Search

We study the type of distributions that Restricted Boltzmann Machines (RBMs) with different activation functions can express by investigating the effect of the activation function of the hidden nodes on the marginal distribution they impose on observed binary nodes. We report an exact expression for these marginals in the form of a model of interacting binary variables with the explicit form of the interactions depending on the hidden node activation function. We study the properties of these interactions in detail and evaluate how the accuracy with which the RBM approximates distributions over binary variables depends on the hidden node activation function and on the number of hidden nodes. When the inferred RBM parameters are weak, an intuitive pattern is found for the expression of the interaction terms which reduces substantially the differences across activation functions. We show that the weak parameter approximation is a good approximation for different RBMs trained on the MNIST dataset. Interestingly, in these cases, the mapping reveals that the inferred models are essentially low order interaction models., Comment: Supplemental material is available as ancillary file and can be downloaded from a link on the right

The random vector functional link (RVFL) network is suitable for solving nonlinear problems from transformer fault symptoms and different fault types due to its simple structure and strong generalization ability. However, the RVFL network has a disadvantage in that the network structure, and parameters are basically determined by experiences. In this paper, we proposed a method to improve the RVFL neural network algorithm by introducing the concept of hidden node sensitivity, classify each hidden layer node, and remove nodes with low sensitivity. The simplified network structure could avoid interfering nodes and improve the global search capability. The five characteristic gases produced by transformer faults are divided into two groups. A fault diagnosis model of three layers with four classifiers was built. We also investigated the effects of the number of hidden nodes and scale factors on RVFL network learning ability. Simulation results show that the number of implicit layer nodes has a large impact on the network model when the number of input dimensions is small. The network requires a higher number of implicit layer neurons and a smaller threshold range. The size of the scale factor has significant influence on the network model with larger input dimension. This paper describes the theoretical basis for parameter selection in RVFL neural networks. The theoretical basis for the selection of the number of hidden nodes, and the scale factor is derived. The importance of parameter selection for the improvement of diagnostic accuracy is verified through simulation experiments in transformer fault diagnosis.

Long-Term Evolution in unlicensed spectrum (LTE-U) is a promising solution to address the spectrum scarcity in the cellular networks. However, the LTE-U and WiFi systems may interfere with each other and degrade the performance. Thus, many works have focused on coexistence design and proposed various solutions. Nevertheless, the hidden terminal problem has not been well studied in the LTE-U system. In this paper, we investigate joint interference detection and resource allocation in the LTE-U system to maximize the overall system throughput while guaranteeing each user's quality-of-service (QoS) requirements. Firstly, to solve the hidden terminal problem and avoid the co-channel interference between the WiFi and LTE-U users, we adopt a cluster-based algorithm to find the potential WiFi interference for each LTE-U user according to their channel quality indicator (CQI) feedback and location information. Secondly, we formulate a joint subcarrier and power allocation problem over the licensed and unlicensed bands with each user's QoS provision. To make the original problem more tractable, we separate it into licensed and unlicensed subproblems and then derive the optimal solution. Moreover, we propose a low-complexity resource allocation scheme. The simulation results show that the proposed schemes can significantly improve the system performance and outperform the existing schemes.

The Listen-Before-Talk (LBT) is the main procedure for Licensed Assisted Access (LAA) to accomplish fair and friendly coexistence with other operators or technologies operating over unlicensed spectrum. However, in LBT, the lack of coordination with other existing systems brings challenges in sustaining performance in the coexistence of LAA and WiFi networks. Specifically, the hidden node problem (HNP) and exposed node problem (ENP) cannot be effectively handled when both LAA and WiFi nodes attempt to access the unlicensed spectrum at the same time. Thus, transmission failure might occur and the network performance would be degraded. In order to mitigate the influences caused by HNP and ENP, based on LBT, we firstly analyze HNP and ENP by means of mathematical approach. The analytical results surprisingly reveal that the hidden node and exposed node probabilities are as high as 41% and 39.33%, respectively. Then, a Give And p-persistent Take (GAT) mechanism with the Listen-Before-Receive (LBR) procedure, namely LBR-GAT, is proposed to cope with LBT to reduce the collision caused by HNP as well as to retrieve the bandwidth sacrificed by the ENP. With LBR-GAT, the LAA sender conditionally gives up or takes back transmission opportunities, and thus the unlicensed spectrum could be efficiently shared between LAA and WiFi. Evaluation results show that the proposed LBR-GAT could conditionally obtain better network performance comparing to legacy LBT.

Wireless Multimedia Sensor Networks (WMSNs) have emerged as the new class of wireless sensor networks (WSNs) to meet the stringent Quality of Service (QoS) requirements of emerging applications. Multipath routing with cross-layer approach appears to be a potential solution for supporting the distinct characteristics of WMSNs. However, due to the broadcast nature of the underlying medium, multiple paths are exposed to inter-path interference. In addition, low-power wireless links are asymmetric, error-prone and unreliable in nature. Consequently, an accurate and stable link quality estimation is essential to guarantee the performance of routing protocol. This paper proposes Triangle link quality metric and minimum inter-path Interference based Geographic Multipath Routing (TIGMR) protocol which finds multiple node-disjoint paths in IEEE 802.15.4 compliant network. This cross-layer routing protocol selects forwarding node based on a triangle link quality metric, remaining energy, and distance while anticipating minimum adjacent path interference effect. In addition, TIGMR protocol avoids Hidden Node Problem (HNP) at the sink node without using Request-To-Send/Clear-To-Send (RTS/CTS) handshake mechanism. Simulation results indicate TIGMR protocol optimizes overall performance and improves network lifetime as compared with state-of-the-art Two-Phase Geographic Forwarding (TPGF) and Link Quality and Energy-Aware Routing (LQEAR) protocols.

