To p ics for the MSCSP 14.0 7.2014 P a g e 1 Advanced Resear c h

of Communication Networks

Head: Prof. Dr. rer. nat. habil. Jochen Seitz
15.08.2013
Page

3

Communications Research Laboratory

Head: Prof. Dr.-Ing. Martin Haardt
15.08.2013
Page 16

guard intervals which are usually implemented by cyclic prefixes (CP).

CP is a random sequence and is solely created in time domain by copying the last part of the output of IDFT. Instead of using a random sequence as CP, a known deterministic sequence called as unique word (UW) can be used which, in result, can be very helpful for channel estimation and synchronization. Therefore, such UW systems does not require dedicated pilot carriers as compared to systems using CP. In [1], two different methods are introduced to construct UW-OFDM. In this project, we will compare the performance of UW-OFDM against the CP-OFDM for the LTE-A downlink scenario.
Requirements: Programming skills in MATLAB.
Tasks
Review of the literature
Implementation of UW-OFDM in Matlab
Performance comparison of UW-OFDM and CP-OFDM for LTE-A downlink
Literature:
[1]. A. Onic and M. Huemer, “Direct vs. two-step approach for unique word generation in UW-OFDM,” in Proc. 2010 Int. OFDM Work., pp. 145–149 Hamburg (Germany), September 2010.
[2]. Huemer, M., Hofbauer, C., Huber, J.B “The Potential of Unique Words in OFDM” in Proceedings of the 15th International OFDM Workshop. Hamburg (Germany), September 2010.
Focus 1/2 students theory / programming

Unique Word OFDM for LTE-A Downlink

Responsible Professor:
Supervisor:

Prof. Martin Haardt
Sher Ali Cheema

guard intervals which are usually implemented by cyclic prefixes (CP).

CP is a random sequence and is solely created in time domain by copying the last part of the output of IDFT. Instead of using a random sequence as CP, a known deterministic sequence called as unique word (UW) can be used which, in result, can be very helpful for channel estimation and synchronization. Therefore, such UW systems does not require dedicated pilot carriers as compared to systems using CP. In this project, we will investigate the channel estimation techniques for UW-OFDM.
Requirements: Programming skills in MATLAB.
Tasks
Review of the literature
Channel estimation using UW
Literature:
[1]. A. Onic and M. Huemer, “Direct vs. two-step approach for unique word generation in UW-OFDM,” in Proc. 2010 Int. OFDM Work., pp. 145–149 Hamburg (Germany), September 2010.
[2]. Huemer, M., Hofbauer, C., Huber, J.B “The Potential of Unique Words in OFDM” in Proceedings of the 15th International OFDM Workshop. Hamburg (Germany), September 2010.
Focus 1 students theory / programming

Channel Estimation Techniques for UW-OFDM

Responsible Professor:
Supervisor:

Prof. Martin Haardt
Sher Ali Cheema

used to compensate the channel dispersion in direct detection optical

systems. DMT schemes also take advantage of the CP, as in CP-OFDM, to mitigate the effect of channel dispersion. CP is a random sequence and is solely created in time domain by copying the last part of the output of IDFT. Instead of using a random sequence as CP, a known deterministic sequence called as unique word (UW) can be used which, in result, can be very helpful for channel estimation and synchronization. In this work, we will investigate the UW-OFDM structure for the DMT schemes and, based on this, propose new UW based DMT schemes for optical communication systems.
Requirements: Programming skills in MATLAB.
Tasks
Review of the literature
Implementation of these schemes in Matlab
Literature:
[1]. Huemer, M., Hofbauer, C., Huber, J.B “The Potential of Unique Words in OFDM” in Proceedings of the 15th International OFDM Workshop. Hamburg (Germany), September 2010.
[2]. M. Wolf, L.Grobe, M. R. Rieche, A. Koher and J. Vucic, “ Block transmission with linear frequency domain equalization for dispersive optical channels” in 12th International Conference on Transparent Optical Networks (ICTON) 2010, Munich, Germany.
[3]. M. Wolf, M. Haardt, “ Comparison of OFDM and frequency domain equalization for dispersive optical channels with direct detection” in 14th International Conference on Transparent Optical Networks (ICTON) 2012, Coventry, England.
Focus 1 students theory / programming

Unique Word based DMT Schemes

Responsible Professor:
Supervisor:

Prof. Martin Haardt
Sher Ali Cheema

used to compensate channel dispersion in direct detection optical systems.

