No abstract

Often in the design process of an engineer, the design specifications of the system are not completely known initially. However, usually there are some physical constraints which are already known, corresponding to a region of interest in the design space that is called feasible. These constraints often have no analytical form but need to be characterised based on expensive simulations or measurements. Therefore, it is important that the feasible region can be modeled sufficiently accurate using only a limited amount of samples. This can be solved by using active learning techniques that minimize the amount of samples w.r.t. what we try to model. Most active learning strategies focus on classification models or regression models with classification accuracy and regression accuracy in mind respectively. In this work, regression models of the constraints are used, but only the (in)feasibility is of interest. To tackle this problem, an information-theoretic sampling strategy is constructed to discover these regions. The proposed method is then tested on two synthetic examples and one engineering example and proves to outperform the current stateof-the-art.

Many smart home applications rely on indoor human activity recognition. This challenge is currently primarily tackled by employing video camera sensors. However, the use of such sensors is characterized by fundamental technical deficiencies in an indoor environment, often also resulting in a breach of privacy. In contrast, a radar sensor resolves most of these flaws and maintains privacy in particular. In this paper, we investigate a novel approach towards automatic indoor human activity recognition, feeding high-dimensional radar and video camera sensor data into several deep neural networks. Furthermore, we explore the efficacy of sensor fusion to provide a solution in less than ideal circumstances. We validate our approach on two newly constructed and published data sets that consist of 2347 and 1505 samples distributed over six different types of gestures and events, respectively. From our analysis, we can conclude that, when considering a radar sensor, it is optimal to make use of a three-dimensional convolutional neural network that takes as input sequential range-Doppler maps. This model achieves 12.22% and 2.97% error rate on the gestures and the events data set, respectively. A pre-trained residual network is employed to deal with the video camera sensor data and obtains 1.67% and 3.00% error rate on the same data sets. We show that there exists a clear benefit in com-Baptist Vandersmissen

In this paper, we tackle the task of multi-target tracking of humans in an indoor setting using a low power 77 GHz MIMO CMOS radar. A drawback of such a highresolution and low-power device is the higher sensitivity to noise, which makes the analysis of signals more challenging. Therefore, a pipeline is proposed to address both pre-processing of the radar signal and multi-target tracking. In the pre-processing phase, we focus on handling the low Signal-to-Noise Ratio (SNR) and eliminating so-called ghost targets. The tracking method we propose is based on Markov Chain Monte Carlo Data Association (MCMCDA), thus taking a combinatorial approach towards the task of tracking. The pipeline is tested on a number of real-world scenarios and shows promising results, overcoming the significant amount of noise associated with embedded radar devices.

Finding the optimal working conditions for nonlinear electrical components under large signal stimuli can be challenging, mainly due to the high number of input dimensions and multiple local minima of the goal function. In this paper a Bayesian optimization method is applied in order to limit the number of evaluations by a commercial harmonic balance simulator. The method is applied to amplifier optimization utilizing Wolfspeed CGH40010F GaN HEMT, for which input power, bias voltages and load at fundamental harmonic frequencies are changed in order to maximize for combined efficiency, gain, and output power. The optimum is found already after 80 iterations.