Cyber Physical System Approach For Design Of Power Grids: A Survey

1

Cyber Physical System Approach For Design Of Power Grids: A Survey Siddhartha Kumar Khaitan and James D. McCalley, Fellow, IEEE, Dept. of Electrical and Computer Engineering, Iowa State University, Ames, IA, USA. {skhaitan, jdm}@iastate.edu

Abstract—Cyber physical systems (CPSs) refer to the class of systems which offer close integration of computation, networking, and physical processes. CPS approach to system design has been conventionally used in several domains, such as smart homes and health-care systems, however, its use in the design of power systems is relatively new. The unique features of CPSs are expected to greatly benefit the smart power grids of tomorrow. In this paper, we review several recent advancements made in the field of CPS approach in design and operation of power grids. We also discuss the application of CPSs in other domains to gain insights into the techniques and design approaches which could also be beneficial for power systems. We present a classification of research work and identify the challenges in wide-scale adoption of CPSs. This survey is intended to enable the researchers and power system operators to get insights into working of CPSs and understand their potential in transforming the future power grids. Index Terms—Power system, power grid, cyber physical systems, survey, review, classification, security.

on several issues such as security and reliability of cyberphysical power grids, architecture design and applications such as reducing the environmental impact of power grids (Section III). Since it is practically infeasible to cover all the aspects of CPSs in a review of this length, we focus on key ideas of different works. Further, we review the cyberphysical system techniques in related area and discuss ways in which these advancements can be useful for power systems also (Section IV). Finally, we highlight the research-challenges which need to be addressed for using cyber-physical system approach to designing power grids and present concluding remarks (Section V). The aim of this paper is to help power system operators and researchers in gaining valuable insights into the CPS approach to design of power grids and address the challenges to architect the reliable and cost-effective power grids of tomorrow. II. OVERVIEW OF CPS R ESEARCH W ORK

I. I NTRODUCTION Recent years have witnessed an increased demands of electricity supply and complex usage patterns of power systems. These challenges demand a significant improvement in design and operation of power systems. To address these issues, researchers have proposed cyber-physical systems (CPSs) approach to the design of power systems [1]. CPS refers to the system which offer close integration between cyber and physical components at all levels. The close interaction between cyber and physical components in the CPSs presents significant challenges in their development [2]. CPSs are composed of complex and heterogeneous systems consisting of computing devices, interfaces, distributed sensors and actuators. Further, the requirement of timely, high-precision communication and coordination between cyber and physical components necessitates efficient co-design approach. To enable seamless integration, the events and decisions in cyber-world need to be communicated to physical-world and vice-versa. Thus, for making wide-scale adoption of cyber-physical systems a reality, several issues need to be addressed. This makes study and understanding of CPSs extremely important. In this paper we survey recent research works in the field of cyber-physical power and energy systems. We first present an overview of the advancements in CPS and classify the works along several characteristics (Section II). We then focus

Several researchers have studied the CPS approach for the design of power systems [3–17], especially for ensuring their security and reliability [4, 12, 14, 15, 17–19]. The security and dependability of CPSs have also been discussed in the context of other areas [8, 20–28]. Several research efforts have focused on issues related to modeling, architecture design, simulation and verification of CPS [5, 29–45]. Other research works discuss issues such as QoS [46–50] and real-time requirements [51–53] in CPS. CPSs have applications in wide-variety of domains, such as data centers [54, 55], social network and gaming [56–58], thermal management [59], smart homes and buildings [60, 61], cloud computing [62], surveillance [63–65], scheduling [66], video processing [67, 68], environmental monitoring [69], medical and health care systems [70, 71], aerospace and air-traffic management [72], vehicular systems [73, 74], and networking systems [75, 76]. III. CPS A PPROACH

IN

P OWER S YSTEMS

Power systems have 24x7 availability requirement and the events such as power outages and blackouts have serious economic and social impacts. For example, the service cost of one hour of downtime in home shipping channels is $113,000, while for credit car authorization, it can be as high as $2,600,000 [77]. Hence, secure operation of power systems is extremely crucial [18]. Further, because of the connected

