Ubiquitous Smart Environment based on Small Distributed Sensors

在这项研究中，我们提出了一种智能传感器，它可以方便地、成本更低地利用声音事件识别。我们还提出了一种即时学习方法，该方法通过评估几种参数组合，在基于DP匹配的目标事件识别过程中自动配置合适的参数集。基于我们的方案，我们设计并实现了一种基于小型DSP处理器的即时学习型声音传感器。通过评测实验，验证了该智能传感器能够自动选择合适的参数来处理各种声音事件，准确率几乎达到80%以上。我们还确认使用多个传感器来识别楼梯的上/下，或桌子或墙壁上敲击位置。

In this research, we propose a smart sensor which can easily and less costly utilize a sound event recognition. We also propose an Instant Learning method which automatically configures an appropriate set of parameters in a DP matching based recognition process for the target event by evaluating several combinations of parameters. Based on our proposal, we have designed and implemented an Instant Learning Sound Sensor using small DSP processor. By evaluation experiment, we confirmed the smart sensor can automatically choose a proper set of parameters for various sound event with almost over 80% accuracy. We also confirm the utilization of multiple sensors to recognize the up/down of stairs, or positions of knocking place on the desks or walls.