The IEEE 802.15.4-2015 TSCH (Time Slotted Channel Hopping) standard provides high-reliability, deterministic delay, and energy efficiency with the channel hopping and the TDMA. The standard does not specify the scheduling method, which is an essential part of the solution. One of the representative studies in TSCH scheduling is Orchestra. It is an autonomous scheduling method with minimal overhead that does not require a negotiation process between nodes. However, with high traffic load, performance of the Orchestra is greatly degraded because of its fixed schedule. If a node has children, continuous collisions occur in the slot of the node. We propose a novel algorithm called SRCA (Slot Reallocation for Collision Avoidance), as a solution for this problem. The SRCA autonomously updates the schedule. When a child requests a slot reallocation, the parent allocates a new slot with a low probability of collision. Another algorithm called SJCA (Slot reJection for Collision Avoidance) is further proposed in this work. It improves the SRCA to operate in an environment with interference where a sending node can disturb a transmission from a hidden node. In the SJCA, a node determines whether a collision due to interference will occur in the slot newly allocated by the parent. If so, it rejects the slot and requests again. We compare the performance of the SRCA and the SJCA against the Orchestra and e-TSCH-Orch through simulations. It is confirmed that the SJCA is robust to interference. In most cases, the SJCA shows better performance of PAR (Packet Acknowledgement Ratio), PLR (Packet Loss Rate), and latency than others. It is also verified that the SJCA provides high-reliability and low latency compared to existing technologies.

Wireless technologies are continuously evolving, including features such as the extension to mid- and long-range communications and the support of an increasing number of devices. However, longer ranges increase the probability of suffering from hidden terminal issues. In the particular case of Wireless Local Area Networks (WLANs), the use of Quality of Service (QoS) mechanisms introduced in IEEE 802.11e compromises scalability, exacerbates the hidden node problem, and creates congestion as the number of users and the variety of services in the network grow. In this context, this paper presents a configurable Colored Petri Net (CPN) model for the IEEE 802.11e protocol with the aim of analyzing the QoS support in mid- and long-range WLANs The CPN model covers the behavior of the protocol in the presence of hidden nodes to examine the performance of the RTS/CTS exchange in scenarios where the QoS differentiation may involve massive collision chains and high delays. Our CPN model sets the basis for further exploring the performance of the various mechanisms defined by the IEEE 802.11 standard. We then use this CPN model to provide a comprehensive study of the effectiveness of this protocol by using the simulation and monitoring capabilities of CPN Tools.

Cooperative spectrum sensing (CSS) is a technique where multiple cognitive radio users cooperate among themselves to make binary decisions about the presence of a primary user. The single cognitive user often faces the hidden terminal problem. However, CSS tackles this problem by sending local sensing-based decisions to the fusion centre. A major drawback of conventional energy detection is the poor performance at low SNR regime. In this work, likelihood ratio statistics is considered as a test-statistic due to its highest statistical power. An improved likelihood ratio statistic-based CSS scheme is proposed by considering several past sensing events. The proposed scheme mitigates the poor detection at low SNR regime and misdetections arising due to sudden drops in signal energy. Furthermore, the generalised Byzantine attack is taken into account considering a security aspect. The proposed scheme is also shown to outperform Anderson Darling-based malicious user detection in CSS at a low SNR regime. The proposed scheme is verified and validated over empirical spectrum data. The performance improvement is at the cost of computational time, which in practice is very low and is justified by the significant performance improvements of the proposed scheme at low SNR regime.

Extreme learning machine (ELM) application to model the shear stress of magnetorheological (MR) fluids has superiority over the existing methods, such as Herschel-Bulkley. Although the shear stress has been successfully predicted, the hidden node numbers are too high reaching up to 10,000 that will hinder the application of the models. Furthermore, the existing works have tried to determine the hidden node number only by trial and error method. Therefore, this paper aims to reduce the hidden node number by employing the particle swarm optimization (PSO) considering the accuracy and the hidden node numbers. The ELM based-shear stress model was firstly defined by treating the magnetic field and shear rate as the inputs and shear stress as output. The objective function optimization method was then formulated to minimize the normalized error and the hidden node numbers. Finally, the proposed methods were tested at various ELM activation functions and samples. The results have shown that the platform has successfully reduced the hidden node numbers from 10,000 to 571 while maintaining the error of less than 1%. In summary, the proposed objective function for PSO optimization has successfully built the optimum shear stress model automatically.