2

dynamics of the CPS, a failure in one component may have cascading effects [24]. Moreover, a multi-pronged attack may take advantage of weaknesses of the separate components of the system, which when combined together, may have a catastrophic consequence. Ten et al. [78] present an approach to evaluate the vulnerabilities of both cyber and power (physical) system of the control centers. By integrating CPS attack/defense modeling with system simulation capability, they quantify the impact of an attack. Their vulnerability assessment method is performed at three levels, viz. system, scenarios, and access points. They evaluate the impact of a potential electronic intrusion by estimating the consequential loss of load in the power system. Their work helps in improving the cyber-security of the CPS based on the vulnerability assessment results. Sun et al. [5] use a model-checking approach for verification of the cyber-physical system, which is a part of an advanced electric power grid. Their approach specifies the model of the system and its desired properties. Using a decomposition approach, the system is logically divided into smaller modules, which can be efficiently checked, since only a subset of the properties is checked for each decomposed small model. The resulting proofs are checked for noninterference with the correctness of the overall system. Ilic et al. [30] propose a technique for modeling cyberphysical energy systems (CPES). They represent all physical components as modules interconnected by means of an electric network. Further, each component is characterized by both physical and cyber input-output signals, internal dynamics, local sensing, and actuation and thus, is modeled as a cyberphysical module. Using this representation, the modular components are integrated according to the network constraints. The energy system is composed of many non-uniform subsystems, such as diverse energy sources and different classes of energy users. Using their model, the system operator can take decision about sensing, the level of data mining required for different physical modules to achieve desired performance for CPES. VIKING (vital infrastructure, networks, information and control system management) project is designed to promote secure and resilient power transmission [7]. In VIKING, the physical part is the power transmission and distribution system and the cyber part is the IT system. The authors in [7] propose models and techniques for understanding vulnerabilities of control systems. They also discuss the impact of vulnerabilities on the electric power transmission and distribution system and devise solutions to mitigate these vulnerabilities. Mohajerani et al. [79] present a technique to detect vulnerability of the power systems against the cyber attacks. Their method iteratively finds the most vulnerable substation inside the grid; then the most critical asset inside that substation and finally places a security agent on the most vulnerable positions based on computation of risk of each asset. Saber et al. [11] present an infrastructure for integrated electricity and transportation to promote use of renewable energy sources (RESs). They model the energy system as a CPES, where the on-board system in a gridable vehicle (GV) acts as a cyber resource, which communicates with utility and

vehicle owner’s preferences. Further, RESs, gridable vehicles (GVs) and thermal power plants form the physical resources. The authors study the effectiveness of RESs and GVs for making a sustainable CPES. They also propose techniques for maximizing the utilization of RESs with a view to reduce carbon emission and cost of operation. Li et al. [80] tackle the task of scheduling in cyber-physical network systems. Such systems have utility in wireless ad hoc networks and the coordinated Electric Vehicle (EV) charging in power grids. By using a combination of Lyapunov optimization and Markov Chain Monte Carlo (MCMC) sampling techniques, the authors solve the problem of distributed EV charging scheduling. Susuki et al. [9] describe an approach to design and analyze power grid dynamic performance based on the theoretical foundations and computational methods developed for hybrid systems. In a power grid, the joint dynamics of physical processes and cyber elements leads to a hybrid of continuous and discrete behaviors. Using hybrid dynamics theory, the authors address the stability issue using reachability analysis. Liu et al. [19] address the issue of malicious data attack under a deterministic model of network state variables and arbitrary attack patterns. They obtain the condition under which detecting the malicious data becomes impossible and the attack increases the state estimation error arbitrarily. Kosut et al. [15] discuss the method to quantify the impact of a malicious data attack on power system state estimation and also suggest counter measures. They discuss the attack strategies of the adversary for malicious data injection in a power grid. They also present the ways to distinguish between the conventional “bad data” due to natural causes (such as meter malfunction and communication outage) from the malicious data which is intentionally injected. Neuman et al. [14] characterize the interactions in cyberphysical power system based on the domain in which threat has originated and the domain where its impact is observed. Based on it, they classify the threats into several kinds, such as cyber-physical threats, cyber-cyber threats, physical-cyber threats and physical-physical threats. For example, physicalcyber threats are those that originate with physical actions and affect the information or networking components of the system. The authors also discuss the architectural changes to mediate the impact of threat propagation. Schneider et al. [4] utilize marked Petri net models to study the interactions between the power and telecommunications infrastructures. The authors discuss the example of 1988 blackout in the Hydro-Quebec system, where a failure in communication system led to serious impact on power system. To capture the behavior of both communication and power infrastructures, the authors create a model using Petri net and use model reduction techniques to simplify the model. Using this model, they estimate the approximate probability of proper operation, an approximate time for proper operation, and also identify potential single point failures. Zhu et al. [12] propose a theoretical framework for resilient and robust control design. In their design, the stochastic switching between structure states models unanticipated events. The uncertainties in each structure represent the known

