IoT as a Service

In this paper, we present a parallel Sliding-Window Belief Propagation algorithm to decode Q-ary Low-Density-Parity-Codes. The bottlenecks of sequential algorithm are carefully investigated. We use MATLAB platform to develop the parallel algorithm and run these bottlenecks simultaneously on thousands of threads of GPU. The experiment results show that our parallel algorithm achieves 2.3$$\times $$× to 30.3$$\times $$× speedup ratio than sequential algorithm.

In recent years, the global low-orbit communication and Internet constellation have entered a period of vigorous development. Low-orbit satellites have the advantages of high landing level and fast satellite geometry change, and the enhancement function of low-orbit navigation can complement the current mid-high-rail GNSS implementation and promote the deep integration of navigation communication, which is an important direction for the future development of the navigation system. Hongyan Navigation System is a low-orbit mobile communication and broadband internet constellation independently developed by China, which integrates navigation enhancement functions, and has real-time communication capability in full-time and complex terrain conditions to provide users with real-time global data communication and integrated information services. However, the Hongyan satellite has the characteristics of fast moving speed, short single-visible time, and greater doppler dynamics received by the users, which is not suitable for the traditional capturing methods. Therefore, it is of great research value and practical significance to study the Hongyan navigation signal. This paper studies the spread spectrum modulation and the synchronous acquisition of the transmitting signal based on the Hongyan constellation. On the Matlab platform, the excellent characteristics of the transmission signal of Hongyan system and the time-sequence capture method are verified.

The wireless universal short-distance network refers to a heterogeneous network that combines multiple wireless short-distance networks such as wireless local area networks and wireless personal area networks. Users can choose to use different multiple access protocols to access the network according to the characteristics of the service. This paper proposes an FPGA-based reconfigurable MAC architecture, which uses a combination of software and hardware to select the working mode of the universal MAC in the FPGA through ARM, so as to achieve fast switching between different networks. The proposed universal MAC module includes a control module based on finite state machine and a data frame transceiver module, it implements the function of switching between different networks by configuring the control module as different finite state machine of different MAC protocols. A universal MAC module combining ALOHA protocol and CSMA/CA protocol is designed and implemented in this paper. The simulation results show that the designed universal MAC module is equivalent to the CSMA/CA protocol in terms of resource utilization, and can be flexibly switching between ALOHA protocol and CSMA/CA protocol.

BOC (Binary-Offset-Carrier) modulation is widely used in the new generation of satellite navigation system. It has been used in GPS, Galileo and BeiDou systems. Compared with the traditional BPSK (Binary Phase Shift Keying) modulation, it has sharper correlation characteristics and stronger anti-multipath performance. The spectrum relocation caused by its splitting frequency can effectively solve the current frequency band congestion problem. But BOC modulation has multiple peak characteristics in the time domain, and the existence of sub-peak increases the difficulty of acquisition. Adding a pseudo-code correlation branch to the original BOC code correlation branch by using autocorrelation side-peak cancellation technique (ASPeCT) can effectively weaken the sub-peak of the correlation function and avoid wrong acquisition. This paper applies this ASPeCT acquisition algorithm to the parallel pseudo-random code FFT acquisition strategy to acquire the BeiDou-3 navigation signals. And the research object is the BOC (1,1) signal of the pilot component in BeiDou-3. Experimental results show that the method adopted in this paper can ensure the acquisition time and the acquisition accuracy reaches half a chip.

Wireless directional ad hoc networks has attracted much attention from academic area in recent years due to its large antenna gain, small multipath interference, long propagation distance and space reusability. However, since the signal in other directions cannot be well sensed, the environment of the communication channel cannot be predicted, thus causing interference between multiple concurrent transmission links. Therefore, it is extremely important to study the wireless directional ad hoc networks and use its advantages to overcome its shortcomings. In this paper, the concurrent link interference problem in the wireless directional ad hoc networks is analyzed. Firstly, this paper analyzes the advantages and disadvantages of the classical Push mechanism and Pull mechanism. Then, a network resource allocation protocol, CSPC (Collision Scattering and Push-Pull Cascading), is proposed for the transmission and reception of network resources. CSPC protocol mainly uses the idea of collision scattering to increase the proportion of data transmission success by using frequency division of the time slot in the time slot where data transmission fails. The simulation results show that using CSPC protocol can efficiently increase the network throughput and solve the collision problem between concurrent transmission links.

In this paper, an acoustic filter with low frequency bandstop and a broadband (0.3–3 kHz) bandstop filter are presented. In this paper, control algorithm and related equipment are integrated with piezoelectric ceramic thin film, ADC and FPGA to study a system which can mix acoustically in the process of acoustic transmission and only allow one-way transmission of sound waves.

