Joint Multi-User Detection with Weighting Factors for Unsourced Multiple Access

Multi-user detection techniques are currently being studied as highly promising technologies for improving the performance of unsourced multiple access systems. In this paper, we propose joint multi-user detection schemes with weighting factors for unsourced multiple access. First, we introduce bidirectional weighting factors in the extrinsic information passing process between the multi-user detector based on belief propagation (BP) and the LDPC decoder. Second, we incorporate bidirectional weighting factors in the message passing process between the MAC nodes and the user variable nodes in BP- based multi-user detector. The proposed schemes select the optimal weighting factors through simulations. The simulation results demonstrate that the proposed schemes exhibit significant performance improvements in terms of block error rate (BLER) compared to traditional schemes. .

An intelligent active probing and trace-back scheme for IoT anomaly detection

Due to their simple hardware,sensor nodes in IoT are vulnerable to attack,leading to data routing blockages or malicious tampering,which significantly disrupts secure data collection.An Intelligent Active Probing and Trace-back Scheme for IoT Anomaly Detection(APTAD)is proposed to collect integrated IoT data by recruiting Mobile Edge Users(MEUs).(a)An intelligent unsupervised learning approach is used to identify anomalous data from the collected data by MEUs and help to identify anomalous nodes.(b)Recruit MEUs to trace back and propose a series of trust calculation methods to determine the trust of nodes.(c)The last,the number of active detection packets and detection paths are designed,so as to accurately identify the trust of nodes in IoT at the minimum cost of the network.A large number of experimental results show that the recruiting cost and average anomaly detection time are reduced by 6.5 times and 34.33%respectively,while the accuracy of trust identification is improved by 20%.

Multi-Panel Extra-Large Scale MIMO Based Joint Activity Detection and Channel Estimation for Near-Field Massive IoT Access

The extra-large scale multiple-input multiple-output(XL-MIMO)for the beyond fifth/sixth generation mobile communications is a promising technology to provide Tbps data transmission and stable access service.However,the extremely large antenna array aperture arouses the channel near-field effect,resulting in the deteriorated data rate and other challenges in the practice communication systems.Meanwhile,multi-panel MIMO technology has attracted extensive attention due to its flexible configuration,low hardware cost,and wider coverage.By combining the XL-MIMO and multi-panel array structure,we construct multi-panel XL-MIMO and apply it to massive Internet of Things(IoT)access.First,we model the multi-panel XL-MIMO-based near-field channels for massive IoT access scenarios,where the electromagnetic waves corresponding to different panels have different angles of arrival/departure(AoAs/AoDs).Then,by exploiting the sparsity of the near-field massive IoT access channels,we formulate a compressed sensing based joint active user detection(AUD)and channel estimation(CE)problem which is solved by AMP-EM-MMV algorithm.The simulation results exhibit the superiority of the AMP-EM-MMV based joint AUD and CE scheme over the baseline algorithms.

A Study of Ensemble Feature Selection and Adversarial Training for Malicious User Detection

The continuously booming of information technology has shed light on developing a variety of communication networks,multimedia,social networks and Internet of Things applications.However,users inevitably suffer from the intrusion of malicious users.Some studies focus on static characteristics of malicious users,which is easy to be bypassed by camouflaged malicious users.In this paper,we present a malicious user detection method based on ensemble feature selection and adversarial training.Firstly,the feature selection alleviates the dimension disaster problem and achieves more accurate classification performance.Secondly,we embed features into the multidimensional space and aggregate it into a feature map to encode the explicit content preference and implicit interaction preference.Thirdly,we use an effective ensemble learning which could avoid over-fitting and has good noise resistance.Finally,we propose a datadriven neural network detection model with the regularization technique adversarial training to deeply analyze the characteristics.It simplifies the parameters,obtaining more robust interaction features and pattern features.We demonstrate the effectiveness of our approach with numerical simulation results for malicious user detection,where the robustness issues are notable concerns.