3

range of disturbances. They present coupled optimality criteria for a robust and resilient design for CPSs and use their method to study the design of a voltage regulator for a synchronous machine with infinite bus. McMillin et al. [17] present an approach to use formal methods of security specification to describe confidentiality in CPSs. Due to the integrated nature of cyber-physical systems, an observation about physical information flow may permit an observer to infer about the cyber actions. For example, the observable parameters of a wind turbine are physical size and velocity of wind which affect its operation; and these parameters may reveal about the cyber features of the system. To address these issues and ensure security, the authors present an approach for information flow verification. Modern computing systems, such as data centers and supercomputers contribute significantly to the world-wide electricity consumption [81–83]. For example, the peak power consumption of each of the 10 most powerful supercomputers in the TOP500 list of supercomputers is up to 10 megawatts [84], which is sufficient to sustain a city of 40,000 people. Hence, a cyber-physical system approach to managing power consumption of data centers is important for reducing the electricity demands from power grid. In a data-center, the online applications for communication and computation constitute the cyber part and hardware components such as racks and cooling equipments constitute the physical part. Li et al. [54] present a thermal forecasting model to predict the temperatures near the data-center servers based on the continuous streams of temperature and airflow measurements. The modeling is based on the physical laws and sensor observations in the data center. Using the data obtained from the sensors, their models learns the parameters of a data center’s cyber-physical system. Qian et al. [59] present a method for managing temperature inside 3D multi-core processor system. Their method models the heat consumption of the 3D chip using a thermal model, and utilizes this model to predict future power consumption of the chip. Further, their method senses the temperature of the processor-chip and adjusts the fluidic flow-rate to continuously maintain the system temperature. The cyber-physical nature of operation of their method provides real-time, prediction-andcorrection and control of the temperature of the processor. Kleissl et al. [60] examine the possibility of joint optimization of energy use by the occupants and IT equipments in the modern buildings. Their method aims at designing zero net energy buildings (ZNEB), i.e. buildings which have zero net annual energy consumption. They model the information and communication infrastructure of the building as energy consumer as well as energy/operations optimizers. They examine the energy consumption of several buildings and explore the opportunity of saving electrical energy by using various means such as renewable energy sources and efficient computing and lighting. The dynamics of power systems evolve very fast and hence, the CPS approach to design of power systems can facilitate handling the complexities posed by temporal variations and designing situation-specific control actions. Singh et al. [85] address the issue of situation-based control in cyber-physical environments using the situation calculus. Their method en-

ables effective handling of temporal events and creating suitable actions in response of the events. IV. CPS A PPROACH

IN

R ELATED A REAS

We now review advancements made in CPSs in other areas and discuss the manner in which these advancements can be useful in power systems also. Huang and Tidwell et al. [53] study the challenges in the design of cyber-physical system for real-time hybrid structural testing. Several real-life systems (e.g. vibration suppression system) cannot be tested before on-site deployment, without using destructive testing. To address this challenge, the authors propose a reusable middleware architecture which allows integrating both cyber and physical components. Using this framework, the designer can obtain insights into real-time hybrid structural testing, e.g. tolerance towards violations of timing constraints and the effects of interactions between physical and cyber components on the overall system characteristics. Their approach can be extended to non-destructive testing of power systems before actual use. Ahmadi et al. [43] propose a hierarchical modeling scheme for designing open CPSs. Their method combines estimation techniques with data mining techniques to capture the complex behavior of the system at different levels of abstraction. The authors demonstrate their approach with the example of green transportation. The goal of this is to reduce fuel consumption and carbon footprint of the vehicles. In their method, the cars traveling on different roads under different traffic conditions form the system. The routing optimizations software uses physical models of cars, streets, and traffic conditions to enable green transportation. The authors have shown that their modeling techniques offers significant improvement in the accuracy of fuel consumption predictions. The optimization approach used in their work can be applied in power system for jointly managing multiple power grids for reducing their carbon footprint while maintaining desired level of power supply. Dabholkar et al. [44] present an approach to specialize general-purpose middleware to fulfill the demands of CPSs used in multiple domains. Their approach is based on featureoriented software development principles, which employs an algebraic structure of existing middleware. Their approach raises the level of abstraction to the level of features which are offered by the middleware, instead of the details of source code. Using this approach, their method specializes the general-purpose middleware to the needs of domainspecific CPS. This promotes code-reuse and also alleviates the excessive expenditure on development, maintenance and testing of domain-specific CPS. Their approach can be useful for specializing the existing CPS-based software for use in power systems. Miller et al. [45] present a co-design framework for development of cyber-physical devices where the real devices or prototypes are connected to the real-time models which simulate the interacting environment. Their framework supports co-design of hardware/software and enables the models of varying speed and accuracy to be implemented within an

