While software defined network (SDN) brings more innovation to the development of future networks, it also faces a more severe threat from DDoS attacks. In order to deal with the single point of failure on SDN controller caused by DDoS attacks, we propose a framework for detection and defense of DDoS attacks in the SDN environment. Firstly, we deploy a trigger mechanism of DDoS attack detection on data plane to screen for abnormal flows in the network. Then, we use a combined machine learning algorithm based on K-Means and KNN to exploit the rate characteristics and asymmetry characteristics of the flows and to detect the suspicious flows determined by the detection trigger mechanism. Finally, the controller will take corresponding actions to defense against the attacks. In this paper, we propose a new framework of cooperative detection methods of control plane and data plane, which effectively improve the detection accuracy and efficiency, and prevent DDoS attacks on SDN. INDEX TERMS Software Defined Network, Distributed Denial of Service (DDoS), collaborative detection, traffic characteristics, detection trigger.

Background Droplet digital PCR (ddPCR) has emerged as a promising tool of pathogen detection in bloodstream infections (BSIs) in critical care medicine. However, different ddPCR platforms have variable sensitivity and specificity for diverse microorganisms at various infection sites. There is still a lack of prospective clinical studies aimed at validating and interpreting the discrepant ddPCR results for diagnosing BSI in intensive care unit (ICU) practice. Methods A prospective diagnostic study of multiplex ddPCR panels was conducted in a general ICU from May 21, 2021, to December 22, 2021. Paired blood cultures (BCs) and ddPCRs (2.5 h) were obtained synchronously to detect the 12 most common BSI pathogens and three antimicrobial resistance (AMR) genes. Firstly, ddPCR performance was compared to definite BSI. Secondly, clinical validation of ddPCR was compared to composite clinical diagnosis. Sensitivity, specificity, and positive and negative predictive values were calculated. Thirdly, the positive rate of AMR genes and related analysis was presented. Results A total of 438 episodes of suspected BSIs occurring in 150 critical patients were enrolled. BC and ddPCR were positive for targeted bacteria in 40 (9.1%) and 180 (41.1%) cases, respectively. There were 280 concordant and 158 discordant. In comparison with BCs, the sensitivity of ddPCR ranged from 58.8 to 86.7% with an aggregate of 72.5% in different species, with corresponding specificity ranging from 73.5 to 92.2% with an aggregate of 63.1%. Furthermore, the rate of ddPCR+/BC− results was 33.6% (147/438) with 87.1% (128 of 147) cases was associated with probable (n = 108) or possible (n = 20) BSIs. When clinically diagnosed BSI was used as true positive, the final sensitivity and specificity of ddPCR increased to 84.9% and 92.5%, respectively. In addition, 40 blaKPC, 3blaNDM, and 38 mecA genes were detected, among which 90.5% were definitely positive for blaKPC. Further, 65.8% specimens were predicted to be mecA-positive in Staphylococcus sp. according to all microbiological analysis. Conclusions The multiplexed ddPCR is a flexible and universal platform, which can be used as an add-on complementary to conventional BC. When combined with clinical infection evidence, ddPCR shows potential advantages for rapidly diagnosing suspected BSIs and AMR genes in ICU practice.

No abstract

The paper studied controlling problem of an underactuated surface vessel with unknown interferences. It proved that the control problem of underactuated surface vessel can be transformed into the stabilization analysis of two small subsystems. This controller was designed by backstepping method and adaptive sliding mode, was suitable for solving the problem of the control of higher systems, can keep the system global asymptotic stability, and can inhibit unknown interference, and boundary layer can weaken the buffeting generated by sliding mode. The unknown interference was estimated by adaptive function. Finally, the simulation results are given to demonstrate the effectiveness of the proposed control laws.