Search

• This work proposes a parallel and distributed deep learning framework for IoT data analytics. • The proposed method is implemented and deployed with a real-world problem of energy consumption forecasting. • Comprehensive comparative study has been conducted to show the outperformance of the proposed method. Energy consumption forecasting based on IoT data and deep learning algorithm inheriting distributed and collaborative learning is a widely studied topic both in engineering and computer science fields. For different households with drastically different energy consumption patterns, the traditional centralized machine learning (ML) and deep learning (DL) methods suffer problems including inaccuracy, inefficiency and laggings of the prediction performance. In this study, we propose a sophisticated multi-channel bidirectional nested LSTM framework (MC-BiNLSTM) combined with discrete stationary wavelet transform (SWT) for highly accurate and efficient energy consumption forecasting. The main contributions of this study include the decomposition using SWT for accuracy improvement and the collaborative BiNLSTM structure for efficiency improvement. A real-world IoT energy consumption dataset, named UK-DALE, is adopted for the comparative study. The experimental results showed the outperformance of the proposed method from various perspectives over the cutting-edge methods existed in the literature. [ABSTRACT FROM AUTHOR]

Blockchain has received a widespread attention because of its decentralized, tamper-proof, and transparent nature. Blockchain works over the principle of distributed, secured, and shared ledger, which is used to record, and track data within a decentralized network. This technology has successfully replaced certain systems of economic transactions in organizations and has the potential to overtake various industrial business models in future. Blockchain works over peer-to-peer (P2P) phenomenon for its operation and does not require any trusted-third party authorization for data tracking and storage. The information stored in blockchain is distributed throughout the decentralized network and is usually protected using cryptographic hash functions. Since the beginning of blockchain technology, its use in different applications is increasing exponentially, but this increased use has also raised some questions regarding privacy and security of data being stored in it. Protecting privacy of blockchain data using data perturbation strategy such as differential privacy could be a novel approach to overcome privacy issues in blockchain. In this article, we cover the topic of integration of differential privacy in each layer of blockchain and in certain blockchain based scenarios. Moreover, we highlight some future challenges and application scenarios in which integration of differential privacy in blockchain can produce fruitful results. • We provide brief discussion about differential privacy and motivation behind using it as a privacy preserving strategy. • We provide the importance of privacy preservation in blockchain and its applications. • We provide detailed analysis over integration of differential privacy in blockchain scenarios. • We presented future research directions for differential privacy integration with blockchain applications. [ABSTRACT FROM AUTHOR]

IoT Mashup applications allow developer to compose existing Web APIs to create value-added composite Web services. The rapid growth of large-scale and complex services makes it difficult to find suitable Web APIs to build IoT Mashup applications for developers. Even if the existing service recommendation methods show improvements in service discovery, the accuracy of them can be significantly improved due to overlooking the impact of sparsity and multiple-dimension information of QoS between Mashup and services on recommendation accuracy. In this paper, we propose a QoS-aware service recommendation based on relational topic model and factorization machines for IoT Mashup applications. This method first uses relational topic model to characterize the relationships among Mashup, services, and their links, and mine the latent topics derived by the relationships. Second, it exploits factorization machines to train the latent topics for predicting the link relationship among Mashup and services to recommend adequate relevant top-k Web APIs for target IoT Mashup creation. Finally, we conduct a comprehensive evaluation to measure performance of our method. Compared with other existing recommendation approaches, experimental results show that our approach achieves a significant improvement in terms of precision, recall, and F-measure. • Use RTM to model the historical interactions between IoT Mashups and services. • Exploit FMs to predict the interactions relationship among IoT Mashups and services. • Develop a real-world dataset from ProgrammableWeb and conduct a set of experiments. [ABSTRACT FROM AUTHOR]