4

embedded processor. Using their approach, an applicationspecific testing platform can be automatically generated. Their work can be useful in designing testbeds for software-tools used for decision support in power systems. Zhu et al. [12] present a framework for control design which aims at achieving both robustness and resilience in CPS. The authors adopt a hybrid framework where the cyber-level policy affects the control system design at physical level and the control policy affects the decisions taken at cyber level. Utilizing this framework, the designer can achieve resilient control of the cyber system along with robust control of the physical system. Their research can be used for enhancing security of power systems. Talcott et al. [40] present an event-based semantics for CPSs. They classify different events based on their characteristics such as punctual v/s durative, single v/s stream. Use of event-based semantics provides a natural way to specify components of open systems in terms of interfaces and observable behavior. It also provides a foundation for development, monitoring and implementation of CPSs. Tan et al. [33] present an architecture for CPS which utilizes the temporal and spatial properties of events. They represent an event as a function of attribute, spatial and temporal event conditions. Using the logical operators, different types of event conditions can be combined to capture composite events. Using this framework, complex relationships of CPSs can be modeled. These research-works can be gainfully employed in design of controllers in power systems which provide situational awareness of the operator. Hnat et al. [86] present a macroprogramming framework for programming CPS, which is called MacroLab. MacroLab allows a user to write a single program for the entire CPS. This program is then decomposed into a set of microprograms that are loaded into each node. MacroLab decomposes a macroprogram in the manner which is suitable for a particular deployment (i.e. hardware topology, network etc.) and also allows easy manipulation of data from sensors and actuators. Their framework can be used to facilitate the design of cybertools for large-scale power systems. Medical Device Plug-and-Play (MD PnP) Interoperability initiative [87] is an example of medical cyber physical systems which aims to provide a framework for safe interconnectivity of medical devices. In addition to developing interoperability standards, MD PnP initiative collects and demonstrates the clinical scenarios where interoperability can be helpful in presenting improvements over the existing practice. The insights gained from their work of developing interoperability standards can be used in management of connected powergrids. V. C ONCLUSION AND F UTURE C HALLENGES As CPSs move from research-prototypes to real-world systems (such as wide area monitoring and control systems (WAMCS)), several challenges arise which need to be addressed for enabling their wide-scale use. The design of CPSs is more challenging than either of isolated physical and cyber systems alone. For CPSs, the desired behavior of computational elements needs to be specified in terms of their influence