Objectives: To implement energy detection and eigenvalue based cooperative spectrum sensing in NI-USRP hardware platform and to obtain its performance.Cooperative spectrum sensing is to be implemented using O and AND fusion rules. Methodology: The hardware is implemented using one primary user transmitter and two cognitive radio users. The implementation is done using LABVIEW and detection performance is analyzed. In cooperative spectrum sensing (CSS), CR system shares its own sensing information with other sensing nodes and utilizes the sensing outcomes of others to give a decision. Findings: Energy detection performs well in high SNR region and eigenvalue based detection performs well even with low SNR region. Energy detector detects the signal completely with SNR greater than 10dB. Eigenvalue detector detects signal completely at SNR of -9dB. Detection probability with OR Cooperative rule is better than AND rule. It increases with number of samples and signal to noise ratio. It is found that cooperative spectrum sensing performs well than the noncooperative sensing by decreasing the chances of miss detection. Novelty: CSS is robust in sensing errors due to hidden node or fading channels and also it decreases the probabilities of false alarm as well as probabilities of miss detection. Keywords: Cognitive radio; Cooperative spectrum sensing; Primary users; Secondary users; Energy detection

We propose three different methods to determine the optimal number of hidden nodes based on $L_{1}$ regularization for a multilayer perceptron network. The first two methods, respectively, use a set of multiplier functions and multipliers for the hidden-layer nodes and implement the $L_{1}$ regularization on those, while the third method equipped with the same multipliers uses a smoothing approximation of the $L_{1}$ regularization. Each of these methods begins with a given number of hidden nodes, then the network is trained to obtain an optimal architecture discarding redundant hidden nodes using the multiplier functions or multipliers. A simple and generic method, namely, the matrix-based convergence proving method (MCPM), is introduced to prove the weak and strong convergence of the presented smoothing algorithms. The performance of the three pruning methods has been tested on 11 different classification datasets. The results demonstrate the efficient pruning abilities and competitive generalization by the proposed methods. The theoretical results are also validated by the results.

Electricity theft behavior has serious influence on the normal operation of power grid and the economic benefits of power enterprises. Intelligent anti-power-theft algorithm is required for monitoring the power consumption data to recognize electricity power theft. In this paper, an adaptive time-series recurrent neural network (TSRNN) architecture was built up to detect the abnormal users (i.e., the electricity theft users) in time-series data of the power consumption. In fusion with the synthetic minority oversampling technique (SMOTE) algorithm, a batch of virtual abnormal observations were generated as the implementation for training the TSRNN model. The power consumption record was characterized with the sharp data (ARP), the peak data (PEA), and the shoulder data (SHO). In the TSRNN architectural framework, a basic network unit was formed with three input nodes linked to one hidden neuron for extracting data features from the three characteristic variables. For time-series analysis, the TSRNN structure was re-formed by circulating the basic unit. Each hidden node was designed receiving data from both the current input neurons and the time-former neuron, thus to form a combination of network linking weights for adaptive tuning. The optimization of the TSRNN model is to automatically search for the most suitable values of these linking weights driven by the collected and simulated data. The TSRNN model was trained and optimized with a high discriminant accuracy of 95.1%, and evaluated to have 89.3% accuracy. Finally, the optimized TSRNN model was used to predict the 47 real abnormal samples, resulting in having only three samples false predicted. These experimental results indicated that the proposed adaptive TSRNN architecture combined with SMOTE is feasible to identify the abnormal electricity theft behavior. It is prospective to be applied to online monitoring of distributed analysis of large-scale electricity power consumption data.

Recent applications in large-scale wireless mesh networks (WSN), e.g., Advanced Metering Infrastructure (AMI) scenarios, expect to support an extended number of nodes with higher throughput, which cannot be sufficiently supported by the current WSN protocols. Two prior protocols, Wi-SUN Field Area Network (Wi-SUN FAN) and IETF 6TiSCH standards, are popularly used that are respectively based on asynchronous Carrier Sense Multiple Access / Collision Avoidance (CSMA/CA) and IEEE 802.15.4 Time Scheduled Channel Hopping (TSCH). However, the former one with CSMA/CA can be prone to the Hidden Node Problem (HNP) that leads to a degradation of reliability, while the latter one with TSCH avoids HNP by using synchronously scheduling but requires massive control signalling that degrades the upper-bound of throughput. Accordingly, this paper tackles the challenge of how to improve the upper-bound of throughput without loss of reliability. To do so, we first present an in-depth evaluation of reliability and throughput between the two existing standards via system-level simulation. We then propose a self-configurable grouping (SCG) method to cluster nodes without using location information of each node. This SCG method eliminates around 99\% HNP in CSMA/CA, thus greatly improves its reliability with maintaining the relatively high upper-bound of throughput. Our results show that the SCG method almost doubles the network throughput compared with both 6TiSCH and Wi-SUN FAN in heavy traffic scenarios, while providing extremely high reliability of more than 99.999\%.

The number of vehicles on the roads has increased proportionally over the last couple of years and this number is likely to rise due to the increase in population growth and the number of vehicles that are manufacturing every day. This high traffic density leads to several problems, from which effectively disseminating the emergency messages is a major concern. Keeping in view the dynamic characteristics of VANETs, significant challenges are faced in disseminating the message across the network. The major challenges are the broadcast storm problem, hidden node problem and the packet collision. Many studies have been performed to devise an effective and reliable mechanism for disseminating emergency messages in a Vehicular ad-Hoc Network (VANET). Researchers have proposed different models to tackle various types of scenarios for emergency message dissemination. This paper not only reviews some recent contributions to emergency message dissemination in vehicular networks but also discusses various proposed methods based on Intelligent Transportation System (ITS), Internet of Things (IoT), Priority messaging, Clustering approach, Software Defined Network (SDN) and Fog Computing. We have also tried to explore the latest developments in emergency message dissemination using 5G networks.

