Embedded Systems

7
The design of innovative software and hardware is the core of technological and industrial
progress. Both the departments of Mathematics and Computer Science and Electrical Engi-
neering play an active role in the development of new, innovative technology. The Master of
Science program in Embedded Systems at the TU/e is illustrative of this active role, as it is a
co-production of these two departments, awaiting students with a background in computer
science, as well as graduates from the field of electrical engineering.
The program rests on a sound theoretical foundation, with an emphasis on the design
of quality embedded systems. As a graduate of this program, you will have developed
a scientific attitude and an engineering approach to the field. Your position will be the
design of embedded systems from a high-level architecture viewpoint, via requirements
and behavioral specifications and using platforms, hardware and silicon. You will be able
to play a leading role in the development of embedded systems, either in scientific research,
in industry or governmental organizations.
After taking some courses, you will probably have a more clear picture of the academic
direction you want to pursue in your studies. If not, you may want to talk to several staff
members or the study advisor. In the specialization for your subject, there are people that
you may want to be involved with for your final master project (see Section 7.4). In order
to compose a well-balanced program that provides adequate prerequisites for this project,
it is advisable to first choose and consult a project supervisor in the specialization of your
interest before choosing elective courses. As a rule of thumb, you should start your search
for a supervisor and the construction of your individual program not later than at the end
of your first year.
The Embedded Systems program focuses on the design of efficient and reliable systems.
In order to be able to compose dependable protocols for the behavior of such systems, you
need knowledge of algorithms, performance, hardware, methods of design and documen-
tation, and an insight into the variability and maintainability of these protocols. All these
aspects are addressed in the compulsory part of the program.
The masters degree program in Embedded Systems is a joint program of the 3TU fed-
eration. The programs offered at each location share the common core part (25 credits).
Furthermore, some specialisation courses are exchanged by 3TU telefacilities.

55
56 Embedded Systems

7.1 Admission
A Bachelor degree in Computer Science or in Electrical Engineering obtained at a Dutch
university provides direct admission to the ES program. Students with a different degree
and from foreign universities have to apply for admission via the admission committee.
Dutch HBO graduates have to take a pre-master program before they can be admitted, see
Section 2.6.
The admission procedure is described in Section 1.5, and the requirements are listed in
the Teaching and Examination Regulations (see Appendix A.3).

7.2 Learning outcomes

Graduates from the Embedded Systems master program

 are qualified to degree level in the domain of science, engineering and technology;
 are capable of acquiring knowledge independently;
 are capable of critical thinking, can reason logically and form opinions;
 have design skills, presentation skills, and communication skills;
 have a holistic view on embedded systems, their development, and their embedding
in larger systems;
 are able to master complex embedded systems;
 can describe and study structure and behaviour of embedded systems;
 possess knowledge of contemporary techniques;
 are proficient in the Design of embedded systems;
 possess knowledge of requirement engineering, modelling, testing and implementa-
tion techniques;
 have a flexible and inquisitive mind with regard to developments in the field;
 invent own specific tools, theories and techniques if unavailable;
 are aware of their own position and that of embedded systems in society;
 can present and communicate their ideas and visions on embedded systems;
 can work in a multidisciplinary design team.

7.3 Curriculum 2016

The Master program Embedded Systems is a two-year program of 120 ECTS in total. The
academic year is subdivided into two semesters, the fall semester starting in September, and
the spring semester starting in February. It is possible to enter the program at the start of
either semester; however, starting in September is preferred. The program is only offered as
a full time study program.
The curriculum consists of courses offered by the Computer Science division of the de-
partment Mathematics and Computer Science and the department of Electrical Engineering
This section describes the curriculum at the TU/e; the programs of the two partner univer-
sities can be found in the program guides of the TUDelft and the University of Twente.
Embedded Systems 57

The ES curriculum is structured in streams, which provide an organized set of courses in

particular subject areas within Embedded Systems. The purpose of streams is, on the one
hand, to provide guidance to students in composing coherent individual study programs,
and on the other hand, to preserve the multidisciplinary nature of each individual ES study
program.
When you enroll to the ES master you should choose a stream right away because this
determines the list of stream courses and the assignment of a mentor. The streams are:
 Systems on Chip
 Embedded Software
 Embedded Networking
 Cyber-Physical Systems

Each stream has the same structure, as summarized in the following table.
Units EC
Mandatory courses 25
Stream courses 15
Stream electives 20
Free electives 20
Master thesis preparation 10
Master project 30