on the physical environment. Hence, a unifying framework is required for modeling them, which allows consistency and low-overhead design. Secondly, the mutual coordination among cyber and physical worlds brings challenge of increased vulnerability to failures and attacks. In power systems, security and reliability are of utmost importance since it is a critical infrastructure system. Finally, high performance and timeliness is important for cyber-physical power system, along with vehicular, aerospace and gaming systems. Similarly, high energy usage efficiency is important to enhance cost-effectiveness of CPSs in smart buildings and homes and data centers; which will also reduce the electricity demands placed on power grid. To address these challenges, research progress is required in both theoretical foundations and application tools for CPSs. Cyber-physical systems offer close interaction and coordination between computational and physical resources of the system and hence, they are expected to play a major role in the design and development of next-generation smart grid. In this paper, we surveyed several research works in CPSs which are useful in design and operation of power systems. We also discussed the ways in which the advancements made in CPS approach in other fields can be extended to power systems. We highlighted the challenges which need to be addressed for enabling large-scale use of CPSs. It is hoped that CPS technologies will play key role in reshaping the power grids of tomorrow. R EFERENCES [1] I. Horv´ath and B. Gerritsen, “Cyber-physical systems: Concepts, technologies and implementation principles,” in TMCE, 2012, pp. 19–36. [2] E. Lee, “Cyber physical systems: Design challenges,” in ISORC. IEEE, 2008, pp. 363–369. [3] C. Davis et al., “Scada cyber security testbed development,” in NAPS. IEEE, 2006, pp. 483–488. [4] K. Schneider et al., “Assessment of interactions between power and telecommunications infrastructures,” IEEE TPWRS, 2006. [5] Y. Sun et al., “Verifying noninterference in a cyber-physical system the advanced electric power grid,” in QSIC. IEEE, 2007, pp. 363–369. [6] S. Karnouskos, “Cyber-physical systems in the smartgrid,” in INDIN. IEEE, 2011, pp. 20–23. [7] A. Giani et al., “The viking project: An initiative on resilient control of power networks,” in ISRCS. IEEE, 2009, pp. 31–35. [8] Y. Mo et al., “Cyber–physical security of a smart grid infrastructure,” Proceedings of the IEEE, vol. 100, no. 1, pp. 195–209, 2012. [9] Y. Susuki et al., “A hybrid system approach to the analysis and design of power grid dynamic performance,” Proc. of the IEEE, 2012. [10] O. Yagan et al., “On allocating interconnecting links against cascading failures in cyber-physical networks,” in INFOCOM WKSHPS, 2011. [11] A. Saber and G. Venayagamoorthy, “Efficient utilization of renewable energy sources by gridable vehicles in cyber-physical energy systems,” Systems Journal, IEEE, vol. 4, no. 3, pp. 285–294, 2010. [12] Q. Zhu et al., “Robust and resilient control design for cyber-physical systems with an application to power systems,” in CDC-ECC, 2011. [13] N. Hadjsaid et al., “Modeling cyber and physical interdependenciesapplication in ict and power grids,” in IEEE/PES PSCE, 2009, pp. 1–6. [14] C. Neuman and K. Tan, “Mediating cyber and physical threat propagation in secure smart grid architectures,” in SmartGridComm, 2011. [15] O. Kosut et al., “On malicious data attacks on power system state estimation,” in UPEC. IEEE, 2010, pp. 1–6. [16] J. Zhao et al., “Cyber physical power systems: architecture, implementation techniques and challenges,” Dianli Xitong Zidonghua(Automation of Electric Power Systems), vol. 34, no. 16, pp. 1–7, 2010. [17] B. McMillin and R. Akella, “Verification of information flow properties in cyber-physical systems,” in FDSCPS, 2011, p. 37. [18] S. Gorman, “Electricity grid in US penetrated by spies,” Wall Street Journal, vol. 8, 2009.

