Topics

Job-ID: WS2020-1

We plan to inte­grate and combine deep learning methods into/with conven­tional physical models in order to improve nonlinear mode­ling for elec­tronic devices like e.g. commu­ni­ca­tions and radio frequency (RF) trans­cei­vers. In parti­cular, we will enhance or replace conven­tional signal process­ing models like e.g. Volterra series, Wiener-models, Hammer­stein models or memory poly­no­mials with new tech­ni­ques of machine learning. Involved re­search fields are deep learning, deep unfol­ding, model –based neural networks, few shot learning, meta learning (transfer models from one device to another), sequence analysis using LSTMs and trans­for­mers/BERT. Rein­force­ment learning methods should be considered for impro­ving models that have delayed signals, there­fore also delayed error or reward signals. The goal is to enhance exis­ting physical models by new deep learning tech­ni­ques to make model simu­la­tion faster without losing preci­sion and mode­ling capa­city or to improve exis­ting models using new machine learning approa­ches. Also online and tracking models should be inves­ti­gated, that is, models that adapt to the current envi­ron­ment like tempe­ra­ture, humi­dity, power consump­tion, voltage devia­tions, fast chan­ging system para­me­ters etc. These online methods might be combined or substi­tuted by few shot and meta-learning methods. 

Further­more, we also plan to enhance and/or substi­tute model-based signal process­ing blocks like e.g. pre-coding, digital pre-distor­tion, nonlinear distor­tion miti­ga­tion, or adap­tive channel esti­ma­tion/equa­liza­tion, by machine learning approa­ches. 

  • Tech­nical skills: Master’s degree (for PhD posi­tion) in computer science, mecha­tro­nics, electrical engi­nee­ring, mathe­ma­tics, physics, or similar subject; know­ledge in one or more of the follo­wing appli­ca­tion domains: signal process­ing, machine learning 

  • Social skills: willing­ness and ability to work in a team, solu­tion-orien­ta­tion, persis­tent rese­ar­cher, well-struc­tured and goal-oriented working style, relia­bi­lity and excel­lent inter­per­sonal skills, hands-on menta­lity 

  • Personal skills: English level C1 

Supervising team: Mario Huemer, Sepp Hoch­reiter, Andreas Springer, Robert Wille

Job-ID: WS2020-2

Due to the ever incre­a­sing comple­xity of circuits and systems, the design of System on Chips starts at high levels of abstrac­tion. Model-based design is a state-of-the-art design metho­do­logy which addresses this deve­lop­ment. Here, initial drafts of the design are defined using mode­ling languages such as UML, SysML, etc. However, already at this stage severe design choices are made which will have a substan­tial impact, e.g., at the even­tual cost of the design. But properly esti­ma­ting those effects is hard to impos­sible using conven­tional methods, which is why those problems are thus far tackled, e.g., by “trial and error”, “gut feeling”, and “expe­ri­ence of the design engi­neer”. This frequently yields situa­tions where, after months of imple­men­ta­tions, design choices turn out to not satis­fying the desired objec­tives—a severe threat for time to market which is crucial in the EDA industry.  

Machine Learning re­search in the fields of deep learning, few shot learning, meta learning, rein­force­ment learning, sequence analysis methods like LSTM or Trans­former/BERT should be performed and new methods to provide improved cost esti­ma­tions should be deve­l­oped. Onto these esti­mates, design choices can be based to obtain opti­mized SoCs, and even­tually to auto­mate further steps of the design process. Possible industry part­ners (provi­ding corre­spon­dingly needed data and use cases) are avail­able. Results gene­rated by this project will directly be applied in indus­trial prac­tice. 

  • Tech­nical skills: Master’s degree in computer science, mecha­tro­nics, electrical engi­nee­ring, mathe­ma­tics, physics, or similar subject; know­ledge in one or more of the follo­wing appli­ca­tion domains: circuit and system design, machine learning 
  • Social skills: willing­ness and ability to work in a team, solu­tion-orien­ta­tion, persis­tent rese­ar­cher, well-struc­tured and goal-oriented working style, relia­bi­lity and excel­lent inter­per­sonal skills, hands-on menta­lity 
  • Personal skills: English level C1
Super­vi­sing team: Mario Huemer, Sepp Hoch­reiter, Andreas Springer, Robert Wille

Job-ID: WS2020-3

Re­search will be focused on buil­ding blocks for mm-wave appli­ca­tions like commu­ni­ca­tions and radar. The candi­date will inno­vate in either receive, transmit, or frequency synthesis in coor­di­na­tion with other rese­ar­cher in the SAL mm-wave labo­ra­tory. 

