With applications demanding increased amount of data, edge computing, where data are processed on the edge, is gaining popularity. Edge devices can store a limited amount of data and need to extract important data from the collected data. On-line interpolation-based data management (O-IDM) is an efficient method for managing time series data on an edge. O-IDM extracts and stores data to represent its overall shape. However, using this method becomes computationally expensive when the amount of the retained data is large. In this paper, we propose fast-O-IDM (FO-IDM), which can perform the same data extraction as O-IDM at a lower computational cost. Furthermore, we propose dynamic representative data management (D-RDM), which is a multiple time series management method that applies FO-IDM. D-RDM can efficiently hold the characteristics of multiple time series by dynamically controlling the capacity based on the data features. The extraction accuracy and computational costs of D-RDM are evaluated via comparison with other management methods. The experimental results confirmed that D-RDM could extract the data more accurately than the other methods, reducing the data amount required to convey the feature of the data by up to 25.0 percentage points compared to naive O-IDM. In addition, when the number of time series and the amount of the retained data were 3 and 30,000 respectively, the processing time of D-RDM was reduced by 91.2 percentage point compared to O-IDM, indicating that the computational cost of D-RDM is sufficiently small for use in edge devices.

Owing to the widespread adoption of the Internet of Things, a vast amount of sensor information is being acquired in real time. Accordingly, the communication cost of data from edge devices is increasing. Compressed sensing (CS), a data compression method that can be used on edge devices, has been attracting attention as a method to reduce communication costs. In CS, estimating the appropriate compression ratio is important. There is a method to adaptively estimate the compression ratio for the acquired data using reinforcement learning (RL). However, the computational costs associated with existing RL methods that can be utilized on edges are often high. In this study, we developed an efficient RL method for edge devices, referred to as the actor–critic online sequential extreme learning machine (AC-OSELM), and a system to compress data by estimating an appropriate compression ratio on the edge using AC-OSELM. The performance of the proposed method in estimating the compression ratio is evaluated by comparing it with other RL methods for edge devices. The experimental results indicate that AC-OSELM demonstrated the same or better compression performance and faster compression ratio estimation than the existing methods.

In recent years, a vast amount of sensor information has been generated as a result of the widespread use of IoT devices. However, with the explosion of data volume, it has become difficult to process all the data centrally on cloud servers. Therefore, it has become important to process data at edge devices where it is acquired. As edge devices cannot store all the acquired data, it is necessary to select and store the important data from the acquired data. In this paper, we propose an online data management method for time-series data called O-IDM, which is designed to extract data to preserve characteristics. O-IDM can extract data necessary to approximate the original data with a small computational cost. We evaluated the extraction performance of the proposed method by comparing it with other data extraction methods using the degree of deviation between the input data and polygonal line approximation base on the extracted data. The experimental results show that the proposed method was able to extract data with high accuracy. Further, the proposed approach was able to reduce the amount of data transmission required to convey the characteristics of the data by up to 66.7 percentage point compared to the naïve time-based sampling method.

Wide spread of IoT devices has made it possible to acquire enormous amounts of realtime sensor information. Due to the explosive increase in the sensing data volume, it becomes difficult to collect and process all the data in one central place. On one hand, storing and processing data on edge devices, so called edge computing, is becoming important. On the other hand, edge devices usually have only limited computing and memory resources, and hence it is not practical to process and save all the acquired data. There is a great demand of effectively selecting data to process on an edge device or to transfer it to a cloud server. In this paper, we propose an efficient density-based data selection and management method called O-D2M by which edge devices store the data representing inherent data distribution. We use a low cost graph algorithm to analyze input data trend and its density. We evaluate effectiveness of the proposed O-D2M comparing to other methods in terms of the accuracy of machine learning models trained by the selected data. Throughout the evaluation, we confirm that O-D2M obtains higher accuracy and lower computation cost while it can reduce the amount of data to be processed or transferred by up to 20 points.