5

[19] Y. Liu, P. Ning, and M. Reiter, “False data injection attacks against state estimation in electric power grids,” ACM TISSEC, 2011. [20] C. Singh and A. Sprintson, “Reliability assurance of cyber-physical power systems,” in IEEE PES-GM. IEEE, 2010, pp. 1–6. [21] A. Cardenas et al., “Challenges for securing cyber physical systems,” in Workshop on Future Directions in Cyber-physical Systems Security. DHS, 2009. [22] L. Tang et al., “Tru-alarm: Trustworthiness analysis of sensor networks in cyber-physical systems,” in ICDM. IEEE, 2010, pp. 1079–1084. [23] T. Gamage, B. McMillin, and T. Roth, “Enforcing information flow security properties in cyber-physical systems: A generalized framework based on compensation,” in COMPSACW. IEEE, 2010, pp. 158–163. [24] M. Anand et al., “Security challenges in next generation cyber physical systems,” Beyond SCADA: Networked Embedded Control for Cyber Physical Systems, 2006. [25] M. Sun et al., “Addressing safety and security contradictions in cyberphysical systems,” in CPSSW, 2009. [26] C. Zimmer, B. Bhat, F. Mueller, and S. Mohan, “Time-based intrusion detection in cyber-physical systems,” in ICCPS, 2010, pp. 109–118. [27] S. Bak, K. Manamcheri, S. Mitra, and M. Caccamo, “Sandboxing controllers for cyber-physical systems,” in ICCPS, 2011, pp. 3–12. [28] A. Banerjee et al., “Ensuring safety, security, and sustainability of mission-critical cyber–physical systems,” Proc. of the IEEE, 2012. [29] T. Morris et al., “Engineering future cyber-physical energy systems: Challenges, research needs, and roadmap,” in NAPS. IEEE, 2009. [30] M. Ilic, L. Xie, U. Khan, and J. Moura, “Modeling future cyberphysical energy systems,” in IEEE PES GM-Conversion and Delivery of Electrical Energy in the 21st Century, 2008, pp. 1–9. [31] P. Zhao et al., “A conceptual scheme for cyber-physical systems based energy management in building structures,” in INDUSCON, 2010. [32] R. Thacker et al., “Automatic abstraction for verification of cyberphysical systems,” in ICCPS, 2010, pp. 12–21. [33] Y. Tan, M. Vuran, and S. Goddard, “Spatio-temporal event model for cyber-physical systems,” in ICDCS. IEEE, 2009, pp. 44–50. [34] J. Lin, S. Sedigh, and A. Miller, “Modeling cyber-physical systems with semantic agents,” in COMPSACW. IEEE, 2010, pp. 13–18. [35] Y. Tan et al., “A concept lattice-based event model for cyber-physical systems,” in ICCPS. ACM, 2010, pp. 50–60. [36] J. Kim and D. Mosse, “Generic framework for design, modeling and simulation of cyber physical systems,” ACM SIGBED Review, 2008. [37] B. Bonakdarpour, “Challenges in transformation of existing real-time embedded systems to cyber-physical systems,” ACM SIGBED Review, vol. 5, no. 1, pp. 1–2, 2008. [38] P. Derler, E. Lee, and A. Vincentelli, “Modeling cyber–physical systems,” Proceedings of the IEEE, vol. 100, no. 1, pp. 13–28, 2012. [39] H. Woo et al., “Design and development methodology for resilient cyberphysical systems,” in ICDCS. IEEE, 2008, pp. 525–528. [40] C. Talcott, “Cyber-physical systems and events,” Software-Intensive Systems and New Computing Paradigms, pp. 101–115, 2008. [41] P. Vicaire et al., “Bundle: A group-based programming abstraction for cyber-physical systems,” IEEE Trans. Ind. Informat, 2012. [42] J. Huang et al., “Toward a smart cyber-physical space: a contextsensitive resource-explicit service model,” in COMPSAC. IEEE, 2009. [43] H. Ahmadi et al., “The sparse regression cube: A reliable modeling technique for open cyber-physical systems,” in ICCPS, 2011. [44] A. Dabholkar and A. Gokhale, “An approach to middleware specialization for cyber physical systems,” in ICDCS. IEEE, 2009, pp. 73–79. [45] B. Miller et al., “Application-specific codesign platform generation for digital mockups in cyber-physical systems,” in ESLsyn, 2011, pp. 1–6. [46] F. Xia et al., “Network QoS management in cyber-physical systems,” in ICESS. IEEE, 2008, pp. 302–307. [47] ——, “Fuzzy logic control based QoS management in wireless sensor/actuator networks,” Sensors, 2007. [48] S. Gupta et al., “Guaranteed QoS with MIMO Systems for Scalable Low Motion Video Streaming Over Scarce Resource Wireless Channels,” in International Conference On Information Processing, 2008. [49] F. Xia et al., “Cyber-physical control over wireless sensor and actuator networks with packet loss,” Wireless Networking Based Control, Springer, 2010. [50] A. Chan et al., “Temporal quality assessment for mobile videos,” in MobiCom, 2012. [51] K. Kang and S. Son, “Real-time data services for cyber physical systems,” in ICDCS, 2008, pp. 483–488. [52] K. Lin and M. Panahi, “A real-time service-oriented framework to

