Embedded Information Systems

A tantárgy neve magyarul / Name of the subject in Hungarian: Beágyazott információs rendszerek

Last updated: 2018. április 27.

Budapest University of Technology and Economics
Faculty of Electrical Engineering and Informatics
Software Engineering, BSc
Course ID Semester Assessment Credit Tantárgyfélév
VIMIAD00 7 2/1/0/f 3  
3. Course coordinator and department Dr. Péceli Gábor,
Web page of the course http://www.mit.bme.hu/eng/oktatas/targyak/vimiad00
4. Instructors Gábor Péceli, professor, Department of Measurement and Information Systems (MIT)
5. Required knowledge Operating systems, Software technology, Computer networks, Artificial intelligence, Data bases.
6. Pre-requisites
NEM ( TárgyEredmény( "BMEVIMIA359" , "jegy" , _ ) >= 2
TárgyEredmény("BMEVIMIA359", "FELVETEL", AktualisFelev()) > 0)

ÉS (Training.Code=("5N-A8") VAGY Training.Code=("5NAA8"))

A fenti forma a Neptun sajátja, ezen technikai okokból nem változtattunk.

A kötelező előtanulmányi rendek grafikus formában itt láthatók.

7. Objectives, learning outcomes and obtained knowledge

Embedded systems can be characterized as computer-based systems performing intensive information exchange with their physical, biological, chemical environment.

This subject on one hand presents the information technology related aspects of these systems, while on the other improves knowledge and capabilities of the students to be able to create such systems via exploring the details of practical applications. Further objective is the introduction of design principles related to real-time autonomous and dependable systems.

The students completing this course will be familiar with the properties of embedded systems, and will make acquaintance with the basic methods of real-time program execution and task synchronization, with the basics of communication in distributed systems, and some of their dependability issues.

This set of knowledge and capabilities enable creative contribution to problems related to the integration of computers and their networks with physical systems, typical in ubiquitous computing, in internet of things, cyber-physical systems or ambient assisted living technologies.

8. Synopsis

1. week (2 hours lecture, 1 hour seminar)

Lecture: 1. Introduction: Recent and future embedded systems. Embedding environments-embedded devices. Embedded system functions. Embedded software as universal system integrator. Cooperation of embedded systems: systems of systems. Embedded devices and the internet. Trends and terms. European initiatives. Challenges.

Seminar: Examples of timing specialties in embedded systems. The relativistic effect. Characterization of HRT-SRT systems.

2. week (2 hours lecture, 1 hour seminar)

Lecture: Introduction cont.: The significance of agreement protocols. Response time requirements. 2. Scheduling in embedded systems: Cyclic, time-sharing, priority-based. The Deadline Monotonic Analysis (DMA) method.

Seminar: DMA of CAN bus communication: calculation of worst-case response time.

3. week (2 hours lecture, 1 hour seminar)

Lecture: Schedulability, schedulability tests. Rate Monotonic (RM), Earliest Deadline First (EDF) schedules, proof of EDF schedulability. Scheduling of non-independent tasks. Priority inheritance. Priority ceilings protocols  (PCP, IPCP).

Seminar: Application of PCP and IPCP protocols.

4. week (2 hours lecture, 1 hour seminar)

Lecture: Processor demand method. Scheduling if deadline is smaller than period. Fault-tolerant scheduling. Simultaneous scheduling periodic and aperiodic tasks.

Seminar: Background scheduling. Fix priority servers: Polling Server, Deferrable Server, Priority Exchange Server and Sporadic Server. Slack Stealing. Dual priority Scheduling.

5. week (2 hours lecture, 1 hour seminar)

Lecture: 3. Time dependency of memory management: Memory management in multi-tasking systems. Time dependencies of resource handling. 4. Measurement of time, time as a service, synchronization of clocks: The concepts of digital time measurement. Measurement of short time duration.

Seminar: Accuracy of time measurement.

6. week (2 hours lecture, 1 hour seminar)

Lecture: Clocks as given accuracy sources of the real time. The reference clock, correct clock, accurate clock. Clock drift, clock offset. Precision, accuracy. Measuring time interval in distributed systems. Time dependency of make. Types of clock systems. Standards of time. Berkeley algorithm. Cristian algorithm. Master-slave algorithms.