With the rapid development of the technology and wireless communication, the user cannot support the computation-intensive applications, owing to the restricted computation resources, energy supply, limited memory space and communication resources. The emerging computation mode, called mobile edge computing (MEC), provides a solution that the user can unload parts of tasks to edge servers. This communication process should be finished in the wireless network. However, computation offloading in the wireless network can encounter many kinds of attacks. Specifically, edge servers located in the edge of network are vulnerable to these security threats, such as spoofing, jamming and eavesdropping. Moreover, the computation offloading has much time latency and energy consumption. Then, how to minimize this consumption is the another problem to be solved. To improve the security and minimize the consumption, we formulate a system containing a primary user (PU), a second user (SU), an attacker and several edge servers. They communicate with each other by multiple input multiple output (MIMO) technology. In this system, the SU chooses an MEC server from the set of not being occupied by PU, determines an offloading rate and a transmission power, then the attacker selects the action of attack. The aim of this system is to optimize the utility of SU. To solve this problem, a Q-learning based optimal offloading strategy is proposed in dynamic environments. Simulation results show that our proposed scheme can improve the capacity of SU and efficiently decrease the attack rate of the attacker.

Network slicing is a novel technology to effectively provide solutions for heterogeneous mobile service requirements in 5G network. Meanwhile, the shortage of spectrum resources becomes more severe with massive access requirement of Internet-of-Things (IoT) applications. In this paper, we study how to allocate spectrum resources to satisfy the diversified traffic requirements with network slicing and improve the utilization of spectrum resources. A spectrum resource allocation model with three layers is considered, including operator layer, slice layer and user layer. At the mobile operator layer, mobile operators can borrow frequency bandwidth from one another to improve the spectrum efficiency. Then, the mobile operator allocates its frequency bandwidth to the slices according to users’ demand. At last, the slice assigns bandwidth to users. A network utility maximization problem is formulated and a distributed resource allocation algorithm is proposed based on alternating direction method of multipliers (ADMM). Simulation results show that the proposed algorithm can quickly converge to the optimal solution. In addition, the overall network utility can be effectively improved by lending or borrowing spectrum resources among operators. By adding the slice layer, the network can meet different types of service requirements.

This paper studies a unmanned aerial vehicle (UAV)-enabled wireless relay communication system, where a UAV provides relay services to a source node and a destination node. To minimize the total power consumption of UAV, we jointly design the trajectory of UAV and transmission power of both UAV and source node, under the constraints on minimum data rate, transmission power, UAV’s mobility, and information-causality. The obtained problem is non-convex and difficult to solve in general. Hence, a successive convex approximation based optimization method is proposed to solve the problem approximately. Numerical simulations are performed to show effectiveness of the proposed method.

Higher order modulation and spatial modulation schemes improve the data rate for differential chaos shift keying (DCSK) systems by transmitting more bits in one time slot, however, the reliability performances become worse respectively due to more dense signal distributions and bad channel condition of active antenna. In this paper, we propose to utilize the antenna and symbol indexes to improve both the data rate and the reliability performances for DCSK systems with aid of antenna selection. Therein the information is modulated cooperatively using the antenna index and the symbol index. Then with the aid of the feedback channel state information (CSI) from receivers, the antenna with the best CSI is selected to transmit modulated symbols, which helps to improve the reliability performance. At the receiver, reverse operations are performed. Furthermore, we provide the theoretical symbol error rate (SER) and bit error rate (BER) over the multi-path Rayleigh fading channel. Simulation results demonstrate that the proposed scheme achieves better reliability performances than the counterpart schemes with the same data rate.

Vehicular edge computing (VEC) has emerged as a promising paradigm to provide low-latency service by extending the edge computing to vehicular networks. To meet the ever-increasing demands of computation and communication resources, utilizing vehicles as augmented infrastructure for computation offloading is an appealing idea. However, due to the lack of effective incentive and task allocation mechanism, it is challenging to exploit vehicles as infrastructure for computation offloading. To cope with these challenges, we first propose a container-based VEC paradigm by using efficient, flexible and customized resources of the vehicles. Then, we present a contract-based incentive mechanism to motivate vehicles to share their resources with service requesters (SRs). The optimal contract items are designed for multiple types of vehicles while maximizing the expected utilities of the SRs. Numerical results demonstrate that the proposed contract-based incentive mechanism is efficient compared with conventional schemes.

With the rapid development of flash memory technology, the buffer capacity of device becomes higher. In this case, infinite buffer could be introduced to derived the upper bound of system performance. In this paper, a new method which considers Enqueue rate and Dequeue rate of buffer is proposed to derive the performance of OCR-TDMA when buffer length of relay $$L = \infty $$L=∞. In simulation part, buffer lengths from 1 to $$\infty $$∞ are considered as parameters to compare system performances, and the theoretical results and simulation results match well. Therefore, the proposed method can be applied in other similar model to simplify the derivation of system performance.

For edge devices, pushing the task to other near devices has become a widely concerned service provision paradigm. However, the energy-constrained nature of edge devices makes optimizing for Quality of Service (QoS) difficult. We choose three factors as QoS: the delay limitation, the CPU usage of terminal and energy consumption. Due to the delay limitation of different tasks for edge computing and the different rates in heterogeneous wireless networks, we propose a network selection task migration algorithm based on Q-learning that captures the trade-off between QoS and energy consumption. Our approach can automatically choose a suitable network to perform task migration reasons about the task’s QoS requirements and computing rate in 4G network, Wi-Fi, Device-to-device (D2D). We demonstrate a working prototype using the YOLOv3 on the Vivo X9 devices. Based on real hardware and software measurements, we achieve 27.79% energy saving and 35% reduction in delay.

