Pub Date : 2018-04-11DOI: 10.1109/ICCPS.2018.00042
Daichi Watari, Ittetsu Taniguchi, T. Onoye
The battery degradation is serious problem for the modern electrical systems. This paper proposes State-of-Health (SoH) aware battery management optimization method on decentralized energy network. The power distribution problem is often solved with mixed integer programming (MIP), and proposed formulation takes into account the SoH model. Our poster shows the details and the effectiveness of proposed method.
{"title":"SoH Aware Battery Management Optimization on Decentralized Energy Network","authors":"Daichi Watari, Ittetsu Taniguchi, T. Onoye","doi":"10.1109/ICCPS.2018.00042","DOIUrl":"https://doi.org/10.1109/ICCPS.2018.00042","url":null,"abstract":"The battery degradation is serious problem for the modern electrical systems. This paper proposes State-of-Health (SoH) aware battery management optimization method on decentralized energy network. The power distribution problem is often solved with mixed integer programming (MIP), and proposed formulation takes into account the SoH model. Our poster shows the details and the effectiveness of proposed method.","PeriodicalId":199062,"journal":{"name":"2018 ACM/IEEE 9th International Conference on Cyber-Physical Systems (ICCPS)","volume":"87 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129005391","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-04-01DOI: 10.1109/iccps.2018.00040
Jiajun Shen, Xueli Fan, Qixin Wang
The aim of this paper is to propose an novel underwater localization scheme considering acoustic refractions. Simulation results shows that our scheme is superior than some other schemes in deep water contexts.
本文的目的是提出一种考虑声折射的水下定位方案。仿真结果表明，该方案在深水环境下优于其他方案。
{"title":"WiP Abstract: Underwater AUV Localization with Refraction Consideration","authors":"Jiajun Shen, Xueli Fan, Qixin Wang","doi":"10.1109/iccps.2018.00040","DOIUrl":"https://doi.org/10.1109/iccps.2018.00040","url":null,"abstract":"The aim of this paper is to propose an novel underwater localization scheme considering acoustic refractions. Simulation results shows that our scheme is superior than some other schemes in deep water contexts.","PeriodicalId":199062,"journal":{"name":"2018 ACM/IEEE 9th International Conference on Cyber-Physical Systems (ICCPS)","volume":"114 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123366364","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-04-01DOI: 10.1109/iccps.2018.00043
Yuchang Won, B. Yu, Jaegeun Park, I. Park, Haegeon Jeong, Jeanseong Baik, Kyungtae Kang, Insup Lee, Kyung-Joon Park, Y. Eun
This paper presents an architecture for cyber-physical systems resilient against external attacks and internal faults. The target CPS consists of multiple physical systems equipped with local embedded systems, a supervision module that oversees operations of the physical systems, and communication network connecting each physical system to the supervision module. We give an instantiation of the architecture on a radio-based train control system to demonstrate the resiliency under various safety critical scenarios. The testbed entitled KRS-DGIST includes a commercial train control and supervision software deployed in the Philippines. The railway installed sensors and other mechanisms are implemented reflecting the actual train control and supervision system, and the dynamics of the trains is computer simulated. Demonstrations are given with attacks on sensors, communication network, and embedded systems.