  • Tech­nical skills: Strong back­ground in circuit design and layout of inte­grated circuits, and RF and commu­ni­ca­tions theory required. Candi­date must be able to handle the Cadence tool set and electro-magnetic field solvers (2.5D and/or 3D). Know­ledge of measu­re­ment equip­ment incl. RF and expe­ri­ence in HW RF veri­fi­ca­tion strongly desired
  • Social skills: Candi­date must be both team player and inde­pen­dent rese­ar­cher. Profi­ci­ency in English required, German language skills not required but a plus. 
  • Personal skills: Highly moti­vated, orga­nised, high energy level. Strong analy­tical skills required, suffi­cient back­ground in electrical engi­nee­ring a must. 
Super­vi­sing team: Harald Pretl, Andreas Stelzer

Job-ID: WS2020-4

Focus a: MIMO Trans­ceiver for Inte­grated Sub-THz Sensor Systems: Buil­ding Blocks and Inte­grated Circuits for MIMO Trans­cei­vers above 100 GHz. 
Focus b: Next Gene­ra­tion/Alter­na­tive mmW-Sensing and -Imaging Concepts: Alter­na­tive and situa­tion aware modu­la­tion schemes as well as imaging concepts, realiza­t­ions, and algo­rithms. 

Re­search will be focused on buil­ding blocks for sub-THz-sensing appli­ca­tions at 120, 230, 330 and 480 GHz, respec­tively, focused on RX-TX sepa­ra­tion at highest frequen­cies as well as array opti­miza­tion under the given cons­traints. The candi­date will inno­vate in either core func­tional blocks, the full sensor system, array concepts toge­ther with rese­ar­chers in the SAL mm-wave labo­ra­tory. 

  • Tech­nical skills: Good back­ground in RF-design and mmW-sensor evalua­tion and depen­ding on work either strong back­ground in inte­grated circuit design for inte­grated design-focused work or strong back­ground in RF-system concepts and mathe­ma­tics/simu­la­tion tools for concep­tual work. 
  • Social skills: Candi­date must be self-contained and moti­vated to explore new topics, and at the same time team player for the realiza­tion of complex systems toge­ther with others. Profi­ci­ency in English required, German language skills not required but welcome. 
  • Personal skills: Highly moti­vated, well orga­nized, high energy level. Strong analy­tical skills required, suffi­cient back­ground in electrical engi­nee­ring or physics a prere­qui­site for all candi­dates. 
Super­vi­sing team: Harald Pretl, Andreas Stelzer

Job-ID: WS2020-5

Re­search in the area of auto­mata learning and veri­fi­ca­tion in order to further auto­mate the testing process of computer-based systems. Novel combi­na­tions of model learning, model-based test-case gene­ra­tion and runtime veri­fi­ca­tion in order to test selected depen­da­bi­lity proper­ties of a given system-under-test. The re­search shall go beyond the logical correct­ness of a system’s func­tio­na­lity and also consider other quality charac­te­ris­tics, like latency and energy consump­tion. 

  • Tech­nical skills: Computer scien­tist with strong back­ground in formal methods, veri­fi­ca­tion or auto­mated test-case gene­ra­tion. Know­ledge of auto­mata-learning algo­rithms is an advan­tage. 
  • Social skills: Candi­date must be both team player and inde­pen­dent rese­ar­cher. Profi­ci­ency in English required, German language skills not required but a plus. 
  • Personal skills: Highly moti­vated, orga­nised, high energy level. 
Super­vi­sing team: Bern­hard Aichernig, Thomas Pock

Job-ID: WS2020-6

Deve­lop­ment of novel and theo­re­ti­cally sound machine learning tech­ni­ques based on varia­tional tech­ni­ques, with a focus on robust­ness and imple­men­ta­bi­lity on embedded/low power devices.  

