ABCDEFGHIJKLMNOPQRSTUVWXYZ
1
ATTENTION: THE PROPOSALS PRESENTED HERE ARE SUGGESTIONS ONLY. DON'T HESITATE TO CONTACT THE ADVISORS IN CASE OF AMENDMENTS OR SUGGESTIONS TO A PROPOSAL.
FOR A COMPLETE LIST OF ADVISORS IN THE PG/EEC-I, VISIT: http://www.comp.ita.br/ensino/pos.html
2
TitleLevel (master/doctorate)Advisor (name and e-mail)Co-advisor (name and e-mail)Project abstractResearch theme
3
Novel approaches to the design of runtime reconfigurable embedded systemsDoctorate (preferably)Denis Loubach - dloubach@ita.br | http://www.comp.ita.br/~dloubachThe graduate student will conduct research concerning the proposal of novel approaches to the design of runtime reconfigurable embedded systems. This should regard the use of formal models of computation in a systematic design methodology. The candidate is expected to be highly motivated to conduct research and must have the ability to work individually and in teams. A good English level is required.

Related research project:
FAPESP Research 2019/27327-6
Runtime Reconfigurable Hardware Platform Model
https://bv.fapesp.br/en/auxilios/107286/runtime-reconfigurable-hardware-platform-model

Related papers:
Ricardo Bonna, Denis S. Loubach, George Ungureanu, and Ingo Sander. 2019. Modeling and Simulation of Dynamic Applications Using Scenario-Aware Dataflow. ACM Trans. Des. Autom. Electron. Syst. 24, 5, Article 58 (October 2019), 29 pages. doi: https://doi.org/10.1145/3342997

Horita, A.Y.; Loubach, D.S.; Bonna, R. Analysis and Identification of Possible Automation Approaches for Embedded Systems Design Flows. Information 2020, 11, 120. doi: https://doi.org/10.3390/info11020120

D. S. Loubach, "A runtime reconfiguration design targeting avionics systems," 2016 IEEE/AIAA 35th Digital Avionics Systems Conference (DASC), Sacramento, CA, USA, 2016, pp. 1-8, doi: http://dx.doi.org/10.1109/DASC.2016.7778089

LOUBACH, DENIS S.; CARDOSO MARQUES, JOHNNY ; Marques da Cunha, Adilson. Considerations on Domain-Specific Architectures Applicability in Future Avionics Systems. In: The 10th Aerospace Technology Congress, October 89, 2019, Stockholm, Sweden, 2019, 2019. p. 156-161. doi: http://dx.doi.org/10.3384/ecp19162018
Embedded Systems Design & Reconfigurable Computing
4
Investigations on open source initiatives related to ISA and FPGA to achieve heterogeneous processingMaster or DoctorateDenis Loubach - dloubach@ita.br | http://www.comp.ita.br/~dloubachThe graduate student will conduct research concerning the study of various open source initiatives including RISC-V and Open FPGA. The idea is to compare them and assess their applicability on different domains, eg machine learning algorithms. The candidate is expected to be highly motivated to conduct research and must have the ability to work individually and in teams. A good English level is required.

Related research project:
Real Time Electronics and Computing for Highly Computer Intensive Real Time Systems
https://sucupira.capes.gov.br/sucupira/public/consultas/coleta/projetoPesquisa/viewProjetoPesquisa.jsf?popup=true&idProjeto=465710
Embedded Systems Design & Reconfigurable Computing
5
Drone navigation without GNSS informationMaster or DoctorateFilipe Verri - verri@ita.br | http://www.comp.ita.br/~verri/ UAV autonomous flight depends on position information (latitude, longitude and altitude) usually from GNSS.

This project aims to develop computer vision algorithms to aid flights with GPS restrictions. Other sensor (besides imaging) will also be used to estimate position.