{"title":"WiP Abstract: KRS-DGIST: A Resilient CPS Testbed for Radio-Based Train Control","authors":"Yuchang Won, B. Yu, Jaegeun Park, I. Park, Haegeon Jeong, Jeanseong Baik, Kyungtae Kang, Insup Lee, Kyung-Joon Park, Y. Eun","doi":"10.1109/iccps.2018.00043","DOIUrl":"https://doi.org/10.1109/iccps.2018.00043","url":null,"abstract":"This paper presents an architecture for cyber-physical systems resilient against external attacks and internal faults. The target CPS consists of multiple physical systems equipped with local embedded systems, a supervision module that oversees operations of the physical systems, and communication network connecting each physical system to the supervision module. We give an instantiation of the architecture on a radio-based train control system to demonstrate the resiliency under various safety critical scenarios. The testbed entitled KRS-DGIST includes a commercial train control and supervision software deployed in the Philippines. The railway installed sensors and other mechanisms are implemented reflecting the actual train control and supervision system, and the dynamics of the trains is computer simulated. Demonstrations are given with attacks on sensors, communication network, and embedded systems.","PeriodicalId":199062,"journal":{"name":"2018 ACM/IEEE 9th International Conference on Cyber-Physical Systems (ICCPS)","volume":"281 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120879870","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-04-01DOI: 10.1109/ICCPS.2018.00018
Thomas Gurriet, Andrew W. Singletary, Jake Reher, L. Ciarletta, E. Feron, A. Ames
This paper presents initial results towards a realizable framework for the safety critical controlled invariance of cyber-physical systems. The main contribution of this paper is the development of a control barrier function based methodology which can be used to enforce set invariance on systems in the presence of non-linear disturbances and uncertainty. The first part of this work is a review of the current methods available for finding viable sets and how they are linked to practical choices regarding safety. Their limitations and directions towards improvements when it comes to handling model uncertainty are also highlighted. The second part of this work is the formulation of a condition which can guarantee set invariance in the presence of generic uncertain in the dynamics. An associated optimization problem to enforce that condition is proposed and a method to convexify the problem and make it solvable in real-time is formally presented. The effectiveness of the proposed framework is illustrated experimentally on a two-wheeled inverted pendulum.
{"title":"Towards a Framework for Realizable Safety Critical Control through Active Set Invariance","authors":"Thomas Gurriet, Andrew W. Singletary, Jake Reher, L. Ciarletta, E. Feron, A. Ames","doi":"10.1109/ICCPS.2018.00018","DOIUrl":"https://doi.org/10.1109/ICCPS.2018.00018","url":null,"abstract":"This paper presents initial results towards a realizable framework for the safety critical controlled invariance of cyber-physical systems. The main contribution of this paper is the development of a control barrier function based methodology which can be used to enforce set invariance on systems in the presence of non-linear disturbances and uncertainty. The first part of this work is a review of the current methods available for finding viable sets and how they are linked to practical choices regarding safety. Their limitations and directions towards improvements when it comes to handling model uncertainty are also highlighted. The second part of this work is the formulation of a condition which can guarantee set invariance in the presence of generic uncertain in the dynamics. An associated optimization problem to enforce that condition is proposed and a method to convexify the problem and make it solvable in real-time is formally presented. The effectiveness of the proposed framework is illustrated experimentally on a two-wheeled inverted pendulum.","PeriodicalId":199062,"journal":{"name":"2018 ACM/IEEE 9th International Conference on Cyber-Physical Systems (ICCPS)","volume":"97 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123020360","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-04-01DOI: 10.1109/iccps.2018.00022
Achin Jain, Truong X. Nghiem, M. Morari, R. Mangharam
Building physics-based models of complex physical systems like buildings and chemical plants is extremely cost and time prohibitive for applications such as real-time optimal control, production planning and supply chain logistics. Machine learning algorithms can reduce this cost and time complexity, and are, consequently, more scalable for large-scale physical systems. However, there are many practical challenges that must be addressed before employing machine learning for closed-loop control. This paper proposes the use of Gaussian Processes (GP) for learning control-oriented models: (1) We develop methods for the optimal experiment design (OED) of functional tests to learn models of a physical system, subject to stringent operational constraints and limited availability of the system. Using a Bayesian approach with GP, our methods seek to select the most informative data for optimally updating an existing model. (2) We also show that black-box GP models can be used for receding horizon optimal control with probabilistic guarantees on constraint satisfaction through chance constraints. (3) We further propose an online method for continuously improving the GP model in closed-loop with a real-time controller. Our methods are demonstrated and validated in a case study of building energy control and Demand Response.