Seminar: Distributed clock synchronization algorithms. The jitter of synchronization message. Fault tolerant averaging algorithm.  

7. week (2 hours lecture, 1 hour seminar)

Lectures: 5. Real-time entities and images: RT entities. RT images. Observations: state observation, event observation, indirect observation. RT objects. Temporal accuracy. Permanence. Action delay. Idempotency.

Seminar: Modelling the embedding environment. The measurement process and the observer. Convergent processes.

8. week (2 hours lecture, 1 hour seminar)

Lecture: Modelling the embedding environment, absorbing the model into the program. The basics of model fitting. Sampling, polling and interrupt. Replica determinism.

Seminar: Regression schemes, parameter fitting, parameter adaptation. Representing clocks via regression.

9. week (2 hours lecture, 1 hour seminar)

Lecture: 6. Real-time communication: Time dependencies of communications: flow control in time. PAR protocols. The Time Triggered Architecture (TTA). Main features of the Time Triggered Protocols. Performance limits. Fundamental conflicts of protocol design.

Seminar: Time conditions of the physical layer. Synchronization capabilities of the coding at the physical level. Time synchronization in wireless networks.

10. week (2 hours lecture, 1 hour seminar)

Lecture: 7. Embedded operating system: RT kernels. RT extensions of standard operating systems. RT Linux. Embedded virtualization. Microkernel technology.

Seminar: Application of embedded operating system services.

11. week (2 hours lecture, 1 hour seminar)

Lecture: 8. Sensor networks: Structures and features of sensor networks. The TinyOS development environment. Communication in sensor networks. CSMA problems. Rooting.

Seminar: Power and energy consumption in embedded systems. Power consumption of CMOS processors. Dynamic Voltage Scaling. Dynamic Power Management..

12. week (2 hours lecture, 1 hour seminar)

Lecture: 9. Efficient implementation: Non-conventional methods of modelling and control in embedded systems: basics qualitative and fuzzy modelling. Hybrid systems.

Seminar: Simple hybrid systems: thermostat with timing. Automated Guided Vehicle.

13. week (2 hours lecture, 1 hour seminar)

Lecture: 10. Safety critical systems: Safety requirements. Reliability measures. Application of redundancies. Handling hardware errors. Handling software errors.

Gyakorlat: Reliability block diagram. Methods of calculation system reliability.

1. week (2 hours lecture, 1 hour seminar)

Spare for holidays.

9. Method of instruction

2 hours lecture + 1 hour seminar/week

10. Assessment

During the semester: Two mid-term exams must be written with satisfactory result (40%), and one home-work of appropriate quality. The minimum requirement is 40% of the total number of points assigned to the home-work. The final mark is computed from the average of the results of the mid-term exams (1/3-1/3) and the home-work (1/3):

  • between 40% - 52.5: pass (2)
  • between 52.5%- 65%: satisfactory(3)
  • between 65%- 80%: good (4)
  • from 80%: excellent (5)
11. Recaps The mid-term exams can be repeated on an organized repeated mid-term occasion. The home work must be submitted before the specified deadline.  Delayed submission is possible till the last working day of the replacement week.

12. Consultations Consultations are organized upon the explicit request of students.

13. References, textbooks and resources

[1]  Péceli G., Embedded information systems. Lecture notes.

[2]  E. A. Lee and S. A. Seshia, Introduction to Embedded Systems - A Cyber-Physical Systems Approach, Second Edition, LeeSeshia.org, 2015.

[3] H. Kopetz: Real-Time Systems, Design Principles for Embedded Applications, Kluwer Academic Publishers, 1997.

[4]  Giorgio Butazzo, Hard Real-Time Computing Systems: Predictable Scheduling Algorithms and Applications, Kluwer, 1997.

[5] Jane W.S. Lu, Real-Time Systems, Prentice-Hall, 2000.

14. Required learning hours and assignment
Studying before lectures 14
Studying before seminars7
Studying before labs
Studying before mid-term exams
 Studying additional topics
15. Syllabus prepared by Gábor Péceli, professor, Department of Measurement and Information Systems (MIT)