Power profiling and monitoring in embedded systems: A comparative study and a novel methodology based on NARX neural networks

Power consumption in electronic systems is an essential feature for the management of energy autonomy, performance analysis, and the aging monitoring of components. Thus, several research studies have been devoted to the development of power models and profilers for embedded systems. Each of these models is designed to fit a specific usage context. This paper is a part of a series of works dedicated to modeling and monitoring embedded systems in airborne equipment. The objective of this paper is twofold. Firstly, it presents an overview of the most used models in the literature. Then, it offers a comparative analysis of these models according to a set of criteria, such as the modeling assumptions, the necessary instrumentation necessary, the accuracy, and the complexity of implementation. Secondly, we introduce a new power estimator for ARM-Based embedded systems, with component-level granularity. The estimator is based on NARX neural networks and used to monitor power for diagnosis purposes. The obtained experimental results highlight the advantages and limitations of the models presented in the literature and demonstrate the effectiveness of the proposed NARX, having obtained the best results in its class for a smartphone (An online Mean Absolute Percentage Error = 2.2%).

[1]  Tohru Ishihara,et al.  A Run-Time Power Analysis Method using OS-Observable Parameters for Mobile Terminals , 2010 .

[2]  Juan E. Tapiador,et al.  Power-aware anomaly detection in smartphones: An analysis of on-platform versus externalized operation , 2015, Pervasive Mob. Comput..

[3]  Sunghyun Choi,et al.  BattTracker: Enabling energy awareness for smartphone using Li-ion battery characteristics , 2016, IEEE INFOCOM 2016 - The 35th Annual IEEE International Conference on Computer Communications.

[4]  Linwei Niu,et al.  Reliability-aware scheduling for reducing system-wide energy consumption for weakly hard real-time systems , 2017, J. Syst. Archit..

[5]  Hiroshi Inamura,et al.  A model-based energy profiler using online logging for Android applications , 2014, 2014 Seventh International Conference on Mobile Computing and Ubiquitous Networking (ICMU).

[6]  Shivakant Mishra,et al.  Optimizing power consumption in multicore smartphones , 2016, J. Parallel Distributed Comput..

[7]  Kun Wang,et al.  Kalman Predictor-Based Proactive Dynamic Thermal Management for 3-D NoC Systems With Noisy Thermal Sensors , 2017, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems.

[8]  Yunxin Liu,et al.  Towards Accurate GPU Power Modeling for Smartphones , 2015, MobiGames@MobiSys.

[9]  Kathryn S. McKinley,et al.  The model is not enough: Understanding energy consumption in mobile devices , 2012, 2012 IEEE Hot Chips 24 Symposium (HCS).

[10]  Nacer K. M'Sirdi,et al.  Data-Driven Approach for Feature Drift Detection in Embedded Electronic Devices , 2018 .

[11]  Klaus David,et al.  Energy consumption of the sensors of Smartphones , 2013, ISWCS.

[12]  Minyong Kim,et al.  Enhancing online power estimation accuracy for smartphones , 2012, IEEE Transactions on Consumer Electronics.

[13]  Minyong Kim,et al.  A Novel GPU Power Model for Accurate Smartphone Power Breakdown , 2015 .

[14]  Sameer Alawnah,et al.  Modeling of smartphones’ power using neural networks , 2017, EURASIP J. Embed. Syst..

[15]  Yiran Chen,et al.  Demystifying energy usage in smartphones , 2014, 2014 51st ACM/EDAC/IEEE Design Automation Conference (DAC).

[16]  Naehyuck Chang,et al.  FEPMA: Fine-grained event-driven power meter for android smartphones based on device driver layer event monitoring , 2014, 2014 Design, Automation & Test in Europe Conference & Exhibition (DATE).

[17]  Hyungshin Kim,et al.  Smart phone power model generation using use pattern analysis , 2012, 2012 IEEE International Conference on Consumer Electronics (ICCE).

