A Promising Technology to Overcome Spectrum ... - IEEE Xplore

A Promising Technology to Overcome Spectrum Inefficiency

Vehbi Cagri Gungor and Dilan *ahin ith the rapid increase in world population and power demand, the aging infrastructure of the existing power grid has caused many problems to electric utilities and customers in terms of system reliability, power quality, and customer satisfaction. Field tests show that the power grid has harsh and complex environmental conditions, dynamic topology changes, connectivity problems, interference, and fading, which make wireless communication very challenging in power grid environments. Recently, cognitive radio (CR) network is recognized as a promising technology to address the communication and networking problems of next-generation power grid, i.e., smart grid (SG). This article presents a comprehensive review about SG characteristics and CR-based SG applications. Also, architectures to support CR networks in SG applications, major challenges, and open issues have been discussed. The rapid growth of wireless technology has resulted in an exponential growth in the usage of diverse wireless applications and devices. The U.S. Federal Communications Commission (FCC) has shown that some portions of the unlicensed bands are so crowded by emerging wireless services and applications [4] that a spectrum scarcity

W

Digital Object Identifier 10.1109/MVT.2012.2190183 ©ARTVILLE, LLC.

JUNE 2012 | IEEE VEHICULAR TECHNOLOGY MAGAZINE

1556-6072/12/$31.00©2012IEEE

Date of publication: 20 April 2012

||| 41

problem is unavoidable. Increased RF demand, strict spectrum allocation policies, and inefficient utilization of spectrum resources reveal the need for a new communication paradigm to provide opportunistic access to the radio spectrum. To this end, CR networks promise to overcome the spectrum inefficiency problems by offering several benefits to utilize spectrum opportunistically with dynamic spectrum management techniques. CR networks adopt a dynamic spectrum allocation (DSA) technique to efficiently use the spectrum. There are two important actors in DSA: 1) primary user (PU) is the owner of a licensed channel that has the priority to use the spectrum and 2) secondary user (SU) is the opportunistic user that is responsible for sensing the licensed spectrum, identifying the unused channels, and preserving the set of locally available channels in the absence of PU [4]. With these unique features, CR networks have the potential to serve as an integrated sensing, communication, and computation system for many real-world applications including SG [10]. SG is modernization of generation, transmission, and distribution of a power grid system with the integration of advanced information and communication technologies (ICTs) infrastructure. The electrical power grid is the most critical and complex infrastructure of today’s world, and it is vulnerable to tremendous security threats. SG with the decentralized nature enables the integration of renewable energy resources and promises a two-way communication path between consumers and electric utilities, which will improve the efficiency of electric utility programs such as demand response, customer participation, advanced smart metering, and outage detection programs [8]. Recent field tests show that an SG system has harsh and complex environmental conditions, dynamic topology changes, connectivity problems, and interference and fading issues during wireless communications [7]. In addition, an SG system is a large-scale network in which there are many interconnected components, producing a huge amount of data to be transmitted, managed, and analyzed by the ICT. However, the current ICT infrastructures and standards are incapable of meeting all the requirements of SG technology. Many SG services, components, and applications have different requirements in terms of bandwidth, capacity, latency, and security, which makes it difficult to design an appropriate ICT system for the overall power grid. Hence, the choice for communication infrastructure for SG is highly critical to provide secure, reliable, and efficient data delivery between the SG components. Most of the traditional communication technologies, either lack of bandwidth, latency, security, fullcoverage, and reliability capabilities or the cost for investment, maintenance, and operation, are high. To this end, CR networks can improve the overall network performance with its adaptive operations to the existing spectrum conditions in the deployment field, and thus, increase spectrum utilization efficiency in SG environments. Dynamic spectrum access capability of CR networks

42 |||

can be benefited to address the unique challenges of SG, such as reliability and delay requirements, multipath fading, noise, different spectrum characteristics changing over location and time, and harsh environmental conditions. The motivations behind why CR network is a promising technology for SG applications can be outlined as follows. n The RF interference, noise from power equipment, and associated packet collisions in wireless links, which decrease the performance of SG communications, can be minimized with the opportunistic access to the spectrum in CR networks. To this end, CR networks can reduce the communication expenses of SG communication networks and increase spectrum efficiency with advanced spectrum management functionalities. n CR networks can manage the low-latency communication links by using unused local television (TV) broadcast spectra and improve SG network performance [5]. n SG is a distributed power system over a large geographic area, and therefore, different spectrum regulations can exist in these areas. CR network can enable large-scale SG applications and services to use different spectrum regulations with its cognitive capabilities. n In SG environments, multiple communication networks may exist due to the large scale of the system. CR networks can help to achieve fair radio spectrum sharing among multiple networks.