  • Tech­nical skills: Computer scien­tist or mathe­ma­ti­cian with strong back­ground in varia­tional/opti­miza­tion methods, machine learning and image process­ing/computer vision. 
  • Social skills: Candi­date must be both team player and inde­pen­dent rese­ar­cher. Profi­ci­ency in English required, German language skills not required but a plus. 
  • Personal skills: Highly moti­vated, orga­nised, high energy level. 
Super­vi­sing team: Bern­hard Aichernig, Thomas Pock
Job-ID: WS2020-7

With the advan­ce­ment in CMOS tech­no­lo­gies, multiple elec­tronic systems are getting inte­grated on a single chip. The commu­ni­ca­tion between these systems demands a dramatic increase in band­width over on-chip and chip-to-chip trans­mis­sion lines. Energy and area effi­cient high-speed data trans­mis­sion is beco­ming a chal­lenge in multi-processor systems-on-chip (MPSoCs) and multi-input-multi-output systems (MIMOs). 

Simul­ta­neous bidi­rec­tional (SBD) signa­ling, where data streams are trans­mitted and received on both sides of a given channel, is an attrac­tive approach offe­ring higher pin effi­ci­ency at a given data rate. The avail­able Ph.D. posi­tion is related to the deve­lop­ment of SBD trans­cei­vers for MIMO appli­ca­tions. This re­search work focuses on an energy and area effi­cient SBD trans­ceiver with adap­tive echo/cross­talk cancel­la­tion for the support of a wide range of chan­nels (on-chip and chip-to-chip trans­mis­sion lines). The aim is to achieve a beha­vioral model of all-important effects and novel circuit archi­tec­tures with adap­tive solu­tions for atte­nua­tion of echo and cross­talk. 

  • Tech­nical skills: Master’s degree or diploma with a strong academic back­ground in analog inte­grated circuit design, good know­ledge and deep under­stan­ding of mathe­ma­tical analysis in circuit and system design, solid expe­ri­ence in IC design CAD tools (like Cadence) for CMOS circuit design and layout 
  • Social skills: Both team player and inde­pen­dent rese­ar­cher 
  • Personal skills: Highly moti­vated, orga­nized 
Super­vi­sing team: Andreas Springer

Job-ID: WS2020-8

Re­search in the area of signal process­ing and physical layer aspects for indus­trial wire­less sensor networks with the goal to charac­te­rize and rate the relia­bi­lity and depen­da­bi­lity of wire­less links within a network. Addi­tio­nally, centra­lized and coope­ra­tive loca­liza­tion methods, based on infor­ma­tion extracted from the wire­less commu­ni­ca­tion between nodes should be applied for enhan­cing secu­rity, safety and relia­bi­lity of the network 

  • Tech­nical skills: Strong back­ground in signal process­ing, commu­ni­ca­tions engi­nee­ring and/or wire­less sensor networks 
  • Social skills: Both team player and inde­pen­dent rese­ar­cher 
  • Personal skills: Highly moti­vated, orga­nized 
Super­vi­sing team: Andreas Springer

Job-ID: WS2020-9

Factory commu­ni­ca­tion is facing the leap of the new gene­ra­tion wired and wire­less tech­no­lo­gies e.g. TSN, WiFi6, 5G, and more, expan­ding from auto­ma­tion to the IIoT. In this context, this project focuses on theo­retic work, simu­la­tion and empi­rical re­search to use for factory auto­ma­tion 5G and WIFI6 tech­no­lo­gies. This will lead to the inves­ti­ga­tion on requi­re­ments and chal­lenges for commu­ni­ca­tion of sensors and actua­tors used in indus­trial envi­ron­ments and design of proto­cols and metho­do­lo­gies for critical control in factory of the future. 

  • Tech­nical skills: Master of Science degree in a rele­vant field such as Electrical Engi­nee­ring, Physics, Applied Mathe­ma­tics, Computer Science, Commu­ni­ca­tions or similar; excel­lent English commu­ni­ca­tion skills (spoken and written). Advan­ta­geous quali­fi­ca­tions: Know­ledge of mobile commu­ni­ca­tions systems, wire­less commu­ni­ca­tion, simu­la­tion tools such as NS3 and OMNET++. Programming skills: C++, Python, Java 
  • Social skills: Both team player and inde­pen­dent rese­ar­cher
  • Personal skills: Highly moti­vated, orga­nized
Super­vi­sing team: Alois Zoitl, Andreas Springer