Big data sensing system (BDSS) plays an important role in the Internet of Things, in which how to reduce power consumption is one crucial problem. Currently, low energy adaptive clustering hierarchy (LEACH) protocol is one well-known algorithm used in BDSS with low energy cost. In this paper, a new variant of bat algorithm combined with centroid strategy is introduced. Three different centroid strategies with six different designs are introduced. In addition, the velocity inertia-free update equation is also provided. The optimization performance is verified by CEC2013 benchmarks in those designs against standard BA. Simulation results prove that the bat algorithm with weighted harmonic centroid (WHCBA) strategy is superior to other algorithms. By integrating WHCBA into LEACH protocol, we develop a two-stage cluster-head node selection strategy and can save more energy compared to the standard LEACH protocol. • Modified bat algorithm with centroid strategy is designed. • Six different centroid strategies are designed and compared. • Velocity inertia-free update equation is designed. • Bat algorithm with weighted harmonic centroid strategy is used to optimize the performance of LEACH protocol, and compared with other four algorithms. [ABSTRACT FROM AUTHOR]

The increasing energy costs and increase in losses in traditional power grid system triggered the integration of Renewable Energy Resources (RERs) in smart homes. The global desire of consumers to rely on RERs such as solar energy, and wind energy has increased dramatically. Similarly, the IT technologies are also playing their part in smart grid development, such as real time data monitoring. On the other hand, with the advancement of these IT technologies in smart meters, the privacy of customers is also at risk Smart grid utility knows the exact generation of any specific renewable resource in a specific interval of time. Utility need to monitor this real time data for load forecasting and implementation of demand response scenarios. However, the utility may misuse the data and may increase the prices for specific time slots when RERs are not present. Similarly, real time monitoring of data can lead to estimation of life routines of users such as sleeping habits, time of usage of heavy appliances, and lifestyle. In this paper, a Differential Privacy based real time Load Monitoring approach (DPLM) is proposed that preserve the privacy of users by masking the values of load in such a way that utility will not be able to judge the usage of specific RER and the daily routine of any smart meter user. We compare our scheme with Gaussian Noise Differential Privacy (GNDP) strategy. Experimental results validate that our DPLM approach provides a desirable solution to protect smart grid user's privacy by efficient noise addition and peak value protection along with having an error rate of only 1.5%. • We discuss integration of differential privacy in renewable energy based smart grid. • We provide importance of privacy in renewable energy resources based smart meters. • We provide analysis of privacy preservation strategies implemented in smart grid. • We presented simulation work over real-time RER based smart metering data. • We evaluated that our DPLM algorithm outperforms previously proposed strategies [ABSTRACT FROM AUTHOR]

Wireless sensor network (WSN) is an important component of a cyber–physical system. Locating node information is a crucial problem for WSN. Currently, distance vector-hop method (DV-Hop), one of popular range-free algorithms, is widely deployed to estimate the location. However, the estimation precision is challenging. In this paper, a new evolutionary algorithm named oriented cuckoo search algorithm (OCS) is designed. In OCS, the global search capability is dominated by the combination of two different random distributions. To provide a deep investigation, ten different random distributions are employed and compared with CEC2013 test suits. Numerical results show the hybrid distribution combined with Lévy distribution and Cauchy distribution achieves the best performance. Furthermore, OCS with this hybrid distribution is also incorporated into the methodology of DV-Hop algorithm to improve the precision performance. Simulation results demonstrate that our modification achieves better precision performance when compared with three other DV-Hop algorithms. [ABSTRACT FROM AUTHOR]

Abstract: Many scientific workflows are data intensive: large volumes of intermediate datasets are generated during their execution. Some valuable intermediate datasets need to be stored for sharing or reuse. Traditionally, they are selectively stored according to the system storage capacity, determined manually. As doing science on clouds has become popular nowadays, more intermediate datasets in scientific cloud workflows can be stored by different storage strategies based on a pay-as-you-go model. In this paper, we build an intermediate data dependency graph (IDG) from the data provenances in scientific workflows. With the IDG, deleted intermediate datasets can be regenerated, and as such we develop a novel algorithm that can find a minimum cost storage strategy for the intermediate datasets in scientific cloud workflow systems. The strategy achieves the best trade-off of computation cost and storage cost by automatically storing the most appropriate intermediate datasets in the cloud storage. This strategy can be utilised on demand as a minimum cost benchmark for all other intermediate dataset storage strategies in the cloud. We utilise Amazon clouds’ cost model and apply the algorithm to general random as well as specific astrophysics pulsar searching scientific workflows for evaluation. The results show that benchmarking effectively demonstrates the cost effectiveness over other representative storage strategies. [Copyright &y& Elsevier]