When the preambles of a wideband 802.11n/ac frame transmission are affected by a collision with a 20MHz frame on one of the subchannels, a receiver might fail to decode the whole wideband frame. Experimental evaluations with off-the-shelf 802.11n hardware showed that the vulnerability of wideband frame transmission to a narrowband collision/interference is a real problem. To mitigate the severe impact of narrowband collision/interference on wideband transmission, we propose to transmit several 20MHz frames in parallel instead of a single wideband frame so that a narrowband collision on one of the subchannels affects only the parallel frame transmitted on that subchannel. Performance evaluations demonstrated that the proposed parallel frame transmission scheme significantly improves the throughput and delay performance of wideband transmissions under different traffic and channel conditions. While the throughput performance of the proposed scheme for saturated traffic conditions was analytically validated, extensive simulation-based evaluations showed that the proposed scheme achieves great performance improvement for unsaturated traffic conditions and hidden node environment.

Artificial neural networks (ANNs) have been developed and applied to a variety of problems, such as pattern recognition, clustering, function approximation, forecasting, optimization, etc. However, existing ANNs have a high computational cost, since their learning methods are mostly based on a parameter tuning approach. Extreme learning machine (ELM) is a state-of-the-art method that generally dramatically reduces the computational cost. An analysis of the ELM method reveals that there are unsolved key factors, including inefficient hidden node construction, redundant hidden nodes, and unstable results. Therefore, we describe a new learning machine based on analytical incremental learning (AIL) in conjunction with principal component analysis (PCA). This learning machine, PCA-AIL, inherited the advantages from the original one and solved the unsolved key factors of ELM, and also extended AIL capability to serve a multiple-output structure. PCA-AIL was implemented with a single-layer feed-forward neural network architecture, used an adaptive weight determination technique to achieve a compact optimal structure and also used objective relations to support multiple output regression tasks. PCA-AIL has two steps: objective relation estimation and multiple optimal hidden node constructions. In the first step, PCA estimated the objective relations from multiple-output residual errors. In the second step, the multiple optimal nodes were obtained from objective relations and added to the model. PCA-AIL was tested with 16 multiple-objective regression datasets. PCA-AIL mostly outperformed other methods (ELM, EM-ELM, CP-ELM, DP-ELM, PCA-ELM, EI-ELM) in terms of fast testing speed – 0.0017 second, a compact model – 19.9 nodes, an accurate performance – RMSE 0.11261, and a stable result – S.D. of RMSE 0.00911: reported in averaged.

As one of the most promising networking technologies, power-line communication (PLC) networks have been gradually used not only in home networking systems but also in industrial IoT (Internet of things). Current studies of PLC medium access control (MAC) protocol (i.e., IEEE 1901) only focus on the single-hop environment; however, in practical industrial IoT and home access communication systems, PLC networks generally utilize a multi-hop architecture. In addition, due to the difference in the MAC standard between 802 series and 1901, existing analytical models of multi-hop IEEE 802.11 wireless networks are not suitable for multi-hop IEEE 1901 PLC networks. In this paper, we propose a theoretical model for performance analysis of multi-hop IEEE 1901 PLC networks, where the impacts of traffic rate (containing relay traffic), buffer size, deferral counter process of 1901, hidden terminal problem and multi-hop environment are comprehensively considered. The modeling process is divided into two parts. In the local modeling part, we construct a brand-new Markov chain model to investigate the carrier sense multiple access with collision avoidance process of IEEE 1901 protocol under the multi-hop environment. In the coupling queueing modeling part, we employ queueing theory to analyze the packet transmission procedure between successive hops. On the basis, we derive the closed-form expressions of throughput, medium access delay, packet blocking probability, end-to-end successful transmission delay and goodput. Through extensive simulations, we verify that our theoretical model can accurately estimate the MAC performance of IEEE 1901 PLC networks in the multi-hop environment.

To confront the increasing mobile traffic demand and the scarcity of the licensed band, 3GPP has standardized a new type of LTE, called License Assisted Access (LAA) LTE (LAA-LTE), which aggregates licensed bands with unlicensed bands. Since the conventional Wi-Fi devices employ −62 dBm clear channel assessment threshold to LAA-LTE's signals, the appearance of LAA-LTE in the unlicensed band leads to a new phenomenon, asymmetric hidden terminal problem , which severely degrades the throughput and delay performance of LAA-LTE as well as Wi-Fi. We propose a downlink interference control scheme for an LAA-LTE eNB to prevent the asymmetric hidden Wi-Fi APs nearby from recklessly transmitting and thereby interfering the eNB's transmission. Specifically, we analyze the distortion of Wi-Fi's signals received at LAA-LTE and LAA-LTE's signals received at Wi-Fi due to the difference of physical-layer parameters between LAA-LTE and Wi-Fi. Based on this analysis, we develop an AP suppression and power control technique to control incoming and outgoing interference of the LAA-LTE eNB. Extensive simulations confirm that the proposed scheme improves the throughput and delay performance of LAA-LTE while maintaining or even improving the performance of Wi-Fi.

