Journal of Communication Engineering
Volume:11 Issue: 1, Winter-Spring 2022

For Orthogonal Frequency Division multiplexing (OFDM) systems in environments with high mobility and non-stationary channel characteristics, channel estimation is a very challenging task. To handle this issue, a deep learning (DL)-based channel estimation and data extraction algorithm is proposed. The purpose of this paper is to analyse DL-based OFDM data extraction algorithm in time-variant Rayleigh fading channels. Moreover, the model is examined in time-invariant environments. The proposed long short-term memory with projection layer (LSTMP) model, can not only exploit the features of channel variation from previous channel estimations, but also extract more features from pilots and received signals. Moreover, the LSTMP can take advantage of the LSTM estimation to further improve the performance of the channel estimation by reducing the complexity and increasing the accuracy. The LSTMP is first trained with simulated data in an offline manner and then tracks the dynamic channel in an online manner. The simulation results show that the proposed LSTMP model algorithm can be effectively employed to adapt to the characteristics of time-variant channels, compared to the conventional algorithms. Additionally, the trade-off between accuracy and complexity is discussed and compared with that of Convolutional Neural Network (CNN) and LSTM.

Three ultra-compact half-mode substrate integrated waveguide (HMSIW) equal/unequal filtering power dividers (FPDs) based on the metamaterial concept and the evanescent mode technique are recommended in this paper. The spiral technique is a well-known technique in planar microwave circuitry. By combining the spiral technique and the conventional complementary split-ring resonator (CSRR) unit cell, the complementary spiral resonator (CSR) unit cell can be achieved. The resonance frequency of the two-turns CSR unit cell is half the resonance frequency of the conventional CSRR unit cell with the same size and shape. Accordingly, the electrical size of the CSR unit cell is smaller than the conventional CSRR unit cell with the same physical size. According to the evanescent technique, by loading the CSR unit cell on the metal surface of the HMSIW structure, an extra forward passband can be obtained. Therefore, three miniaturized equal/unequal FPDs with arbitrary power-dividing ratios of 1:1, 1:4, and 1:8 have been designed, simulated, fabricated and measured based on the compact HMSIW-CSR structure. A reasonable agreement between simulated and measured results has been achieved. The fractional 3-dB bandwidth of the designed equal/unequal FPDs are approximately 21 % at 2.4 GHz. The whole dimension of the proposed FPDs is about 0.06 λg × 0.06 λg which confirms the small size of the presented structures.

Cyber-physical systems (CPSs) are deeply intertwining and integrating the physical processes with cyber components. In these intelligent systems, a process is monitored and controlled by cyber systems and different types of sensitive information is exchanged in a real-time manner. Nowadays, the security of these systems has been considered increasingly. Connecting physical devices to the cyber network makes the critical infrastructures more vulnerable to the adversarial activities. The primary target of attacks against CPSs is often disrupting physical processes under control. Since, improving the security of CPSs has gained considerable importance nowadays. This paper presents a method for modeling the security of CPSs using stochastic Petri nets (SPNs). The proposed method models the system control loop associated with anomaly detection systems (ADSs) in normal behavior and under security attacks. By using this model, we can investigate the consequences of the integrity and denial of service attacks against CPSs and perform probabilistic and temporal analysis of the system under security attacks. By solving the proposed model, the security of CPSs is estimated in terms of metrics, such as mean-time-to-failure and availability. Finally, the security of a chemical plant is investigated as an illustrative example to represent the effectiveness of the proposed modeling method.

Receiver structure either coherent or non-coherent in a relay channel with amplify and forward (AF) and decode and forward (DF) relaying schemes is studied and a new receiver is proposed for both relaying schemes. In the network with multiple relays the performance of the receiver at the destination node is categorized based on the available source-relay (S-R) channel state information (CSI) including no, partial, and full CSI. Next, an adaptive receiver is proposed where full CSI of the S-R channel is exploited in the receiver of the destination node. It is shown that the proposed receiver outperforms other designed receivers in the sense of bit error rate (BER) criterion.