Cognitive Radio-Based Smart Grid Applications In this section, the major CR-based SG applications are introduced briefly.

Advanced Metering Infrastructure Advanced metering infrastructure (AMI) can be defined as a whole measurement and collection system, which includes smart meters, such as electric, gas, and heat meters, at consumer premises, communication network between consumer and service providers, and data management systems to manage and analyze the data for further processes. The collected data are so huge and important that the communication backbone should be reliable, secure, scalable, and cost-effective enough to meet the requirements in terms of bandwidth and latency. However, providing a robust communication backbone is sometimes hardly achievable because of the characteristics of the communication technologies used for AMI data transmissions. Most of the AMI communication models consist of thousands of smart meters, many access points, and a mesh network, which is formed between smart meters for data routing purposes by using industrial, scientific, and medical (ISM) frequency bands. Moreover, the aggregated data are routed to the electric utility by access points using, mostly, licensed bands. This model comes with some obstacles for realization of AMI in SG. The reliability and security of data communications between AMI components suffer from crowded and noisy ISM bands in urban areas. Performance

IEEE VEHICULAR TECHNOLOGY MAGAZINE | JUNE 2012

degradation, latency, and packet losses are some of the consequences of heterogeneous spectrum characteristics of the crowded wireless links. Furthermore, the choice of licensed bands for data communications between access points and utility brings extra costs to the system, which is another obstacle to obtain a cost-effective AMI deployment. To this end, CR technology can be suitable for AMI communication backhaul system by providing dynamic and opportunistic access to the spectrum to enhance the performance of data communications with increased reliability, security, and efficiency in seamless data communications.

strikes, icing, hurricanes, landslides, bird damage, and overheating—that will influence the safety, reliability, and security of the transmission lines. Furthermore, the distributed and large-scale nature of the transmission lines creates challenges for wireless sensors in terms of maintenance, different spectrum regulations, electromagnetic interference, and fading. To this end, CR networks can improve the network performance of overhead transmission line monitoring systems and increase spectrum efficiency and capacity with advanced spectrum management functionalities such as the ability of using different spectrum regulations, efficient spectrum utilization, and dynamic spectrum access.

Distributed Generation Distributed generation can be described as a small-scale electric generation near customer premises, which relieves the overload of the power grid by offering clean and renewable electricity resources, such as wind and solar, to provide reliability and balance the power demand and supply. The integration of renewable energy resources to the SG system brings many advantages, such as efficient, reliable, and green energy generation and safe connection to the grid, local, or distributed power storage capabilities. Energy storage is also another important opportunity for SG with the growing market for electric vehicles and portable electronic devices, the increasing penetration of distributed energy resources, and the increasing interest in handling the peak demand. Power quality, voltage, and frequency stability are some of the system parameters that need to be measured for the continuity of generation and storage operations. A robust and reliable communication network is required for the overall distributed generation system integrity and safety. Wireless networks might be suitable for these kinds of applications because of their low cost and easy deployment features. However, several challenges exist with the integration of wireless nodes to the power grid, such as network contention, noise, obstructions, and interference. CR networks can improve SG system performance through opportunistic spectrum access techniques by maximizing spectrum utilization.

Wide Area Monitoring

Power Outage Detection

Home Area Network

In the United States, the estimated cost of power outages in 2002 was approximately US$79 billion [9]. Hence, power outages have significant economic consequences. Power outages in the electric system cannot be timely detected because of the lack of automated analysis and poor visibility. Wireless nodes spread over the power distribution system, and consumer premises can sense and monitor the functioning of power flows and detect outages. To this end, advanced CRenabled devices and monitoring systems can help reduce outages by increasing wireless communications reliability in SG.