User Station Security Protection Method Based on Random Domain Name Detection and Active Defense

The power monitoring system is the most important production management system in the power industry. As an important part of the power monitoring system, the user station that lacks grid binding will become an important target of network attacks. In order to perceive the network attack events on the user station side in time, a method combining real-time detection and active defense of random domain names on the user station side was proposed. Capsule network (CapsNet) combined with long short-term memory network (LSTM) was used to classify the domain names extracted from the traffic data. When a random domain name is detected, it sent instructions to routers and switched to update their security policies through the remote terminal protocol (Telnet), or shut down the service interfaces of routers and switched to block network attacks. The experimental results showed that the use of CapsNet combined with LSTM classification algorithm can achieve 99.16% accuracy and 98% recall rate in random domain name detection. Through the Telnet protocol, routers and switches can be linked to make active defense without interrupting services.

Complex field network-coded cooperation based on multi-user detection in wireless networks

Cooperative communication can achieve spatial diversity gains,and consequently combats signal fading due to multipath propagation in wireless networks powerfully.A novel complex field network-coded cooperation（CFNCC） scheme based on multi-user detection for the multiple unicast transmission is proposed.Theoretic analysis and simulation results demonstrate that,compared with the conventional cooperation（CC） scheme and network-coded cooperation（NCC） scheme,CFNCC would obtain higher network throughput and consumes less time slots.Moreover,a further investigation is made for the symbol error probability（SEP） performance of CFNCC scheme,and SEPs of CFNCC scheme are compared with those of NCC scheme in various scenarios for different signal to noise ratio（SNR） values.

Wavelet Packet Domain LMS Based Multi-User Detection

An improved wavelet packet domain least mean square （IWPD-LMS） based adaptive muhiuser detection algorithm is proposed. The algorithm employs the wavelet packet transform to rewhiten the input data, and chooses the best wavelet packet basis according to a novel convergence contribution function rather than the conventional Shannon entropy. The theoretic analyses show that the inadequacy of the eigenvalue spread of the tap-input correlation matrix is ameliorated, thus the convergence performance is improved greatly. The simulation result of convergence performance and bit error rate（BER） performance as a function of the signal power to noise power ratio（SNR） are presented finally to prove the validity of the proposed algorithm.

Study of multi-rate multi-user detection based on supervision decision

Multi-user detection (MUD) based on multirate transmission in code division multiple access (CDMA) system is discussed. Under the requirement of signal interference ratio (SIR) detection at base station and framework with parallel interference cancellation, a supervision decision algorithm based on pre-decision of probabilistic data association (PDA) and hard decision is proposed. The detection performance is analyzed and simulation is implemented to show that the supervision decision algorithm improves the detection performance effectively.

A Linear Interference Cancellation Multi-User Detector for Synchronous CDMA Communications

This paper describes a linear interference cancellation multi user detector for synchronous DS CDMA systems under the condition that all spread spectrum code waveforms have the constant cross correlating coefficients. The basic idea is to get the estimation for total multiple access interference (MAI) of all users using a reference code waveform, then subtract the total MAI from the received signal. The structure of such a detector is nearly similar to the conventional detector. The BER expression obtained in the paper shows significant performance improvement compared to the other detectors.

Active User and Data Detection for Uplink Grant-free NOMA Systems

This paper proposes some low complexity algorithms for active user detection(AUD),channel estimation(CE)and multi-user detection(MUD)in uplink non-orthogonal multiple access(NOMA)systems,including single-carrier and multi-carrier cases.In particular,we first propose a novel algorithm to estimate the active users and the channels for single-carrier based on complex alternating direction method of multipliers(ADMM),where fast decaying feature of non-zero components in sparse signal is considered.More importantly,the reliable estimated information is used for AUD,and the unreliable information will be further handled based on estimated symbol energy and total accurate or approximate number of active users.Then,the proposed algorithm for AUD in single-carrier model can be extended to multi-carrier case by exploiting the block sparse structure.Besides,we propose a low complexity MUD detection algorithm based on alternating minimization to estimate the active users’data,which avoids the Hessian matrix inverse.The convergence and the complexity of proposed algorithms are analyzed and discussed finally.Simulation results show that the proposed algorithms have better performance in terms of AUD,CE and MUD.Moreover,we can detect active users perfectly for multi-carrier NOMA system.