One of the major drawback of OFDM is its high peak to average power ratio (PAPR) of the output signal which seriously limits the power efficiency of transmitter’s high power amplifier. Transmitting a signal with high PAPR requires highly linear power amplifiers with a large back-off to avoid adjacent channel interference due to nonlinear effects. In this work, we will investigate the different PAPR reduction algorithms for DMT schemes such as AC-DMT and DC-biased DMT.
Requirements: Programming skills in MATLAB.
Tasks
Review of the literature
Implementation of different PAPR reduction algorithms in Matlab
Literature:
S.H. Han, J. H. Lee, “An overview of peak-to-average power ratio reduction techniques for multicarrier transmission” in IEEE wireless communication volume 12, issue2, April 2005.
M. Wolf, L.Grobe, M. R. Rieche, A. Koher and J. Vucic, “ Block transmission with linear frequency domain equalization for dispersive optical channels” in 12th International Conference on Transparent Optical Networks (ICTON) 2010, Munich, Germany
M. Wolf, M. Haardt, “ Comparison of OFDM and frequency domain equalization for dispersive optical channels with direct detection” in 14th International Conference on Transparent Optical Networks (ICTON) 2012, Coventry, England.
Focus 1 students theory / programming

Peak to Average Power Ratio Reduction for DMT Schemes

Responsible Professor:
Supervisor:

Prof. Martin Haardt
Sher Ali Cheema

potential for wide range of applications within next generation transmission

and broadcasting technologies. Despite many advantages, the main challenge in VLC systems to date has been in improving data rates while considering the low bandwidths of the commercial LED devices. Many advance modulation schemes such as block transmission with frequency domain equalization, discrete multi-tone transmission (DMT), and carrier less amplitude and phase (CAP) have been suggested for VLC. Moreover, equalization also plays a vital role in improving the spectral efficiency of VLC systems. In this work, we will investigate the performance of CAP modulation scheme with decision feedback equalization and will compare it with the other schemes in indoor multipath environment.
Requirements: Programming skills in MATLAB.
Tasks
Review of the literature
Implementation of CAP with DFE in Matlab
Literature:
M. Wolf, L.Grobe, M. R. Rieche, A. Koher and J. Vucic, “ Block transmission with linear frequency domain equalization for dispersive optical channels” in 12th International Conference on Transparent Optical Networks (ICTON) 2010, Munich, Germany
F.-M.Wu, C.-T. Lin, et al., “1.1-Gb/sWhite-LED-Based Visible Light Communication Employing Carrier-Less Amplitude and Phase Modulation,” IEEE Photonics Technology Letters, pp. 1730 – 1732, Oct. 2012
Focus 1 students theory / programming

Carrierless Amplitude and Phase Modulation with Decision Feedback Equalization for VLC

Responsible Professor:
Supervisor:

Prof. Martin Haardt
Sher Ali Cheema

(BS) however close range users may gain if they communicate

directly between each other. Device-to-device (D2D) networking allows direct communication between cellular users and provide the following benefits:
Offload traffic from the core network
Higher data rates and power efficiency for both users and networks
Optimized spectrum reuse if sharing is allowed (underlay case)
Better coverage
Improved energy efficiency and reduced backhaul demand
Requirements: Strong programming skills in MATLAB and good knowledge about wireless systems.
Tasks
When one should use direct D2D communication instead of cellular?
Comparison between in-band (overlay and underlay) and out-of-band modes
Use of MIMO to provide additional degrees of freedom
How to implement D2D on new generation systems where the transceivers for uplink and downlink are different
(please contact the supervisors for more details)
Focus 1 students theory / programming

Device to Device (D2D) communications

Responsible Professor:
Supervisor:

Prof. Martin Haardt
Bilal Zafar, Sher Ali Cheema

Zhang
jianshu.zhang@tu-ilmenau.de
Description: 60 GHz band provides a large amount of unlicensed bandwidth,

which can be used to boost the data rate. The use of CMOS techniques makes it feasible to produce mmWave arrays. To exploit the benefits of 60 GHz band and massive MIMO arrays, novel physical layer techniques are desired.
Tasks
Literature study of the current mmWave WiFi Standard, i.e., 802.11 ad
Build up a standard compliant link level simulator via simulink
Investigate and / or develop signal processing techniques in one of the following research directions:
MIMO strategies / Channel Estimation / Comparison of single and multi-carrier PHY/
limited feedback
References
[1] Agilent Technologies, “Wireless LAN at 60 GHz – IEEE 802.11 ad Explained: Application Note", White paper, 2013.
[2] O. El Ayach, S. Rajagopal, S. Abu-Surra, Z. Pi, and R. W. Heath, “Spatially Sparse Precoding in Millimeter Wave MIMO Systems”, IEEE
Transactions on Wireless Communications, vol. 13, no. 3, 2014.
Focus
1 or 2 students, theory / programming / hardware / measurements / protocols

The Physical Layer of Future mmWave WiFi

Zhang
jianshu.zhang@tu-ilmenau.de
Description: In the development, standardization as well as the implementation process

of equipment manufacturers of wireless networks, simulations are necessary to test and optimize algorithms and procedures. System level simulations focus on network-related issues and are indispensable for evaluating new mobile network technologies.
Tasks
Understand the concept of system level simulator
Develop a System Level Simulator for LTE Heterogeneous Networks, involving relaying, D2D, and M2M, with a focus on:
Mobility management / radio resource management / PHY abstraction / stochastic modelling
References [1] J. C. Ikuno, M. Wrulich, and M. Rupp, “System level simulation of LTE networks", IEEE 71st Vehicular Technology Conference, Taipei,
Taiwan, May 2010
[2] H. Zhang, Y. Xie, L. Feng, and Y. Fang, “Base Station Design and Siting Based on Stochastic Geometry", Vehicular Technologies
Deployment and Application, InTech, Feb. 2013.
Focus
1 or 2 students, theory / programming / hardware / measurements / protocols

System Level Simulator for SDMA Enhanced LTE Heterogeneous networks

Zhang
jianshu.zhang@tu-ilmenau.de
Description: Compared to traditional FMCW radar, OFDM radar offers several attractive

features for radar applications, e.g., tolerance against Doppler shift. Yet the potential of OFDM radar, or using other multi-carrier techniques, has not been fully exploited, especially when multiple antennas are used.
Tasks
Literature study of current MIMO OFDM radar techniques
Implement and improve the performance of MIMO OFDM radar in range, velocity, and angle estimation, with the focus on
novel estimation techniques \ near-field radar \ alternative multicarrier techniques, e.g., FBMC
References
[1] C. Sturm, E. Pancera, T. Zwick, and W. Wiesbeck, “A novel approach to OFDM Radar processing", Radar Conference, IEEE, May 2009.
[2] M. Braun, C. Sturm, and F. K. Jondral, “Maximum likelihood speed and distance estimation for OFDM radar”, in IEEE Radar
Conference, 2010. [3] Y. L. Sit and W. Wiesbeck, “MIMO OFDM Radar with Communication and Interference Cancellation Features“, in IEEE Radar
Conference, 2014.
Focus
1 or 2 students, theory / programming / hardware / measurements / protocols

The Potential of MIMO Multi-Carrier Radar Systems

heterogeneous data, for that reason we would like to analyse

the so called big data using tensor algebra. Since the big data requires a lot of memory space a first and obvious solution is to perform a compression of the data. For that reason an implementation of the state of the algorithms as well as appropriate modifications should be achieved.
Tasks
Literature study on tensor algebra and tensor decompositions.
Literature study on big data compression
Implementation of the state of the art algorithms
References
[1] T. G. Kolda and B. W. Bader. “Tensor decompositions and applications”. SIAM, 51:455-500, 2009.
[2] N. D. Sidiropoulos, E.E Papalexakis and C. Faloutsos . “A Parallel Algorithm for Big tensor decomposition using randomly compressed cubes (PARACOMP)”, IEEE Internation Conference on Acoustic, Speech and Signal Processing (ICASSP), 2014
Focus
1 student theory / programming / hardware / measurements

Responsible Professor:
Supervisor:

Prof. Martin Haardt
Kristina Naskovska

Compression of a big tensor using tensor decompositions

Page

heterogeneous data, for that reason we would like to analyse

the so called big data using tensor algebra. Moreover, some of this data is either missing or it is corrupted and needs to be estimated. This can be achieved via decomposition of a low-rank tensor with mission entries. An implementation of the state of the algorithms as well as appropriate modifications should be achieved.
Tasks
Literature study on tensor algebra and tensor decompositions.
Literature study on decomposition of low-rank tensor
Implementation of the state of the art algorithms
References
[1] T. G. Kolda and B. W. Bader. “Tensor decompositions and applications”. SIAM, 51:455-500, 2009.
[2] M. Mardani, G. Mateos and G.B. Giannakis. “Imputation of Streaming Low-Rank Tensor data”, 8th IEEE Sensor Array and Multichannel Signal Processing Workshop (SAM), 2014
Focus
1 student theory / programming / hardware / measurements

Responsible Professor:
Supervisor:

Prof. Martin Haardt
Kristina Naskovska

Decomposition of a low-rank tensor with mission entries

Page

part of a tensor and data analysis. Decomposition of a

tensor in its rank-one components is commonly know as the CP decomposition, and its is calculated based on Trilinear-ALS. There exist other algebraic solutions such as SECSI, based on joint matrix diagonalization. In this project we would like to investigate non-symmetric matrix diagonalizations as well as possibility to introduce constrains on the factor matrices.
Tasks
Literature study on tensor algebra and tensor decompositions.
Literature study on decomposition of low-rank tensor
Implementation of the state of the art algorithms
References
[1] T. G. Kolda and B. W. Bader. “Tensor decompositions and applications”. SIAM, 51:455-500, 2009.
[2] P. Tichavsky, A. H. Phan and A. Cichocki “TWO-SIDED DIAGONALIZATION OF ORDER-THREE TENSORS”,
EUSIPCO, 2015 (submitted)
[3] F. Roemer, C. Schroeter and M. Haardt. “A semi-algebraic framework for approximate CP decompositions via joint
matrix diagonalization and generalized unfoldings. ASILOMAR, 2012
Focus 2 student theory / programming / hardware / measurements

Responsible Professor:
Supervisor:

Prof. Martin Haardt
Kristina Naskovska

Decomposition of a tensor in rank-one components

Page

addressed. Besides the increase of data-rates also new aspects need

to be considered, i.e., relaying, cooperation, MIMO. The terminals which access the network are not only powerful mobile phones but also small devices, like sensors with very limited resources. Non-binary LDPC codes achieve a high spectral efficiency with a moderate increase of complexity as compared to their binary counterparts.
Tasks
Get an understanding of non-binary LDPC codes and the basic decoding schemes (Belief propagation, Extended-Min-Sum (EMS) decoding) by considering just a single link.
Extend the scenario by a relay and investigate how the decoding complexity can be decreased, e.g., by cooperation.
References
[1] Esdras Nicoletto da Cunha and Renato Baldini Filho. “A Simple Cooperative LDPC Coding Scheme”, Revista Telecomunicacoes, vol. 15, no. 02, 2013.
[2] Marjan Karkooti and Joseph R. Cavallaro. “Communications Using Scalable, Medium Block-length LDPC Codes”, in Proc. IEEE WCNC, 2008.
Focus 1 student theory / programming / hardware / measurements

Responsible Professor:
Supervisor:

Prof. Martin Haardt
Marko Hennhöfer

Non-binary LDPC codes

Page

interference processes which don’t have a Gaussian distribution. Receivers with

decision schemes based on Gaussian assumptions will show a performance degradation in such scenarios as not all information can be exploited. Modelling the interference as poly-gaussian process allow simplified receiver implementations.
Tasks
Get an understanding of the poly-Gaussian modelling of noise and interference.
Implement a poly-Gaussian receiver structure for a SIMO system.
Evaluate the performance by comparing the results with traditional receiver schemes.
References
[1] Chabdarov Sh.M., Safiullin N.Z., Feoktistov A.Yu. Osnovy statisticheskoi teorii radiosvyazi: Poligaussovy modeli i metody. Kazan': KAI, 1983, 87p. (in Russian)
[2] Gantmakher F.R. Teoriya matrits. M.:Nauka, 1967, 574p. (in Russian)
[3] Levin B.R. Teoreticheskie osnovy statisticheskoi radiotekhniki. M.: Radio i svyaz', 1989, 656p. (in Russian)
[4] Venttsel' E.S. Teoriya veroyatnostei. M.: Nauka, 1969, 576p. (in Russian)

.
Focus 1 student theory / programming / hardware / measurements

Responsible Professor:
Supervisor:

Prof. Adel Nadeev, Prof. Martin Haardt
Damir Rakhimov, Marko Hennhöfer

Poly-Gaussian Modelling in Wireless Systems

Page

Information Technology
RF and Microwave Research Laboratory

Head: Prof. Dr. rer. nat.

habil. Matthias Hein

Measurement Research Lab

Head: Prof. Dr.-Ing. habil. Reiner S. Thomä
15.08.2013
Page 30

Christian Schneider
Responsible Professor : Reiner S. Thomä
Description:
The K-factor plays an

important role in wireless channel analysis and modelling. Different methods to
estimate the K-factor from measured channel sounding data sets are known, e.g. Maximum Likelihood
estimation and different moment based methods[1-3]. For parametric data sets no specific approach
has been discussed so far.
Tasks:
Study the estimation algorithm available in literature and our own papers [3]
Propose extension/application of K-Factor estimation for parametric channel data sets
Getting familiar with the available implementation the estimation of the K-Factor [1],[3], and different channel data sets (measured and parametric)
Implementation of the new algorithms (ML and moment method [2] and extend this to parametric data sets)
Compare and study the different methods, derive conclusion
References:
[1] Greenstein, L.J.; Michelson, D.G.; Erceg, V., "Moment-method estimation of the Ricean K-factor," Communications Letters, IEEE , vol.3, no.6, pp.175,176, June 1999, doi: 10.1109/4234.769521

[2] Tepedelenlioglu, C.; Abdi, A.; Giannakis, G.B., "The Ricean K factor: estimation and performance analysis," Wireless Communications, IEEE Transactions on , vol.2, no.4, pp.799,810, July 2003, doi: 10.1109/TWC.2003.814338
[3] Bottcher, A.; Vary, P.; Schneider, C.; Narandzic, M.; Thoma, R.S., "Estimation of the Radio Channel Parameters from a Circular Array with Directional Antennas," Vehicular Technology Conference (VTC Spring), 2011 IEEE 73rd , vol., no., pp.1,5, 15-18 May 2011, doi: 10.1109/VETECS.2011.5956303
Focus: 1-2 students Theory/programming/simulations/measurements/hardware

Christian Schneider
Responsible

Professor: Reiner S. Thomä

Description:
Multiuser MIMO (MU-MIMO) significantly increase the spectral efficiency of cellular systems. When
a base station (BS) or access point (AP) has fewer RF chains than the number of Tx antennas at
BS/AP and the number of users are comparatively large, then the selection and scheduling of users
plays a vital role in maximizing spectral efficiency inside a cell.
Tasks:
Theoretical understanding and implementation of state of the art user selection techniques.
Analysis of pros and cons of different algorithms from complexity and performance point of view.
References:



Focus: 2 students Theory/programming/simulations/measurements/hardware

Systems Group

Head: Prof. Dr.-Ing. habil. Andreas Mitschele-Thiel
15.08.2013
Page 34

an unmanned aerial vehicle (UAV) is flying over an urban

area, which suffers from a disaster. The purpose of an UAV is to localize ’survived’ devices, enabled with Wi- Fi module. In order to improve localization, a suitable signal propagation model is required. In our unique scenario it is crucial to find a fitting signal model which will result in the improved localization accuracy.
Tasks
Literature study on existing signal propagation models
Implementation of chosen models in Python
Simulation of different signal propagation models in different scenarios
Real time experiments with the most suitable model (optional)
References
[1] Faria, Daniel B.: Modeling Signal Attenuation in IEEE 802.11 Wireless LANs - Vol. 1 / Kiwi Project, Stanford University. 2006 (TR-KP06-0118). – Forschungsbericht
[2] Oleksandr Artemenko, Adarsh Harishchandra Nayak, Sanjeeth Baptist Menezes, Andreas Mitschele-Thiel: Evaluation of Different Signal Propagation Models for a Mixed Indoor-Outdoor Scenario Using Empirical Data , 7th International Conference on Ad Hoc Networks (ADHOCNETS'15), San Remo, Italy, September 2015
Focus
2 students theory / programming / hardware / measurements

Responsible Professor:
Supervisor:

Prof. Andreas Mitschele-Thiel
M.Sc. Alina Rubina

Comparison of Different Signal Propagation Models for a Mixed Indoor-Outdoor Scenario