In this paper, we present a vehicle-to-vehicle (V2V) wireless channel measurement at 5.9 GHz in the tunnel environments. The small scale fading characteristics are analyzed for outside and inside the tunnel, and the conjunction part in between. We evaluate the received signal magnitude inside the tunnel by comparing its distribution with five typical theoretical fading distributions. The best fit among the considered fading distributions is found to be Rician distribution that has the lowest Goodness of Fit (GoF) indicator. The K-factor calculated from measurements data inside the tunnel is lower than the values obtained outside the tunnel. Further, the K-factor is found to be dependent on the transmitter (Tx)-receiver (Rx) distance in the considered scenario.

Wireless Local Area Network (WLAN) has been greatly developed for the last twenty years. Quality of Service (QoS) and Quality of Experience (QoE) in high-dense deployment scenario greatly challenges the next-generation WLAN [1]. To address this challenge, researchers proposed channel reservation mechanism for WLAN, but the premise of channel reservation mechanism is to know the next packets exact arrival time. This assumption is impractical since the next packets arrival time is a stochastic process. In order to solve the above problem, this paper proposes an arrival prediction based channel reservation media access control (MAC) for the next generation WLAN. Simulations show that the protocol can reduce the network collisions generated by concurrency to a certain extent and improve network throughput.

In the next generation Wireless Local Area Network (WLAN), IEEE 802.11 regards improving the throughput of the network as a major technical goal. Distributed Multiple Input Multiple Output (D-MIMO) System is an important issue to improve system capacity efficiency, but D-MIMO technology requires extremely high clock consistency between nodes. Therefore, this paper proposes a synchronization protocol based on Orthogonal Frequency Division Multiple Access (OFDMA) for D-MIMO. Firstly, this paper chooses a new clock synchronization process for the synchronization protocol (Two-Way Message Exchange), which is characterized by high synchronization accuracy. Second, we propose and refine the procedure of the synchronization protocol and design the frame structure of the protocol to make it compatible with IEEE 802.11 frame format. The simulation results show that with the increase of the number of APs in D-MIMO, proposed synchronization protocol reduces the overhead by nearly 50% compared with single user (SU) based synchronization protocol. Moreover, the clock synchronization accuracy increases with the increase of the number of information interactions.

Video-based analysis technology has a wide range of applications in intelligent transportation system (ITS). Vehicle segmentation and behavior analysis has become an important research area in traffic video analysis. To solve the problem of 2D video detection technology in actual traffic video scenes, a bottom-up analysis method is employed to study the related technical problems. Firstly, M-BRISK descriptor algorithm is proposed for describing local feature points, which based on the method of original BRISK. Secondly, a 3D feature analysis method based on rigid motion constraints for vehicle trajectory is proposed. With the result of camera calibration and the preset back-projection plane, the 2D trajectory points can be back-projected to the 3D space, and the back projection data of the 2D image can be reconstructed in 3D space. Thirdly, similarity measure method is proposed for achieving the trajectory clustering. The experimental results show that the proposed method not only accelerates the speed of clustering method, but also improves the accuracy of trajectory clustering at some extent. Moreover, the vehicle motion information contained in the trajectory data can be analyzed to recognize vehicle behavior. All of these provide an important data foundation for vehicle abnormal behavior detection and the identification of traffic status levels in traffic scenes.

This paper investigates optimal distributed opportunistic channel access in wireless cooperative networks with multiple relays deployed. While probing all potential relay channels could result in significant overhead and spectrum efficiency affected, distributed OCA strategies with smart relays probing is studied in this research. To achieve reliable communications of high efficiency, number of probed relays and way to use have to be carefully decided in a dynamic manner. Finding that the sequential channel probing and access are coupled, an optimal distributed OCA is much challenging, and main difficult lies in how to exploit multi-source diversity, multi-relay diversity and time diversity in full manner. To tackle this problem, an analytical framework is built based on theory of optimal sequential observation planned decision. This decision-theoretic approach integrates the design of MAC layer and physical layer, enabling smart probing and cooperative transmissions under multiple relays. Based on it, an optimal DOCA/SP strategy is proposed to maximize average system throughput, and the optimality is rigorously proved. The implementation is described, and through numerical and simulation results effectiveness is validated.

Nowadays, mobile edge computing (MEC) has become a promising technique to provide mobile devices with intensive computation capability for the applications in the Internet of Things and 5G communications. In a MEC system, a mobile device, who has computation tasks to complete, would like to offload part or all the data for computation to a MEC server, due to the limit of local computation capability. In this paper, we consider a MEC system with one MEC server and multiple mobile devices, who access into the MEC server via frequency division multiple access (FDMA). The energy consumption of all the mobile devices is targeted to minimized via optimizing the computation and communication resources, including the amount of data for offloading, the bandwidth for accessing, the energy budget for offloading data, the time budget for offloading, for each mobile device. An optimization problem is formulated, which is non-convex. We decompose it into two levels. In the lower level, a convex optimization problems is formulated. In the upper level, a one-dimensional variable is to be optimized by bisection search method.