A new circularly polarized (CP) dielectric resonator (DR) antenna is addressed in this research. A cross-shaped slot embedded on the ground plane is used as a feed of a sandglass-shaped dielectric resonator (SSDR). An analytical method is also proposed to determine the initial dimensions of the proposed SSDR. The performance of the DR antenna is investigated using two electromagnetic (EM) full-wave simulators, including CST Studio Suite and Ansys HFSS. To this end, the most important antenna parameters such as the antenna gain, reflection coefficient, radiation pattern, total efficiency and axial ratio are reported. The numerical results show that the introduced DR antenna provides 600MHz (12.25%) impedance bandwidth from 4.6-5.2 GHz at the operation frequency of 4.9GHz. Correspondingly, the proposed antenna provides the axial ratio bandwidth (AR< 3dB) range of 320MHz (6.5%) from 4.85-5.17 GHz and high gain around 7.3dBi with more than 94% total efficiency. The designed antenna is suitable for wireless systems such as Wi-Fi and Wi-Max.

Creating a secure communication channel in hierarchical networks is a very important issue and several algorithms have been proposed for it. Unfortunately, due to limited resources in fog networks, which are a special type of hierarchical networks, it is not possible to use conventional algorithms. In this article, we have presented an algorithm to create a secure communication channel on fog networks, which is based on key pre-distribution and is used for multi-cloud fog networks. In this method, by using the SBIBD, blocks are generated to be assigned to the cloud nodes, and by using the residual design on the SBIBD, classes and blocks are created to be assigned to the fog nodes and end devices. The results show that the proposed method increases scalability and reduces communication, memory and computing overheads. The probability of capturing the network in the proposed method is about 0 and its connectivity is about 1.

In the recent research on Vehicular Ad-hoc Networks (VANETs), new practical goals are pursued. They provide real-world communications between vehicles and make them reliable and easily used. The VANETs have a fundamental role in reducing traffic accidents and improving traffic on the roads. Authentication in VANETs is a critical security service, and vehicles should be protected from breaking their personal information. Vehicles can be traced and investigated in the event of an accident or liability arising out of non-repudiation when the vehicle is faced with a rush of incoming messages. Hence, the Roadside Units' (RSUs) efficiency is reduced and causes delays in checking messages. This study presents an authentication framework using proxy vehicles for VANETs. Reducing the computational cost and proficiency increment are the features of the proposed method on the RSUs side. The proposed framework supports managing the revocation list.  The Vehicle-to-Infrastructure (V2I) and Vehicle-to-Vehicle (V2V) authentication protocols are guaranteed and designed in this proposed framework, therefor, a merged V2I and V2V authentication is presented and embedded in the proposed framework. The designed protocol applies offline and online signatures to check messages, the revocation key to prevent malicious messages from being sent, and the time limit to use the network. The analysis shows that the suggested protocol is more feasible and reasonable for use in VANETs.

Today, fog computing plays an essential role in human life. One of the challenges in the fog and cloud environment is the hierarchical service process. Requests are sent to Fog, and if Fog cannot provide service, they are sent to cloud, which is a time-consuming process. This paper provides a framework that specifies when a request is sent, in which environment it can be serviced, and provides interfaces for properly managing nodes and domains and managing the service of requests. Two new architectures have been presented in the management interfaces. In one of these management interfaces, the most appropriate domain is determined using the SAW method of game theory and user expectations for placing the application. Then, in the other management interface specified in the domain gateway, it suggests the most appropriate node using the PSO algorithm. Since the placement of the application is based on the expectations of the users, it increases the quality of the service. The proposed method has been implemented in iFogSim and its results have been evaluated with authentic articles. It was observed proposed method has better performance and better service speed than the state-of-the-art research works and significant improvement in service response time.