The proliferation of IoT devices has piqued the interest of several adversaries looking for a different means to gain unauthorized access to systems or for other illicit reasons. As a result, protecting these devices ...
详细信息
The proliferation of IoT devices has piqued the interest of several adversaries looking for a different means to gain unauthorized access to systems or for other illicit reasons. As a result, protecting these devices is essential. The IDS acts as a second line of defense after the firewall and can be beneficial in the IoT networks. This paper presents a Real Time Intrusion Detection System based on the Machine Learning model Random Forest and has been set up for the IoT node consisting of Arduino, NodeMCU and an Ultrasonic sensor. Unlike most of the systems that train and test the model only on data from the dataset, this has been tested with real time network traffic. The dataset used is self made, created by monitoring the network traffic of our IoT network and not the usual popular dataset that is not IoT specific.
Firefly algorithm has been widely used in many optimization problems since it was proposed because of its good searching ability. However, in the searching process, the standard firefly algorithm can easily fall into ...
详细信息
Firefly algorithm has been widely used in many optimization problems since it was proposed because of its good searching ability. However, in the searching process, the standard firefly algorithm can easily fall into local optimal because a fixed step value is used. To solve this problem, an improved firefly algorithm is proposed in this paper by introducing an adjustable strategy for step size setting to improve the optimization accuracy and a double-bottom map to accelerate the convergence. In order to test the performance of the algorithm, twelve benchmark functions are used. Compared to other algorithms, the experimental results show that the modified firefly algorithm has better convergence capabilities and higher accuracy.
Graph learning is crucial for extracting meaningful information from graph-structured data, enabling effective solutions to various downstream tasks. However, existing methods for solving downstream graph learning tas...
Graph learning is crucial for extracting meaningful information from graph-structured data, enabling effective solutions to various downstream tasks. However, existing methods for solving downstream graph learning tasks often rely on the availability of the graph structure, which may not always be accessible in real-world applications. To overcome this limitation, recent approaches have introduced exploratory learning techniques, which aim to tackle graph learning tasks on graphs with unknown topology. In this article, we provide a comprehensive overview of exploratory graph learning applied to two widely studied graph learning tasks: 1) influence maximization and 2) community detection. We delve into the problem formulation of both tasks concerning graphs with unknown topological information. Additionally, we explore the application of exploratory learning techniques to address these problems effectively.
In this paper we introduce a novel method of estimating romantic, social and sexual attraction between two people by quantifying their bodily coordination using wearable sensors in a speed-date setting. We developed s...
详细信息
In this paper we introduce a novel method of estimating romantic, social and sexual attraction between two people by quantifying their bodily coordination using wearable sensors in a speed-date setting. We developed simple synchrony and convergence features, inspired from the literature and specifically adapted to be extracted from accelerometer data. To our knowledge, this is the first time that motion convergence is used for estimating attraction. Our features could predict one-way social attraction with a 73% Area under the ROC curve (AUC), out-performing previous work in a similar setting. We also showed that prediction performance increased when the male and female data are separated. We could also predict mutual romantic attraction with an AUC of 80%. Finally, we found that social attraction could be predicted better from movement correlation features whereas for romantic and sexual interest mimicry features were better indicators. Additionally, we found that “mimicking of female to male” and “convergence of female's movement to male's” were indicators of sexual and romantic mutual attraction in our data.
The channel estimation is a vital part of turbo equalization. The structure of combined channel estimation and turbo equalization iterative update is presented in this paper. The low computation complexity algorithm c...
详细信息
The channel estimation is a vital part of turbo equalization. The structure of combined channel estimation and turbo equalization iterative update is presented in this paper. The low computation complexity algorithm called soft decision feedback LMS algorithm is used in channel estimation and the data-reused method is used in the initial channel estimation to overcome the problem of low convergence rate for LMS algorithm. The result of simulation shows that the turbo equalizer with this channel estimation method not only has good performance but also has ability to tracking the fast varying channel.
Cloud computing divides a huge program into countless subtasks through the network, which are calculated and analyzed by multiple servers, and then the results are returned to users. Therefore, the strategy of task sc...
