In an Internet of Things (IoT) system, it is essential that the data measured from the sensors are accurate so that the produced results are meaningful. For example, in AgriTalk, a smart farm platform for soil cultivation with a large number of sensors, the produced sensor data are used in several Artificial Intelligence (AI) models to provide precise farming for soil microbiome and fertility, disease regulation, irrigation regulation, and pest regulation. It is important that the sensor data are correctly used in AI modeling. Unfortunately, no sensor is perfect. Even for the sensors manufactured from the same factory, they may yield different readings. This paper proposes a solution called SensorTalk to automatically detect potential sensor failures and calibrate the aging sensors semi-automatically. Numerical examples are given to show the calibration tables for temperature and humidity sensors. When the sensors control the actuators, the SensorTalk solution can also detect whether a failure occurs within a detection delay. Both analytic and simulation models are proposed to appropriately select the detection delay so that, when a potential failure occurs, it is detected reasonably early without incurring too many false alarms. Specifically, our selection can limit the false detection probability to be less than 0.7%.

Several tools provide popular solutions for creating innovative Internet of Things (IoT) applications on single Arduino board. However, interactions among multiple Arduino boards need significant effort to be established. This paper proposes Arduino-IoTtalk (ArduTalk), a graphical user interface (GUI)-based platform aims to develop IoT network applications for interaction among multiple Arduino boards. ArduTalk utilizes IoTtalk, an IoT device management platform for quickly establishing connections and meaningful interactions between IoT devices without concerning the lower layer communication protocols. In this paper, the IoTtalk GUI has been significantly enhanced for ArduTalk. By integrating Arduino with the enhanced IoTtalk, ArduTalk allows a user to arbitrarily link and relink sensors to actuators without or with little programming effort, and quickly generate Arduino applications for different purposes. We conduct measurements to investigate the time complexity for data delivery among multiple Arduino boards, and propose a damping mechanism to address the fairness issue caused by the discrepancies between local and remote delays among the Arduino boards.

In one of the recent trends of Internet of Things (IoT), the IoT data are manipulated by Artificial Intelligence (AI) techniques for smart applications. By including AI into existing IoT application programs, significant coding effort is required. This paper proposes a solution called AItalk to resolve this issue. Unlike traditional AI‐based IoT applications that tightly integrate the AI mechanism within the network applications, the novel idea of AItalk is to treat the machine learning mechanism as a cyber IoT device. Our solution allows seamless inclusion of machine learning capability to the existing IoT applications without any programming effort. The advantage of this approach is that we can decompose a complex AI application into simplified distributed modules connected by using the IoT technology, and therefore the AI solution can be built more effectively. Also, in our approach, data can be easily processed in real time for an AI application. Supervised machine learning naturally fits the IoT applications, where the sensors provide the features to the AI algorithms, and the remote controllers serve as the labels. We show an example that the overhead of the IoT communication in AItalk is less than 30 ms and the AI prediction time is less than 2 ms.

Internet of Things (IoT) has been widely utilized with Artificial Intelligence (AI). However, it requires substantial effort to integrate AI and big data with IoT. To mitigate this problem, AItalk was proposed. By treating AI as a cyber IoT device, we do not need to write code of the AI mechanism in the network applications, as traditional AI-based IoT applications did. This paper describes how the AI tools such as scikit-learn and TensorLayer are accommodated as cyber IoT devices in AItalk, and extends AItalk for non-IoT applications. We use house valuation as an example to show how AItalk can flexibly include the factors that have significant impact on the house price. We show that by adding extra features other than housing profile features, the accuracy for the prediction (valuation) can be improved by 40%. We also investigated the communication overhead of the distributed AItalk structure, which is 3.7% for the computation of one house price valuation.