Abstract The use of the quadcopter‐type drone is now at a mature and practical stage, and many major manufacturers are expanding their range of applications. Because of the high maneuverability and practicality of this flying object, how we ensure it is used for benign purposes in urban life has become a major issue. Most such drones are small enough to avoid many current airborne detection methods and cheap enough to be disposable. We were able to gather enough amount of multimedia sources including drone‐related sources using open online platforms, such as YouTube and other similar services. For a subset of commercial quadcopter drones, we collected sound sample data consisting of recordings and YouTube videos and built a trained simple nonlinear neural network filter to classify them. We used Mel‐frequency cepstral coefficients to reduce the amount of data as the input signal and created two‐dimensional spectrogram images to minimize the loss of temporal information in the sample, as suggested by previous research. Most of the acoustic signal energy in the drone sound existed in specific frequency bands and residual signal patterns were spread up to around 15 kHz. We designed and tested several recurrent and convolutional neural network models and one of the simple neural network models based on the LeNet architecture could classify over 10 different types of drones from acoustic data with over 97% accuracy including an additional path for the detection of general acoustic drone patterns.

This paper proposes some modifications of the COST 231 Walfisch-Ikegami (WI) model to reflect the information related to building and terrain heights of practical environments. Based on the modified WI model, a dual slope propagation model is developed for small cells with electrical and mechanical tilt. To validate the proposed model, propagation measurements are performed in a metropolitan area using commercially deployed small cells. We suggest a procedure to compute the path loss from measurement data considering transmit antenna beam patterns, and determine the break point for the proposed dual slope model using measurement results. Finally, the proposed model is compared through data analysis with conventional propagation prediction models in terms of root mean square deviation (RMSD).

Interpolated second-order polynomials (ISOPs) are proposed to design efficient cascaded integrator-comb (CIC)-based decimation filters for a programmable downconverter. It is shown that some simple ISOPs can effectively reduce the passband droop caused by CIC filtering with little degradation in aliasing attenuation. In addition, ISOPs are shown to be useful for simplifying halfband filters that usually follow CIC filtering. As a result, a modified halfband filter (MHBF) is introduced which is simpler than conventional halfband filters. The proposed decimation filter for programmable downconverter is a cascade of a CIC filter, an ISOP, MHBFs, and a programmable finite impulse response filter. A procedure for designing the decimation filter is developed. In particular, an optimization technique that simultaneously designs the ISOP and programmable FIR filters is presented. Design examples demonstrate that the proposed method leads to more efficient programmable downconverters than existing ones.