详细信息
Cloud computing divides a huge program into countless subtasks through the network, which are calculated and analyzed by multiple servers, and then the results are returned to users. Therefore, the strategy of task scheduling is very important for computing performance. Aiming at the essence of cloud computing task scheduling and the optimization problem of seeking solutions, this paper proposes a hybrid algorithm called MMES algorithm (MA-MIX-ESDA). This algorithm not only guarantees the search space of electrostatic discharge algorithm (ESDA), but also accelerates its convergence speed, and solves the problem that mayfly algorithm (MA) is easy to fall into local optimization. Latin hypercube sampling is used for population initialization, exploration and development are balanced by the direction of the balance vector, and the step size control factor is added to jump out of local optimization. In order to evaluate the performance of the algorithm, 23 groups of test functions commonly used by CEC and 30 benchmark functions of CEC2014 are used to test the global search and local development functions of the algorithm, and the results are compared with the improved algorithm and classical algorithm. Experimental results show that the proposed MMES algorithm is more superior in search space and convergence speed.
Towards the nonlinear underwater acoustic channels with the severe linear distortion, based on the fast de-correlation characteristic of wavelet packet transform, a decision feedback blind equalization algorithm based...
详细信息
Towards the nonlinear underwater acoustic channels with the severe linear distortion, based on the fast de-correlation characteristic of wavelet packet transform, a decision feedback blind equalization algorithm based on momentum and orthogonal wavelet packet transform (MWPT-DFE) is proposed, which is derived by revising the iteration equation of forward weight vector for decision feedback structure blind equalizer. Comparing with normal decision feedback blind equalization algorithm (DFE) and the decision feedback blind equalization algorithm based on orthogonal wavelet Packet transform (WPT-DFE), the new algorithm has fast convergence, good performance for tracking and smaller MSE. The efficiency of the proposed algorithm is proved by computer simulation in underwater acoustic channels.
As the number of Internet users increases rapidly, IPv4 address shortage has been developing. The shortage can be completely solved by using IPv6, but it takes too much time to replace all the equipments in the world....
详细信息
As the number of Internet users increases rapidly, IPv4 address shortage has been developing. The shortage can be completely solved by using IPv6, but it takes too much time to replace all the equipments in the world. Therefore, NAT is used as a temporally solution. NAT reduces the use of address by connecting a local address to a public address only when it is necessary, which can be a good solution for server-client models, but causes NAT-Traversal problem in peer-to-peer communication such as VoIP using SIP in NAT environment. STUN and TURN are used to solve NAT-Traversal problem. This document explains how STUN and TURN are applied to SIP to solve problems when using NAT and points out problems that STUN and TURN cause, and suggests a way to improve.
Federated learning is a state-of-the-art technology used in the fog computing, which allows distributed learning to train cross-device data while achieving efficient performance. Many current works have optimized the ...
详细信息
ISBN:
(数字)9781728150895
ISBN:
(纸本)9781728150901
Federated learning is a state-of-the-art technology used in the fog computing, which allows distributed learning to train cross-device data while achieving efficient performance. Many current works have optimized the federated learning algorithm in homogeneous networks. However, in the actual application scenario of distributed learning, data is independently generated by each device, and this non-homologous data has different distribution characteristics. Therefore, the data used by each device for local learning is unbalanced and non-IID, and the heterogeneity of data affects the performance of federated learning and slows down the convergence. In this paper, we present a dynamic sample selection optimization algorithm, FedSS, to tackle heterogeneous data in federated learning. FedSS dynamically selects the training sample size during the gradient iteration based on the locally available data size, to settle the expensive evaluations of the local objective function with a massive amount of dataset. We theoretically analyze the convergence and present the complexity estimates of our framework when learning large data from unbalanced distribution. Our experimental results show that the use of dynamic sampling methods can effectively improve the convergence speed with heterogeneous data, and keep computational costs low while achieving the desired accuracy.
Recently, enormous growth of mobile data traffic can be load to operator's core network. Offloading traffic is a promising solution to alleviate the load. The 3GPP proposes the offloading solutions such as local I...
详细信息
Recently, enormous growth of mobile data traffic can be load to operator's core network. Offloading traffic is a promising solution to alleviate the load. The 3GPP proposes the offloading solutions such as local IP access (LIPA) and selected IP traffic offload (SIPTO). Based on the LIPA/SIPTO, many relate works provides the offloading procedures and schemes but it doesn't take the network condition and QoS requirement into account. Thus, we propose the new offloading algorithm according to the network condition and QoS requirement. By simulation, we show that the proposed algorithm select the offloading application that can meet the QoS requirement.
暂无评论