{"title":"Learning and Control Using Gaussian Processes","authors":"Achin Jain, Truong X. Nghiem, M. Morari, R. Mangharam","doi":"10.1109/iccps.2018.00022","DOIUrl":"https://doi.org/10.1109/iccps.2018.00022","url":null,"abstract":"Building physics-based models of complex physical systems like buildings and chemical plants is extremely cost and time prohibitive for applications such as real-time optimal control, production planning and supply chain logistics. Machine learning algorithms can reduce this cost and time complexity, and are, consequently, more scalable for large-scale physical systems. However, there are many practical challenges that must be addressed before employing machine learning for closed-loop control. This paper proposes the use of Gaussian Processes (GP) for learning control-oriented models: (1) We develop methods for the optimal experiment design (OED) of functional tests to learn models of a physical system, subject to stringent operational constraints and limited availability of the system. Using a Bayesian approach with GP, our methods seek to select the most informative data for optimally updating an existing model. (2) We also show that black-box GP models can be used for receding horizon optimal control with probabilistic guarantees on constraint satisfaction through chance constraints. (3) We further propose an online method for continuously improving the GP model in closed-loop with a real-time controller. Our methods are demonstrated and validated in a case study of building energy control and Demand Response.","PeriodicalId":199062,"journal":{"name":"2018 ACM/IEEE 9th International Conference on Cyber-Physical Systems (ICCPS)","volume":"126 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124340116","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The encrypted controller is a novel and promising approach for enhancing the security of industrial control systems [1]. This approach encrypts both the signals that transmitted over communication channels and the control law, so as to protect all information of a plant and its controller from attackers. However, in spite of considerable research efforts [2]–[12], the encrypted controller is still quite far from its implementation and application in industrial control systems. In-depth experimental studies are necessary for bridging the gap between theory and application of the encrypted controller. In this demonstration, we present an industrial control system testbed for experimental studies of the encrypted controller. Moreover, we carry out an experimental study that improves and evaluates the security of the industrial control system which comprises the encrypted controller using the testbed.
{"title":"Demo Abstract: An Industrial Control System Testbed for the Encrypted Controller","authors":"Xing Li, Mengxiang Liu, Rui Zhang, Peng Cheng, Jiming Chen","doi":"10.1109/iccps.2018.00045","DOIUrl":"https://doi.org/10.1109/iccps.2018.00045","url":null,"abstract":"The encrypted controller is a novel and promising approach for enhancing the security of industrial control systems [1]. This approach encrypts both the signals that transmitted over communication channels and the control law, so as to protect all information of a plant and its controller from attackers. However, in spite of considerable research efforts [2]–[12], the encrypted controller is still quite far from its implementation and application in industrial control systems. In-depth experimental studies are necessary for bridging the gap between theory and application of the encrypted controller. In this demonstration, we present an industrial control system testbed for experimental studies of the encrypted controller. Moreover, we carry out an experimental study that improves and evaluates the security of the industrial control system which comprises the encrypted controller using the testbed.","PeriodicalId":199062,"journal":{"name":"2018 ACM/IEEE 9th International Conference on Cyber-Physical Systems (ICCPS)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125638189","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-03-27DOI: 10.1109/ICCPS.2018.00034
Gaurav Gupta, S. Pequito, P. Bogdan
Brain interfaces are cyber-physical systems that aim to harvest information from the (physical) brain through sensing mechanisms, extract information about the underlying processes, and decide/actuate accordingly. Nonetheless, the brain interfaces are still in their infancy, but reaching to their maturity quickly as several initiatives are released to push forward their development (e.g., NeuraLink by Elon Musk and `typing-by-brain' by Facebook). This has motivated us to revisit the design of EEG-based non-invasive brain interfaces. Specifically, current methodologies entail a highly skilled neuro-functional approach and evidence-based a priori knowledge about specific signal features and their interpretation from a neuro-physiological point of view. Hereafter, we propose to demystify such approaches, as we propose to leverage new time-varying complex network models that equip us with a fractal dynamical characterization of the underlying processes. Subsequently, the parameters of the proposed complex network models can be explained from a system's perspective, and, consecutively, used for classification using machine learning algorithms and/or actuation laws determined using control system's theory. Besides, the proposed system identification methods and techniques have computational complexities comparable with those currently used in EEG-based brain interfaces, which enable comparable online performances. Furthermore, we foresee that the proposed models and approaches are also valid using other invasive and non-invasive technologies. Finally, we illustrate and experimentally evaluate this approach on real EEG-datasets to assess and validate the proposed methodology. The classification accuracies are high even on having less number of training samples.