For Space-based internet of things (S-IoT) application scenario, a Dynamic Maximum Iteration Number (DMI) scheduling decoder for LDPC codes is proposed. Distinct from traditional Static Maximum Iteration Number (SMI) scheduling LDPC decoder using fixed maximum iteration number, our DMI LDPC decoder has extra circuit to obtain the dynamic maximum iteration number, which could improves BER performance only at the expense of a slightly logic resources and a small ratio of memories, compared with conventional SMI scheme. Therefore, the DMI decoder is very suitable for the fluctuation of signal-to-noise ratio of S-IoT link.

In order to manage real-time ambient temperature values at all times and places, this paper realizes a 4G-based temperature sensor system for the internet of things. Sensor platform gets the temperature values from wireless terminal nodes through the ZigBee coordinator, and upload it to the remote MySQL database server via 4G. In addition, an application is developed on mobile phone for users to obtain real-time temperature values, and a website for PC users is set up to realize the temperature values searching and displaying in real time.

The point cloud obtained by the 3D laser scanner contains a very large amount of data, in order to transmit the point cloud data as much as possible with the limited bandwidth, the effective compression of point cloud data has become a problem that needs to be solved urgently nowadays. In this paper, we use the compressive sensing theory to compress and reconstruct one of the point features, that is, the Y color component, served as the signal. We also use the K-SVD algorithm to explore the signal’s sparsity according to its unique structural features, the K-SVD algorithm can learns a sparse basis matrix that is common to all point cloud models used in our experiments. For experimental results, we use rate-distortion metric. The results show that for each point cloud model, our method can achieve a higher probability to reconstruct the original data after compressed.

In this paper, the fundamental problem of multiple secondary users (SUs) contending for opportunistic spectrum sensing and access over multiple channels in cognitive radio networks is investigated, when sensing is imperfect and each SU can access up to a limited number of channels at a time. For each channel, the busy/idle state is independent from one slot to another. The availability information of channels is unknown and has to be estimated by SUs during channel sensing and access process. Learning loss, also referred as regret, is thus inevitable. To minimize the loss, we model the channel sensing and access process as a multi-armed bandit problem, and contribute to proposing policies for spectrum sensing and access among multiple SUs under both centralized and distributed framework. Through theoretical analysis, our proposed policies are proved with logarithmic regret asymptotically and in finite time, and their effectiveness is verified by simulations.

In this paper, we study the distributed DOS problem for wireless multiple relay networks. Formulating the problem as an extended three-level optimal stopping problem, an optimal strategy is proposed guiding distributed channel access for multiple source-to-destination communications under the help of multiple relays. The optimality of the strategy is rigorously proved, and abides by a tri-level structure of pure threshold. For network operation, easy implementation is presented of low complexity. The close-form expression of the maximal expected system throughput is also derived. Furthermore, numerical results are provided to demonstrate the correctness of our analytical expressions, and the effectiveness is verified.

The application of satellite technology in the Internet of Things (IOT) can just make up for the defects of the ground system for its wide coverage and anti-damage. More and more satellites will participate in IOT. Due to the environmental protection exhaust and high specific impulse of cryogenic propellants like liquid hydrogen and liquid oxygen, they will play an important role in satellite applications. Cryogenic liquid storage is difficult and self-pressurization phenomenon often occurs. Pressure rise prediction with high accurate is necessary when designing tank for storage. Numerical calculation of computational fluid dynamic model and experiments are always time and financial consuming. A theoretical thermal diffusion model is investigated in the paper by using a concentration parameter model in the vapor and a one-dimensional heat conduction model in the liquid. The validation of the predictive capability is conducted by comparing the predictions with experimental data. Favorable agreement is found for both the experimental cylindrical and oblate spheroidal tanks. The effect of fill level and tank size is also studied.

Due to limited resource and changing environments, wireless sensor networks are susceptible to device failures. In this paper, we evaluate network’s vulnerability under potential device failures or attacking. Specifically, we model wireless sensors and their operating procedure as an S-T network, where the information rate regarding the network performance is defined. The network robustness is evaluated via considering how network capacity varies when network changes. The evaluation process turns out to be a maximum flow interdiction problem, which is then solved by transforming into a dual formation and approximating with a linear programming. Lastly, via numerical simulation, the proposed scheme is shown to be well suitable for evaluating network’s robustness.

This paper designs and implements a management system of agricultural water and fertilizer integration based on Internet of Things (IOT). The design of the management system, which solves the problems of energy consumption, accuracy, stability and expansibility, is mainly composed of modules with monitoring management, water and fertilizer control, data management and remote control, while realizing greening, intellectualization, networking and digitization. The system takes SK-S7G2 microcontroller as the core to build a bidirectional data channel between the client and temperature sensor, humidity sensor and light intensity sensor by using 4-G wireless network and Ethernet communication modes. Then, the sensor parameters are sent to the client and the control instructions of the client are received to realize the remote real-time control of the motor, fan and other equipment. the production environment data collected by SK-S7G2 are transferred into the database of cloud server to construct an agricultural database of real parameters of crop growth environment. Finally, the feasibility and validity of the management system have been verified by the real environmental test, which proves that the design of the management system proposed in this paper is an effective solution of the intelligent agricultural water and fertilizer irrigation system with the characteristics of energy saving, high efficiency, accuracy and scalability.

