Earticle

The computer system is becoming more complex and massive network data, which brings great difficulties to the traditional intrusion detection system. Intrusion detection system is an important part of the network and information security architecture, which is mainly used to distinguish the normal activities of the system and the suspicious and intrusion patterns. But the challenge is how to effectively detect network intrusion behavior in order to reduce the false alarm rate and false negative rate. Based on the shortcomings of existing intrusion detection methods, the fuzzy C- means clustering method is proposed to analyze the intrusion detection data, so as to find out the abnormal network behavior patterns. By testing the CUP99 data set, the results show that the IFCA is not only feasible but also accurate and efficient. The improved fuzzy clustering algorithm proposed in this paper can improve the detection rate of intrusion detection and reduce the false detection rate, and can be widely used in intrusion detection system.

The internet of things (IoT) is a technology exchanging and sharing information without an operation and a help of a human. The technology is commercialized rapidly. However, IoT processes and transmits various forms of data at complex and different networks and it is exposed to the risks of information leakage, manipulation, and falsification. Moreover, the security vulnerability of IoT increased even further because IoT is composed of many low-performance devices and it is hard to apply the conventional security technologies to these low-performance devices. This study proposed a security method using a hash function and a symmetric encryption method to enhance confidentiality, mutual authentication, and data privacy under the IoT environment with limited resources. The proposed method authenticated a device by using a secret key, a device ID, and a random number and produced a session key, which was necessary to encrypt the transmitted data. This method could properly respond to the counterfeit, falsification, disguise, and replay attacks and secure the confidentiality of the data.

Network big data is an important driving force for the upgrading of information technology, and network data has brought great opportunities to the economic survey. The openness and freedom of the network also produced the possibility of private information and data to be destroyed or violated, and the security of the Internet is becoming more and more important. In this paper, the author analyzes the impact of fiscal expenditure on the new urbanization and income gap by using data mining technology. The result shows that the public expenditure increase 1% will lead urbanization rate increase 0.0462%; also, the effect of public finance expenditure on urbanization has a certain time lag, and the effect will be more obvious in the long term. At the same time, public expenditure will help to narrow the income gap in the long term, public expenditure at lag 2 increase 1% will lead income gap decrease 0.758%. So that, increasing public expenditure is an important way to promote the equalization of basic public services between urban and rural areas.

With the popularity of large data network, data mining technology can be used in human resource management and provide strong decision support. In this paper, the author put forward the application of BIRCH clustering algorithm in HRM database, and analyze the associated reasoning methods. Based on the multiple regression analysis, the result shows that strategic human resource management effectiveness has significant positive correlation with organizational innovation (p<0.01). At the same time, the model M6 show that the nature of the enterprise has a significant negative impact on organizational learning, and the industry has positive impact on organizational learning. The conclusion shows that organizational learning plays a complete intermediary role between strategic human resource management and management innovation.

In this work, a quick authentication scheme is implemented to prevent Phishing and DNS spoofing. In DNS spoofing, attackers inject the fake DNS server by duplicating the IP addresses and fake server redirect network traffic to wrong destinations. In phishing, phishers clone the legitimate website and user think that it is original website and users giving away their username and passwords to attacker’s website and attackers hack their confidential information and they can misuse it for financial gain, identity theft, gaining fame, malware distribution and industrial espionage. We host the phishing website but we cannot pass the link through common hosting websites like Google and Facebook. So, phishers force the legitimate users to open a phished link with the DNS spoofing through fake DNS server then user directly redirect to a fake server. So our proposed work is to prevent DNS spoofing, to prevent the Phishing attacks by isolating it using 64-bit time synchronized public key encryption.

In this paper, we investigate the secrecy outage probability (SOP) for the cooperative amplify-and-forward (AF) two-way relay networks (TWRN) with channel estimation (CE) errors under Rayleigh fading channels. In the proposed system, two sources simultaneously broadcast their respective information to all the relays firstly. Then, the optimal relay performs AF protocol on the received signals sent from the two sources and forward them to both sources. In both phases, An eavesdropper receives the signals from sources and relay nodes illegally. The performance of the proposed system is quantified by deriving the upper bound of the overall outage probability. Impact of CEs on the system is illustrated by simulations.

Covert channels are among the serious and most complicated security flaws that attempt to broke the access control policies, since they allow an unplanned communication medium within an information system. The important impact of this vulnerability gives it a wide interest by the researchers. In this paper, we propose a new approach based on an enhanced version of the shared resources matrix method, where the principle of mutation analysis is integrated to allow an efficient and fast detection of covert channels. We have conducted an evaluation within a real system and the obtained results show very promising performances.

With the rapid development of the Internet, especially the development of mobile Internet, network size and network infrastructure changes, especially the more random access node to detect network attacks challenges traditional detection algorithm using neural networks, so that the calculated volume increased dramatically, and will detection accuracy is not high, for this problem, we use the stochastic process theory hidden Markov chain transformation neural network, so that the convergence of a substantial decline, while improving network attack detection accuracy degree.

The detector performance depends on the clock accuracy and the model, the clock parameters estimation error will lead to a decline in the performance of the detectors. Therefore, in order to reach the detector achieves a set of performance, it need to adjust the receiver of the crystal quality. In this paper, the method of detecting the clock states of the satellite navigation receiver is used to detect the receiver. The clock state model of the receiver is established in the static and motion mode, and the time variation of the clock state is analyzed.