HAN is a subnetwork of SG, which creates a communication path between home appliances, in-home displays, energy management systems, and energy dashboards. HAN creates a technology platform for many SG applications, such as demand response, home energy management, load management, and smart metering, and each of them has its own operational goals toward the realization of SG, such as peak-demand shaving, power cost reduction, customer satisfaction, power supply and demand match, and accurate power measurements. The heterogeneity in wireless technology options in HAN structure imposes great challenges to the performance of the communication between SG components. The challenge is the RF interference from heterogeneous wireless technologies, since many of them,

Overhead Transmission Line Monitoring The transmission line is one of the most critical parts of the power grid. There exist many problems—such as lightning


Wide area monitoring in SG is a very common application applied by electric utilities to monitor transmission, generation, and distribution parts of the grid and retrieve the information about the status of the grid components. For example, transformer monitoring, capacitor bank monitoring, and network protection monitoring can be achieved via wide area monitoring. Many communication technologies are used to provide the visibility of the grid and broadband connectivity, such as licensed and unlicensed communication technologies. However, the choice of the communication technology brings its own challenges. Unlicensed technology can be cost-effective and easy to deploy over the transmission and distribution power lines, but it may not perform well in harsh RF conditions of the power grid. To this end, CR networks can be suitable for wide area monitoring applications, since they enable opportunistic usage of available frequency bands and provide quality of service (QoS) with the ability to coexist in the same area on the same channel with other wireless networks [6].

Architectures to Support CR Networks in Smart Grid Applications Basically, the SG network can be divided into three segments: home area networks (HANs), neighborhood area networks (NANs), and wide area networks (WANs).

||| 43

e.g., ZigBee, Bluetooth, and WiFi, operate in the 2.4-GHz license-free ISM band. The adoption of advanced CR technology capabilities, such as the ability of sensing and changing its transmission and reception parameters according to spectrum availability at different channels, will add intelligence to the existing HAN to manage adaptive transmission scheduling mechanisms minimizing interference and provide optimal data rate with low interference levels. The CR technology can also be adapted to the HAN architecture with the integration of the CR-based home gateway, as proposed in [11]. In HAN, the CR-based home gateway can have self-configuration capabilities to intelligently sense unused frequencies, adaptively connect and change the transmitters’ parameters, and utilize interference constraints.

Neighborhood Area Network NAN can be described as the second layer of SG communication architecture, which collects the metering and service information from multiple HANs and transmits them to the data collectors connecting NAN to the upper layer, i.e., WAN layer. The interference and spectrum inefficiency problems, due to the crowdedness of the available bands, impose great challenges to the communication performance of NAN components. Hence, Yu et al. proposes the integration of CR-based NAN gateway to the system to connect multiple CR-based home gateways and distribute spectrum band among them according to the transmission demand. This can improve spectrum efficiency, ensure QoS requirements, and improve throughput of CRbased NAN architectures [11].

Wide Area Network WAN is the upper layer of the SG communication architecture that provides broadband communication between SG substations, NANs, distributed grid devices, and the electric utility. Both wireless and wired communication technologies can be used for communication requirements of the grid elements in WAN. However, traditional WAN management methods do not create economical, efficient, and adaptive operations for the communication infrastructure. Specifically, in traditional WAN management approaches, the separate optimization of WAN and NAN can create additional challenges to the system reliability. For example, NANs in the same WAN may operate at the same range of spectrum bands, which causes spectrum limitations among different NANs, and each NAN demands a different amount of spectrum bands, which makes finding the optimum spectrum allocation challenging [11]. On the other hand, the WAN architecture in SG can include many CR-based NAN gateways with the capability to communicate with the control centers, which are connected to CR base stations distributed over a wide area and a spectrum broker managing the licensed spectrum sharing between different NANs [11].