Mobility management is crucial for mobile Internet services. Mobile IPv6 and many subsequent variants aim to provide network layer mobility support for mobile nodes or mobile networks, whose performance is highly dependent on user mobility model, network topology and traffic model. Several previous researches have evaluated their performance by simulations, analytical models, and experiments. However, most of them adopt classic mobility models such as RWP without considering more realistic human mobility characteristics since most of mobile devices are carried by human being. Therefore, the performance evaluation of these solutions under realistic human mobility models becomes an important issue. In this paper we investigate human mobility characteristics and evaluates their impacts on existing mobility management protocols based on a unified simulation platform which abstracts movement parameters from realistic traces and uses them to tune other mobility models. The final results show: (1) The mobility model has an important impact on performance evaluation, and the delivery costs of RD and RWP models are different from the realistic trace which may mislead the protocols design; (2) The current mobility management protocols little consider the human mobility characteristics which cannot benefit the positive of human mobility characteristics, and they should consider the impacts of human mobility and absorb human mobility characteristics in future.

To handle the worldwide problem of aging, one of the most successful and cost-effective solutions is an ambient assisted living system. These systems integrate a collection of sensors, the Internet of Things, health management, human-computer interaction, offline medical entities and nursing services. The key technological component is a human activity recognition and anomaly detection system. We designed a platform framework that defines activity at three levels: atomic, basic and complex. Our framework process uses separate modelling with classical algorithms, so that four kinds of anomalies (point, set, scene and trend) can be detected. We implemented a real-world system and used it over two years within a scenario with nearly 200 users, thus proving the validity of the system and identifying certain deficiencies in the user’s experience. Our system has the characteristics of practicability, compatibility, cost-effectiveness and robustness.

The message transmission reliability is an important performance of CAN communication. The message transmission reliability of the CAN refers to the ability of the message to be successfully transmitted within its deadline. In order to assess the message transmission reliability of CAN, a four-node CAN communication model is set up based on the deterministic and stochastic Petri net in this paper, which is used to simulate the arbitration mechanism and error handling mechanism of CAN. The model is used to demonstrate the operation of the CAN bus in an interference environment with transient bursts which is used to simulate the external electromagnetic interference. Message transmission failure probability under the interference is used as the indicator of the message transmission reliability of CAN, and a solution method for proposed model based on the queuing theory is given to acquire the stationary value of the message transmission failure probability. The simulation gives the analysis of reliability of message transmission under different interference arrival intervals, different message priorities, different message periods and different number of nodes, which verifies the validity and feasibility of the proposed model.

This paper presents a Sliding-Window Belief Propagation with Unequal Window Size (SWBP-UWS) algorithm to deal with the nonstationary heterogeneous source. In this algorithm, the entire source is divided into several sections according to its variation and each optimum window size is individually determined by each section. The experimental results show this algorithm outperforms the SWBP algorithm.

Traffic lights detection is an important task for intelligent vehicles. It is non-trivial due to variance backgrounds and illumination conditions. Therefore, a traffic lights detection system that can apply to different scenes is necessary. In this paper, we research the traffic lights detection based on deep learning, which can extract features with representation and robustness from input image automatically and avoid using artificial features. The approach of traffic lights detection proposed in this paper includes two stages: (1) region proposal and (2) classification of traffic lights. Firstly, we propose a region proposal method based on intensity, color, and geometric information of traffic lights. Secondly, convolutional neural network (CNN) was introduced for the traffic lights classification, obtaining 99.6% average accuracy. For detection, we evaluate our system on 6804 images of different scenes, the recall and accuracy of detection achieve 99.2% and 98.5% respectively.

Compared with traditional methods, deep neural networks can extract deep information of targets from different aspects in range resolution profile (HRRP) radar automatic target recognition (RATR). This paper proposes a new convolutional neural network (CNN) for target recognition based on the full consideration of the characteristics (time-shift sensitivity, target-aspect sensitivity and large redundancy) of radar HRRP data. Using a convolutional layer with the large convolution kernel, large stride, and large grid size max-pooling, the author built a streamlined network, which can get better classification accuracy than other methods. At the same time, in order to make the network more robust, the author uses the center loss function to correct the softmax loss function. The experimental results show that we have obtained a smaller feature within the class and the classification accuracy is also improved.

CAN bus is widely used in automotive distributed embedded systems. Its protocol makes it reliable and efficient in in-vehicle communication system and industrial control system. However, the limited communication bandwidth limits its transmission efficiency. In order to improve the transmission efficiency of CAN bus, the influence of CAN bus structure to the transmission performance is analyzed in this paper. A typical CAN bus network is set and the message transmission is simulated under different CAN bus structure based on CANoe, which is widely used to study the CAN bus. The influence of different network structure on the transmission performance is analyzed. The simulation result demonstrated that the multi-level-bus CAN network using gateway reduces busload and minimize message delay effectively. It is also demonstrated that the simulation method is able to be used to find the appropriate CAN bus network structure according the busload or message delay requirement.