{"title":"Re-Thinking EEG-Based Non-Invasive Brain Interfaces: Modeling and Analysis","authors":"Gaurav Gupta, S. Pequito, P. Bogdan","doi":"10.1109/ICCPS.2018.00034","DOIUrl":"https://doi.org/10.1109/ICCPS.2018.00034","url":null,"abstract":"Brain interfaces are cyber-physical systems that aim to harvest information from the (physical) brain through sensing mechanisms, extract information about the underlying processes, and decide/actuate accordingly. Nonetheless, the brain interfaces are still in their infancy, but reaching to their maturity quickly as several initiatives are released to push forward their development (e.g., NeuraLink by Elon Musk and `typing-by-brain' by Facebook). This has motivated us to revisit the design of EEG-based non-invasive brain interfaces. Specifically, current methodologies entail a highly skilled neuro-functional approach and evidence-based a priori knowledge about specific signal features and their interpretation from a neuro-physiological point of view. Hereafter, we propose to demystify such approaches, as we propose to leverage new time-varying complex network models that equip us with a fractal dynamical characterization of the underlying processes. Subsequently, the parameters of the proposed complex network models can be explained from a system's perspective, and, consecutively, used for classification using machine learning algorithms and/or actuation laws determined using control system's theory. Besides, the proposed system identification methods and techniques have computational complexities comparable with those currently used in EEG-based brain interfaces, which enable comparable online performances. Furthermore, we foresee that the proposed models and approaches are also valid using other invasive and non-invasive technologies. Finally, we illustrate and experimentally evaluate this approach on real EEG-datasets to assess and validate the proposed methodology. The classification accuracies are high even on having less number of training samples.","PeriodicalId":199062,"journal":{"name":"2018 ACM/IEEE 9th International Conference on Cyber-Physical Systems (ICCPS)","volume":"205 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132152281","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-10-07DOI: 10.1109/ICCPS.2018.00012
Sagong Uk Sang, Xuhang Ying, Andrew Clark, L. Bushnell, R. Poovendran
Automobiles are equipped with Electronic Control Units (ECUs) that communicate via in-vehicle network protocol standards such as the Controller Area Network (CAN). These protocols were designed under the assumption that separating in-vehicle communications from external networks is sufficient for protection against cyber attacks. This assumption, however, has been shown to be invalid by recent attacks in which adversaries were able to infiltrate the in-vehicle network. Motivated by these attacks, intrusion detection systems (IDSs) have been proposed for in-vehicle networks that attempt to detect attacks by exploiting physical properties such as clock skew of an ECU. In this paper, we propose the cloaking attack, an intelligent masquerade attack in which an adversary modifies the timing of transmitted messages to match the clock skew of a targeted ECU. The attack leverages the fact that, while the clock skew is a physical property of each ECU that cannot be changed by the adversary, the estimation of the clock skew by other ECUs is based on the timing of network traffic, which, being a cyber component only, can be modified by an adversary. We implement the proposed cloaking attack and test it on two IDSs, namely, the current state-of-the-art IDS and its adaptation to the widely-used Network Time Protocol (NTP). We implement the cloaking attack on two hardware testbeds, a prototype and a real vehicle, and show that it is able to deceive both IDSs. We also introduce a new metric called the Maximum Slackness Index to quantify the effectiveness of a clock skew-based IDS in detecting masquerade attacks when the adversary is unable to precisely match the clock skew of the targeted ECU.
