The detection of primary user signals is essential for optimum utilization of a spectrum by secondary users in cognitive radio (CR). The conventional spectrum sensing schemes have the problem of missed detection/false alarm, which hampers the proper utilization of spectrum. Spectrum sensing through deep learning minimizes the margin of error in the detection of the free spectrum. This research provides an insight into using a deep neural network for spectrum sensing. A deep learning based model, “DLSenseNet”, is proposed, which exploits structural information of received modulated signals for spectrum sensing. The experiments were performed using RadioML2016.10b dataset and the outcome was studied. It was found that “DLSenseNet” provides better spectrum detection than other sensing models.

The global demand for energy is increasing daily with the expansion of energy infrastructure and the addition of new appliances. Efficient Energy Management System (EMS) is the need of the day. All residential and commercial buildings can achieve better energy efficiency and consumption with the use of EMS. Load forecasting is one of the methods to enable EMS to work efficiently. The accuracy of load forecast depends on many factors. The load forecast model must consider the weather forecast for the region in developing an accurate forecast. This paper develops Artificial Neural Network (ANN) and Bagged Regression Trees to generate and predicted load forecast in Urban area using Meteorological data. ANN model is compared with Bagged Regression Trees for prediction accuracy. Good agreement was observed by comparing these results with those available in the literature. It has been observed through analysis that Bagged Regression Trees produce better load prediction for the day ahead load in the urban area.

Real time data communication in Smart Grid is a big challenge due to its vastness and complexities. Wireless Regional Area Network (WRAN) - IEEE 802.22 permits opportunistic use of television white space (TVWS) by secondary user, if it is not being utilized by primary user. This standard has the potential to transfer big data of smart grid in real time to control center. The IEC 61850 is the standard for electrical substation automation that uses TCP/IP for data transfer among different `intelligent electronic devices' (IEDs). IEC 61850 consists of different message types which belong to different performance classes that are mapped to different communication protocols. The different performance classes are GOOSE, SV, GSSE (Generic substation status event) and MMS (Manufacturing Message Specification). GOOSE and SV are directly mapped on Ethernet for fast data transfer. The performance classes of Generic Substation State Event (GSSE) and Manufacturing Message Specification (MMS) can be used for teleprotection and transmitting synchrophasors over Wide Area Network (WAN). This paper reviews IEC 61850 and explores the possibility of IEEE 802.22 for teleprotection and transmitting synchrophasors over Wide Area Network (WAN) using GSSE and MMS. The superframe structure of IEEE 802.22 consists of 16 frames each of 10 ms. This 10 ms sub-frame in IEEE 802.22 is capable to send data on uplink frequency using upstream burst.

The existing electricity grid is subjected to numerous challenges that are expected to be overcome by Smart Grid. It is expected to provide utility companies with better visibility and pervasive control over their assets and services. There is no full-proof technology that can provide real-time communication in Smart Grid. This paper explores the possible infrastructure for real-time communication in Smart Grid with modern techniques. Reliable communication can be maintained with better Quality of services. The paper also proposes the innovative algorithm to decongest the network for optimum bandwidth utilization. It is possible to have real-time communication of Big data in Smart Grid through right messages priority and better bandwidth utilization technique of Cognitive Radio.

Author(s): Vasudev Dehalwara, Mohan Kolheb*, Sunita Kolheba Department of Computer Science and Engineering, MA Natio