Conference : IEEE Embedded and Ubiquitous Computing

Venue : Melbourne, Australia

Dates : 24-26 October 2011

IEEE Embedded and Ubiquitous Computing conference was held in Melbourne Australia where I was able to present a full paper on my AmbiKraf technology titled “AmbiKraf: A Nonemissive Fabric Display for Fast Changing Textile Animation”. Here, my presentation was mainly focused on addressing the technical issues of the system. I presented few prototypes that we have done including the table cloth system, byobu system, wall hangers etc and spoke of the system characteristics in terms of temperature control and power consumption in detail.

The presentation was well received. However, my presentation was rather different from most of the presentations at the conference as they focused more on architecture level embedded computing issues and less on the application and integration layer. Thus this raised many questions on the system from the audience. One of the main questions I received was how the system, if it was wearable, would react to the body temperature. I clarified here by stating that, since we use temperature controllers the body heat would have a less effect since the controller would re adjust the cloth’s temperature to suit the required temperature. Also there were questions about selecting the actuation temperatures, and power consumption issues where I talked about our current works with the more efficient smaller peltier elements.

Since I also mentioned about a potential application area being for the military for camouflaging according to the surroundings, there was a researcher who was interested in the sensing mechanism. Even though AmbiKraf does not focus on the input to the system we were able to discuss this work in terms of the current sensing techniques.

In addition to this, there were couple of other talks that I found to be interesting at the conference. Even though they are not related directly I found them to be interesting in the Embedded Computing field. The abstracts of the papers are as follows.

Energy Efficient and Robust Multicast Routing for Large Scale Sensor Networks : In this paper we present RE2MR, an energy efficient and robust multicast routing protocol suitable for large scale real-world WSN deployments. RE2MR, a hybrid multicast protocol, builds on the strengths of existing topology-based, hierarchical and geographic multicast solutions, and addresses their limitations. RE2MR establishes a network topology in which multicast member nodes are connected to the root node via near-optimal multicast routing paths. RE2MR discovers deployment area irregularities (e.g., holes) that affect the optimality of multicast routing and considers them when recomputing the near-optimal solution. RE2MR incurs little computational overhead on forwarding nodes, a negligible communication overhead and ensures reliable multicast packet delivery. We implement RE2MR in Tiny OS and evaluate it extensively using TOSSIM. RE2MR reduces the energy consumption by up to 57% and the end-to-end delay by up to 8%, when compared with state of art solutions.

Thermal-Aware Code Transformation across Functional Units : Various functional units (FUs) have been designed in modern embedded processors to perform different functions when running an application. For many applications the occurrences of different instructions are not the same after they are compiled. As a consequence, the temperature of the processor is very high arising from the major heating contribution of the special structure active functional unit and thus the system will suffer severe damage. Thus, to remedy this hurdle, this paper provides a solution by shifting the loading from heavy-loading FUs to light-loading FUs. Our approach first identifies all FUs that can exchange the loading among them and then presents a thermal model for these exchangeable FUs to estimate the temperature impact on shifting loading. Finally, the loading shifting has been performed by transforming code under the consideration of limited performance loss without hardware cost. The result shows that our approach can reduce the temperature at the small cost of performance degradation and code expansion.

Augmented Reality Marker for Operating Home Appliances : The popularity of network home appliances that can interconnect with other home appliances through networks increases. The operation and function of these appliances are complex since they can share contents and data with other network home appliances. However, it remains difficult for users to identify network home appliances, since their locations are not installed and cannot be displayed easily when operating them. Because these problems, more obvious and intuitive operations are needed for controlling network home appliances. For this solution, we have implemented augmented reality (AR) technology in our past proposed system for easier, intuitive operation to control multiple network home appliances connected by home networks. AR technology can provide virtual graphics and apply additional information to specific areas through camera displays. In current AR technology, an AR marker made from a graphical image (an image marker) is necessary to display the virtual AR graphics to the devices's identified location. Unfortunately, such image markers are not suitable when they are implemented on network home appliances due to poor recognition when lighting environments or distances are changed. In this paper, we explain the LED augmented reality marker (LED marker), implement a new marker, and evaluate it, and consider a collaboration system for controlling multiple network home appliances connected to home networks.