Yanfeng Shu

Hydrological Sensor Web for the South Esk Catchment in the Tasmanian state of Australia

The sensor Web is a distributed sensing system in which information is shared globally. The emergence of this technology will enable the integration of different sensing platforms with temporal and spatial variability. This has a potential to revolutionise hydrological monitoring and forecasting. Our project will establish a sensor Web test bed in the South Esk river catchment, which is located in the North East of Tasmania. The test bed will allow us to evaluate the emerging open geospatial consortium standards and specifications for sensor Web enablement (SWE) and provide a research platform for developing next-generation hydrological and water resource management tools. We intend to use short-term river flow forecasting as our use case for the SWE test bed.

Performance evaluation of the impact of mobile base station on clustered wireless sensor networks

Base station mobility can be exploited to minimise the energy consumption in a wireless sensor network. This paper investigates the impact that base station movement has upon the performance of cluster-based wireless sensor networks. Three types of base station movement are considered: movement influenced by the position of cluster-heads, random movement and movement partially influenced by the position of cluster-heads. In spite of the overhead associated with finding the base station, the wireless sensor network was shown to gather at least 15% more data packets per Joule of energy with a mobile base station compared to a static base station.

Salad leaf disease detection using machine learning based hyper spectral sensing

In this paper a novel application of salad leaf disease detection has been developed using a combination of machine learning algorithms and Hyper Spectral sensing. Various field experiments were conducted to acquire different vegetation reflectance spectrum profiles using a portable high resolution ASD FieldSpec4 Spectroradiometer, at a farm located in Richmond, Tasmania, Australia, (-42.36, 147.29), A total of 105 spectral samples were collected through three different experiments with baby salad leaves. In this study, Principal Component Analysis (PCA), Multi-Statistics Feature ranking and Linear Discriminant Analysis (LDA) Classifiers were used to classify disease affected salad leaves from the healthy salad leaves with 84% classification accuracy. This study concluded that the machine learning based approach along with a high resolution hyper Spectroradiometer could potentially provide a novel mechanism to use in the farm for rapid detection of salad leaf disease.

Semantic water data translation: a knowledge-driven approach

In order for the Bureau of Meteorology (BOM), Australia, to build and maintain an integrated national water information system, over 240 organisations are required to provide their data to BOM. These organisations use a wide range of systems and data formats. To ensure robust and reliable data delivery, BOM has established Water Data Transfer Format (WDTF) as a standard format for data transfer. Meanwhile, the Water Regulations 2008 were enacted to specify the water information required from organisations. This paper analyses semantic gaps between data from organisations, WDTF, and the Regulations requirements, and proposes a knowledge-driven approach in which these gaps are captured in a way that facilitates data translation and validation. Throughout the paper, real data examples are used to illustrate the details of the approach and its feasibility.

Energy Adaptive Sensor Scheduling for Noisy Sensor Measurements

In wireless sensor network applications, sensor measurements are corrupted by noises resulting from harsh environmental conditions, hardware and transmission errors. Minimising the impact of noise in an energy constrained sensor network is a challenging task. We study the problem of estimating environmental phenomena (e.g., temperature, humidity, pressure) based on noisy sensor measurements to minimise the estimation error. An environmental phenomenon is modeled using linear Gaussian dynamics and the Kalman filtering technique is used for the estimation. At each time step, a group of sensors is scheduled to transmit data to the base station to minimise the total estimated error for a given energy budget. The sensor scheduling problem is solved by dynamic programming and one-step-look-ahead methods. Simulation results are presented to evaluate the performance of both methods. The dynamic programming method produced better results with higher computational cost than the one-step-look-ahead method.