Over the last two decades, a lot of work has been done in improving network security, particularly in intrusion detection systems (IDS) and anomaly detection. Machine learning solutions have also been employed in IDSs to detect known and plausible attacks in incoming traffic. Parameters such as packet contents, sender IP and sender port, connection duration, etc. have been previously used to train these machine learning models to learn to differentiate genuine traffic from malicious ones. Generative Adversarial Networks (GANs) have been significantly successful in detecting such anomalies, mostly attributed to the adversarial training of the generator and discriminator in an attempt to bypass each other and in turn increase their own power and accuracy. However, in large networks having a wide variety of traffic at possibly different regions of the network and susceptible to a large number of potential attacks, training these GANs for a particular kind of anomaly may make it oblivious to other anomalies and attacks. In addition, the dataset required to train these models has to be made centrally available and publicly accessible, posing the obvious question of privacy of the communications of the respective participants of the network. The solution proposed in this work aims at tackling the above two issues by using GANs in a federated architecture in networks of such scale and capacity. In such a setting, different users of the network will be able to train and customize a centrally available adversarial model according to their own frequently faced conditions. Simultaneously, the member users of the network will also able to gain from the experiences of the other users in the network. The training is performed in a coordinated and localized setting, which will also eliminate the need to collect data from users into a central storage and ensure privacy. On a whole, the proposed work introduces an architecture and mechanism for a well-coordinated intelligent intrusion detection system, addressing concerns of large-scale shared model updates, date privacy and model tampering.
The Nellore Schist Belt (NSB) is a curvilinear Archaean schist belt, approximately 350 km long and 8-50 km wide. The Nellore Schist Belt is considered to be Neoarchean in age and stratigraphically NSB is classified as the western Udayagiri group (dominated by metasediments) and underlying eastern Vinjamuru group (dominated by metabasalts). There is a long controversy regarding the contact relationship between Udayagiri and Vinjamuru groups. Earlier researchers regarded the contact between two groups as tectonic on the basis of metamorphism. A shear zone and a possible thrust contact between the two groups have also been reported. On the basis of present study, an NNW-SSE trending, westerly dipping inclined transpressional zone is found at the contact between Udayagiri and Vinjamuru groups in the central western part of the NSB. Kinematic analysis of both the hanging wall and foot wall of the westerly dipping thrust zone shows presence of strong S1 schistosity, shear bands and S-C fabric in both strike and dip section along with east-verging overturned fold, westerly dipping inverted beds, suggesting partitioning of non-coaxial deformation in strike-slip and dip-slip component along with a pure shear component. Strike-slip is more prominent in the northern part of the contact than the southern part. The presence of steep to moderate northerly plunging non-orthogonal stretching/mineral elongation lineation all along the contact and clockwise shift of plot of the same in stereo net from its orthogonal position and presence of other kinematic indicators in plan suggests a right lateral strike-slip component. As a whole, it is suggested that Udayagiri group is thrusted over Vinjamuru group along a westerly dipping thrust plane with a right lateral strike-slip motion and simultaneous E-W contraction.
Intensely deformed Proterozoic Nallamalai fold belt (NFB) occupies the eastern part of undeformed, Proterozoic Cuddapah basin and is separated from Cuddapah basin by easterly dipping Rudravaram thrust line. Igneous activity in NFB is represented by intrusive syenite and lamproite dykes. This study brings out the petrological and geochemical character of unreported intrusive dolerite dykes and sills within the Cumbum Formation of NFB. The dykes are undeformed, whereas the sills suffered D1 phase of deformation, although both are petrologically and geochemically similar. The rocks are highly fractionated sub-alkaline tholeiite, consist mainly of clino-pyroxene and plagioclase. Chondritenormalised rare earth element (REE) plot shows enriched light rare earth element (LREE) and flat heavy rare earth element (HREE). A primitive mantle-normalised multi-element spider diagram shows troughs in Nb, Ta, Ti and Zr indicating subduction zone character. Tectonic discrimination plots indicate both within-plate and subduction zone environment of formation. Geochemical modelling also indicates enriched spinel peridotite as a probable source for dykes and sills. We suggest that the lithospheric mantle beneath NFB got enriched by subduction-derived fluid around 1.6 Ga, and in later stage, these dykes and sills were generated by extension of the enriched lithosphere at different phases in turn inheriting the subduction zone geochemical characters.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.