Each stream has a three mandatory courses, and a selected set of elective courses from which
at least 20 credit points should be chosen. The remaining part of your individual study pro-
gram can be used for free electives, i.e., courses taken from any master program offered
at the TU/e. We give some recommendations in Section 7.3.6. In addition you may also
choose courses from the ES programs in Delft or Twente, provided that there is no content-
wise overlap with other courses in your individual study program. You need to obtain
permission from the study advisor before enrolling in courses from Delft or Twente to con-
firm there is no overlap issue. Within your study program you should include 15 credit
points in international experience if you do not yet have that and are not doing your master
project abroad. To prepare for the master project, you start with a master thesis preparation
in quarters 5 and 6, followed by the master project in quarters 7 and 8. In the following
sections, more details for each of the streams are given.

7.3.1 Mandatory program elements

The mandatory program elements for all students are listed below:

Quarter Code Unit EC Exam

1 2IMF30 System Validation 5 w+a
2 5SIA0 Embedded Computer Architecture 5 a+o
2 2IMN25 Quantitative Evaluation of Embedded Systems 5 w+a
3 2IMN20 Real-time Systems 5 w+a
4 5LIB0 Embedded Systems Laboratory 5 a+o
5-6 2IMC05/5T514 Preparation graduation project ES1 10 a
7-8 Master thesis project 30
58 Embedded Systems

1
Preparation for graduation project ES consists of a literature survey and feasibility study for the
graduation project. The preparation has to be finished and graded before the start of the final
project.

7.3.2 Systems on Chip stream

Modern chips are rapidly evolving into complete systems on chip (SoCs). Heterogeneous
SoCs with one or more general-purpose embedded processors (e.g., ARM cores), dedicated
accelerators, reconfigurable fabric, and advanced memory and interconnect structures are
typically used in devices where energy efficiency and performance are both important. Ho-
mogeneous SoCs are characterized by replication of regular structures, leading to multi-
and many-core architectures that target high-performance computing. The emergence of
SoCs leads to new challenges in VLSI design, design automation, programming and code
generation, task and communication mapping and scheduling, memory management, and
model-driven design-space exploration.

Quarter Code Unit ECTS Exam

Mandatory courses 65
1 2IMF20 Hardware verification 5 w+a
3 5LIE0 Multiprocessors 5 a+o
4 5LID0 Systems on silicon 5 w+a
Stream electives 1 25
1 5LIN0 Video processing 5 w
1 5LIL0 Intelligent architectures 5 a
2 5LIH0 Digital integrated circuit design 5 w+a
2 5LIF0 Advanced digital circuit design 5 w/o
2 2IMF25 Automated reasoning 5 w+a
2 2IMNT1 Embedded computer architectures 2 5 w+a
2 5KK07 Energy efficient embedded systems 5 a
2 5LIG0 Applied combinatorial algorithms 5 a+o
3 5LIM0 Parallelism, compilers and platforms 5 w+a
3 5SIB0 Electronic design automation 5 w+a
3 5LIJ0 Embedded control systems 5 w
4 2IMN35 VLSI programming 5 a
6 2IMF00 Seminar formal system analysis 5 a
1
You have to choose at least 20 credit points from this list

7.3.3 Embedded Software stream

The behaviour and functionality of embedded systems is largely determined by the software
that it runs. This software must be of excellent quality because embedded systems must of-
ten perform safety critical tasks without human supervision for long periods of time with
guaranteed performance. Parallelism (in tasks, data and instructions), distribution, and re-
activity pose additional challenges. As the software generally outlives the hardware, the
software must be such that it is easily maintainable and adaptable to always changing cir-
cumstances. Model-driven design, formal proof and advanced code generation techniques
are very important to achieve this.
Embedded Systems 59

Quarter Code Unit EC Exam

Mandatory courses 65
2 2IMF25 Automated reasoning 5 w+a
3 5LIM0 Parallelization, compilers and platforms 5 w+a
4 2IMP20 Generic language technology 5 w+a
Stream electives 1 25
1 5LIN0 Video processing 5 w
1 2IMN10 Architecture of distributed systems 5 w+a
1 2IMA10 Advanced algorithms 5 w+a
1 5LIL0 Intelligent architectures 5 a
2 2IMP10 Program verification techniques 5 w+a
2 5LIG0 Applied combinatorial algorithms 5 a+o
3 2IMP25 Software evolution 5 w+a
3 2IMF35 Algorithms for model checking 5 w+a
3 5LIE0 Multiprocessors 5 a+o
3 5LIJ0 Embedded control systems 5 w
4 2IMN35 VLSI programming 5 a
4 2IMNT2 Real-time software development 5 a
4 5LIK0 Embedded signal processing systems 5 w+a
6 2IMF00 Seminar formal system analysis 5 a
6 2IMN00 Seminar systems architecture and networking 5 a
1
You have to choose at least 20 credit points from this list