44 |||

Challenges in Smart Grids Recent field tests show that an SG system has harsh and complex environmental conditions, dynamic topology changes, connectivity problems, and interference and fading issues during wireless communications [7]. Hence, traditional SG communication architectures with diverse communication technologies are subject to several communication problems, e.g., spectrum inefficiency and inefficient spectrum allocations, which degrade the performance of data transmissions between SG components and pose great challenges to meet the QoS requirements of SG applications. In the following, we try to summarize the major challenges in traditional SG architectures without the adoption of CR technology. n Limited Spectrum: The available spectrum is limited for wireless technologies in SG environments. There is a big competition for spectrum usage, especially in the bands below 3 GHz, which leads to the spectrum scarcity problem. With CR technology, the dynamic and opportunistic spectrum utilization can be provided. n Scalability: The scalability feature of the wired communication technologies for WAN connections in SG is limited because of high installation and maintenance costs. Hence, for wide area communication, wireless technologies are preferred because of its flexibility. However, scalability in wireless technologies is provided by adding more wireless access points and routers to the network, which will also increase the installation costs. With CR technology, the base station coverage area for the IEEE 802.22 can be 33 km if the power level of the customer-premises equipment (CPE) is 4 W, and it can be extended to 100 km if higher power levels are allowed [6]. IEEE 802.22 standard is developed to bring broadband wireless access to wide-range rural areas operating in TV white spaces from 54 to 862 MHz, on a noninterfering basis with the primary users. It contains unique features, e.g., spectrum sensing, geolocation and intrasystem coexistence for CR-based operations [15]. n Limited Bandwidth: Some of the SG applications, e.g., AMI, real-time pricing, and power network protection and control, need greater aggregated bandwidths for data transmissions; however, wireless communication channels can provide limited bandwidth in harsh SG environments. To this end, CR networks can improve the overall network performance with its adaptive operations to existing spectrum conditions in the deployment field, and thus, increase spectrum utilization efficiency in SG environments. However, the realization of CR networks for SG mainly requires efficient spectrum management functionalities for SG applications.

Spectrum Management Functionalities in CR-Based Smart Grid Applications The spectrum management functionalities, such as spectrum sensing, spectrum decision, spectrum sharing, and spectrum mobility, are vital for the cognitive cycle,


which is the required task to determine the accurate communication parameters of SG communication and adapt to the dynamic radio environments [2].

Spectrum Sensing Spectrum sensing is the process of becoming aware of the available spectrum bands and the presence of the PUs and detection of the spectrum holes [12]. Spectrum sensing operation is an extensive power-consuming approach and poses great challenges to provide cost-effective communications in large-scale SG deployments. Hence, some solutions need to be deployed to achieve feasible CR-based SG communications. Minimum hardware, e.g., using single radio, and nonadvanced spectrum sensing functionalities can be used to degrade the complexity level of sensing operations and minimize energy consumption. Moreover, minimizing the sensing durations to an optimum level can be a good solution until it affects the sensing accuracy. On the other hand, spectrum sensing can be a challenging task because of the uncertainty of the several resources for the realization of SG. For instance, CR can detect primary transmitters even if there are no primary transmitters at all. This misunderstanding is known as the false alarm that results in the low utilization of the spectrum, which is unpreferable for SG communications. Hence, successful differentiation at this point is critical for the success of CR-based SG applications. Furthermore, the other obstacle for SG communication is the noise power, which is an important measurement to be aware of the required detection sensitivity. An SG has harsh environmental conditions with multipath, fading, and environmental noise, which will create additional challenge to CR technology to be sensitive in detection mechanisms.

(MAC) protocols, which add redundant challenges, e.g., time synchronization [2]. For instance, power control methods are necessary for spectrum sharing process in wide-range SG deployments for the adoption to radio environments and maximize the network life-time [1]. The opportunistic spectrum access capability can be used to adjust the transmission parameters to reduce redundant power consumption of sensor nodes and, hence, prevent the performance degradation of SG communications. Furthermore, accurate time measurements and time synchronization may be required for some SG applications, e.g., phasor measurement monitoring applications and equipment fault diagnostics; however, maintaining network-wide time synchronization may be difficult to adopt.

Spectrum Mobility In spectrum mobility, spectrum handoff is used to overcome the interference caused by SG components. Spectrum handoff can be accomplished by changing the physical regions where the existing path passes or switches the currently used spectrum band [3]. In both cases, the QoS requirements of the current SG transmission will be affected. Hence, the decision of the switching activities should be made based on the requirements of different SG applications.

Open Research Issues in CR-Based Smart Grid Applications The implementation of reliable, secure, and efficient communication infrastructure for transmission of critical SG data is a fundamental step for the realization of SG applications. However, there are still open research issues and challenges that need to be resolved.