8/24/15

Page

the network of wireless static nodes. Different selection algorithms of

reference data improve the localization information.
Tasks
Implementation of existing algorithms in Python
Simulation of different algorithms in different scenarios
Implementation of a new algorithm (optional)
References
[1] Artemenko, Oleksandr ; Simon, Tobias ; Mitschele-Thiel, Andreas; Schulz, Dominik ; Ta, Muhammad Rheza S.: Comparison of Anchor Selection Algorithms for Improvement of Position Estimation During the Wi-Fi Localization Process in Disaster Scenario. In: The 37th IEEE Conference on Local Computer Networks (LCN). Clearwater, Florida, USA, 10 2012

Focus
2 students theory / programming / hardware / measurements

Responsible Professor:
Supervisor:

Prof. Andreas Mitschele-Thiel
M.Sc. Alina Rubina

Implementation and Comparison of Reference Selection Algorithms for Localization in Wireless Networks

8/24/15

Page

mobile anchor (UAV). Usage of multiple UAV will improve the

speed of localization process. Time is a crucial factor in the disaster scenario.
Tasks
Literature research on the application of multiple UAVs in 2D and 3D scenarios
Implementation of one of the strategies in Python
Comparison of the scenario with one or multiple UAVs
References
[1] Artemenko, Oleksandr ; Rubina, Alina ; Golokolenko, Oleg ; Mitschele-Thiel, Andreas: How Different Trajectories of Moving Beacons Influence the Localization of Nodes in Disaster Scenarios Using Wireless Communication. In: The 17th International Symposium on Wireless Personal Multimedia Communications (WPMC’2014). Sydney, Australia, 09 2014
[2] Besada-Portas, E.; De La Torre, L.; de la Cruz, J.M.; de Andrés-Toro, B., "Evolutionary Trajectory Planner for Multiple UAVs in Realistic Scenarios," Robotics, IEEE Transactions on , vol.26, no.4, pp.619,634, Aug. 2010
Focus
2 students theory / programming / hardware / measurements

Responsible Professor:
Supervisor:

Prof. Andreas Mitschele-Thiel
M.Sc. Alina Rubina

Development of the Trajectory for Multi-UAV Scenario in Wireless Networks

8/24/15

Page

One of the most challenging tasks is a obstacle/collision avoidance.

Ultrasonic-based obstacle avoidance algorithms are studied in this work.
Tasks
Literature study on existing approaches on obstacle avoidance and requirements for them
Implementation of several different algorithms on UAV
Comparison of results
Extension of the best approach (optional)
References

[1] Borenstein, J.; Koren, Y., "Obstacle avoidance with ultrasonic sensors," Robotics and Automation, IEEE Journal of , vol.4, no.2, pp.213,218, Apr 1988. doi: 10.1109/56.2085
[2] Frew, Eric, and Raja Sengupta. "Obstacle avoidance with sensor uncertainty for small unmanned aircraft." Decision and Control, 2004. CDC. 43rd IEEE Conference on. Vol. 1. IEEE, 2004.
Focus
1/2 student theory / programming / hardware / measurements

Responsible Professor:
Supervisor:

Prof. Dr.-Ing. Habil. Andreas Mitschele-Thiel
M.Sc. Oleksandr Andryeyev

Advanced Ultrasonic-Based Obstacle Avoidance on Small UAVs

Page

today. UAV should be able to safely take-off, land and

follow some route. During these operations, UAV should consider the current state of environment. In addition, operator should be able to takeover control in case of emergency.
Tasks
Literature study on existing approaches on safe take-off, landing and route following approaches
Implementation and comparison between chosen approaches on UAV
Extension of existing approaches (optional)
References

[1] Jahn, B.; Barth, A.; Wulff, K.; Simon, T.; Romisch, J., "Rate control and flight stabilization for a quadrotor system," Unmanned Aircraft Systems (ICUAS), 2013 International Conference on , vol., no., pp.642,649, 28-31 May 2013; doi: 10.1109/ICUAS.2013.6564744
[2] Eendebak, P. T., A. W. M. van Eekeren, and R. J. M. den Hollander. "Landing spot selection for UAV emergency landing." SPIE Defense, Security, and Sensing. International Society for Optics and Photonics, 2013.
Focus
1 student theory / programming / hardware / measurements

Responsible Professor:
Supervisor:

Prof. Dr.-Ing. Habil. Andreas Mitschele-Thiel
M.Sc. Oleksandr Andryeyev

Safe Flight Routines For Small UAVs

Page