Artificial Intelligence (AI) is one of the hottest research directions nowadays. Machine learning is an important branch of AI. It allows the machine to make its own decisions without human telling the computer exactly what to do. At the same time, Media Access Control (MAC) is also an important technology for the next generation Wireless Local Area Network (WLAN). However, due to transmission collision, noise, interference, channel fading and other reasons, the transmission between access point (AP) and station (STA) may fail. This is limiting the overall performance. If the node can obtain the real-time failure reasons, it can adjust protocol parameters accordingly such as Modulation and Coding Scheme (MCS) and Contention Window (CW). Then, the overall performance of WLAN is improved. Therefore, a machine learning based failure reason identification approach is proposed for the next generation WLAN. In this paper, access environment is divided into four categories: nice, severe collision, deep fading and both deep fading. Different training models are used to train the data. Through our experiments, the accuracy can reach 83%, while that of Random Forest model can reach 99%.

Traffic vehicle detection and recognition is a core technology of advanced driver assistant system (ADSD) for the intelligent vehicle. In this paper, we employ the convolution neural network (CNN) to perform the end-to-end vehicle detection and recognition.Two vehicle classification CNNs are proposed. One is a convolution neural network consisting of four convolution layers and another is a multi-label classification network. The first networks can achieve the accuracy more than 95% while the second can achieve the accuracy more than 98%. Due to the multiple constraints, the proposed multi-label classification network is able to converge fast and achieve higher accuracy.The vehicle detection model proposed in this paper is a model on the basis of the network model single shot multibox detector (SSD). Our network model employs the network proposed for vehicle classification as a basis network for feature extraction and design a multi-label loss for detection. The proposed network structure can achieve 77.31% mAP on the vehicle detection dataset. Compared with that of SSD network model, the obtained mAP is improved by 2.17%. The processing speed of proposed vehicle detection network can reach 12FPS, which can meet the real-time requirements.

With the emergence of the Industrial Internet of Things (IIoT), a huge amount of devices in the factory need be connected to the Internet. The use cases being considered include data exchange for inter-factory manufacturing, integrity of data collection and real-time monitoring, etc. It is important to allow seamless communications among IIoT devices of different protocols and standards. In this research, we investigate how to achieve IIoT interoperability via two popular global IIoT standards: oneM2M and OPC-UA by adopting the OPC-UA system in the field domain but connecting it to the oneM2M system in the infrastructure domain. We propose an optimized design of oneM2M Interworking Proxy Entity for resources mapping and procedures mapping between OPC-UA and oneM2M. The design contains both (1) interworking functions and (2) OPC-UA client APP, installed on the oneM2M Middle Node in the field domain, to handle the data exchange.

In the case that the GNSS satellite signal is interfered and the satellite constellation visibility is affected, an independent pseudolite navigation and positioning system can be constructed to achieve the positioning operation of the target user. This paper proposes an independent navigation and positioning system consisting of four pseudolites based on the principle of Time Difference of Arrival (TDOA). In this paper, the geometric dilution of precision (GDOP) expression of the positioning system under the TDOA technology is derived. In view of the influence of pseudolites geometric layout on GDOP, this paper proposes two layout schemes, Y-type and T-type, and simulates the distribution of GDOP values under each layout scheme. After comparing and analyzing the simulation graphs, it is concluded that the T-shaped geometric layout can significantly reduce the GDOP value compared with the Y-shaped layout.

In this paper, we study the physical layer (PHY) interference abstraction. First, we analyse the physical layer abstraction methods for the next generation wireless local area network (WLAN), including high frequency WLAN such as institute of electrical and electronics engineers (IEEE) 802.11ay and low frequency WLAN such as IEEE 802.11ax, and describe their implementation details. Then the received bit mutual information rate (RBIR) and the mean mutual information (MMIB) of PHY abstraction methods are studied separately. We design process of two PHY abstraction methods and implement them in the simulation platform. Some corresponding simulation results are gived. The simulation results show that the RBIR method accurately predicts link level simulation performance in a simple mapping method.

At present, Wi-Fi plays a very important role in the fields of online media, daily life, industry, military and etc.Exactly predicting the traffic arrival time is quite useful for WiFi since the access point (AP) could efficiently schedule uplink transmission. Thus, this paper proposes a machine learning-based traffic arrival prediction method by using random forest regression algorithm. The results show that the prediction accuracy of this model is about 95$$\%$$%, significantly outperforming the linear prediction flow. Through prediction, resources can be reserved in advance for the arrival of data traffic, and the channel can be optimally configured, thereby achieving better fluency of the device and smoothness of the network.

As increasingly more IoT/M2M devices are connected to Internet, they will cause serious congestion to IoT/M2M systems normally deployed in the cloud. Although Cloud can scale out to support more data requests, it may not be able to satisfy the low latency demanded by certain IoT/M2M applications. Fog, as an edge of Cloud, can alleviate the congested problem in the cloud and provide low latency for critical IoT/M2M applications due to its proximity to IoT/M2M devices. In this research, we propose (1) utilizing oneM2M, a global IoT/M2M standard, as the middleware to connect the cloud and the fog, (2) using Traffic Classifiers to intercept and divert IoT/M2M traffic requiring low latency to Fog and (3) deploying independent scalability mechanisms for Cloud and Fog. We demonstrate and verify our scalability design using a smart hospital use case and show that our proposed system can achieve better scalability results in terms of latency, CPU usage and power consumption compared to those with only Fog or Cloud.

