INTRODUCTION BINOMIAL DATA Challenger Disaster Example Binomial Regression Model Inference Tolerance Distribution Interpreting Odds Prospective and Retrospective Sampling Choice of Link Function Estimation Problems Goodness of Fit Prediction and Effective Doses Overdispersion Matched Case-Control Studies COUNT REGRESSION Poisson Regression Rate Models Negative Binomial CONTINGENCY TABLES Two-by-Two Tables Larger Two-Way Tables Matched Pairs Three-Way Contingency Tables Ordinal Variables MULTINOMIAL DATA Multinomial Logit Model Hierarchical or Nested Responses Ordinal Multinomial Responses GENERALIZED LINEAR MODELS GLM Definition Fitting a GLM Hypothesis Tests GLM Diagnostics OTHER GLMS Gamma GLM Inverse Gaussian GLM Joint Modeling of the Mean and Dispersion Quasi-Likelihood RANDOM EFFECTS Estimation Inference Predicting Random Effects Blocks as Random Effects Split Plots Nested Effects Crossed Effects Multilevel Models REPEATED MEASURES AND LONGITUDINAL DATA Longitudinal Data Repeated Measures Multiple Response Multilevel Models MIXED EFFECT MODELS FOR NONNORMAL RESPONSES Generalized Linear Mixed Models Generalized Estimating Equations NONPARAMETRIC REGRESSION Kernel Estimators Splines Local Polynomials Wavelets Other Methods Comparison of Methods Multivariate Predictors ADDITIVE MODELS Additive Models Using the gam Package Additive Models Using mgcv Generalized Additive Models Alternating Conditional Expectations Additivity and Variance Stabilization Generalized Additive Mixed Models Multivariate Adaptive Regression Splines TREES Regression Trees Tree Pruning Classification Trees NEURAL NETWORKS Statistical Models as NNs Feed-Forward Neural Network with One Hidden Layer NN Application Conclusion APPENDICES Likelihood Theory R Information Bibliography Index

Extending the linear model with R : generalized linear, mixedeffects and nonparametric regression models

With the expeditious development of Internet, the security of communication is one of the most essential elements.Internet Key Exchange(IKE), resolves problems of building and updating key that is shared in the unsafe environment such as Internet.

Internet Key Exchange

The use of the DNS as the underlying technology of new resolution name services can lead to privacy violations. The exchange of data between servers and clients flows without protection. Such an information can be captured by service providers and eventually sold with malicious purposes (i.e., spamming, phishing, etc.). A motivating example is the use of DNS on VoIP services for the translation of traditional telephone numbers into Internet URLs. We analyze in this paper the use of statistical noise for the construction of proper DNS queries. Our objective aims at reducing the risk that sensible data within DNS queries could be inferred by local and remote DNS servers. We evaluate the implementation of a proof-of-concept of our approach. We study the benefits and limitations of our proposal. A first limitation is the possibility of attacks against the integrity and authenticity of our queries by means of, for instance, man-in-the-middle or replay attacks. However, this limitation can be successfully solved combining our proposal together with the use of the DNSSEC (DNS Security extensions). We evaluate the impact of including this complementary countermeasure.

Anonymous Resolution of DNS Queries

Authentication is an important mechanism for the reliable operation of any Voice over IP (VoIP) infrastructure. Digest authentication has become the most widely adopted VoIP authentication protocol due to its simple properties. However, even this lightweight protocol can have a significant impact on the performance and scalability of a VoIP infrastructure. In this paper, we present Proxychain - a novel VoIP authentication protocol based on a modified hash chain construction. Proxychain not only improves performance and scalability, but also offers additional security properties such as mutual authentication. Through experimental analysis we demonstrate an improvement of greater than 1700% of the maximum call throughput possible with Digest authentication in the same architecture. We show that the more efficient authentication mechanisms of Proxychain can be used to improve the overall security of a carrier-scale VoIP network.

/pdf/proxychain-developing-a-robust-and-efficient-authentication-3k5qycousl.pdf

Proxychain: developing a robust and efficient authentication infrastructure for carrier-scale VoIP networks

Graph perturbations and corresponding spectral changes in Internet topologies

IPSec provides security solutions to various network based web applications through its protocol suite. Implementing IPSec in the network layer includes significant bandwidth and processing time overhead to application traffic. The upper layer protocols such as TCP and UDP behavior on network traffic is also influenced by these overheads. In our work, we have studied the impact of IPSec on the metrics: Number of Transactions, Transmission Time, Round Trip Time, Average Burst Size and Available Bandwidth for IPv4 and IPv6 network traffic. Authentication algorithm: Hashed Message Authentication Code - Message Digest 5(HMAC-MD5), encryption algorithm: Advanced Encryption Standard(AES) and compression algorithm deflate are considered for Authentication Header(AH), Encapsulated Security Payload(ESP) and IP Compression(IPComp) protocols of IPSec. Using the characterized metrics, computed overheads and user behavior on web page requests, we have analyzed the performance impact of IPSec on web application servers.