Real-Time Multi Fractal Trust Evaluation Model for Efficient Intrusion Detection in Cloud

Handling service access in a cloud environment has been identified as a critical challenge in the modern internet world due to the increased rate of intrusion attacks.To address such threats towards cloud services,numerous techniques exist that mitigate the service threats according to different metrics.The rule-based approaches are unsuitable for new threats,whereas trust-based systems estimate trust value based on behavior,flow,and other features.However,the methods suffer from mitigating intrusion attacks at a higher rate.This article presents a novel Multi Fractal Trust Evaluation Model(MFTEM)to overcome these deficiencies.The method involves analyzing service growth,network growth,and quality of service growth.The process estimates the user’s trust in various ways and the support of the user in achieving higher service performance by calculating Trusted Service Support(TSS).Also,the user’s trust in supporting network stream by computing Trusted Network Support(TNS).Similarly,the user’s trust in achieving higher throughput is analyzed by computing Trusted QoS Support(TQS).Using all these measures,the method adds the Trust User Score(TUS)value to decide on the clearance of user requests.The proposed MFTEM model improves intrusion detection accuracy with higher performance.

A Hybrid DNN-RBFNN Model for Intrusion Detection System

Intrusion Detection Systems (IDS) are pivotal in safeguarding computer networks from malicious activities. This study presents a novel approach by proposing a Hybrid Dense Neural Network-Radial Basis Function Neural Network (DNN-RBFNN) architecture to enhance the accuracy and efficiency of IDS. The hybrid model synergizes the strengths of both dense learning and radial basis function networks, aiming to address the limitations of traditional IDS techniques in classifying packets that could result in Remote-to-local (R2L), Denial of Service (Dos), and User-to-root (U2R) intrusions.

An Efficient Noisy-ICA Based Approach to Multiuser Detection in IDMA Systems

Interleaved Division Multiple Access (IDMA) is a new access scheme that has been proposed in the literature to increase the capacity of wireless channels. The performance of such systems depends on the accuracy of the channel state information at the receiver. In this paper, a Noisy-Independent Component Analysis (N-ICA) based IDMA receiver for multiple access communication channels is proposed. The N-ICA component is applied as a post processor. Unlike other IDMA receivers, the proposed scheme detects and separates the transmitted symbols without channel state information tracking. The performance of the proposed technique is presented in terms of raw bit error rate (BER) without channel coding for different signal to noise ratios (SNR). Simulation results demonstrate that N-ICA post processor provides an improvement in performance in terms of bit error rate (BER) in loaded systems. When the system is not loaded, the proposed post processor presents the same performance as conventional IDMA receiver with less iterations leading to a complexity reduction.

A Turbo Decoder Included in a Multi-User Detector: A Solution to be Retained

This paper deals with the presentation of different multi-user detectors in the Universal Mobile Telecommunications System (UMTS) context. The challenge is always to optimize the compromise between performance and complexity. Compared with the solution commonly used today, the rake detector, successive interference cancellation (SIC) detector has better performance despite its higher complexity. Our innovative solution proposes joining detector and channel turbo decoder to get a significant gain in terms of performance. Furthermore, when detection and decoding are implemented in a single function, complexity does not increase much.

MIMO Multi Carrier Interleave Division Multiple Access System with Multiuser Detection: Performance Results

This paper provides the performance analysis of multi input multi output (MIMO) Multi carrier (MC) interleaver based multiple-access system with multiuser detection. Interleave-division multiple access (IDMA) is a scheme in which users are separated by employing different interleavers instead of different signatures as in a conventional code-division multiple-access (CDMA) scheme. The basic principle of IDMA is that the interleaver is different for different users. Interleavers are generated independently and randomly. At the receiver, ZF, LLSE, VBLAST/ZF/MAP and VBLAST/LLSE/MAP detectors are employed. The performance of the system is analyzed for different channel conditions using extensive simulation runs based on Monte Carlo simulation trials. Simulation results divulge that VBLAST/LLSE/MAP detector results in significant performance amelioration compared to other detectors as applied to MC/IDMA system.