7.3.4 Embedded Networking stream

Embedded Networking (EN) refers to the powerful trend of the last twenty years of connect-
ing embedded systems into networks. For electronic systems in automotive, for example,
the network is often the point of integration. More recently we have seen the concept of the
Internet of Things, the vision that everyday objects get enriched with embedded electron-
ics, that these objects are uniquely identified and communicate using a unified protocol and
naming scheme (indeed, the Internet protocols). EN includes the fields of sensor networks,
but also networked systems that represent a platform and are not identical to the applica-
tion. From the sensor network domain, concerns of effective resource management (like
size, energy, memory, communication bandwidth) are derived. Quality metrics for EN in-
clude performance (latency, throughput), dependability (quality of service) and scalability.
Besides these, EN challenges lie in the architecture of system and software, in management
and sharing of distributed resources, in interoperability and semantics, in security and pri-
vacy, and in application development. While it is fairly easy to sketch advanced applica-
tions, it is not straightforward to realize these in a cost-effective manner. Relevant topics for
this stream are: distributed systems (architecture and protocols), networked systems, data
semantics, network security, system correctness, and resource management.

Quarter Code Unit ECTS Exam

Mandatory courses 65
1 2IMN10 Architecture of distributed systems 5 w+a
1 5LIC0 Networked embedded systems 5 a+o
2 2IMN15 Internet of things 5 w+a
continued on next page
60 Embedded Systems

continued from previous page

Quarter Code Unit ECTS Exam
Stream electives 1 25
1 2IMA10 Advanced algorithms 5 w+a
1 2IMF20 Hardware verification 5 w+a
2 5LIH0 Digital integrated circuit design 5 w+a
2 5LIF0 Advanced digital circuit design 5 w/o
2 2IMF25 Automated reasoning 5 w+a
2 5KK07 Energy efficient embedded systems 5 a
2 2IMS20 Hackers hut 5 w+a
3 5SIB0 Electronic design automation 5 w+a
3 2IMS15 Verification of security protocols 5 w+a
4 5LIA0 Embedded visual control 5 a
4 5LID0 Systems on silicon 5 w+a
4 5LIK0 Embedded signal processing systems 5 w+a
6 2IMF00 Seminar formal system analysis 5 a
6 2IMN00 Seminar systems architecture and networking 5 a
1
You have to choose at least 20 credit points from this list

7.3.5 Cyber-Physical Systems stream

Cyber-Physical Systems (CPS) are characterized by a tight coupling between embedded
computer (cyber) systems and physical processes, monitored and controlled by those com-
puter systems. Next generation embedded systems in automotive, professional printing,
electron microscopy, semiconductor lithography, and medical imaging, are rapidly evolving
into CPS. CPS require integral, multidisciplinary design, involving computer engineering,
control, mechatronics, networking, signal processing, and mathematical modeling. Embedded-
system design needs to take into account constraints and requirements from other domains;
cross-domain design paradigms are needed to optimize the embedded system in combina-
tion with the monitored and controlled system.

Quarter Code Unit ECTS Exam

Mandatory courses 65
2 2IMN15 Internet of things 5 w+a
3 5LIJ0 Embedded control systems 5 w
4 5LIK0 Embedded signal processing systems 5 w+a
Stream electives 1 25
1 2IMN10 Architecture of distributed systems 5 w+a
1 2IMA10 Advanced algorithms 5 w+a
1 5LIC0 Networked embedded systems 5 a+o
1 5LIL0 Intelligent architectures 5 a
1 2IMF20 Hardware verification 5 w+a
2 5LIF0 Advanced digital circuit design 5 w/o
2 5LIV0 Video Health Monitoring 5 w/o
2 2IMNT1 Embedded computer architectures 2 5 w+a
2 5LIG0 Applied combinatorial algorithms 5 a+o
3 2IMP25 Software evolution 5 a
continued on next page
Embedded Systems 61

continued from previous page

Quarter Code Unit ECTS Exam
3 5LIE0 Multiprocessors 5 a+o
3 5LIM0 Parallelism, compilers and platforms 5 w+a
3 5SIB0 Electronic design automation 5 w+a
4 2IMP20 Generic language technology 5 w+a
4 5LIA0 Embedded visual control 5 a
4 2IMN35 VLSI programming 5 a
4 2IMNT2 Real-time software development 5 a