{"title":"Cloaking the Clock: Emulating Clock Skew in Controller Area Networks","authors":"Sagong Uk Sang, Xuhang Ying, Andrew Clark, L. Bushnell, R. Poovendran","doi":"10.1109/ICCPS.2018.00012","DOIUrl":"https://doi.org/10.1109/ICCPS.2018.00012","url":null,"abstract":"Automobiles are equipped with Electronic Control Units (ECUs) that communicate via in-vehicle network protocol standards such as the Controller Area Network (CAN). These protocols were designed under the assumption that separating in-vehicle communications from external networks is sufficient for protection against cyber attacks. This assumption, however, has been shown to be invalid by recent attacks in which adversaries were able to infiltrate the in-vehicle network. Motivated by these attacks, intrusion detection systems (IDSs) have been proposed for in-vehicle networks that attempt to detect attacks by exploiting physical properties such as clock skew of an ECU. In this paper, we propose the cloaking attack, an intelligent masquerade attack in which an adversary modifies the timing of transmitted messages to match the clock skew of a targeted ECU. The attack leverages the fact that, while the clock skew is a physical property of each ECU that cannot be changed by the adversary, the estimation of the clock skew by other ECUs is based on the timing of network traffic, which, being a cyber component only, can be modified by an adversary. We implement the proposed cloaking attack and test it on two IDSs, namely, the current state-of-the-art IDS and its adaptation to the widely-used Network Time Protocol (NTP). We implement the cloaking attack on two hardware testbeds, a prototype and a real vehicle, and show that it is able to deceive both IDSs. We also introduce a new metric called the Maximum Slackness Index to quantify the effectiveness of a clock skew-based IDS in detecting masquerade attacks when the adversary is unable to precisely match the clock skew of the targeted ECU.","PeriodicalId":199062,"journal":{"name":"2018 ACM/IEEE 9th International Conference on Cyber-Physical Systems (ICCPS)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125350983","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-10-05DOI: 10.1109/ICCPS.2018.00029
Jyotirmoy V. Deshmukh, Xiaoqing Jin, R. Majumdar, Vinayak S. Prabhu
Embedded controllers for cyber-physical systems are often parameterized by look-up maps representing discretizations of continuous functions on metric spaces. For example, a non-linear control action may be represented as a table of pre-computed values, and the output action of the controller for a given input computed by using interpolation. For industrial-scale control systems, several man-hours of effort are spent in tuning the values within the look-up maps. %and sub-optimal performance is often associated with %inappropriate values in look-up maps. Suppose that during testing, the controller code is found to have sub-optimal performance. The parameter fault localization problem asks which parameter values in the code are potential causes of the sub-optimal behavior. We present a statistical parameter fault localization approach based on binary similarity coefficients and set spectra methods. Our approach extends previous work on (traditional) software fault localization to a quantitative setting where the parameters encode continuous functions over a metric space and the program is reactive. We have implemented our approach in a simulation workflow for control systems in Simulink. Given controller code with parameters (including look-up maps), our framework bootstraps the simulation workflow to return a ranked list of map entries which are deemed to have most impact on the performance. On a suite of industrial case studies with seeded errors, our tool was able to precisely identify the location of the errors.