This paper addresses the spectrum sensing problem in an orthogonal frequency-division multiplexing (OFDM) system based on machine learning. To adapt to signal-to-noise ratio (SNR) variations, we first formulate the sensing problem into a novel SNR-related multi-class classification problem. Then, we train a naive Bayes classifier (NBC), and propose a class-reduction assisted prediction method to reduce spectrum sensing time. We derive the performance bounds by translating the Bayes error rate into spectrum sensing error rate. Compared with the conventional methods, the proposed method is shown by simulation to achieve higher spectrum sensing accuracy, in particular at critical areas of low SNRs. It offers a potential solution to the hidden node problem.

As the main enablers of cognitive radio (CR), numerous spectrum sensing techniques have been proposed to date. Despite the numerous techniques, the existing spectrum sensing techniques tend to fail in rendering an efficient spectrum sensing whenever a hidden terminal problem arises. Meanwhile, this problem can happen at any time in any severely fading primary-to-secondary channels resulting in very low primary signal-to-noise ratios (SNRs) and hence ineffective detection of the primary user. Toward overcoming this problem, by introducing a tensor-based hypothesis testing framework, this paper proposes an efficient tensor-based detector (TBD) for a multiple-input multiple-output (MIMO) CR networks over multi-path fading channels. For the proposed spectrum sensing technique, insightful asymptotic performance analyses are provided and Monte-Carlo simulations that assess its performance have been conducted. These simulations corroborate that TBD outperforms the generalized likelihood ratio test (GLRT) detector and the maximum-minimum eigenvalue (MME) detector, especially in the very low SNR regime which is a manifestation of the hidden terminal problem. Moreover, the simulations validate the derived asymptotic performance characterizations.

The concept of SCN offers a fast framework with universal approximation guarantee for lifelong learning of non-stationary data streams. Its adaptive scope selection property enables for proper random generation of hidden unit parameters advancing conventional randomized approaches constrained with a fixed scope of random parameters. This paper proposes deep stacked stochastic configuration network (DSSCN) for continual learning of non-stationary data streams which contributes two major aspects: 1) DSSCN features a self-constructing methodology of deep stacked network structure where hidden unit and hidden layer are extracted automatically from continuously generated data streams; 2) the concept of SCN is developed to randomly assign inverse covariance matrix of multivariate Gaussian function in the hidden node addition step bypassing its computationally prohibitive tuning phase. Numerical evaluation and comparison with prominent data stream algorithms under two procedures: periodic hold-out and prequential test-then-train processes demonstrate the advantage of proposed methodology., This paper has been published in Information Sciences

The energy efficiency and delivery robustness are two critical issues for low duty-cycled wireless sensor networks. The asynchronous receiver-initiated duty-cycling media access control (MAC) protocols have shown their effectiveness through various studies. In receiver-initiated MACs, packet transmission is triggered by the probe of receiver. However, it suffers from the performance degradation incurred by packet collision, especially under bursty traffic. Several protocols have been proposed to address this problem, but their performance is restricted by the unnecessary backoff time and long negotiation process. In this article, we present CD-MAC, an energy-efficient and robust contention-detectable mechanism for addressing the collision-catching problem in receiver-initiated MACs. By exploring the temporal diversity of the acknowledgments, a receiver recognizes the potential senders and subsequently polls individual senders one by one. On that basis, CD-MAC can successfully avoid packet collision even though multiple senders have data packets to transmit to the same receiver. We implement CD-MAC in TinyOS and evaluate its performance on an indoor testbed with single-hop and multi-hop network scenarios. The results show that CD-MAC can significantly improve throughput by 1.72 times compared with the state-of-the-art receiver-initiated MAC protocol under bursty traffic loads. The results also demonstrate that CD-MAC can effectively mitigate the influence of hidden terminal problem and adapt to network dynamics well.

Hidden node problem sometimes referred to as frequent packets collision that mostly leads to loss of packets is no longer new in wireless networks because it affects the previous IEEE802.11 standards. The new IEEE802.11ah standard which is also a sub-standard of IEEE 802.11 is no exemption. As a matter of fact, IEEE802.11ah suffers from a hidden node problem more than networks (IEEE 802.11a/b/n/ac) due to their wider coverage which is up to 1km, high number of devices they can support (over 8000 nodes to one AP) and frequent simultaneous sleeping and sending of the nodes (power saving mode). A few researchers have worked on this hidden node problem in IEEE802.11ah but could not get a lasting solution to it. Therefore, this paper proposes an algorithm which detects hidden nodes and also proposes a theoretical solution based on previous works which was also experimentally verified through the BIHD-CM.

The recent developments in self-interference (SI) cancellation techniques have led to the practical realization of FD radios that can perform simultaneous transmission and reception. FD technology is attractive for various legacy communications standards. In this article, after discussing the opportunities of FD technology at the network layer, we present a cross-layer aided routing protocol, called X-FDR, for multihop FD wireless networks. X-FDR exploits a physical (PHY) layer model capturing the imperfection of SI cancellation. At the medium access control (MAC) layer, X-FDR adopts an optimized MAC protocol that implements a power control mechanism without creating the hidden terminal problem. X-FDR exploits the unique characteristics of FD technology at the network layer to construct energy-efficient and low end-to-end latency routes in the network. Performance evaluation demonstrates the effectiveness of X-FDR in achieving the gains of FD at higher layers of the protocol stack.