1
You have to choose at least 20 credit points from this list

7.3.6 Free Electives

In principle all master courses offered at the TU/e can be chosen as free electives. Also
courses from the Embedded Systems programs in TU Delft and the University of Twente
may qualify as suitable free electives. Your complete study program including the free
electives always requires approval by the examinations committee. Your mentor will help
you to compose a balanced program that is very likely to be approved.
Some courses may be prescribed as homologation courses (up to 10 credits) to make
up for deficiencies in former education (see Section 7.3.7). This prescription is part of the
admission decision.
If you do not (yet) have a degree from another country or at least 15 credit points in
international experience you should reserve 15 credits from the room for free electives to
take courses abroad or to do an internship of 15 credits abroad, see Section 1.8.

7.3.7 Homologation units

Students entering Embedded Systems program have very diverse backgrounds. When you
are admitted to the program, or directly at the start of the program, it will be evaluated
to what extent you satisfy the BSc-level knowledge and skills required for the Embedded
Systems master program in general, and the stream of your choice in particular. Deficiencies
should be repaired by taking one or more homologation modules. Homologation packages
may consist of one or more homologation modules and may comprise at most 10 EC in
total. The credits obtained for homologation modules will count as free electives in your
study program. We may prescribe homologation modules, depending on your background
knowledge and skills and the stream of your choice.
We will offer the homologation modules below. You will generally be expected to do the
homologation modules by self-study, with some guidance by a responsible lecturer. In most
cases, the homologation modules are associated with existing BSc-level courses, so that you
can attend lectures, tutorials and lab sessions if needed.
62 Embedded Systems

Quarter Code Unit ECTS

1 5LFK0 Circuit analysis 2.5
1 5LIQ0 Linear Systems, signals and control 2.5
1 5LIS0 Computer architecture and C programming 2.5
1 2IHT10 Logic and set theory 2.5
1 5LIP0 Digital integrated circuits: fundamentals 2.5
2 5LIR0 Linear Systems, signals and control (DBL) 2.5
3 2IHS10 Software specification 2.5

If you feel that you lack knowledge and skills not addressed by one of the homologation
modules above, then contact the study advisor.

7.3.8 Seminars, Internship and multi-disciplinary design project

Since the master project can already be performed in an industrial environment, it is usually
advised to spend the rest of the study program on regular courses. In some cases, however,
an internship (code 2IMC10 or 5L990) may be a valuable addition to the program, provided
that it enhances practical experience, provides deepening of knowledge, contributes to the
specialization, and perhaps provides international experience. See Section 1.8 for details
on internships. Instead of an internship, a multi-disciplinary design project (code 2IW70
or 5KK05) can be performed, in which students work in groups of 2 to 3 on an embedded
design under the supervision of a member of staff. We refer to the course page for more
information. Note that it is not possible to do both an internship and the multi-disciplinary
design project.
The Computer Science division of the department Mathematics and Computer Science
offers seminars and capita selecta that may help you prepare for your master project by
studying topics not covered by regular courses. The table below lists the possibilities and
codes for seminars, capita selecta, internship and multi-disciplinary design project.

Quarter Code Unit EC Exam

Seminars
4 2IMS00 Seminar information security technology 5 a
4 2IMA00 Seminar algorithms 5 a
6 2IMM00 Seminar data mining 5 a
6 2IMG00 Seminar applied geometric algorithms 5 a
6 2IMW00 Seminar web engineering 5 a
6 2IMI00 Seminar architecture of information systems 5 a
6 2IMN00 Seminar systems architecture and networking 5 a
6 2IMP00 Seminar software engineering and technology 5 a
6 2IMV00 Seminar visualization 5 a
6 2IMF00 Seminar formal system analysis 5 a
6 2IMA00 Seminar algorithms 5 a
Capita Selecta
2IMS05 Capita selecta security 5 a
2IMW05 Capita selecta web engineering 5 a
2IMF05 Capita selecta formal system analysis 5 a
2IMI05 Capita selecta architecture of information systems 5 a
2IMA05 Capita selecta algorithms 5 a
continued on next page
Embedded Systems 63