Quality of Service Spectrum Decision Spectrum decision process consists of two steps: spectrum characterization and spectrum selection, which are the vital steps to characterize the spectrum band in terms of the received signal strength, interference, energy efficiency, and transmission power, number of users, QoS, and security requirements of SG applications [2]. Hence, providing QoS-aware communication is highly important in choosing the appropriate spectrum band to meet the specific requirements of SG communications in spectrum decision process. However, SG system has a distributed nature, and the radio interference, network density, and channel characteristics vary over a wide-range geographical area, which limits obtaining enough knowledge about spectrum availability and network topology. As a result, this problem poses great challenges in making accurate spectrum decisions and meeting QoS requirements of SG applications.

CR-based SG applications have different QoS requirements in terms of reliability, latency, and data rate [11]. In addition, SG is a heterogeneous network, and it contains electric equipment that has dramatically different limitations, e.g., resource limitations in terms of computing power and storage capability. However, it is still an open research issue to design QoS-aware communication protocols capable of meeting application requirements in a scalable way. To meet the QoS requirements for SG applications, the communication technology should provide high bandwidth and low-latency and low-cost communications between SG components. In CR networks, the data rate, acceptable error rate, delay bound, transmission mode, and bandwidth of the transmission can be determined based on the application requirements; hence, spectrum management functions can be determined accordingly to choose the appropriate spectrum band providing QoS requirements [4].

Interoperability Spectrum Sharing Spectrum sharing process focuses on the selection of the best channel and power allocation, and some of the functionalities resemble the core functionalities of medium access control


SG requires advanced communication protocols among each of its component to exchange information independent from manufacture or any type of physical device. Hence, several communication technologies and different standards will be

||| 45

used to meet the specific QoS requirements of SG components and applications [14]. These communication technologies may demand operating on different spectrum bands. Hence, CR technology should accomplish the overall coordination between complex, different, faraway SG components, which is a very difficult task and imposes a great challenge to CR technology for SG applications. Standard-based and interoperable communication protocols can be good options to help CR technology to realize such a complex communication infrastructure.

Interference In SG environments, interference avoidance schemes should be applied to the CR networks [13]. The spectrum management cycle can eliminate this problem by providing spectrum sharing functionality. The interference between overlapped cells may lead to coexistence problem, which degrades the performance of the network. Hence, coordination between base stations is a must to overwhelm this problem [4]. The data transmission for SG applications should be highly reliable; however, in case of interference problem, the communication cannot be achieved with full bandwidth capacity, or it can be achieved at lower data rates. Adaptive power allocation and dynamic spectrum access schemes of CR networks can be used to overcome RF interference.

Dynamic Spectrum Usage After the selection of the best available channel for the required SG application, the next step is to make the network protocols adaptive to the chosen spectrum. However, the conventional communication protocols are designed considering the static spectrum allocations, and their performance might be adversely affected by the dynamic spectrum usage [1]. Hence, new methods and protocols should be developed for efficient spectrum usage for SG applications.

Conclusions Recently, CR network is recognized as a promising technology to address SG communication and networking problems. Dynamic spectrum access capability of CR networks can be benefited to address the unique challenges of SG, such as reliability and delay requirements, multipath fading, noise, different spectrum characteristics changing over location and time, and harsh environmental conditions. This article presents a comprehensive review of SG characteristics and CR-based SG applications. Also, architectures to support CR networks in SG applications, major challenges, and open issues have been discussed. We expect that this article will motivate the research community to further explore this promising research area.

Acknowledgments This work was supported by the European Union Seventh Framework Programme (FP7) Marie Curie International Reintegration Grant (IRG) under Grant PIRG05-GA-2009-249206.

46 |||

Author Information Vehbi Cagri Gungor received his Ph.D. degree in electrical and computer engineering from Broadband and Wireless Networking Laboratory, Georgia Institute of Technology, Atlanta. He worked as a project leader at Eaton Corporation, Innovation Center, Wisconsin. Currently, he is the codirector of Computer Networks and Mobile Communications Laboratory and the graduate programs (Ph.D. and M.S.) coordinator at the Department of Computer Engineering, Bahcesehir University, ´Istanbul, Turkey. He is the recipient of the IEEE International Symposium on Computer Networks 2006 Best Paper Award, the European Union FP7 Marie Curie IRG Award in 2009, Tu ¨ rk Telekom Research Grant Awards in 2010 and 2012, and the San-Tez Project Awards supported by Alcatel-Lucent Teleta+ and the Turkish Ministry of Science, Industry, and Technology in 2010. His research interests include SG communications, nextgeneration wireless networks, wireless ad hoc and sensor networks, CR networks, and Internet protocol networks. Dilan S¸ahin received her B.Sc. degree in computer engineering and software engineering (as a double major) from Bahcesehir University, _Istanbul, Turkey, in 2010. She is pursuing her M.Sc. degree at Computer Engineering Department, Bahcesehir University. She is currently a research assistant at Computer Networks and Mobile Communications Laboratory. Her research interests include SG communication networks, wireless sensor networks, and CR networks.