{"title":"Parameter Optimization in Control Software Using Statistical Fault Localization Techniques","authors":"Jyotirmoy V. Deshmukh, Xiaoqing Jin, R. Majumdar, Vinayak S. Prabhu","doi":"10.1109/ICCPS.2018.00029","DOIUrl":"https://doi.org/10.1109/ICCPS.2018.00029","url":null,"abstract":"Embedded controllers for cyber-physical systems are often parameterized by look-up maps representing discretizations of continuous functions on metric spaces. For example, a non-linear control action may be represented as a table of pre-computed values, and the output action of the controller for a given input computed by using interpolation. For industrial-scale control systems, several man-hours of effort are spent in tuning the values within the look-up maps. %and sub-optimal performance is often associated with %inappropriate values in look-up maps. Suppose that during testing, the controller code is found to have sub-optimal performance. The parameter fault localization problem asks which parameter values in the code are potential causes of the sub-optimal behavior. We present a statistical parameter fault localization approach based on binary similarity coefficients and set spectra methods. Our approach extends previous work on (traditional) software fault localization to a quantitative setting where the parameters encode continuous functions over a metric space and the program is reactive. We have implemented our approach in a simulation workflow for control systems in Simulink. Given controller code with parameters (including look-up maps), our framework bootstraps the simulation workflow to return a ranked list of map entries which are deemed to have most impact on the performance. On a suite of industrial case studies with seeded errors, our tool was able to precisely identify the location of the errors.","PeriodicalId":199062,"journal":{"name":"2018 ACM/IEEE 9th International Conference on Cyber-Physical Systems (ICCPS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129863611","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-07-07DOI: 10.1109/ICCPS.2018.00028
Eva Darulova, Einar Horn, Saksham Sharma
Finite-precision arithmetic, widely used in embedded systems for numerical calculations, faces an inherent tradeoff between accuracy and efficiency. The points in this tradeoff space are determined, among other factors, by different data types but also evaluation orders. To put it simply, the shorter a precision's bit-length, the larger the roundoff error will be, but the faster the program will run. Similarly, the fewer arithmetic operations the program performs, the faster it will run; however, the effect on the roundoff error is less clear-cut. Manually optimizing the efficiency of finite-precision programs while ensuring that results remain accurate enough is challenging. The unintuitive and discrete nature of finite-precision makes estimation of roundoff errors difficult; furthermore the space of possible data types and evaluation orders is prohibitively large. We present the first fully automated and sound technique and tool for optimizing the performance of floating-point and fixed-point arithmetic kernels. Our technique combines rewriting and mixed-precision tuning. Rewriting searches through different evaluation orders to find one which minimizes the roundoff error at no additional runtime cost. Mixed-precision tuning assigns different finite precisions to different variables and operations and thus provides finer-grained control than uniform precision. We show that when these two techniques are designed and applied together, they can provide higher performance improvements than each alone.
{"title":"Sound Mixed-Precision Optimization with Rewriting","authors":"Eva Darulova, Einar Horn, Saksham Sharma","doi":"10.1109/ICCPS.2018.00028","DOIUrl":"https://doi.org/10.1109/ICCPS.2018.00028","url":null,"abstract":"Finite-precision arithmetic, widely used in embedded systems for numerical calculations, faces an inherent tradeoff between accuracy and efficiency. The points in this tradeoff space are determined, among other factors, by different data types but also evaluation orders. To put it simply, the shorter a precision's bit-length, the larger the roundoff error will be, but the faster the program will run. Similarly, the fewer arithmetic operations the program performs, the faster it will run; however, the effect on the roundoff error is less clear-cut. Manually optimizing the efficiency of finite-precision programs while ensuring that results remain accurate enough is challenging. The unintuitive and discrete nature of finite-precision makes estimation of roundoff errors difficult; furthermore the space of possible data types and evaluation orders is prohibitively large. We present the first fully automated and sound technique and tool for optimizing the performance of floating-point and fixed-point arithmetic kernels. Our technique combines rewriting and mixed-precision tuning. Rewriting searches through different evaluation orders to find one which minimizes the roundoff error at no additional runtime cost. Mixed-precision tuning assigns different finite precisions to different variables and operations and thus provides finer-grained control than uniform precision. We show that when these two techniques are designed and applied together, they can provide higher performance improvements than each alone.","PeriodicalId":199062,"journal":{"name":"2018 ACM/IEEE 9th International Conference on Cyber-Physical Systems (ICCPS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-07-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129452233","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：