Search

The growth of the number of connected devices in Internet of Things (IoT) systems causes a huge increase in network traffic. Thus, there is a significant demand for systems that can predict the measurements of the distributed IoT-based applications to mitigate the increasing network traffic. Existing methods utilized a distributed scheme, i.e., dual prediction schemes (DPS), to achieve this task. The idea of this scheme is based on deploying a predictive model on the data sources (i.e., sensors) and the fusion center in a distributive manner. The state-of-the-art results can be simply reached using predictive models, e.g., adaptive filters. Deep learning-based approaches are not well utilized to address this problem. In this context, we proposed a distributed scheme utilizing fog and edge computing technologies. The proposed DPS includes an LSTM predictive model to reduce the data transmission instances of connected IoT devices; it is called LSTM-DPS. In addition, we proposed an updating mechanism that updates the LSTM model according to a set of tracking parameters that observes the model behavior during the deployment due to the changes of data properties through time. The proposed updating mechanism guarantees that the deployed LSTM model is identical at the data source and fusion center. The LSTM-DPS is evaluated using two real datasets. The obtained results show that the LSTM-DPS outperforms state-of-the-art methods in terms of communication reduction ratios.

A cooperative molecular communication (CMC) system is considered inside a cylindrical-shaped channel where a few cooperative nodes (CNs) are intermediately placed between a transmitter (TX) and a fusion center (FC). The expressions for the maximum achievable rate and probability of error at the FC considering AND and OR rules are derived. Furthermore, the performance of the CMC system in a cylindrical channel is compared with the direct and CN-assisted systems. The CMC system with randomly-placed CNs is also analyzed and compared with the uniformly-placed CNs, and it is found that a lower probability of error is obtained in the case of uniform placement\ud of CNs. Furthermore, the system performance as a function of radial displacement of TX and FC under constant flow is compared with that under laminar flow and a higher probability of error is observed under laminar flow. The increased probability of error under laminar flow occurs due to the fact that the drift velocity decreases towards the walls of the cylindrical channel. The analytical expressions are verified using Monte-Carlo simulations.

Spectrum sensing is a fundamental component in a cognitive radio network to utilize the frequency bands effectively. In this article a fast Riemannian distance-based K-medoids (FRDK)-based cooperative spectrum sensing (CSS) method is developed to identify the state of primary user (PU). In particular, two CSS scenarios are considered, one is secondary users (SUs) with a single antenna and the other is SUs with multiple antennas. In the multiantenna case, a Riemannian mean-based data fusion method is proposed to fuse sensing data from SUs with multiple antennas. To implement CSS, we propose a FRDK-based framework, where SUs collect sensing data, preprocess, and upload these data to an appointed fusion center (FC). Then, the FC transforms these data as samples on a manifold and uses the FRDK algorithm to train a classifier for identifying the state of PU. Furthermore, the convergence and the complexity analysis of training process are presented. Finally, the effectiveness of the proposed FRDK-based CSS method is verified under different conditions.

Deep cooperative sensing is a cooperative spectrum sensing (CSS) algorithm based on a deep neural network (DNN). Since training DNN requires a large amount of sample data, what is worse existing algorithms directly send training data to the fusion center (FC), which greatly occupies the transmission channel. Motivated by this, in this paper, we introduce the federated learning framework to CSS and propose a federated learning-based spectrum sensing (FLSS) algorithm. In the federated learning framework, there is no need to gather data together. Each secondary user (SU) uses local data to train the neural network model and sends the gradient to FC to integrate the parameters. This framework can perform collaborative training while ensuring local data privacy and greatly reducing the traffic load between SU and FC. Besides, we adopt an efficient network model ShufflenetV2 to reduce the number of parameters and improve training efficiency. Simulation results demonstrate that the FLSS can achieve a detection probability of 98.78% with a false alarm probability of 1% at SNR = -15dB.

In this paper, we detect illegal spectrum access behaviors in the dynamic spectrum sharing of the cognitive wireless networks, in which multiple authorized users (AUs) work in the same channel. First, we detect whether there are users occupy the channel, and if the channel is occupied, then recognize whether there is the illegal user (IU) and infer the number of AUs roughly. In the light of the thorny challenge that legal spectrum utilization behaviors and illegal spectrum access behaviors coexist probabilistically, the spectrum sensing problem is formulated as a mathematical model of ternary hypothesis test. Moreover, when multiple AUs are working at the same time, it is extremely difficult to detect IU. To tackle the problem, taking into account the various potential combinations of AUs and IU, we exploit a two-step detector and derive related detection schemes based on the generalized likelihood ratio test (GLRT), the Rao test, and the Wald test following the generalized multi-hypothesis Neyman-Pearson (GMNP) criterion respectively. Additionally, we design a cooperative spectrum sensing (CSS) scheme on the basis of the global GMNP criteria, in which the distributed detection framework composed of K sensing nodes and a fusion center (FC). Finally, to verify the performance of the proposed detection scheme in a variety of parameter configurations, we provide comprehensive simulations and find that there will be at most (2N-1) detection valleys when N AUs work at the same time. Compared with single-sensing node detection, the CSS scheme can significantly improve the detection performance of the proposed detection schemes, especially in the upper bound and valley of the detection performance.

