On improving dependability of analog and mixed-signal SoCs: A system-level approach

M.A. Khan

    Research output: ThesisPhD Thesis - Research UT, graduation UT

    306 Downloads (Pure)


    Dependability of electronic systems, being an indispensable part of our civilian, industrial and military applications, has become increasingly important as a result of continuous technology scaling. The dependability or human reliance on these electronic systems has decreased as a result of new technologies which are far less mature as compared to older technologies. The electrical characteristics of the transistors and the wires will vary statistically in a spatial and temporal manner, directly translating into design uncertainty during fabrication and even during operational life. This combined impact of manufacturing uncertainty (e.g. process variability) and temporal degradation (aging) results in time-dependent variability and hence the means to impact the functionality and dependability of electronic systems. Unfortunately, traditional worst-case design slacks or margins are not sufficient anymore to capture the time-dependent system variability, especially in new technology nodes, and would result in over-pessimistic implementations with significant penalties in terms of area/delay/energy. As a result, time-dependent uncertainties become a great threat to the design of complex systems-on-chip (SoC) implementations and their dependability. It becomes extremely important in case of safety- or mission-critical systems because the dependability failure of these systems may result in enormous cost damage or even loss of human lives. Therefore, maintaining or achieving high system dependability in safety- or mission-critical systems is the most important property. Analog and mixed-signal front/back ends, being an important part of most critical systems especially in safety-critical (e.g. automotive, medical etc.) and mission-critical (e.g. military, space etc.) systems have received relatively little attention with regard to dependability. The dependability of these analog and mixed-signal front/back ends is essential in order to have a dependable interface between the real world and digital world. This is the main goal of the current research where new system-level strategies have been explored and investigated in order to enhance the dependability of analog and mixed-signal front ends especially during their operational life.
    Original languageEnglish
    QualificationDoctor of Philosophy
    Awarding Institution
    • University of Twente
    • Smit, Gerardus Johannes Maria, Supervisor
    • Kerkhoff, Hans G., Advisor
    Thesis sponsors
    Award date7 Nov 2014
    Place of PublicationEnschede
    Print ISBNs978-90-365-3777-3
    Publication statusPublished - 7 Nov 2014


    • METIS-306179
    • IR-92381
    • Analog and mixed-signal dependability
    • Markov analysis
    • Degradation modelling
    • Dependable systems
    • Sensitivity analysis
    • offset voltage
    • design space exploration for dependable system
    • behavioral modeling
    • input signal monitoring
    • charge-redistribution SAR ADC
    • temperature monitoring
    • lifetime prediction
    • self-diagnosis
    • dependability characterization
    • dependability recovery scheme
    • dependable IPs
    • dependable design
    • EWI-25253
    • CAES-TDT: Testable Design and Test
    • self-calibration hardware redundancy
    • time befor failure


    Dive into the research topics of 'On improving dependability of analog and mixed-signal SoCs: A system-level approach'. Together they form a unique fingerprint.

    Cite this