The popularity of smart home has added a lot of convenience to people’s lives. However, while users use these smart products, users’ privacy data has also been leaked and it may cause some risks. Besides, because of untrusted third-party servers, we simply use traditional privacy-preserving methods could no longer protect users’ private information effectively. In order to solve these problems, this paper proposes a privacy-preserving method for multi-private data: We first determine the privacy data format that needs to be protected, such as audio or text. Secondly, if the data format is text, we will use the local differential privacy method. We first obtain the key attributes of the user from the key information chain, and then select the appropriate localized differential privacy method according to the text characteristics of the key attributes. The user realizes the local disturbance of the data and then uploads it to the data collection center– the cloud platform. Finally, when an attacker attempts to obtain user information from the cloud platform, it uses the central differential privacy method to add noise and the noise-added data is transmitted to the attacker. If the data format is voice frequency, we first convert the voice information into binary code, then chaotically encrypt the binary code, and upload the encrypted binary code to the cloud platform. We verify the effectiveness of our methods by experiments, and it can protect users’ privacy information better.

The orbital-angular-momentum (OAM) mode multiplexing is one of the promising ways to improve the efficiency of the spectrum utilization of the network in the Internet of Things (IoT). In this article, the propagation properties and the communication performance of the OAM modes in radio frequency are studied. The transverse patterns for the single mode, two/three superimposed modes are discussed, while the influences of the mode number and the propagation distance on the beam width are analyzed. Based on 2FSK and 2PSK modulations the bit error rate (BER) of OAM modes are found varying with both the mode and the receiving radius of the array antennas. By analyzing the BER, it is found that when both the transmitting antenna and the transmitting power are fixed, also the noise power is the same, an OAM mode can have different optimal receiving radii in the single mode transmission and in the mode multiplexing transmission. These results will be helpful in optimizing the OAM mode receiving system and may have applications in the network of the IoT.

Vehicle re-identification (Re-ID) technology plays an important role in intelligent video surveillance systems. Due to various factors, e.g., resolution variation, viewpoint variation, illumination changes, occlusion, etc., vehicle Re-ID is a very challenging computer vision task. In order to solve this problem, a joint pyramid feature representation network (JPFRN) is proposed in this paper. Based on the consideration that various convolution blocks with different depths hold various resolution and semantic information of the vehicle image, which can help to effectively identify the vehicle, the proposed JPFRN method obtains four vehicle feature blocks with different depths by designing pyramidal feature fusion of each convolution block in a basic network. After that, a joint representation of these pyramidal features is feed into the loss function for learning discriminative features for vehicle Re-ID. We validated the proposed approach on a commonly used vehicle database i.e., VehicleID. Extensive experimental results show that the proposed method is superior to multiple state-of-the-art vehicle Re-ID methods.

Sensor networks are groups of specialized transducers that have a communications infrastructure which is intended to record and monitor conditions at different locations. Sensor network are widely adopted in distributed interference because they are lightweight, small and extremely portable. In order to ensure the safety of civil aviation, it is necessary to control and disperse the unmanned aerial vehicle (UAV). This paper proposed the precise distributed interference for the UAV management and control by aggregating interference signal energy at the specific point and in the specific area. First, the distance between the distributed jammers and the target point is given, and the time of different wavelets reach the target point is calculated. Then the initial phases of all the wavelets are adjusted so as to ensure all the wavelets reaching the peak value at the specific point. Finally, the interference generated in specific area is proposed to meet practical application requirements. The experiment results demonstrate that the proposed distributed interference method can be used for precise interference under UAV management and control and even in the dynamic warfare environments.

With the development of vehicle sensors, artificial intelligence and vehicle network technique, Advanced Driving Assistant System (ADAS) technology is now experiencing a rapid development. However the interaction of the sensors’ data in the real scene is seldom discussed. This paper firstly describes the application scenarios and typical working mechanism of ADAS. Then it analyses the advantages and deficiencies of the environment perception only based on vehicle-self sensors. Secondly, it describes the main research of vehicle networking. Lastly through two typical scenes it analyses the possibility and problem of communicating the sensors’ data via vehicle networking. Through the application scenarios analysis, it proposes a new potential research route for the ADAS.

As one of the hot techniques, the Internet of Things (IoT) is gradually penetrating all aspects of human life. The limitation of the spectrum resources has limited the development of the IoT, which forces us to look for new ways to increase the efficiency of the spectrum utilization. The Electromagnetic (EM) wave with orbital angular momentum (OAM), also called the EM vortex wave is a promising method to solve this problem. In this article, the basic theory of EM wave with OAM in radio frequency (RF) is introduced and the main techniques in the OAM radio beam, including the generation of the EM with OAM, the receive, the multiplexing based on OAM mode are summarized. Based on the main properties of EM wave with OAM in RF, the potential applications of EM vortex beam in the IoT are discussed.

As one of the promising techniques to improve the spectral efficiency of the network in the Internet of Things (IoT), the orbital-angular-momentum (OAM) multiplexing optical wireless communication has been studied a lot. In the optical beams with multi-OAM modes, the vortices deviating from the beam center sometimes cannot be avoidable, and they will strongly influence the transverse patterns during the beam propagation. In this article, the expressions of optical beams with off-axis vortices are derived in a commonly used focusing system, and the effect of the vortices deviating from the beam center on the propagation properties of the optical beam is discussed. We find that the number of the bright spots in the transverse patterns of the superposition of two OAM modes with off-axis vortices is not always equal to the absolute value of the mode difference which is observed in the field with only on-axis vortices. The bright spot can also be found to rotate during the beam propagating, and as the number of the modes increases in the overlapping, the superposition patterns become more complicated and these patterns can be adjusted by the off-axis distance and the topological charge of the vortices. Our result will be helpful in developing the network of the IoT.