[18]  Axel Jantsch,et al.  Accuracy-Aware Power Management for Many-Core Systems Running Error-Resilient Applications , 2017, IEEE Transactions on Very Large Scale Integration (VLSI) Systems.

[19]  Sagar Naik,et al.  A Computing Profiling Procedure for Mobile Developers to Estimate Energy Cost , 2015, MSWiM.

[20]  Pierre-André Cornillon,et al.  Régressión :: théorie et applications , 2007 .

[21]  Nikil D. Dutt,et al.  Quality-aware mobile graphics workload characterization for energy-efficient DVFS design , 2014, 2014 IEEE 12th Symposium on Embedded Systems for Real-time Multimedia (ESTIMedia).

[22]  Timo Hämäläinen,et al.  MARTE profile extension for modeling dynamic power management of embedded systems , 2012, J. Syst. Archit..

[23]  Chandra Krintz,et al.  A run-time, feedback-based energy estimation model For embedded devices , 2006, Proceedings of the 4th International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS '06).

[24]  Abel G. Silva-Filho,et al.  On the use of nonlinear methods for low-power CPU frequency prediction based on Android context variables , 2016, 2016 IEEE 15th International Symposium on Network Computing and Applications (NCA).

[25]  Alessio Merlo,et al.  Measuring and estimating power consumption in Android to support energy-based intrusion detection , 2015, J. Comput. Secur..

[26]  Aziz Naamane,et al.  Constructing an Accurate and a High-Performance Power Profiler for Embedded Systems and Smartphones , 2018, MSWiM.

[27]  Ranveer Chandra,et al.  Empowering developers to estimate app energy consumption , 2012, Mobicom '12.

[28]  Rosarium Pila,et al.  Utilization-based power consumption profiling in smartphones , 2016, 2016 2nd International Conference on Contemporary Computing and Informatics (IC3I).

[29]  Markus Wagner,et al.  Deep parameter optimisation on Android smartphones for energy minimisation: a tale of woe and a proof-of-concept , 2017, GECCO.

[30]  F. Massey.  The Kolmogorov-Smirnov Test for Goodness of Fit , 1951 .

[31]  Aziz Naamane,et al.  Modular Modelling of an Embedded Mobile CPU-GPU Chip for Feature Estimation , 2017, ICINCO.

[32]  Young Geun Kim,et al.  Stabilizing CPU Frequency and Voltage for Temperature-Aware DVFS in Mobile Devices , 2015, IEEE Transactions on Computers.

[33]  Hojung Cha,et al.  DevScope: a nonintrusive and online power analysis tool for smartphone hardware components , 2012, CODES+ISSS.

[34]  Abram Hindle,et al.  GreenMiner: a hardware based mining software repositories software energy consumption framework , 2014, MSR 2014.

[35]  Matti Siekkinen,et al.  Modeling, Profiling, and Debugging the Energy Consumption of Mobile Devices , 2015, ACM Comput. Surv..

[36]  Joan Daniel Prades,et al.  The Power of Models: Modeling Power Consumption for IoT Devices , 2015, IEEE Sensors Journal.

[37]  GaniAbdullah,et al.  A Review on mobile application energy profiling , 2015 .

[38]  R. Engle.  Autoregressive conditional heteroscedasticity with estimates of the variance of United Kingdom inflation , 1982 .

[39]  Hojung Cha,et al.  Accurate Prediction of Available Battery Time for Mobile Applications , 2016, ACM Trans. Embed. Comput. Syst..

[40]  Markus Wagner,et al.  Optimising Energy Consumption Heuristically on Android Mobile Phones , 2016, GECCO.

[41]  Naehyuck Chang,et al.  Online estimation of the remaining energy capacity in mobile systems considering system-wide power consumption and battery characteristics , 2013, 2013 18th Asia and South Pacific Design Automation Conference (ASP-DAC).

