The introduction of high powered LEDs has brought extensive challenges with regard to thermal design, from chip level to system level thermal management. With the increased efficiency of modern high power LEDs, it is important that the thermal modeling of an LED application calculate the actual thermal load produced by the component. In the case of high power LEDs; this will differ significantly from the total power consumed. In the following paper the advantage of implementing a detailed and computationally efficient LED model is assessed. The assessment compares the detailed model with thermal resistance models and considers its ability to model transient applications as well as an optimization tool.

Automatic livestock tracking is necessary for countries facing stock theft problems, like South Africa and Kenya. This paper presents a conceptual design of architecture for real-time wireless livestock tracking based on Internet of Things paradigm. It is a hierarchical model consisting of three building blocks, where the first block is represented with wireless sensor network. Additionally, we have developed a low-power device for livestock tracking in an outdoor environment. The animal tracking device (AnTrack) is self-sustainable with a watertight solar panel(s), designed as a collar to be worn by the animals. A detailed analysis of the AnTrack power consumption proves that the device is capable to generate enough supply power, even when there is no sunshine for a week. This device can be used as a robust building block of future real-time Internet of Things livestock tracking solutions. Keywords-livestock tracking architecture, Internet of Things, animal collar.

No abstract