In this paper, an Unequally Weighted Sliding-Window Belief Propagation (UW-SWBP) algorithm was proposed to decode the binary LDPC code. We model the important of overall beliefs of variable nodes in a sliding window as Gaussian distribution, which means central nodes play a more importance role than the nodes on both sides. The UW-SWBP demonstrates better performance than SWBP algorithm in both BER and FER metrics.

The sensor networking based on multi-source information fusion can significantly improve direction accuracy, and the sensor networking is always used in direction of spread spectrum signals. Aimed at eliminating the phase ambiguity of the phase-comparison method, this paper proposed a direction-finding method based on ISM (Incoherent Signal Subspace Method) and CSM (Coherent Signal Subspace Method) algorithms. Firstly, the wideband spread spectrum signal is divided into narrowband at different time points. Then perform the DCT (Discrete Cosine Transform) and obtain the covariance matrix at different frequency points. Finally, the narrowband signal power spectrums at independent frequency points are synthesized to obtain the total power spectrum of spread spectrum signal. The simulation results demonstrate that the ISM and CSM algorithms can accurately determine the direction of the spread spectrum signal, and the direction error is kept within 1° when the signal-to-noise ratio (SNR) is higher than 5 dB, which satisfy the accurate direction-finding requirement. Therefore, the ISM and CSM algorithms based on sensor networking is a necessary solution in high-precision direction of spread spectrum signals.

In the high-density deployment scenario of the next generation wireless local area network (WLAN), the intensification of conflict makes spectrum utilization low. In order to improve the spectrum efficiency, the academia and industry will introduce Co-frequency Co-time Full Duplex (CCFD) technology into MAC as a key technology. However, the existing full-duplex Medium Access Control (MAC) protocol based on access point (AP) scheduling has the problem of low success rate in establishing full-duplex links. In order to solve this problem, a dynamic full-duplex link matching algorithm based on Binary-Graph is proposed, which is based on the author’s earlier research on FD-OMAX [16]. This algorithm uses bipartite graph to establish the relationship model between full-duplex link and Resource Unit (RU). In each round of full-duplex transmission, AP establishes the optimal full-duplex link transmission based on the user’s dynamic interference information on RU resources. In order to improve the success probability of establishing full-duplex links and spectrum efficiency, an enhanced trigger-free full-duplex MAC protocol, EnFD-OMAX, is designed on the basis of FD-OMAX protocol. The simulation results show that compared with FD-OMAX protocol, MuFuPlex protocol, OMAX protocol and FuPlex protocol, the throughput distribution of EnFD-OMAX protocol increases by 26.5%, 56.60%, 88.37% and 118.4% under saturated traffic. In high-density deployment scenarios, the probability of full duplex link to successful transmission and MAC efficiency are increased by 88.98% and 149.9% respectively compared with OMAX protocol.

Aiming at the large delay characteristics of space-terrestrial integrated network, an efficient MAC protocol based on TDMA adaptive block ACK is proposed in this paper. Among them, the adaptive block ACK can effectively reduce the delay of the space-terrestrial integrated network by reducing the probability of retransmitting, at the same time, it can reduce the overhead of the MAC layer and help to achieve high throughput of the Space-Terrestrial integrated network. The simulation results with different access strategies show that the proposed efficient MAC access protocol has great advantages over the traditional ACK and block ACK access methods in system delay and network throughput, and has certain guiding significance for Space-Terrestrial integrated network.

Tunnel is the basic construction project to construct underground, underwater, and mountain buildings, such as mines, subways, and so on. Effective monitoring of tunnel environment plays on important part in the tunnel management, it can provide early warning before the accidents and minimize the hazard of the accident. Compared with the traditional IoT technologies such as ZigBee, Wi-Fi, and so on, the LoRa (Long Range) technology emerging in recent years performs better in terms of coverage range, connection number, and energy consumption, etc. It is foreseeable that the LoRa technology will certainly make a tremendous promotion in the field of tunnel environment monitoring. In this paper, A Lora-based tunnel environment monitoring system is designed and implemented. Firstly, the system architecture is described. Then, the detailed design and implementation of LoRa terminal hardware, LoRa terminal embedded software, LoRa server and monitoring App is given respectively. Finally, the experimental results show that the system has good network coverage capacity and communication reliability, can accurately monitor the tunnel environment and have low power consumption. This system has good promotion and application value.

This paper presents a 2D Q-ary Sliding-Window Belief Propagation (2DQSWBP) algorithm to decode Low-density Parity-check (LDPC) code which is utilized to compress highway traffic surveillance video under Wyner-Ziv video framework. This framework is beneficial for camera device with limited memory and computing ability. The differences of successive frames is modeled as Truncated Discrete Laplace (TDL) distribution. The experimental result shows the 2DQSWBP outperforms the Q-ary Belief Propagation (QBP) algorithm in both Bit-Error-Rate and computing time.