support sustainable cyber-physical systems,” in INDIN, 2010, pp. 15–21. [53] H. Huang et al., “Cyber-physical systems for real-time hybrid structural testing: a case study,” in ICCPS, 2010, pp. 69–78. [54] L. Li et al., “Thermocast: A cyber-physical forecasting model for data centers,” in Proc. KDD, vol. 11, 2011. [55] L. Rao et al., “Mec-idc: joint load balancing and power control for distributed internet data centers,” in ICCPS, 2010, pp. 188–197. [56] F. Wu, F. Chu, and Y. Tseng, “Cyber-physical handshake,” in ACM SIGCOMM Computer Communication Review, 2011, pp. 472–473. [57] C. Wu et al., “Multi-screen cyber-physical video game: an integration with body-area inertial sensor networks,” in PERCOM Workshops, 2010. [58] E. Miluzzo et al., “Sensing meets mobile social networks: the design, implementation and evaluation of the cenceme application,” in SenSys, 2008, pp. 337–350. [59] H. Qian et al., “Cyber-physical thermal management of 3d multi-core cache-processor system with microfluidic cooling,” JOLPE, 2011. [60] J. Kleissl and Y. Agarwal, “Cyber-physical energy systems: focus on smart buildings,” in DAC. ACM, 2010, pp. 749–754. [61] C. Lai et al., “OSGi-based services architecture for Cyber-Physical Home Control Systems,” Computer Communications, 2011. [62] S. Craciunas et al., “Information-acquisition-as-a-service for cyberphysical cloud computing,” in HotCloud, 2010. [63] N. Ozay et al., “Distributed synthesis of control protocols for smart camera networks,” in ICCPS. IEEE, 2011, pp. 45–54. [64] J. Chen et al., “Fidelity-aware utilization control for cyber-physical surveillance systems,” TDPS, vol. 23, no. 9, pp. 1739–1751, 2012. [65] L. Ma et al., “A high-confidence cyber-physical alarm system: Design and implementation,” in GreenCom & CPSCom. IEEE, 2010. [66] F. Zhang et al., “Task scheduling for control oriented requirements for cyber-physical systems,” in RTSS, 2008. IEEE, 2008, pp. 47–56. [67] M. Lindberg et al., “Feedback control of cyber-physical systems with multi resource dependencies and model uncertainties,” in RTSS, 2010. [68] S. Mittal et al., “FPGA: An Efficient And Promising Platform For RealTime Image Processing Applications,” in CSI-RDHS, 2008. [69] H. Lim et al., “The national weather sensor grid: a large-scale cybersensor infrastructure for environmental monitoring,” International Journal of Sensor Networks, 2010. [70] I. Lee and O. Sokolsky, “Medical cyber physical systems,” in DAC, 2010, pp. 743–748. [71] A. Cheng, “Cyber-physical medical and medication systems,” in ICDCS, 2008, pp. 529–532. [72] W. Zhang et al., “A hierarchical flight planning framework for air traffic management,” Proceedings of the IEEE, pp. 179–194, 2012. [73] A. Wagh et al., “Human centric data fusion in vehicular cyber-physical systems,” in INFOCOM WKSHPS. IEEE, 2011. [74] J. Yuan et al., “Driving with knowledge from the physical world,” in ACM SIGKDD. ACM, 2011, pp. 316–324. [75] M. Li and W. Zhao, “Visiting power laws in cyber-physical networking systems,” Mathematical Problems in Engineering, vol. 2012, 2011. [76] M. Kim et al., “An application framework for loosely coupled networked cyber-physical systems,” in EUC. IEEE, 2010, pp. 144–153. [77] W. Feng, “Making a case for efficient supercomputing,” Queue, vol. 1, no. 7, p. 54, 2003. [78] C. Ten et al., “Vulnerability assessment of cybersecurity for scada systems,” TPWRS, vol. 23, no. 4, pp. 1836–1846, 2008. [79] Z. Mohajerani et al., “Cyber-related risk assessment and critical asset identification within the power grid,” in IEEE PES TDCE, 2010. [80] Q. Li and R. Negi, “Distributed scheduling in cyber-physical systems: The case of coordinated electric vehicle charging,” in GLOBECOM Workshops (GC Wkshps), 2011 IEEE. IEEE, 2011, pp. 1183–1187. [81] S. Mittal, Z. Zhang, and Y. Cao, “CASHIER: A Cache Energy Saving Technique for QoS Systems,” IEEE VLSID, 2013. [82] S. Mittal, “A Survey of Architectural Techniques for DRAM Power Management,” Int. J. of High Performance System Architecture, 2013. [83] S. Mittal et al., “EnCache: Improving cache energy efficiency using a software-controlled profiling cache,” in IEEE EIT, May 2012. [84] W. Feng and K. Cameron, “The green500 list: Encouraging sustainable supercomputing,” Computer, 2007. [85] V. Singh and R. Jain, “Situation based control for cyber-physical environments,” in MILCOM. IEEE, 2009, pp. 1–7. [86] T. Hnat et al., “Macrolab: a vector-based macroprogramming framework for cyber-physical systems,” in ACM SenSys, 2008. [87] J. Goldman, R. Schrenker, J. Jackson, and S. Whitehead, “Plug-and-play in the operating room of the future,” 2005.