analyzed in the soil respiration monitoring sensor networks.