Neighbor discovery plays a major role in the wireless self-organizing network, which is the cornerstone for the autonomous coverage extension and performance enhancement such as cooperative transmission and multi-hop transmission. However, the hidden node problem is frequently encountered in neighbor discovery procedure based on the simple flooding due to the uncoordinated transmission, which gives rise to unnecessary energy waste. In this paper, the hidden node problem in the conventional neighbor discovery procedure is analyzed. A novel neighbor discovery procedure with the ability of node distance estimation and power allocation for the packet retransmission after collision is proposed, which utilizes fuzzy location information to properly configure the transmission power of the collided packets for better reception. The proposed algorithm improves the power efficiency of the neighbor discovery procedure by reducing the number of transmissions required for the successful reception with negligible induced complexity. Through the Monte-Carlo simulation, the proposed procedure with fuzzy location information and the power allocation method outperforms the conventional procedure in different node population scenarios, and the proposed procedure is insensitive to the distance estimation error, displaying strong robustness.

Feedforward neural networks are widely used as universal predictive models to fit data distribution. Common gradient-based learning, however, suffers from many drawbacks making the training process ineffective and time-consuming. Alternative randomized learning does not use gradients but selects hidden node parameters randomly. This makes the training process extremely fast. However, the problem in randomized learning is how to determine the random parameters. A recently proposed method uses autoencoders for unsupervised parameter learning. This method showed superior performance on classification tasks. In this work, we apply this method to regression problems, and, finding that it has some drawbacks, we show how to improve it. We propose a learning method of autoencoders that controls the produced random weights. We also propose how to determine the biases of hidden nodes. We empirically compare autoencoder based learning with other randomized learning methods proposed recently for regression and find that despite the proposed improvement of the autoencoder based learning, it does not outperform its competitors in fitting accuracy. Moreover, the method is much more complex than its competitors., International Joint Conference on Neural Networks IJCNN 2021

The authors have requested that this preprint be removed from Research Square.

LTE-Licensed Assisted Access (LAA) networks are beginning to be deployed widely in major metropolitan areas in the US in the unlicensed 5 GHz bands, which have existing dense deployments of Wi-Fi. This provides a real-world opportunity to study the problems due to hidden-node scenarios between LAA and Wi-Fi. The hidden node problem has been well studied in the context of overlapping Wi-Fi APs. However, when Wi-Fi coexists with LAA, the hidden node problem is exacerbated since LAA cannot use the well-known Request-to-Send (RTS)/Clear to-Send (CTS) mechanism to resolve contentions, resulting in throughput degradation for Wi-Fi. In this paper, we describe detailed measurements and conclusions from experiments on the campus of the University of Chicago which presents a perfect hidden node scenario where Wi-Fi access points (APs) controlled by us and an LAA base-station (BS) deployed by AT&T are hidden from each other, but the clients are not. We performed careful experiments in three different regions of the coexistence area: (i) clients midway between LAA & Wi-Fi; (ii) clients close to the Wi-Fi AP; and (iii) clients close to the LAA BS. Our results show that in a situation where LAA uses an aggregate of three unlicensed channels (60 MHz bandwidth) which overlap with an 80 MHz Wi-Fi transmission, the Wi-Fi throughput at client devices suffers considerably. Overall, Wi-Fi performance is impacted by the hidden node problem more severely than LAA. In the best outdoor conditions, the throughput of LAA and Wi-Fi is reduced by 35% and 97% respectively when coexisting with each other as compared when the other system is not present. Furthermore, we conclude that when both LAA and Wi-Fi use multiple 20 MHz channels and there are multiple Wi-Fi APs coexisting with LAA on the same set of channels, the choice of Wi-Fi primary channels can have a significant impact on LAA throughput., IEEE ICC 2021 Workshop on Spectrum Sharing Technology for Next-Generation Communications

To address time unfairness access problems in the dynamic channel, an underwater adaptive contention window (CW) adjustment backoff algorithm (QL-UACW) based on Q-learning is presented in this paper. That algorithm employs the concept of reinforcement learning to adaptively adjust the competition window by autonomously learning the network communication process. Based on QL-UACW and in order to address space unfairness access issues of nodes location, a load adaptive hybrid media access control (MAC) protocol with competition for the same directional nodes and time slot allocation for different directional nodes called SDHA is also presented. The protocol divides the nodes into co-nodes and counter-nodes according to the relative communication directions of the nodes. The co-nodes use competition protocol and the counter-nodes use time slot allocation protocol to solve the hidden terminal problem of node communication in order to prevent space access unfairness. Theoretical and simulation analysis show that the QL-UACW algorithm effectively improves the fairness of node access channels, reduces the collision rate of data frames, and increases network throughput. Additionally, the SDHA protocol is proven to provide advantages in normalized network throughput, average end-to-end delay, and the success rate of data frame transmission. The work is supported by “the National Key Research and Development Program, China under Grant No. 2018YFC0407900 ”, supported by “the National Natural Science Foundation of China under Grant No. 61971206 and 62072072 ”, supported by “the Open fund of State Key Laboratory of Acoustics, China under Grant No. SKLA201901 ”, supported by “ China Academy of Military Sciences Fund under Grant No. 2019 ”, supported by “ Liaoning BaiQianWan Talents Program, China under grant No. 2016 ”, supported by “ Natural Science Foundation of Liaoning Province Project, China under Grant No. 20170540793 ”.

