The recent advances in big data and cloud computing have deeply changed people's daily lives. The implementation of such techniques addresses various fields and makes things ‘smart’ in different domains, such as data mining, security, and web technologies. The concept of smart cloud refers to the intelligent controls over cloud servers to enable resource optimisations and quality cloud services. The application of a smart cloud can be also attached to the environment of smart computing that can be attached to data and security subjects. Therefore, it is necessary to address these research issues for the effective and efficient design and development of smart computing and communications.
To fill this gap, this special issue, entitled ‘Special Issue on Smart Computing and Communication’, tackles these research issues by exploring novel and bold ideas of emerging technologies for the smart computing and communication techniques in the big data and cloud computing systems. We received nearly 70 submitted manuscripts for this SI, reflecting the importance of the SI theme in the big data and cloud systems. After a rigorous peer review done by at least 2 or 3 reviewers and at least one round revision (many papers have undergone the two-round or three-round revision), only 14 papers have been accepted. We summarised the accepted papers in this editorial as follows.
The first article studies a secure file sharing mechanism for the cloud with proxy re-encryption implemented with the disintegration protocol to secure storage data in storage and in the flight. The second article proposes a simplified particle algorithm based on the nonlinear decrease extreme disturbance and Cauchy mutation.
Nowadays, supercomputers are more and more powerful but the power consumption is also rapidly increasing. In order to investigate this issue, the third article presents the energy measurement tool experimented by a software framework that eases the energy collection in a cluster environment and the authors also show how to optimise the parallel I/O operation's energy efficiency. As a widely used diagnosis of macular disease, Optical Coherence Tomography (OCT) often needs to check and analyse lots of OCT images which is time-consuming and inefficient. To solve this issue, the fourth article proposes an approach to detect the normal or abnormal status of the retina by OCT images that can differentiate several kinds of macular diseases from the normal image, including macular oedema, macular hole, and age-related macular degeneration. The fifth article reviews various networks used real-time media distribution includes the average delay for multiple video streams over Flexilink using a simulation model and the explanation for the recommendation for testing Flexilink using hardware.
Machine learning techniques are also playing an important role in smart computing for big data and cloud computing. To address this research topic, the sixth article studies the Deep Learning Firefly Algorithm (DLFA) to optimise the parameters of Support Vector for Regression (SVR). The seventh article explores the advantages of Sentinel-2A RENDVI in crop identification, and then uses the supervised classification SVM algorithm, object-oriented classification method and assisted with field measured data to extract the winter wheat planting area.
Big data generates urgent needs for a large amount of data retrieval in the distributed systems like clouds. To fill this gap, the eighth article studies the usage of fog computing to efficiently manage the Internet service and resources. The cost and resources can be reduced and the billing can be attained through a calculation of the utilised resources. The ninth article presents a research on the total amount of literature in the treatment of cardiovascular diseases by traditional Chinese medicine is on the rise. The tenth article proposes the distributed dynamic cuckoo filter system based on Redis cluster and an efficient hash indexing structure – dynamic cuckoo filter was presented. The eleventh article presents a new attributes recognition model using Feature Nets that can improve emotion recognition performance and generalisation capabilities. Meanwhile, in order to accelerate the big data process, the twelfth article studies a new NMF-based mixed data analysis algorithm with maximum overall coverage constraints introduced in traditional NMF. The thirteenth article studies an unsupervised tool for automatically reverse the format of the industrial protocol from network trace that applies an extended voting expert algorithm to infer the boundaries of industrial protocol fields. The fourteenth article gives the new design of the real-time operating system scheduling mechanism based on time deterministic cyclic scheduling for live multimedia tasks processing in the real-time hyper-physical system.
We would like to express our sincere gratitude to the reviewers who voluntarily participated in the reviewing process. We also thank all the authors who submitted their manuscripts to this special issue. Finally, we sincerely appreciate the support from the Editor-in-Chief, Professor Andrew I. Adamatzky of International Journal of Parallel, Emergent and Distributed Systems.