As a new type of network structure, the Software Defined Network (hereinafter referred to as SDN) provides a new solution for network flow management and optimization, which has made the accurate detection of anomaly SDN flows a hot research topic. This paper presents an SDN-based flow detection method, builds structures for detecting anomaly SDN flows and performs classification detection on the flows using the DPTCM-KNN (Double P-value of Transductive Confidence Machines for K-Nearest Neighbors) algorithm. The experimental results show that the algorithm proposed achieves lower a false positive rate, higher precision and better adaptation to the SDN environment than do other algorithms of the same type.

As the important elements of the Internet of Things system, wireless sensor network (WSN) has gradually become popular in many application fields. However, due to the openness of WSN, attackers can easily eavesdrop, intercept, and rebroadcast data packets. WSN has also faced many other security issues. Intrusion detection system (IDS) plays a pivotal part in data security protection of WSN. It can identify malicious activities that attempt to violate network security goals. Therefore, the development of effective intrusion detection technologies is very important. However, many dimensions of the datasets of IDS are irrelevant or redundant. This causes low detection speed and poor performance. Feature selection is thus introduced to reduce dimensions in IDS. At the same time, many evolutionary computing (EC) techniques were employed in feature selection. However, these techniques usually have just one Candidate Solution Generation Strategy (CSGS) and often fall into local optima when dealing with feature selection problems. The self-adaptive differential evolution (SaDE) algorithm is adopted in our paper to deal with feature selection problems for IDS. The adaptive mechanism and four effective CSGSs are used in SaDE. Through this method, an appropriate CSGS can be selected adaptively to generate new individuals during evolutionary process. Besides, we have also improved the control parameters of the SaDE. The K-Nearest Neighbour (KNN) is used for performance assessment for feature selection. KDDCUP99 dataset is employed in the experiments, and experimental results demonstrate that SaDE is more promising than the algorithms it compares.

Overhead crane is regarded as a kind of complex nonlinear dynamic system in which the underactuated characteristic makes the controller hardly control the space of load swing effectively. In addition, the system internal and external uncertain factors also give rise to the adverse effect for control performance. Therefore, aiming at these problems, a type-2 fuzzy theory-based sliding mode controller (type-2 FSMC) is designed for damping the oscillation of the payload in which the proposed type-2 FSMC inherits the benefits of type-2 fuzzy logic theory and sliding mode control theory for improving anti-swing control performance. Furthermore, in order to configure the type-2 FSMC parameter efficiently, an adaptive differential evolutionary algorithm-based parameter tuning mechanism is presented for acquiring the optimal control performance. Through computer simulation and comparisons under different operation conditions, the optimal type-2 FSMC demonstrates the effectiveness in eliminating the payload residual vibration.

Channel estimation is a fundamental problem for downlink transmission in millimeter wave (mmWave) multiple input multiple output (MIMO) systems. This paper proposes a channel estimation algorithm by exploiting the separable structured sparsity of mmWave massive MIMO channel. The mmWave downlink channel is firstly formulated as a two dimensional (2D) separable compressive sensing (CS) model according to the sparsity structure of the channel in angle of arrivals (AoAs) and angle of departures (AoDs) domains. Then a separable compressive sampling match pursuit (SCoSaMP) algorithm is proposed to solve the separable CS recovery problem for channel estimation. Based on the separable sparsity structure of the channel, we design the precoding and combining matrices under the metric of mutual information to further improve the performance of channel estimation. Simulations demonstrate the advantages of the proposed algorithm over the traditional CS-based channel estimation methods.

For a long time, it has been frequently reported that outdoor adventurers get lost and die. During the course of an outdoor adventure, it is a matter of life and death to locate the lost adventurer and start the emergency rescue after determining that someone was lost. In this paper, a new wireless sensor network localization algorithm for outdoor adventures is proposed. This algorithm is notably easy to implement, with rather low implementation overhead. In the proposed localization algorithm, first, the Fermat point model is used to divide the original triangular pyramid constructed by the neighbor anchor nodes of the unknown node into four smaller sub-triangular pyramids. Next, the available sub-triangular pyramid in which the unknown node is located can be further divided into a set of irregularly shaped subspaces by the mid-perpendicular plane model. Finally, the centroid of the available subspace can be computed and regarded as the location of the unknown node. Simulation results show that the proposed localization algorithm can remarkably improve the localization accuracy.

Intelligent decision is the key technology of smart systems. Data mining technology has been playing an increasingly important role in decision making activities. Frequent itemset mining, as an important step of association rule analysis, is becoming one of the most important research fields in data mining. Weighted frequent itemset mining in uncertain databases should take both the existential probability and importance of items into account in order to find frequent itemsets of great importance to users. However, the introduction of weight makes the weighted frequent itemsets not satisfy the downward closure property any longer. As a result, the search space of frequent itemsets cannot be narrowed according to downward closure property which leads to a poor time efficiency. In this paper, the weight judgment downward closure property for weighted frequent itemsets and the existence property of weighted frequent subsets are introduced and proved first. Based on these two properties, the WD-FIM (Weight judgment Downward closure property based Frequent Itemset Mining) algorithm is proposed to narrow the searching space of weighted frequent itemsets and improve the time efficiency. Moreover, the completeness and time efficiency of WD-FIM algorithm are analyzed theoretically. Finally, the performance of the proposed WD-FIM algorithm is verified on both synthetic and real-life datasets.