Economic power dispatch (EPD) is a significant managing problem in power systems. Control and management programs in large-scale systems, especially the power system, can be run centrally or in a distributed manner, each with its own advantages and disadvantages. If the power system is managed as a central fusion center, it will be vulnerable to natural disasters and cyber-attacks on the control center, despite its fast processing speed. On the other hand, if power system is managed in a distributed manner, information will be exchanged between all buses, and there is a risk that the security of the system will be compromised and intercepted by spreading information on all buses. The combination of security and system speed at the same time can ensure that power system needs are met. This paper addresses the solution of the EPD by defining a novel computational architecture, namely hybrid network architecture that merges both distributed and centralized architectures to speed up obtaining the solution of the EPD in large-scale power systems, while maintaining the necessary security level. This proposed method allows adjusting the convergence speed of the operating system along with setting the cybersecurity of the distributed system. This can play a key role in operating system reliability, which includes speed and security. To perform the optimization step, using the alternating direction method of multipliers (ADMM) framework and hybrid network architecture, a decentralized algorithm, named hybrid ADMM (HADMM), is developed to be used in solving the EPD problem. Besides, an acceleration scheme that exploits the relationship between ADMM and Douglas–Rachford splitting is proposed, which results in a faster algorithm called accelerated hybrid ADMM (AHADMM). The simulation results of the proposed algorithm applied to solve the EPD problem on the IEEE 118-bus system show the acceptable performance of the proposed algorithm under different scenarios.

In this paper, we consider a two-way network-coded relaying system consisting of two Primary Users (PUs) communicating via the relay node as the primary network with licensed spectrum. Multiple IoT devices capable of harvesting energy from a power beacon, utilize the same spectrum as Secondary Users (SUs). The SUs cooperate with the primary network and then transmit their own data. The devices harvest their required energy from a power beacon (PB), located near the primary relay node and store the energy in finite-size batteries. In phase one, the primary network communicate using the network coding technics and in the second phase, the SUs cooperate with the relay node by retransmitting the relay’s data to PUs. In the last phase, SUs transmit their gathered information to a fusion center (FC). We apply TDMA and Signals Combining (SC) schemes in cooperation phase and derive the closed-form expressions for the outage sumrate of the secondary network. To increase the outage sumrate of the secondary network, a new NOMA-based simultaneous cooperate and transmit (NBSCT) scheme with a new time schedule is proposed. The corresponding expressions are derived and compared with that of the mentioned methods for various network parameters. The numerical results verifies that the outage sumrate of secondary network in proposed NBSCT scheme outperforms the TDMA and SC schemes under Rician fading and improves the outage sumrate by approximately $$9.8\%$$ .

In this article, we investigate the analog–digital hybrid transceiver optimization for distributed Internet-of-Things (IoT) sensing networks consisting of a multiantenna fusion center (FC) and several multiantenna sensor nodes. Analog–digital hybrid transceiver is an economic way to realize tradeoffs between hardware cost and performance for multiantenna communications. Under the nonconvex unit modulus constraints and transmit power constraint at each sensor, two synchronization schemes are considered for the hybrid linear minimum mean-square error (LMMSE) transceiver optimization. First, a centralized algorithm is proposed, in which the hybrid transceivers are computed at the FC. Based on the framework of alternating direction method of multipliers (ADMMs), the unit modulus constraints can be satisfied by projecting the elements of analog transceivers onto the unit modulus circle. However, the centralized algorithm usually suffers from strict synchronous requirements and high communication overhead. In order to accommodate the inevitable computing and communication delays in distributed IoT sensing networks, an asynchronous distributed ADMM (AD-ADMM) algorithm is proposed. By using the aged information, the hybrid transceivers are computed at the sensors without the coordination of the FC. Thus, the AD-ADMM algorithm can greatly reduce the computation overhead of the FC and improve the scalability of IoT sensing networks. Simulation results are presented to show that both the centralized ADMM and AD-ADMM algorithms perform closely to the fully digital counterpart.

This article is concerned with the distributed networked fusion estimation problems for unstable systems with limited communication capacity, where the system and measurement noises are unknown but bounded. To overcome the unboundedness of measurement information for unstable systems, it is proposed to quantize the innovations that are sent to the fusion center over bandwidth constrained channels. By using the bounded recursive optimization idea, the design problems of local stable estimators and distributed fusion criterion under the quantized innovations are converted into two different convex optimization problems that can be easily solved by the standard software. The target tracking system is employed to demonstrate the effectiveness of the proposed methods.