Data augmentation aims to generate new and synthetic features from the original data, which can identify a better representation of data and improve the performance and generalizability of downstream tasks. However, data augmentation for graph-based models remains a challenging problem, as graph data is more complex than traditional data, which consists of two features with different properties: graph topology and node attributes. In this paper, we study the problem of graph data augmentation for Graph Convolutional Network (GCN) in the context of improving the node embeddings for semi-supervised node classification. Specifically, we conduct cosine similarity based cross operation on the original features to create new graph features, including new node attributes and new graph topologies, and we combine them as new pairwise inputs for specific GCNs. Then, we propose an attentional integrating model to weighted sum the hidden node embeddings encoded by these GCNs into the final node embeddings. We also conduct a disparity constraint on these hidden node embeddings when training to ensure that non-redundant information is captured from different features. Experimental results on five real-world datasets show that our method improves the classification accuracy with a clear margin (+2.5%–+84.2%) than the original GCN model.

Contention-based wireless channel access methods like CSMA and ALOHA paved the way for the rise of the Internet of Things in industrial applications (IIoT). However, to cope with increasing demands for reliability and throughput, several mostly TDMA-based protocols like IEEE 802.15.4 and its extensions were proposed. Nonetheless, many of these IIoT-protocols still require contention-based communication, e.g., for slot allocation and broadcast transmission. In many cases, subtle but hidden patterns characterize this secondary traffic. Present contention-based protocols are unaware of these hidden patterns and can therefore not exploit this information. Especially in dense networks, they often do not provide sufficient reliability for primary traffic, e.g., they are unable to allocate transmission slots in time. In this paper, we propose QMA, a contention-based multiple access scheme based on Q-learning, which dynamically adapts transmission times to avoid collisions by learning patterns in the contention-based traffic. QMA is designed to be resource-efficient and targets small embedded devices. We show that QMA solves the hidden node problem without the additional overhead of RTS / CTS messages and verify the behaviour of QMA in the FIT IoT-LAB testbed. Finally, QMA's scalability is studied by simulation, where it is used for GTS allocation in IEEE 802.15.4 DSME. Results show that QMA considerably increases reliability and throughput in comparison to CSMA/CA, especially in networks with a high load., Comment: 23 pages, 26 figures

The critical nature of vehicular communications requires their extensive testing and evaluation. Analytical models can represent an attractive and cost-effective approach for such evaluation if they can adequately model all underlying effects that impact the performance of vehicular communications. Several analytical models have been proposed to date to model vehicular communications based on the IEEE 802.11p (or DSRC) standard. However, existing models normally model in detail the MAC (Medium Access Control), and generally simplify the propagation and interference effects. This reduces their value as an alternative to evaluate the performance of vehicular communications. This paper addresses this gap, and presents new analytical models that accurately model the performance of vehicle-to-vehicle communications based on the IEEE 802.11p standard. The models jointly account for a detailed modeling of the propagation and interference effects, as well as the impact of the hidden terminal problem. The model quantifies the PDR (Packet Delivery Ratio) as a function of the distance between transmitter and receiver. The paper also presents new analytical models to quantify the probability of the four different types of packet errors in IEEE 802.11p. In addition, the paper presents the first analytical model capable to accurately estimate the Channel Busy Ratio (CBR) metric even under high channel load levels. All the analytical models are validated by means of simulation for a wide range of parameters, including traffic densities, packet transmission frequencies, transmission power levels, data rates and packet sizes. An implementation of the models is provided openly to facilitate their use by the community.

With the emergence of new technologies and standards for wireless communications and an increase in application and user requirements, the number and density of deployed wireless ad hoc networks is increasing. For deterministic ad hoc networks, Time-Division Multiple Access (TDMA) is a popular medium access scheme, with many distributed TDMA scheduling algorithms being proposed. However, with increasing traffic demands and the number of wireless devices, proposed protocols are facing scalability issues. Besides, these protocols are achieving suboptimal spatial spectrum reuse as a result of the unsolved exposed node problem. Due to a shortage of available spectrum, a shift from fixed spectrum allocation to more dynamic spectrum sharing is anticipated. For dynamic spectrum sharing, improved distributed scheduling protocols are needed to increase spectral efficiency and support the coexistence of multiple co-located networks. Hence, in this paper, we propose a dynamic distributed multi-channel TDMA (DDMC-TDMA) slot management protocol based on control messages exchanged between one-hop network neighbors and execution of slot allocation and removal procedures between sender and receiver nodes. DDMC-TDMA is a topology-agnostic slot management protocol suitable for large-scale and high-density ad hoc networks. The performance of DDMC-TDMA has been evaluated for various topologies and scenarios in the ns-3 simulator. Simulation results indicate that DDMC-TDMA offers near-optimal spectrum utilization by solving both hidden and exposed node problems. Moreover, it proves to be a highly scalable protocol, showing no performance degradation for large-scale and high-density networks and achieving coexistence with unknown wireless networks operating in the same wireless domain.