Mean Field Annealing Neural Network for the Optimal DS/CDMA Multiuser Detector *

In this paper, an optimal multi user detector in DS/CDMA communication systems based on the mean field annealing (MFA) neural network is proposed. It is shown that the NP complete problem of minimizing the objective function of the optimal multi user detector can be translated into minimizing an MFA network energy function. Numerical results show that the proposed detector offers significant performance gain relative to the conventional detector and decorrelating detector while it can be implemented easily in analog hardware.

Masquerade Detection Using Support Vector Machine

A new method using support vector data description (SVDD) to distinguishlegitimate users from mas-queradcrs based on UNIX user command sequences is proposed Sliding windowsare used to get low detection delay. Experiments demonstrate that the detection effect usingenriched sequences is better than that of using truncated sequences. As a SVDD profile is composedof a small amount of support vectors, our SVDD-based method can achieve computation and storageadvantage when the detection performance issimilar to existing method.

一种基于XGBoost的用户行为异常检测方法

伴随企业业务的不断扩增和电子化发展,企业自身数据和负载数据都开始暴增。然而,作为企业核心资产之一的内部数据,却面临着日益严峻的安全威胁。越来越多以周期长、频率低、隐蔽强为典型特征的非明显攻击绕过传统安全检测方法,对大量数据造成损毁。对此,根据用户、实体、行为关联,提出整合可以反映用户行为基线的各类数据,提取几种最能反映用户异常的基础特征,利用XGBoost的特征选择策略与FCM聚类算法结合,通过异常打分,定位异常风险最大的一批用户。结果显示,新提出的异常检测算法有着82%以上的准确率,以及86%以上的召回率。

ResNet-UAN-AUD:基于深度学习的水声上行非正交多址通信系统活动用户检测方法

水下声学网络(Underwater Acoustic Networks,UAN)是探测未知水域的重要技术手段。非正交多址(Non-Orthogonal Multiple Access,NOMA)是一种新颖的移动通信技术,支持时域、频域、空域/编域的非正交分配,可有效地提高网络容量和用户接入数,为性能和电量受限的UAN提供创新解决方案。活动用户检测(Active User Detection,AUD)是NOMA通信系统的基础支撑,对于NOMA系统消除信号干扰和提高接收性能至关重要。ResNet是基于残差模块跳跃连接的神经网络,解决了深度学习的梯度消失和网络退化问题。提出了一种基于深度学习的水声上行NOMA通信系统AUD检测方案。首先,构建水声上行NOMA通信系统基本模型;其次,实施NOMA活动用户检测问题的数学表征;接着,开发基于ResNet网络的水声NOMA系统活动节点检测方法(ResNet-UAN-AUD);最后,执行仿真实验。结果表明,ResNet-UAN-AUD的检测性能接近基于长短期记忆网络的活动用户检测(LSTM-UAN-AUD)方案,而复杂度略高于基于卷积神经网络的活动用户检测(CNN-UAN-AUD)技术,实现了次优目标,适合水声上行NOMA系统使用。

基于联合情感的多任务谣言检测方法

情感分析在社交媒体谣言检测中有重要作用,现有的谣言检测方法侧重于使用文章的情感特征,忽略了用户评论与文章的联合情感,单任务深度学习谣言检测方法缺少足够的标签数据导致准确率难以提升.为解决上述问题,基于谣言检测与联合情感检测两个任务的相关性,提出了多任务联合学习的谣言检测方法,在同一模型中实现联合情感检测以及谣言检测两个任务.首先,构建编码器提取文章及对应评论的语义特征并映射到同一语义空间,并通过注意力机制加权融合语义特征;其次,通过基于公共情感分类器与情感词典共同构建的情感提取网络提取文章与对应评论的情感特征;最后,将语义特征与联合情感特征融合后输入到两个共享损失函数的分类器中,分别得到谣言检测和联合情感检测的分类结果.实验结果表明,多任务模型的效果强于单任务模型,在公开的中文Weibo-16数据集、英文Twitter-15数据集上相较于对比方法中最好的方法,提出的模型在准确率上分别提升了3.5和2.9个百分点,F1值提高了3.1和3.9个百分点.