论文标题
部分可观测时空混沌系统的无模型预测
Sub-monolayer Biolasers: Lower Gain, Higher Sensitivity
论文作者
论文摘要
储层计算是预测湍流的有力工具,其简单的架构具有处理大型系统的计算效率。然而,其实现通常需要完整的状态向量测量和系统非线性知识。我们使用非线性投影函数将系统测量扩展到高维空间,然后将其输入到储层中以获得预测。我们展示了这种储层计算网络在时空混沌系统上的应用,该系统模拟了湍流的若干特征。我们表明,使用径向基函数作为非线性投影器,即使只有部分观测并且不知道控制方程,也能稳健地捕捉复杂的系统非线性。最后,我们表明,当测量稀疏、不完整且带有噪声,甚至控制方程变得不准确时,我们的网络仍然可以产生相当准确的预测,从而为实际湍流系统的无模型预测铺平了道路。
Biomarker detection is the key to identifying health risks. However, designing sensitive biosensors in a single-use mode for disease diagnosis remains a major challenge. Here, we report sub-monolayer biolasers with remarkable repeatability for ultrasensitive and disposable biomarker detection. The biolaser sensors are designed by employing the telecom optical fibers as distributed optical microcavities and pushing the gain molecules down to the sub-monolayer level. We observe a status transition from the monolayer biolaser to the sub-monolayer biolaser by tuning the specific conjugation. By reducing the fluorophores down to the threshold density (~ 3.2 x 10-13 mol/cm2), we demonstrate an ultimate sensitivity of sub-monolayer biolaser with six orders of magnitude enhancement compared with the monolayer biolasers. We further achieved ultrasensitive immunoassay for Parkinson's disease biomarker, alpha-synuclein, with a lower limit of detection of 0.32 pM in serum. This biosensor with massive fabrication capability at ultralow cost provides a general method for the ultrasensitive disposable biodetection of disease biomarkers.