Search

The paper investigated the feasibility of Hyper spectra to determine the concentration of soil organic matter (SOM) in Harbin. The 95 soil samples were collected to a depth from 0 to 20cm. Reflectance measurements from 350nm to 2500nm were collected in a controlled laboratory environment. Three multivariate techniques (stepwise multiple linear regression(SMLR), artificial neural network(ANN), partial least-squares regression(PLSR)) and pre-processing transform nations of spectral data were compared with the aim of identifying the best combination to predict soil organic matter. The coefficient of determination (R2), the root mean square error (RMSE) were used to evaluate the models. compared three multivariate methods of inferential modeling, based on R2 and RMSE, partial least-squares regression performed best (the highest averageR2=0.826, the lowest average RMSE = 0.161).

This paper presents a multi-agent based architecture of context-aware systems. Our objective is to make the architecture more extensible and can be easily applied to developing most context-aware applications. We also discussed how to add context-aware abilities to the existing applications. In our research, we divide the context-aware system into several parts, and we use multiple agents to process corresponding information respectively. Each agent is highly independent of the others, and we nearly don't need to make any changes to other agents when we want to modify one of them.

Context-aware computing are gaining increasing attention due to the value they add to wireless networks. We described the innards of IGS (intelligent guide system), an 802.11 location-based tour guide system. Our objective is to empower mobile devices to recognize the context when the tourists visit some scenic spots and provide corresponding services. In our work, we proposed BAFO (be away from obstacles) algorithm to deploy access points (APs), and proposed a received signal strength (RSS) based algorithm MNNSS (modified nearest neighbor in signal space) to position mobile terminals. We divided the whole IGS system into four parts, and each part can do its job independently. We also used a friendly user interface so that the user can easily get the information they need. These proposed techniques can be further extended to the indoor guiding and tracing system.

Context-aware computing is emerging as a promising technology for mobile and pervasive environments. Unfortunately, complex context-aware applications are still hard to build. In this paper, we describe an 802.11 based indoor context-aware system used in hospital. Our objective is to empower mobile devices to recognize the context in which hospital workers perform their tasks. This system is an accessory of the management system in the hospital, and it can add Context-aware ability to the existing system by providing Context-aware communications. In our work, we use BAFO (Be Away From Obstacles) algorithm to deploy APs (Access Points), and use a Received Signal Strength (RSS) based Algorithm MNNSS (Modified Nearest Neighbor in Signal Space) to position mobile terminals, then describe the implementation of this multi-agent based system