[42]  James Won-Ki Hong,et al.  FSM-based Wi-Fi power estimation method for smart devices , 2015, 2015 IFIP/IEEE International Symposium on Integrated Network Management (IM).

[43]  Hojung Cha,et al.  AppScope: Application Energy Metering Framework for Android Smartphone Using Kernel Activity Monitoring , 2012, USENIX Annual Technical Conference.

[44]  Lei Yang,et al.  Accurate online power estimation and automatic battery behavior based power model generation for smartphones , 2010, 2010 IEEE/ACM/IFIP International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS).

[45]  Abram Hindle,et al.  Deep Green: Modelling Time-Series of Software Energy Consumption , 2017, 2017 IEEE International Conference on Software Maintenance and Evolution (ICSME).

[46]  Jean-Charles Grégoire,et al.  Modelling and improving the battery performance of a mobile phone application: A methodology , 2015, 5th International Conference on Energy Aware Computing Systems & Applications.

[47]  Geoff V. Merrett,et al.  Accurate and Stable Run-Time Power Modeling for Mobile and Embedded CPUs , 2017, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems.

[48]  Marco D. Santambrogio,et al.  Adaptive and Flexible Smartphone Power Modeling , 2013, Mob. Networks Appl..

[49]  XianPing Tao,et al.  Improing Screen Power Usage Model on Android Smartphones , 2015, 2015 Asia-Pacific Software Engineering Conference (APSEC).

[50]  Feng Xia,et al.  A Review on mobile application energy profiling: Taxonomy, state-of-the-art, and open research issues , 2015, J. Netw. Comput. Appl..

[51]  Massoud Pedram,et al.  ThermTap: An online power analyzer and thermal simulator for Android devices , 2015, 2015 IEEE/ACM International Symposium on Low Power Electronics and Design (ISLPED).

[52]  Noor Zaman,et al.  Energy efficient middleware: Design and development for mobile applications , 2017, 2017 19th International Conference on Advanced Communication Technology (ICACT).

[53]  Aleksandar Milenkovic,et al.  An Environment for Automated Measuring of Energy Consumed by Android Mobile Devices , 2016, PECCS.

[54]  Minyong Kim,et al.  Accurate GPU power estimation for mobile device power profiling , 2013, 2013 IEEE International Conference on Consumer Electronics (ICCE).

[55]  Seokjun Lee,et al.  Accurate power modeling of modern mobile application processors , 2017, J. Syst. Archit..

[56]  James Won-Ki Hong,et al.  User-Centric Prediction for Battery Lifetime of Mobile Devices , 2008, APNOMS.

[57]  Aziz Naamane,et al.  A Novel Easy-to-construct Power Model for Embedded and Mobile Systems - Using Recursive Neural Nets to Estimate Power Consumption of ARM-based Embedded Systems and Mobile Devices. , 2018 .

[58]  Sagar Naik,et al.  A framework for detecting energy bugs in smartphones , 2015, 2015 6th International Conference on the Network of the Future (NOF).

[59]  Hyungshin Kim,et al.  Automated power model generation method for smartphones , 2014, IEEE Transactions on Consumer Electronics.

[60]  Peris-LopezPedro,et al.  Power-aware anomaly detection in smartphones , 2015 .

[61]  Li Sun,et al.  Modeling WiFi Active Power/Energy Consumption in Smartphones , 2014, 2014 IEEE 34th International Conference on Distributed Computing Systems.

[62]  Ramesh Govindan,et al.  Estimating mobile application energy consumption using program analysis , 2013, 2013 35th International Conference on Software Engineering (ICSE).

[63]  Yaser Jararweh,et al.  Energy Optimisation for Mobile Device Power Consumption: A Survey and a Unified View of Modelling for a Comprehensive Network Simulation , 2016, Mob. Networks Appl..