continued from previous page

Quarter Code Unit EC Exam
2IMM05 Capita selecta data mining 5 a
2IMG05 Capita selecta applied geometric algorithms 5 a
2IMN05 Capita selecta systems architecture and networking 5 a
2IMP05 Capita selecta software engineering and technology 5 a
2IMV05 Capita selecta visualization 5 a
Internship / Multi-disciplinary design project1 (second year)
2IMC10/5L990 Internship 15 a
2IMC25/5KK05 Multi-disciplinary design project 10 a

1
Note that only one of these may be chosen, and that the Internship requires permission of the
coordinator. See Section 1.8 for the details.

7.4 Final project

Prior to starting the final project you should first choose and consult the intended final
project supervisor. The project can be completed in any of the specializations listed in Sec-
tion 10.1 and Section 10.3, provided that a staff member of the associated group has the
supervision. In case the project is carried out under supervision of a staff member of the
Mathematics and Computer Science department, the course code is 2IMC00. In case the
supervisor is from the Electrical Engineering department, the code is 5T746. If you are in
doubt how to choose the graduation supervisor please consider contacting your mentor.
The start of your master project is marked by submitting a completed graduation plan
containing the necessary information on the project (name, place, period, supervisor, and so
on), and stating the fact that you obtained approval of your study program (see Section 1.10)
and that you fulfill all requirements listed in Section 7.4.1. The form must be accompanied
by a project description and signed by you, your supervisor, the head of the relevant spe-
cialization, and the study advisor.

7.4.1 Admission
During the master project, you should be able to work and concentrate on your project
full-time. In practice, however, it turns out to be rather difficult to plan all your curricular
activities and, especially, their success:

 If all courses have been completed, permission to start the master project will be
granted.
 If more than two courses or 10 EC (whichever is lower) have not been completed, such
permission will not be granted.
 In other cases (no more than two courses or 10 EC not yet completed) it depends on
the status of the uncompleted courses whether permission to start the master project
will be granted, and if so, whether it is feasible to work on the project full time.

Courses that are to be taken as homologation units must be completed and passed before
you are allowed to start the master project. Note that the preparation for graduation ES project
(2IMC05 or 5T514) must be finished and graded before you can start the master project.
64 Embedded Systems

Also, be aware that you are not allowed to finish your project (with presentation and de-
fense) before you have completed all your courses. For more information, please contact
the study advisor.

7.4.2 Planning
Together with your supervisor, you decide on a description of your topic and a global plan-
ning. You also arrange the supervision method, including how often you and your supervi-
sor will meet to discuss progress. The project is concluded with a thesis and a presentation
followed by a defense.
In general, the master project should be completed within 6 months from the start. An
extension to 9 months is possible. In exceptional cases, and only if it is clear that the project
can be finished, the exam committee may allow for an additional 3 months period. It is
important to note that the project must be finished within 1 year. A project not finished
within 1 year is automatically cancelled and graded as fail. You then have to start a new
project with a different supervisor. The complete examination regulations can be found on
the website.

7.4.3 Assessment
Your final project is graded by an assessment committee. The committee usually consists
of your supervisor, a staff member from your specialization area, and a staff member from
one of the other research groups of the division of Computer Science of the department of
Mathematics and Computer Science or of the department of Electrical Engineering. The
supervisor is responsible for forming this committee at least one month before graduation.
The assessment committee takes the following criteria into account:
Results: Significance of the results versus difficulty of the problem or project
goals.
Report: Structure, completeness, correctness, readability, argumentation.
Graduation presentation: Structure, contents, clarity, contact with audience.
Defense: Argumentation, demonstration of knowledge, competency in dis-
cerning main aspects from details of the project.
Execution of the project: Level of independence, planning, organization, handling dead-
lines and setbacks, level of own contribution.
The actual assessment form used by the committee has a more fine-grained list of criteria
for evaluating the work. Not all criteria are equally important. The assessment committee
decides the relative importance of each criterion to arrive at a final grade. The motivation
for the grade is documented in an assessment report, see the MSc assessment form.

7.4.4 Checklist
The graduation checklist (Appendix B) summarizes all the steps required from having your
study program approved to the graduation ceremony.