References [1] O. B. Akan, O. B. Karli, and O. Ergul, ‘‘Cognitive radio sensor networks,’’ IEEE Network, vol. 23, no. 4, pp. 34–40, Aug, 2009. [2] I. F. Akyildiz, W. Y. Lee, and K. Chowdhury, ‘‘Spectrum management in cognitive radio ad hoc networks,’’ IEEE Network, vol. 23, no. 4, pp. 6–12, July–Aug, 2009. [3] I. F. Akyildiz, L. Won-Yeol, M. C. Vuran, and S. Mohanty, ‘‘A survey on spectrum management in cognitive radio networks,’’ IEEE Commun. Mag., vol. 46, no. 4, pp. 40–48, Apr, 2008. [4] I. F. Akyildiz, W. Y. Lee, M. C. Vuran, and S. Mohanty, ‘‘Next generation/dynamic spectrum access/cognitive radio wireless networks: A survey,’’ Comput. Netw. J., vol. 50, pp. 2127–2159, Sept. 2006. [5] O. Fatemieh, R. Chandra, and C. A. Gunter, ‘‘Low cost and secure smart meter communications using the TV white spaces,’’ in Proc. 3rd Int. Symp. Resilient Control Systems (ISRCS), Aug. 2010, pp. 37–42. [6] A. Ghassemi, S. Bavarian, and L. Lampe, ‘‘Cognitive radio for smart grid communications,’’ in Proc. IEEE Smart Grid Communications, Oct. 2010, pp. 297–302. [7] V. C. Gungor, B. Lu, and G. P. Hancke, ‘‘Opportunities and challenges of wireless sensor networks in smart grid,’’ IEEE Trans. Ind. Electron., vol. 57, no. 10, pp. 3557–3564, Oct. 2010. [8] V. C. Gungor, D. Sahin, T. Kocak, S. Ergut, C. Buccella, C. Cecati, and G. P. Hancke, ‘‘Smart grid technologies: Communications technologies and standards,’’ IEEE Trans. Ind. Informatics, vol. 7, no. 4, pp. 529–539, Nov. 2011. [9] K. Moslehi and R. Kumar, ‘‘Smart grid: A reliability perspective,’’ in Proc. Innovative Smart Grid Technologies (ISGT), Jan. 2010, pp. 1–8. [10] R. C. Qiu, Z. Hu, Z. Chen, N. Guo, R. Ranganathan, S. Hou, and G. Zheng, ‘‘Cognitive radio network for the smart grid: Experimental system architecture, control algorithms, security, and microgrid testbed,’’ IEEE Trans. Smart Grid, vol. 2, no. 4, pp. 724–740, Dec. 2011. [11] R. Yu, Y. Zhang, S. Gjessing, C. Yuen, S. Xie, and M. Guizani, ‘‘Cognitive radio based hierarchical communications infrastructure for smart grid,’’ IEEE Network, vol. 25, no. 5, pp. 6–14, Sept.–Oct. 2011. [12] T. Wysocki and A. Jamalipour, ‘‘Spectrum management in cognitive radio: Applications of portfolio theory in wireless communications,’’ IEEE Wireless Commun., vol. 18, no. 4, pp. 52–60, Aug. 2011. [13] W.-Y. Lee and I. F. Akyildiz, ‘‘Optimal spectrum sensing framework for cognitive radio networks,’’ IEEE Trans. Wireless Commun., vol. 7, no. 10, pp. 3845–3857, Oct. 2008. [14] J. Wang, M. Ghosh, and K. Challapali, ‘‘Emerging cognitive radio applications: A survey,’’ IEEE Commun. Mag., vol. 49, no. 3, pp. 74–81, Mar, 2011. [15] IEEE 802.22 Working Group. [Online]. Available: http://www.ieee802.org/22/.