With the development of China’s sports industry, the technical and tactical level of the team is required to be higher and higher. This study mainly discusses the diagnostic model of volleyball technique and tactics based on artificial neural network. With the help of the correlation function in Matlab neural network toolbox, in the training process of volleyball technical and tactical evaluation neural network, the sample data of volleyball technical evaluation index is repeatedly simulated and studied, and finally the network parameters with the minimum error and the highest accuracy are saved as the network model for subsequent verification and evaluation. The middle layer is the hidden layer, which makes the network approach the result of volleyball experts’ evaluation of the technology by adjusting the weights of neurons. The last layer is the output layer, which outputs the actual evaluation results of volleyball experts on the technology. Through repeated training and comparison of input samples, the maximum number of training times of BP network for volleyball technical and tactical evaluation is determined to be 32. Some common experience of estimating hidden node number is provided by trial-and-error method. On this basis, the number of hidden nodes to minimize the network error is finally determined to be 4 through repeated training and comparison. In the process of network diagnosis, the average difference between the evaluation score of network output and the score of actual experts is less than 1%, which reaches a very high precision. It shows that the volleyball skill evaluation model based on BP neural network is feasible in technology and the result is relatively reliable.

The brain is intrinsically organized into large-scale networks that constantly re-organize on multiple timescales, even when the brain is at rest. The timing of these dynamics is crucial for sensation, perception, cognition, and ultimately consciousness, but the underlying dynamics governing the constant reorganization and switching between networks are not yet well understood. Electroencephalogram (EEG) microstates are brief periods of stable scalp topography that have been identified as the electrophysiological correlate of functional magnetic resonance imaging defined resting-state networks. Spatiotemporal microstate sequences maintain high temporal resolution and have been shown to be scale-free with long-range temporal correlations. Previous attempts to model EEG microstate sequences have failed to capture this crucial property and so cannot fully capture the dynamics; this paper answers the call for more sophisticated modeling approaches. We present a dynamical model that exhibits a noisy network attractor between nodes that represent the microstates. Using an excitable network between four nodes, we can reproduce the transition probabilities between microstates but not the heavy tailed residence time distributions. We present two extensions to this model: first, an additional hidden node at each state; second, an additional layer that controls the switching frequency in the original network. Introducing either extension to the network gives the flexibility to capture these heavy tails. We compare the model generated sequences to microstate sequences from EEG data collected from healthy subjects at rest. For the first extension, we show that the hidden nodes ‘trap’ the trajectories allowing the control of residence times at each node. For the second extension, we show that two nodes in the controlling layer are sufficient to model the long residence times. Finally, we show that in addition to capturing the residence time distributions and transition probabilities of the sequences, these two models capture additional properties of the sequences including having interspersed long and short residence times and long range temporal correlations in line with the data as measured by the Hurst exponent.

As Bring Your Own Device (BYOD) policy has become widely accepted in the enterprise, anyone with a mobile device that supports Wi-Fi tethering can provide an active wireless Internet connection to other devices without restriction from network administrators. Despite the potential benefits of Wi-Fi tethering, it raises new security issues. The open source nature of mobile operating systems (e.g., Google Android or OpenWrt) can be easily manipulated by selfish users to provide an unfair advantage throughput performance to their tethered devices. The unauthorized tethering can interfere with nearby well-planned access points (APs) within Wi-Fi networks, which results in serious performance problems. In this paper, we first conduct an extensive evaluation study and demonstrate that the abuse of Wi-Fi tethering that adjusts the clear channel access parameters has strong adverse effects in Wi-Fi networks, while providing the manipulated device a high throughput gain. Subsequently, an online detection scheme diagnoses the network condition and detects selfish tethering devices by passively exploiting the packet loss information of on-going transmissions. Our evaluation results show that the proposed method accurately distinguishes the manipulated tethering behavior from other types of misbehavior, including the hidden node problem.

For smooth integration with middleboxes on the Internet, TCP (Transmission Control Protocol) is favorably considered as a transport-layer protocol for IoT (Internet of Things) networks. In constrained networks, TCP tends to perform well with a small window size. For example, the uIP (micro IP) TCP/IP stack sets the TCP window size to one segment by default. In such a case, managing the retransmission timer is a primary approach to congestion control. In this paper, we examine the congestion control mechanism of TCP in the uIP stack using grid topology networks. In the preliminary test using the Cooja network simulator, the results show that the original uIP TCP causes lots of retransmissions when a radio duty cycling mechanism such as ContikiMAC is used. One main reason is that, once retransmission is deemed to be necessary, the original uIP TCP sets the retransmission timer based on the fixed RTO (retransmission timeout) before performing a retransmission. Since ContikiMAC may cause large RTT (round-trip time) variations due to the hidden terminal problem, the retransmission timer based on the fixed RTO value may cause lots of retransmissions. To address the problem, we propose a new scheme for managing the retransmission timer which adopts the notion of weak RTT estimation of CoCoA, exponential backoffs with variable limits, and dithering. Simulation results show that our proposed scheme reduces retransmissions while enhancing throughput and fairness when an RDC (radio duty cycling) mechanism is used.