See:

- https://github.com/twhui/LiteFlowNet3
- https://arxiv.org/pdf/2007.09319.pdf
- T.-W. Hui, X. Tang and C. C. Loy, "A Lightweight Optical Flow CNN
-- Revisiting Data Fidelity and Regularization," in IEEE Transactions on
Pattern Analysis and Machine Intelligence,
https://doi.org/10.1109/TPAMI.2020.2976928
Artificial Intelligence / Machine Learning / Deep Learning
6
Early Fire Detection using Remote SensingMaster or DoctorateFilipe Verri - verri@ita.br | http://www.comp.ita.br/~verri/ Popular fire detection techniques such as satellite imaging and remote camera-based sensing suffer from late detection and low reliability while early wildfire detection is a key to prevent massive fires. On the other hand, fixed sensors may not be flexible enough to cover the desired area.

This project consists in the development of machine learning models to predict fire using data from sensors embedded in drones (including multispectral imaging).

Suggested readings:

- Wildfire Spread Prediction and Assimilation for FARSITE Using
Ensemble Kalman Filtering:
https://www.sciencedirect.com/science/article/pii/S187705091630727X

- A Data Mining Approach to Predict Forest Fires using Meteorological
Data: http://www3.dsi.uminho.pt/pcortez/fires.pdf
Artificial Intelligence / Machine Learning / Deep Learning
7
Active Learning for Deep LearningMaster or DoctorateFilipe Verri - verri@ita.br | http://www.comp.ita.br/~verri/ Development of a machine learning framework to plan data collection and annotation to improve the performance of deep neural networks in multi-class object-recognition tasks. Suggested reading: Overview of Active Learning for Deep Learning
<https://jacobgil.github.io/deeplearning/activelearning>
Artificial Intelligence / Machine Learning / Deep Learning
8
Parallel and sequential solutions to knapsack problemsDoctorate (preferably)Vitor Curtis - curtis@ita.br | http://www.comp.ita.br/~curtisThe graduate student will research and propose novel sequential and parallel approaches to traditional problems in combinatorial optimization, namely, knapsack programming (KP), multiple KP and subset-sum, commonly solved by dynamic programming and mathematical optimization.
The candidate is expected to be highly motivated to conduct research, publications and must have the ability to work individually and in teams.
Good English and academic writing skills are required.
For more information, check the papers:

- V. V. Curtis, and C.A.A. Sanches. An improved balanced algorithm for the subset-sum problem. EJOR, Vol. 275, Issue 2, 2019, pp. 460-466, doi: 10.1016/j.ejor.2018.11.055.
- V. V. Curtis, and C.A.A. Sanches. A low-space algorithm for the subset-sum problem on GPU. Computers & Operations Research., Vol. 83, 2017, pp. 120-124. doi: 10.1016/j.cor.2017.02.006.
- V. V. Curtis, and C.A.A. Sanches. An efficient solution to the subset-sum problem on GPU. Concurrency Computat.: Pract. Exper., Vol. 28, 2016, pp. 95–113. doi: 10.1002/cpe.3636.
High performance computing, Algorithms, Parallel computing, Optimization
9
Autoencoders for anomaly detection in containersMasterLourenço A Pereira - ljr@ita.brContainers have become a prominent solution to a vast range of computing infrastructures, from small devices in the edge to high-end systems in the cloud. Such a mechanism implements a convenient way to isolate a set of processes execution contexts. Furthermore, understanding traffic behavior and finding anomalous activities improve this solution, meaning protection components benefit a large class of applications. Thereby, in this project, the student will investigate auto-encoders to characterize normal traffic and redirect suspicious packets to a honeynet. Suggested reading: https://accuknox.com/anomaly-detection/ Cybersecurity / Machine Learning / Computer Networks
10
Development of cyber-physical systems: analysis and design techniques that use results of the STPA (System-Theoretic Process Analysis)Doctorate and MasterCelso Hirata - hirata@ita.br
11
Extension and adaptation of STPA (System-Theoretic Process Analysis) for performance evaluationDoctorate and MasterCelso Hirata - hirata@ita.br
12
Ontology-based Big Data IntegrationDoctorate and MasterJosé M Parente de Oliveira - parente@ita.brIn big data, integrating data coming from different sources, formats and sizes is a significant challenge. Normall, integration is based on data syntax aspects, which can lead to consider different data representing the same entity as different. In this proposal, we aim at developing an integration model that takes ontologies for representing data semantic and to allow a way for entity resolution.
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100