Hyperspectral images contain tens to hundreds of bands, implying a high spectral resolution. This high spectral resolution allows for obtaining a precise signature of structures and compounds that make up the captured scene. Among the types of processing that may be applied to Hyperspectral Images, classification using machine learning models stands out. The classification process is one of the most relevant steps for this type of image. It can extract information using spatial and spectral information and spatial-spectral fusion. Artificial Neural Network models have been gaining prominence among existing classification techniques. They can be applied to data with one, two, or three dimensions. Given the above, this work evaluates Convolutional Neural Network models with one, two, and three dimensions to identify the impact of classifying Hyperspectral Images with different types of convolution. We also expand the comparison to Recurrent Neural Network models, Attention Mechanism, and the Transformer architecture.. Furthermore, a novelty pre-processing method is proposed for the classification process to avoid generating data leaks between training, validation, and testing data. The results demonstrated that using 1 Dimension Convolutional Neural Network (1D-CNN), Long Short-Term Memory (LSTM), and Transformer architectures reduces memory consumption and sample processing time and maintain a satisfactory classification performance up to 99% accuracy on larger datasets. In addition, the Transfomer architecture can approach the 2D-CNN and 3D-CNN architectures in accuracy using only spectral information. The results also show that using two or three dimensions convolution layers improves accuracy at the cost of greater memory consumption and processing time per sample. Furthermore, the pre-processing methodology guarantees the disassociation of training and testing data.

This work proposes a novel methodology to evaluate SRAM-based FPGA's susceptibility with respect to Single-Event Upset (SEU) as a function of noise on VDD power pins, TotalIonizing Dose (TID) and TID-imprinted effect on BlockRAM cells. The proposed procedure is demonstrated for SEU measurements on a Xilinx Spartan 3E FPGA operating in an 8 MV Pelletron accelerator for the SEU test with heavy-ions, whereas TID was deposited by means of a Shimadzu XRD-7000 X-ray diffractometer. In order to observe the TID-induced imprint effect inside the BlockRAM cells, a second SEU test with neutrons was performed with Americium/Beryllium ( ²⁴¹ AmBe). The noise was injected into the power supply bus according to the IEC 61.000-4-29 standard and consisted of voltage dips with 16.67% and 25% of the FPGA's VDD at frequencies of 10 Hz and 5 kHz, respectively. At the end of the experiment, the combined SEU failure rate, given in error/bit.day, is calculated for the FPGA's BlockRAM cells. The combined failure rate is defined as the average SEU failure rate computed before and after exposition of the FPGA to the TID.

This paper proposes a channel model to be applied to the communications systems of CubeSats. The model considers the Low Earth Orbit geometry when a satellite is passing over a ground station, and the propagation surrounding multiple paths of the ground station location in urban areas. The geometry was used to define the deterministic factors which contribute to the fading of the communication signal, such as Doppler effect and loss in free space. These are parameters that change as a function of time and elevation angle. Thus, a complete digital communication system, at the link layer level, is presented, using Markov Chains to model the previously cited effects in the form of a Finite-State Markov Channel. The proposed model was used as an uplink channel between a ground station and a CubeSat, both implementing a protocol stack, following the Consultative Committee for Space Data Systems (CCSDS) recommendations. The ground station and the proposed communication channel were implemented through a functional simulation model and a telecommand and telemetry unit, implemented in hardware, was used as a case study for the CubeSat. Through the analysis of the simulated system, with telecommands sent by the station and responses sent by the hardware unit (in a hardware-in-theloop setup), it was possible to demonstrate the operation of the proposed channel together with the retransmission mechanism suggested in the CCSDS recommendations, in order to mitigate communication issues.