Research On Sensing Terminals And Data Management Of Internet Of Things

Preceded by computer, internet, and mobile communication, IoT is taken into consideration of the fourth wave of the Information Communication Technology (ICT) industry, which demonstrates broad prospects and huge market potential. However, the development and implementation of IoT still faces up big challenges, among which several ones are related to sensing terminals and data management such as:1) Development of field devices software to meet the challenges of function specialty, complicated target platforms, high reliability, strict timing constrains, and so on; 2) In addition to the aforementioned challenges, energy loss and wearing out issues of field devices have been aggravated by the power supply constraints and the deployment inconvenience; 3) Efficient storage and fast query of IoT data originated from field devices which are large-scale, heterogeneous, and time-sensitive. In response to these challenges, this dissertation studies several key issues relating to field devices and data management of IoTs, aiming to improve the availability of field devices and data management of IoTs. Specifically, this thesis focuses on the following four parts:1) In order to improve the development of field devices software, it employs an automatic code generation method based on the Esterel modelling language, which is about to rank up the automation level of software development. It employs the Esterel language to model the software, and employs operating system abstract layer to achieve a uniform application programming interface among multiple target platform. Through these means, the automatic code generation for multiple target platform can be obtained.2) In order to improve the availability of and extend the lifetime of field devices, it proposes a compilation method to improve the energy efficiency and wear levelling of field devices. It firstly proposes an evaluation model of green indicators for field devices. Guided by this model, the green indicators of field devices are improved via two aspects. The first one is to improve the energy efficiency and wear levelling of instruction bus via exchanged the positions of operands within an instruction without changing its semantics. The second one is to improve the green indicators of bus and registers at the same time via register re-allocation which preserves the original program semantics.3) In order to improve the efficiency of IoT data management, it proposes a method for efficient storage and fast query of IoT data. A multi-level Meta data standard is proposed based on the Dublin Meta data standard and its extension, in order to specify the semantics of the heterogeneous IoT data. Relational database and file system are employed together for efficient storage of the semi-structural IoT data, and optimization algorithms are proposed to improve the storage efficiency via eliminating redundancy. In addition, a twice-tree-matching scheme is proposed for fast query of IoT data, the first matching according the semantics in the form of Meta data standard, while the second matching according the relational expressions.4) It proposes a preliminary research on the application of IoT technology to evacuation, in purpose of evaluating the integration of aforementioned methods into IoT applications. A novel IoT application framework in evacuation is prosed. Then, a simulation experiment is proposed for preliminary evaluation of this application framework.