[64]  Abram Hindle,et al.  GreenOracle: Estimating Software Energy Consumption with Energy Measurement Corpora , 2016, 2016 IEEE/ACM 13th Working Conference on Mining Software Repositories (MSR).

[65]  Joel J. P. C. Rodrigues,et al.  A case and framework for code analysis-based smartphone application energy estimation , 2017, Int. J. Commun. Syst..

[66]  Ming Zhang,et al.  Where is the energy spent inside my app?: fine grained energy accounting on smartphones with Eprof , 2012, EuroSys '12.

[67]  Yao Guo,et al.  Understanding Application-Battery Interactions on Smartphones: A Large-Scale Empirical Study , 2017, IEEE Access.

[68]  Lin Zhong,et al.  Self-constructive high-rate system energy modeling for battery-powered mobile systems , 2011, MobiSys '11.

[69]  Andrea De Lucia,et al.  Software-based energy profiling of Android apps: Simple, efficient and reliable? , 2017, 2017 IEEE 24th International Conference on Software Analysis, Evolution and Reengineering (SANER).

[70]  Markus Wagner,et al.  In-vivo and offline optimisation of energy use in the presence of small energy signals: A case study on a popular Android library , 2018, MobiQuitous.

[71]  Yuan-Cheng Lai,et al.  Calibrating parameters and formulas for process-level energy consumption profiling in smartphones , 2014, J. Netw. Comput. Appl..

[72]  Luca Ardito,et al.  Profiling Power Consumption on Mobile Devices , 2013 .

[73]  Ding Li,et al.  An Empirical Study of the Energy Consumption of Android Applications , 2014, 2014 IEEE International Conference on Software Maintenance and Evolution.

[74]  Paul Holleis,et al.  A DIY power monitor to compare mobile energy consumption in situ , 2013, MobileHCI '13.

[75]  Hang Xie,et al.  Time series prediction based on NARX neural networks: An advanced approach , 2009, 2009 International Conference on Machine Learning and Cybernetics.

[76]  Jing Huang,et al.  Energy-Efficient Resource Utilization for Heterogeneous Embedded Computing Systems , 2017, IEEE Transactions on Computers.

[77]  Sujata Banerjee,et al.  PowerVisor: a battery virtualization scheme for smartphones , 2012, MCS '12.

[78]  Wojciech Mazurczyk,et al.  Seeing the Unseen: Revealing Mobile Malware Hidden Communications via Energy Consumption and Artificial Intelligence , 2016, IEEE Transactions on Information Forensics and Security.

[79]  Yuan-Cheng Lai,et al.  Semi-online power estimation for smartphone hardware components , 2015, 10th IEEE International Symposium on Industrial Embedded Systems (SIES).

[80]  Andrea De Lucia,et al.  PETrA: A Software-Based Tool for Estimating the Energy Profile of Android Applications , 2017, 2017 IEEE/ACM 39th International Conference on Software Engineering Companion (ICSE-C).

[81]  Gabriele Bavota,et al.  Optimizing energy consumption of GUIs in Android apps: a multi-objective approach , 2015, ESEC/SIGSOFT FSE.

[82]  Paramvir Bahl,et al.  Fine-grained power modeling for smartphones using system call tracing , 2011, EuroSys '11.

[83]  Ombretta Gaggi,et al.  An empirical analysis of energy consumption of cross-platform frameworks for mobile development , 2017, Pervasive Mob. Comput..

[84]  Fengyuan Xu,et al.  V-edge: Fast Self-constructive Power Modeling of Smartphones Based on Battery Voltage Dynamics , 2013, NSDI.

[85]  Minyong Kim,et al.  An online power estimation technique for multi-core smartphones with advanced display components , 2012, 2012 IEEE International Conference on Consumer Electronics (ICCE).

[86]  Jeffrey M. Voas,et al.  Mobile Application and Device Power Usage Measurements , 2012, 2012 IEEE Sixth International Conference on Software Security and Reliability.