1 - 3 of 3
-
PublicationGenerative Machine Learning for Resource-Aware 5G and IoT SystemsExtrapolations predict that the sheer number of Internet-of-Things (IoT) devices will exceed 40 billion in the next five years. Hand-crafting specialized energy models and monitoring sub-systems for each type of device is error prone, costly, and sometimes infeasible. In order to detect abnormal or faulty behavior as well as inefficient resource usage autonomously, it is of tremendous importance to endow upcoming IoT and 5G devices with sufficient intelligence to deduce an energy model from their own resource usage data. Such models can in-turn be applied to predict upcoming resource consumption and to detect system behavior that deviates from normal states. To this end, we investigate a special class of undirected probabilistic graphical model, the so-called integer Markov random fields (IntMRF). On the one hand, this model learns a full generative probability distribution over all possible states of the system-allowing us to predict system states and to measure the probability of observed states. On the other hand, IntMRFs are themselves designed to consume as less resources as possible-e.g., faithful modelling of systems with an exponentially large number of states, by using only 8-bit unsigned integer arithmetic and less than 16KB memory. We explain how IntMRFs can be applied to model the resource consumption and the system behavior of an IoT device and a 5G core network component, both under various workloads. Our results suggest, that the machine learning model can represent important characteristics of our two test systems and deliver reasonable predictions of the power consumption.
-
PublicationAdaptive camouflage panel in the visible spectral range( 2018)In this work an adaptive panel in the visible spectral range is presented. Principal possibilities and basic aspects of adaptive camouflage in the VIS are considered and some details are discussed. The panel consists of modular tiles, each containing several high power four-color-LEDs controlled by a microcontroller and high current power supply and each tile designed to operate autonomously. To control the color and the intensity several color sensors were integrated into the system. The purpose of the panel is to take on a uniform color to best match its appearance to a given reference color, where both the panel and the reference color are subject to the same environmental conditions. The panel was not designed, however, to produce different camouflage patterns. The tiles on the surface were covered by a dark plastic plate in order to provide dark and saturated colors and to guarantee a dark appearance in the passive state of the system. As was to be expected, extreme situations like high ambient brightness and direct solar illumination turned out to be particularly challenging. Substantial tests and some modifications were performed to achieve a satisfactorily uniform color reproduction of a given reference color. Physical measurements as well as observer tests have been performed to demonstrate the capability of the adaptive system.
-
