R & D Collaboration between Industry and Research Institutes. Prof Sunil Maharaj University of Pretoria 31 October PDF Free Download

R & D Collaboration between Industry and Research Institutes Prof Sunil Maharaj University of Pretoria 31 October 2013

AVI AFRIQUE R&D Collaboration between Industry and Research Institutes 31 October 2013 Prof. Sunil Maharaj SENTECH Chair in BWMC HOD Departement Elektriese, Elektroniese & Rekenaar-Ingenieurswese Department of Electrical, Electronic & Computer Engineering Kgoro ya Merero ya Mohlakase, Elektroniki le Khomphutha 2

Presentation Outline 1. Knowledge Economy: R&D and High Level HCD

Presentation Outline 1. Knowledge Economy: R&D and High Level HCD 2. National Development Plan Alignment

Presentation Outline 1. Knowledge Economy: R&D and High Level HCD 2. National Development Plan Alignment 3. Case Study: Industry University Collaboration - USA - Japan

Presentation Outline 1. Knowledge Economy: R&D and High Level HCD 2. National Development Plan Alignment 3. Case Study: Industry University Collaboration - USA - Japan 4. South African Opportunities

Presentation Outline 1. Knowledge Economy: R&D and High Level HCD 2. National Development Plan Alignment 3. Case Study: Industry University Collaboration - USA - Japan 4. South African Opportunities 5. ATNS UP Initiative

Need for Research and Higher Level Skills Need to develop a knowledge based economy Knowledge based economies have shown to have a higher GDP per capita World Bank 2012 Knowledge Economy Index (KEI) ranked SA 67 out of 146 countries The 2009/10 Global Competitiveness (GC) Report shows the correlation between GC and KEI SA s ito GC is ranked 52 out of 144 countries in 2012/13 Hence we need as a nation to increase our Research activities and innovation 10

Correlation between Economic Wealth of Nations and their Citation Intensity Impact of nations output 11

Department of Science and Technology (DST) 10 Year Plan Source: DST Ten Year Plan for SA (2008-2018) 12

PhD Production in South African HE 13

Contribution of doctoral qualifications in Eng. Sc.: 74% 14

Engineering Sc. Skills utilization: 78% Source: ASSAf Consensus Report, Sept. 2010 15

Industry must see the HCD value proposition and buy in! 16

National Development Plan Targets Based on 2010 data SA produces 1420 PhD graduates This implies 28 PhD graduates per million of population Industry must see the HCD value proposition and buy in! 17

National Development Plan Targets Based on 2010 data SA produces 1420 PhD graduates This implies 28 PhD graduates per million of population In Comparison: UK 288 PhD s per million USA 201 PhD s per million Korea 187 PhD s per million Industry must see the HCD value proposition and buy in! SA s target for 2030 is 5000 PhD s This implies approx. 100 PhD s per million of population Majority needs to be in SET for SA to be a leading innovator 19

R & D Investment in South Africa 20

R & D Expenditure in South Africa 21

R & D Intensity (2008/2009 baseline) 22

Are we Investing Enough? 11 billion/year by 2020 23

University-Industry Research Partnerships in the United States Bronwyn H. Hall University of California at Berkeley and NBER February 2004 24

Why Industries Link with Research Institutions? - an industry perspective in priority 1. Access to new research 2. Development of new products 3. Maintaining a relationship with the university 4. Obtaining new patents 5. Solving technical problems 6. Improving products and recruiting students were viewed as less important than these other above stated reasons. Tax Incentive in South Africa! 25

What Universities thought? 1. Obtaining funds for research assistance, lab equipment, and their own research agenda; 2. Obtaining insights into their own research by being able to field test theory and empirical research. 3. They viewed acquiring practical knowledge useful for teaching, student internships and job placement 4. Obtaining patentable inventions and business opportunities as less important motivations for entering into research alliances with industry 26

BENEFITS Influence of university research on industry R&D - university research enhanced and stimulated research and R&D in industry - augumenting the capacity of business to solve complex problems IP Protection and Technology Transfer through licensing Bayh-Dole Act by Congress in 1980 fostered publicprivate collaboration by promoting commercialization of government funded research 27

University Industry Collaborations in Japan Tokyo Institute of Technology Case Study Reference: Kyong-Joo Lee, Korea AIST, R&D Management, 41, 2, 2011, Blackwell Publishing Ltd. 28

Japanese university-industry collaborations ranking in 2006 29

Annual Revenue of joint R&D project at Tokyo Institute of Technology USD million 30

Proportion of Univ.-Industry partners based on firm size Ref: Kazuyuki Motohashi, Elsevier, Research policy, 34 (2005) 31

