Agenda

PRORISC and SAFE 2019

PRORISC and SAFE 2019

PRORISC is an annual conference on Integrated Circuit (IC) design and SAFE is an annual conference on Microsystems, Materials, Technology and RF-devices. Both conferences are organized together within the three technical Dutch universities Twente, Delft and Eindhoven. The conference is organized by PhD students and is intended for PhD candidates to expand their network and share their research ideas, which provides a unique opportunity for future collaborations. Each year, one of the technical universities will be responsible for the organization of the two conferences. In 2019 the PRORISC will be held at at the campus of Delft University of Technology.

Additional information ...


Migrating Target Detection in Wideband Radars

Nikita Petrov

Additional information ...


Radar Micro-Doppler Patterns for Drone's Characterisation

Radar Micro-Doppler Patterns for Drone's Characterisation

Yefeng Cai

Micro-Doppler patterns of multi-propeller drones measured by radar systems are widely used in the classification of different drones, since the micro-Doppler patterns illustrate the velocity and motion properties of the drones. However, on this topic, there are a few issues the current researches have not tackled yet, and these will be discussed in this presentation. First of them is the lack of mathematical description of the micro-Doppler patterns. and most research works are based on real measured data so far. In this thesis, an EM backscattering model in HH plane of drone propeller is developed, simplifying the propeller’s geometry structure as a few cylinder thin wires. Radar signal model and micro-Doppler model are subsequently developed for the thin-wire propeller model when it is rotating. Second, most current researches focus on the micro-Doppler patterns achieved in short CPI cases that are valid for radar systems with PRI much shorter than the rotation period of the drone’s propellers. In this thesis, the drone micro-Doppler patterns in long CPI circumstances are investigated. Features are proposed to characterise the amplitude and frequency distribution of the simulated micro-Doppler spectrum. Applying these features to SVM gives good classification accuracy for the simulated micro-Doppler data. Third, most researches at present are carried out in short range for static or stable hovering drones, while from a practical point of view, it is also of great interest to investigate the drone micro-Doppler patterns in long range and dynamic scenarios. In this thesis, the micro-Doppler patterns of different drones at a distance of 9 kilometres are achieved by S-band radar in long CPI circumstance. Applying the previously proposed features to the real measured micro-Doppler spectra to SVM gives good classification accuracy for drones in hovering and manoeuvring flight modes.


Signal Processing Seminar

Adaptive classification of radar emitters

Aybuke Erol
METU, Turkey

Radar receivers collect interleaved signals from all electromagnetic sources in the environment. The ultimate goal of electronic intelligence is to separate these sources (deinterleaving) and find their types (emitter identification). Knowing the type of a source, it is possible to comment on its mission and operation. All in all, deinterleaving and emitter identification together build a system that solves an adaptive classification problem. One of the biggest challenges in this problem is that the system does not know all emitter types in the world since a great part of this information is confidential within each country. In addition, radar receivers sequentially provide radar pulses to the system. Therefore, the classifier should be able to increase its number of classes whenever an unfamiliar emitter type is encountered. What’s more, it should be able to distinguish between the unfamiliar emitter types, which enforces online learning.

The proposed system solves deinterleaving using fuzzy ARTMAP due to several reasons. First, it is supervised which makes the system able to start with a priori information or data. Secondly, it works with sequential input and enables online learning. Last but not least, it can increase its number of classes. After fuzzy ARTMAP, radar clusters are formed. Next, a representation for each cluster should be found, to be compared with the representations of already known emitter types. The challenge here is that describing an emitter type by single numeric values would not be fair as radar features are generally interval based. For example, emitters today do not operate on a single frequency, they rather have a frequency range in which they can operate. Hence, the representation and comparison of emitter types and radar clusters are considered under symbolic data analysis. Both parts, solved with fuzzy ARTMAP and symbolic data analysis, are improved in terms of classification accuracy from their baseline methods with the use of Jaccard index.