Theodore Laopoulos

Energy Consumption Estimation in Embedded Systems

This paper presents an energy consumption modeling technique for embedded systems based on a microcontroller. The software tasks that run on the embedded system are profiled, and their characteristics are analyzed. The type of executed assembly instructions, as well as the number of accesses to the memory and the analog-to-digital converter, is the required information for the derivation of the proposed model. An appropriate instrumentation setup has been developed for measuring and modeling the energy consumption in the corresponding digital circuits.

Measurement of current variations for the estimation of software-related power consumption [embedded processing circuits]

A current measurement configuration for the estimation of the power consumption of processing systems is presented in this work. The problem addressed is to measure the variations of the power supply current of digital circuits (and especially of embedded processing circuits) and to calculate from these measurements the energy consumption variations associated with certain tasks performed by the system software. Accurate monitoring of the instantaneous variations of the power supply current provides the appropriate information for the estimation of the power consumption at different operating situations of the processor (core) and of the overall processing system, as well (consumption of peripheral units). The proposed instantaneous current measuring approach, along with the execution of special test programs for analysis of inter-instruction effects, is expected to provide clear information of the power behavior of single-chip processing systems for low-power applications.

Measurements analysis of the software-related power consumption in microprocessors

In this paper the measurements taken for the development of instruction-level energy models for microprocessors are presented and analyzed. An appropriate measuring environment and a suitable measuring methodology were developed for taking the necessary measurements. The energy of an instruction is defined as a sum of three components. The pure base energy cost, the inter-instruction cost and the effect of the energy sensitive factors (instruction parameters). These components are characterized for each instruction of the ARM7TDMI embedded processor and their values are analyzed. Using the resulted models estimates of the energy consumption of real software kernels with only up to 5% error was determined.

Instruction level energy modeling for pipelined processors

A new method for creating instruction level energy models for pipelined processors is introduced. This method is based on measuring the instantaneous current drawn by the processor during the execution of the instructions. An appropriate instrumentation set up was established for this purpose. According to the proposed method the energy costs (base and inter-instruction costs) are modeled in relation to a reference instruction (e.g. NOP). These costs incorporate inter-cycle energy components, which cancel each other when they are summed to produce the energy consumption of a program resulting in estimates with high accuracy. This is confirmed by the results. Also the dependencies of the energy consumption on the instruction parameters (e.g. operands, addresses) are studied and modeled in an efficient way.

Measurement of Power Consumption in Digital Systems

A study on measurement configurations for the estimation of power consumption of processing systems is presented in this work. The problem addressed is to measure the energy that a digital system consumes for a numerous number of clock cycles and to assign this consumption to a specific group of instructions. This kind of energy measurement provides the information for determining the energy consumption of specific software routines that run on digital processing systems. The proposed configuration may be integrated in the same chip as a peripheral unit, which will monitor the energy consumption during the different tasks of operation. It is expected to provide clear information of the power behavior of single-chip processing systems for low-power applications, as well as for built-in self test purposes

Instrumentation Set-up for Instruction Level Power Modeling

Energy constraints form an important part of the design specification for processors running embedded applications. For estimating energy dissipation early at the design cycle, accurate power consumption models characterized for the processor are essential. A methodology and the corresponding instrumentation setup for taking current measurements to create high quality instruction level power models, are discussed in this paper. The instantaneous current drawn by the processor is monitored at each clock cycle. A high performance instrumentation setup has been established for the accurate measurement of the processor current, which is based on a current sensing circuit, instead of the conventional solution of a series resistor.

A Camera Based Method for the Measurement of Motion Parameters of IPMC Actuators

A visual measurement technique that is specifically developed to monitor the movements of ionic polymer metal composite (IPMC) materials in underwater conditions is presented. The proposed method is based on the extraction of appropriate information about the movement of an IPMC strip from subsequent picture frames. Since the time difference between these pictures is known, the system calculates both the displacement and the speed of the moving part. The processing algorithm implements a fully automated procedure for the localization of the point of interest (i.e., the edge of a microrobotic wing) and calculates the distance (displacement) between subsequent frames. This method is particularly useful for rather large and slow movements of the IPMC strips, as found in underwater microrobotic applications. The experimental results are compared with concurrent reference measurements by a high-accuracy laser-positioning system. A setup based on a low-cost camera offers a resolution of 0.5 mm or better, which is quite good for the specific application.

Improving sensitivity and selectivity of SnO/sub 2/ gas sensors by temperature variation

A microcontroller-based gas-sensing system is presented in this paper. The analysis presented here exploits the differences in the steady-state performance of SnO/sub 2/ gas sensors at different operating temperatures and the potential use of such differences for improving their selectivity and sensitivity. Sets of experimental measurements of sensitivity versus temperature are used for the detailed presentation of the proposed approach. The results indicate that selective identification and rather accurate measurement of a mixture of CH/sub 4/ and CO gases are quite possible. Finally, a microcontroller-based configuration is presented as a working example of a small-size implementation, which may offer measurements of improved sensitivity and selectivity along with high accuracy and reliability.

Switching from computer to microcomputer architecture education

In the last decades, the technological and scientific evolution of the computing discipline has been widely affecting research in software engineering education, which nowadays advocates more enlightened and liberal ideas. This article reviews cross-disciplinary research on a computer architecture class in consideration of its switching to microcomputer architecture. The authors present their strategies towards a successful crossing of boundaries between engineering disciplines. This communication aims at providing a different aspect on professional courses that are, nowadays, addressed at the expense of traditional courses.

An improved tracking technique for visual measurements of ionic polymer–metal composites (IPMC) actuators using Compute Unified Device Architecture (CUDA)

The implementation of a real-time measurement system based on visual measurements of displacement of an actuator–cantilever is presented in this paper. This work is aimed at accelerating image processing for the fast tracking of small actuators based on ionic polymer–metal composites using the graphics processing unit (GPU) approach. The proposed processing techniques for point tracking are based on the analysis of subsequent images of the moving item. The area-segmentation approach is used which combines region prediction, successive scanning, edge filtering and match processing. The overall implementation uses the CPU and GPU, while the results achieved indicate that the computation process speeds up by more than 40×. This is a quite useful improvement especially for real-time measurement and control applications of closed-loop systems based on IPMC materials.