Initiatives & Trends

On July 6 a technical study titled: Keys Under Doormats: Mandating Insecurity by Requiring Government Access to All Data and Communications has been published. The scientists behind this work are more than domain experts. Among the others: Steven M. Bellovin known for his work in the encrypted keys exchange; Josh Benaloh who invented the Benaloh cryptosystem; Matt Blaze who worked on cryptographic files systems; Whitfield Diffie universally known as one the pioneers of public key cryptography; Peter G. Neumann who worked on Provably Secure Operative Systems; Ronald L. Rivest who is “R” in RSA; Jeffrey I. Schiller known for being one of the creator of Kerberos and Bruce Schneier one the biggest expert in security and cryptography.

For our July 2015 issue, we cover recent patents dealing with various applications of adaptive filtering. The section below covers patents granted recently for reducing blocking effect and ringing noise, radio frequency spektrum analysis, filter bank implementation, signal amplifying, low complexity filtering, magnetic fluid analysis, signal canceling and filter representation.

The special issue of the Proceedings of the IEEE in July is the second issue focusing on reconfigurable systems. In the first special issue (vol. 103, no. 3, 2015) introduced by eNewsletter in June, the foundational concepts are addressed. In this special issue, the editors focus on the basic concepts discussed in the first issue by considering more advanced applications, the extension of reconfigurability to other phenomenologies/domains, and the impacts of selected emerging technologies.

Machine learning, a subfield of computer science, has been widely applied in many areas from science to engineering to many interdisciplinary fields. Nature Reviews Genetics recently published an article that summarized machine learning applications in genetics and genomics, authored by University of Washington researchers Maxwell W. Libbrecht and William Stafford Noble. The topics of interest included supervised versus unsupervised learning, generative versus discriminative modeling, incorporating prior knowledge, handling heterogeneous data, feature selection, imbalanced class sizes, handling missing data, and modelling dependence among examples.