Category of Univ.- Industry collaboration by firm size 32

Problems in R&D collaborations with Univ. by firm size 33

Opportunities In South Africa 1. The DTI THRIP Programme 34

Opportunities In South Africa 1. The DTI THRIP Programme 2. The DST Scientific and Technological Research and Development Tax Incentive (effective 1 October 2012) 35

Example of Global Competitiveness Partnership 36

Rural Broadband Applications 39

Our Proposed Approach 40

Multi-Radio Cognitive Router 41

Internet Cloud Based Cognition 43

Throughput Aggregation Result: WiFi + LTE (USA Trial) 44

Proposed High Level Architecture 45

ATNS UP Partnership Human Capital Development through R & D Project

Engineering @ UP: Global Context Thomson ISI Essential Science Indicators Citations as at 2012 South Africa is 45 out of 97 countries in the world In South Africa UP is the top Engineering School On the world list of 1267 top engineering schools: World rankings Top 1% 47

48 Reference: ATNS: Matlane 2010, Tech. Innovation and Corporate Entrepreneurship

ATNS UP Partnership ATNS related technology projects for efficiency Investigating future technologies and proof of concept Developing high level skills to innovate in the SA and African Aviation Industry Provide continuous professional development to engineers and technicians Develop industry specific training and postgraduate modules Collaborate with ATNS international partners for research and skills development 49

Transponder Position Monitoring using Multilateration 50

Project Overview Multilateration (MLAT) is a mathematical technique that uses time difference of arrival to determine a transmitter s location. A transponder is a transmitter that is fixed on an aircraft and that transmits identity and altitude information when interrogated The goals of this project are to: Investigate the accuracy of MLAT algorithms in the context of O.R. Tambo International Airport Implement a system that is able to accurately locate transponders using data captured at the airport Investigate the effects of multipath on the reception of transponder replies

Multilateration Theory

Transponder Signals Used Mode A/C Transponder Reply 0.45 μs pulses Comprised of 15 pulses Transmits 4 octal values Altitude or identity information transmitted Mode S Transponder Reply 0.5 μs pulses 8 μs preamble 56 or 112 μs PPM data block Information transmitted based on Downlink Format

Multilateration Algorithms An overview of the algorithms that were implemented in this project: No. of Receivers 2D Algorithms 3D Algorithms 3 1. Friedlander 2D N/A 4 1. GS 2D 1. Friedlander 3D 2. Bucher s 5 Not needed 1. GS 3D 5 Not needed 1. Friedlander Least Squares Each algorithm takes the timestamps of the received reply and the positions of the receivers as input. The algorithms that use more receivers are advantageous because they produce unambiguous answers.

Modelling the Problem

Random Position Localization Simulation of the algorithms implemented involved randomly generating positions on the airport surface 10000 locations were generated with a given receiver geometry No. of Receivers Used 4 receiver algorithms 5 receiver algorithms Algorithm Name Bucher 3D Friedlander 3D GS 3D Friedlander LS Mean (m) 0.4 0.3 0.0 0.0 Standard Deviation (m) 3.5 3.0 0.0 0.0 Variance(m 2 ) 12.1 8.8 0.0 0.0 RMSE (m) 3.5 3.0 0.0 0.0 Median (m) 0.0 0.0 0.0 0.0 25th Percentile (m) 0.0 0.0 0.0 0.0 50th Percentile (m) 0.0 0.0 0.0 0.0 75th Percentile (m) 0.0 0.0 0.0 0.0 Interquartile Range (m) 0.0 0.0 0.0 0.0

Accuracy using Real Data The accuracy of the system implemented was measured using the test transmitters/transponders on the airport surface For a given sample taken on the 28 th of August: No. of Receivers Used 4 receiver algorithms 5 receiver algorithms Algorithm Name Bucher 3D Friedlander 3D GS 3D Friedlander LS Mean (m) 13.0 13.0 14.9 14.9 Standard Deviation (m) 10.8 10.8 12.8 12.8 Variance (m 2 ) 116.9 116.9 163.9 163.9 Median (m) 11.2 11.2 12.7 12.7 25th Percentile (m) 4.6 4.6 3.2 3.2 50th Percentile (m) 11.2 11.2 12.7 12.7 75th Percentile (m) 13.1 13.1 29.1 29.1 Interquartile Range (m) 8.4 8.4 25.9 25.9

Conclusions and Further Research The system is able to use MLAT to locate the position of transponders. Numerous sources of inaccuracy: TDOA, receiver geometry, position representation further scope of research Multipath compensation works, better technique for identifying multipath affected replies is required

Audio to text conversion and language Translation system for an Air Traffic Control call centre

The Broadband Opportunity for Enabling Economic Growth The Mobile Economy Where is the digital highway going? A perspective shared by GSM Association 60

Thank you for this opportunity

R & D Collaboration between Industry and Research Institutes. Prof Sunil Maharaj University of Pretoria 31 October 2013