Marius Muresan

Deploying Simulink Models into System-On-Chip Structures

Nowadays design of embedded systems for controlling industrial processes requires higher and higher integration of analog and digital features. In this way, the system-on-chip approach for designing such embedded systems seems to be the most competitive solution in terms of cost-effectiveness and performance (Chindris, 2002 and Pop and Chindris, 2003). The classical solution still holds some of the advantages: most of the software tooling required for designing an embedded controller will consist in a control algorithm simulator (as Matlab Simulink) and code generation software for deploying the algorithm into the embedded controller. For that solution, code generation templates and S-function compilers are widely spread and cover a large family of microcontrollers and microprocessor. The paper will present a new method for using control algorithm simulators along with code generation software in order to deploy applications inside programmable system-on-chip devices. Deployment tests and experimental results will be also covered for a simple PID algorithm as test reference.

Software in the Loop environment reliability for testing embedded code

This paper makes an analysis of the Software in the Loop (SiL) environment reliability for testing embedded system algorithms. The embedded system used during this analysis was developed around a C167CR microcontroller and based on a real-time (RT) operating system (OS). The main function of the control algorithm implemented into the embedded system is to control the speed of a DC motor. The embedded system software was wrapped into a SiL environment where the real-time operating system and the low level drivers where emulated. The obtained results where compared with the ones from Hardware in the Loop (HiL) testing system. Several aspects were followed during analysis: accuracy of the obtained results, proper functioning of the emulated RT OS, the possibility of using the same tools over product life cycle and implementation of automated.

Performances Evaluation of Frequency Measurement Techniques in Automotive Industry

Many automotive sensors have as output a square wave signal, whose frequency is related which measurement parameters. Such sensors are use for measuring the shafts rotation and wheels rotation. The measurement systems are developed basically based on a microcontroller, which timers capture feature is used for frequency measurement. The frequency measurement principle is the same for every microcontroller; the differences consist in peripherals settings. Three frequency measurement techniques were studied; the experimental results were made using a sixteen bits microcontroller - C167 from Infineon. Knowing the performances of each frequency measurement techniques is a step forward to reduce the design costs by choosing the appropriate technique according with design specification.

PSoC dual sensor for pressure and temperature using bipolar junction effects

This paper presents theoretical and experimental aspects of the methodology for embedded sensor design with PSoC controllers for measuring the pressure and temperature using two NPN transistors, enlightening the benefits of the method: a PSoC design reduces the time-to-market, increase flexibility, lower part count, provides in-system performance improvement, design security, and field upgrades. Also, another benefit of a PSoC controller is the dynamical re-configuration. The effect of pressure over an emitter-base junction makes a modification of the saturation current, that means a translation of the input characteristic of the emitter-base junction to higher values. The method consists in measuring the temperature and pressure using two transistors and compensates the dependencies between them. This is done by using the same PSoCs pins and an internal dynamic reconfiguration: one configuration for measuring pressure and one configuration for measuring temperature

Simulating embedded targets for efficient code implementation

Nowadays design of embedded systems for controlling processes requires high complexity software modules. In this way, the automatic code generation approach for designing such embedded systems seems to be the most competitive solution in terms of cost-effectiveness and performance

OrCAD Link Tool for Automated Optical Inspection of Assembled Boards

Automated optical inspection (AOI) of assembled PCBs is one of the most widely spread non-destructive testing method used in mass-production of electronic systems. Fast video processing algorithms and precise mechanical systems are supporting this cost-effective method for investigating electronics assemblies. However, for small production quantities, or for often changes in production line, the AOI bottleneck resides in the effort needed for adapting the pattern recognition files and component locations for each new board. The paper will present an effective method for importing assembly and technology files directly from the CAD environment and generating proper configuration files for the AOI system. The proposed linking tool will not only have the advantage of reducing time-to-test parameter, but will also increase the accuracy of the AOI process by providing test setups directly from CAD files of the device-under-test board (Prasad, 2002).

Modeling bipolar junction effects for high entropy generators

The analog noise produced by semiconductor junctions is considered to be the fundamentally unpredictable sum of astronomical numbers of independent quantum actions. In this way, the unpredictability part of this makes noise attractive for cryptography. On the other hand, real noise often exhibits measurable and predictable correlations which contradict the simple model. Modeling bipolar junction effects can provide some quantitative and qualitative information on the entropy generated by such devices. The paper proposes a flowchart for modeling noise related bipolar junction effects along with an algorithm for building noise models for semiconductor components.

Simulating control algorithms from embedded systems for generation of optimized code

Modern design of embedded systems is based on a microcontroller whos running a real time operating system. The big numbers of tasks that must be performed without affecting the system performances imply a lot of software analysis. A critical part in embedded system software is the part related to control algorithms: it is important to have a correct command in a critical amount of time. The simple way for checking the control algorithms time constrain is to simulate on real-time platforms together with plant models. Simulating embedded systems control software can show us the weak part of the software, the part where code can be simplify in order to improve the execution time and reduce the product time to market. A powerful tool providing us very good result is xPC-Target from MATLAB/Simulink.

Bipolar Junction Effects for High Entropy Data Harvesters

Harvesting high-entropy data from hardware generators is usually a difficult task and most of the applications will take data that is believed to contain enough entropy and use it to ldquoseedrdquo a cryptographic algorithm. The paper proposes an evaluation of such system on chip device designed to harvest high-entropy data from thermal phenomena (thermal noise and avalanche noise) and to analyze ldquowhiteningrdquo algorithms for reducing bias and correlation. Taking into account the features of system on chip devices: complete user management of analog and digital embedded resources, the possibility of creating custom peripherals and the high level of re-usability not only for the software modules but also for embedded hardware peripherals, the new design addresses the main challenges for building true random number generators and proposes a design path and a testing procedure for high-entropy generators.

Calibration parameters principles for MATLAB S-functions using CANApe

In the development of embedded software one of the harder tasks are testing and optimize the ECUs. The basic way to test and optimize is by measuring and calibrating the process. One of the common methods to solve this problem is dasiaHardware-in-the-Looppsila(HIL). With this method, the real ECU is place in a simulative environment for the software under test. The software runs on the real target hardware. The basic idea with dasiaSoftware-in-the-Looppsila(SIL) is the same, except that all runs on standard PC hardware. [1] Both for HIL and SIL many simulated models for automotive industry are made in MATLAB/Simulink. A way how to put ECUs software into MATLAB/Simulink in order to have a complete testing environment will reduce considerably the testing costs and time to market. To have a unitary view both for HIL and SIL we must use the same approach. To keep the consistency we should use the same measurement and application tool, in our case CANApe. This will be done using XCP protocol based on Ethernet TCP/IP transport layer. [2] The entire ECUs software will be wrapped into a MATLAB/Simulink S-Function. The obtained S-Function will be a part of the entire model, with the possibility of having access to the inside variables of the S-Function.