Impact of IPSec Overhead on Web Application Servers

A new process for the partitioning of plutonium and uranium during the reprocessing of spent fuel discharged from fast reactor was optimised using hydroxyurea (HU) as a reductant. Stoichiometric ratio of HU required for the reduction of Pu(IV) was studied. The effect of concentration of uranium, plutonium and acidity on the distribution ratio (Kd) of Pu in the presence of HU was studied. The effect of HU in further purification of Pu such as solvent extraction and precipitation of plutonium as oxalate was also studied. The results of the study indicate that Pu and U can be separated from each other using HU as reductant.

Partitioning of plutonium and uranium in aqueous medium using hydroxyurea as reducing agent

The spent fuel from Fast Breeder Test Reactor of various burnups from 25 to 155 GWd/te is being reprocessed in CORAL (COmpact Reprocessing of Advanced fuels in Lead shielded cell) using a modified PUREX (Plutonium Uranium Recovery by EXtraction) process. Total plutonium (Pu238, 239, 240, 241 & 242) concentration in the sample is analysed by HTTA (Thenoyl Trifluoro Acetone) extraction method wherever interference from other alpha emitting nuclides (Raffinate) and bulk natural uranium (uranium products) are present "as reported by Milyukov et al. (Analytical chemistry of plutonium, 1967) and Natarajan and Subba Rao (BARC, pp. 38–43, 2007)". This method requires the addition of corrosive reagents such as NH2OH.HCl which is a problem in waste disposal for reduction. A salt-free reagent such as Hydroxyurea is studied as a reducing agent which has the ability to reduce both Pu(VI) and Pu(IV) to Pu(III) "as reported by Zhaowu (260(3):601–606, 2004) and Zhaowu (262(3):707–711, 2004)". Pu(III) thus formed can be easily oxidised to Pu(IV) by NaNO2 for the extraction of Pu by HTTA.

Modified TTA method for the determination of plutonium

A spectrophotometric method has been developed for the determination of hydroxyurea. The methodology involves the reduction of standard potassium permanganate (KMnO4) prepared in 1N sulfuric acid medium by hydroxyurea (HU). KMnO4 is having λmax at 525 and 545 nm. The decrease in absorbance for a given KMnO4 solution by the addition of HU is compared to the solution without HU having the same concentration of KMnO4. Molar absorption coefficient (e) value is found to be 957 mol -1 L cm-1 with a precision of 3.8%. The correlation coefficient was 0.9975. The linear dynamic range of this method was found to be 10 to 80 ppm of HU. Effect of acidity and temperature on the stability of HU in nitric acid medium was also studied. Acid dependence on radiation effect of HU solution and dose dependence of stability of HU solution were also studied.

Spectrophotometric Determination of Hydroxyurea and Stability in Nitric Acid Medium

Spent fuel discharged from Fast Breeder Test Reactor (FBTR) in Kalpakkam is being reprocessed by modified plutonium uranium reduction extraction (PUREX) process using 30% TBP (tributylphosphate) as extractant in the presence of heavy normal paraffin (HNP) as diluent. Partitioning of uranium (U) and plutonium (Pu) is carried out using oxalate precipitation method. Uranium oxide product obtained by this method contains appreciable amount of plutonium which has to be recovered. Recovery of plutonium from this uranium oxide product is carried out by reducing Pu to inextractable Pu(III) using hydroxyurea (HU) and then uranium is extracted into 30% TBP. A small amount of Pu which is extracted in the organic phase is stripped back to aqueous phase by scrubbing with scrubbing agent containing 0.1 M HU in 4 M nitric acid. Similarly U and Pu are co-extracted into 30% TBP and then Pu is removed by scrubbing with 0.1 M HU in 4 M nitric acid. Further decontamination from Pu is obtained in the stripping stages. By this method Pu contamination in the uranium oxide is brought from 7300 ppm to 0.4–3 ppm (wt/wt). This uranium product obtained can be handled on table top.

S. Meenakshi

Papers

Impact of IPSec Overhead on Web Application Servers

Partitioning of plutonium and uranium in aqueous medium using hydroxyurea as reducing agent

Modified TTA method for the determination of plutonium

Spectrophotometric Determination of Hydroxyurea and Stability in Nitric Acid Medium

Purification and recovery of plutonium from uranium oxide obtained in the fast reactor fuel reprocessing plant using hydroxyurea as reductant