Search

This paper presents a 2.4 - 2.48 GHz varactor-loaded transmission-line (VLTL) phase shifter used for beamforming in indoor positioning systems (IPS). The phase shifter consists of two main blocks integrated on the same substrate: a single-stage Low-noise amplifier (LNA) block and a 56 cascading low-pass networks block. A control voltage of 4 - 24 V is applied to both of these blocks for operation. The simulated results provide a continuous phase shift range of around 300 degrees with an average gain of 8.9 ± 0.4 dB. Moreover, the noise figure (NF) is maintained lower than 2.8 dB while the reflection coefficient at the input and output of the circuit yields below -15 dB. The design is implemented on 0.18 um RF CMOS technology with a core size of 4.4 mm x 3.4 mm.

Visual sensing has been increasingly employed in industrial processes. This paper presents the design and implementation of an industrial wireless camera system, namely, EFCam, which uses low-power wireless communications and edge-fog computing to achieve cordless and energy-efficient visual sensing. The camera performs image pre-processing (i.e., compression or feature extraction) and transmits the data to a resourceful fog node for advanced processing using deep models. EFCam admits dynamic configurations of several parameters that form a configuration space. It aims to adapt the configuration to maintain desired visual sensing performance of the deep model at the fog node with minimum energy consumption of the camera in image capture, pre-processing, and data communications, under dynamic variations of application requirement and wireless channel conditions. However, the adaptation is challenging due primarily to the complex relationships among the involved factors. To address the complexity, we apply deep reinforcement learning to learn the optimal adaptation policy. Extensive evaluation based on trace-driven simulations and experiments show that EFCam complies with the accuracy and latency requirements with lower energy consumption for a real industrial product object tracking application, compared with four baseline approaches incorporating hysteresis-based adaptation.

Enabled by the increasingly available embedded hardware accelerators, the capability of executing advanced machine learning models at the edge of the Internet of Things (IoT) triggers wide interest of applying the resulting Artificial Intelligence of Things (AIoT) systems in industrial applications. The in situ inference and decision made based on the sensor data containing patterns with certain sophistication allow the industrial system to address a variety of heterogeneous, local-area non-trivial problems in the last hop of the IoT networks, avoiding the wireless bandwidth bottleneck and unreliability issues and also the cumbersome cloud. However, the literature still lacks presentations of industrial AIoT system developments that provide insights into the challenges and offer important lessons for the relevant research and engineering communities, no matter the development is successful or not. In light of this, we present the design, deployment, and evaluation of an industrial AIoT system for improving the quality control of Hewlett-Packard’s ink cartridge manufacturing lines. While our development has obtained promising results, we also discuss the lessons learned from the whole course of the effort, which could be useful to the developments of other industrial AIoT systems.

In this paper, we propose an adaptive perturbation-aided opportunistic hybrid beamforming (AP-OHBF) scheme for temporally correlated millimeter wave (mmWave) channels. The basic idea is that the proposed scheme perturbs the channel parameters to track the channels of the scheduled mobile stations (MSs), whose channels are in favorable conditions, based on only the signal-to-interference-plus-noise ratio (SINR) information of MSs. The BS utilizes these parameters to generate the precoders and adopts the successful perturbation for data transmission, which achieves performance improvements. By contrast, if a perturbation results in performance degradation, it is discarded, whereas the precoders corresponding to the best-known parameters in the memory are instead used.

Intelligent connectivity plays a vital role in improving the crop irrigation, crop yield, and crop quality as well as enhanced crop monitoring, weather prediction, and animal care. In this paper, Fuzzy logic and the Internet of Things (IoT) are integrated to improve the accuracy and power consumption of the farming system. Fuzzy rules are applied to control valve of motor for water irrigation. Firebase real time database is also employed to monitor the field remotely through the Internet. Moreover, XGboost is applied to predict the condition of farming. The results demonstrate that our proposed system is better in terms of accuracy, power consumption, error rate as compared to the conventional farming system.

Applicable potentials of a modular robotic system is dependent on the quality of intercommunication of robotic building blocks. In this paper, we present a gradient-based routing protocol for modular robotic systems. The distributed communication protocol was built up by local directional communication of neighbouring modules. In addition to the key features of neighbourhood discovery and awareness as well as link maintenance and recovery, the distributed communication protocol is leveraged by a gradient-based routing algorithm enabling robotic modules to automatically establish a directional communication route in reconfigurable structures. We demonstrated the effectiveness of our gradient-based algorithm through an education and entertainment application on our newly developed modular robotic system.

In this paper, we propose an improvement of the traditional Weighted K-Nearest Neighbors (WKNN) that is pervasively used in most WiFi fingerprint-based approaches for indoor localization. Instead of using a fixed value of K, we employ Support Vector Regression to build K-NN model that can determine the appropriate value of K in each specific position. The experimental results show that our proposed method gives higher positioning accuracy compared with other methods.

In recent years, the incredible growth and diversity of IoT systems and applications have generated an enormous amount of sensing data which play an essential role in IoT-based smart systems. So far, it has consumed much time and cost to collect these sensing data enough for these intelligent systems, which is the main reason leading to the needs of sharing available data to shorten time-consuming and cost savings for the data collection process. However, there are many challenges in ensuring integrity, security, and fairness in the data sharing process. In this paper, the authors propose a model leveraging the emerging Blockchain technology as an alternative solution to enhance the security of IoT data sharing management in terms of three principal criteria including confidentiality, integrity, and availability. Our prototype on Ethereum Blockchain demonstrates the feasibility of the proposed model.

In this paper, we propose novel mathematical models for effectively choosing a set of Unspent Transaction Outputs (UTXOs) in transaction-based blockchains in terms of two major objectives. The first one is to minimize the transaction size, as a result, minimize the transaction fee for miners paid by users. The second one is to shrink the UTXO set size that consequently reduces the searching space and computation overhead. Our proposed models are evaluated on real transactions collected from Bitcoin network and transactions generated by Highest Value First (HVF) and Lowest Value First (LVF) based approaches. The experimental results show that our proposed models gain better performance compared to other existing approaches.

In this paper, we consider the problem of making an optimal offloading decision for a mobile user in an ad-hoc mobile cloud in which the mobile user can offload his computation tasks to nearby mobile cloudlets via a device-to-device (D2D) communication-enabled cellular network. We propose a deep reinforcement learning (DRL)-based offloading scheme which enables the user to make near-optimal offloading decisions by taking into account uncertainties of user's and cloudlets' movements and the cloudlets' resource availabilities. We first propose a Markov decision process (MDP)-based offloading problem formulation which considers the composite states of the user's and cloudlets' queue states and the distance states between the user and cloudlets as the system state space. The objective of the formulated MDP-based problem is to determine the optimal actions on how many tasks the user should process locally and how many tasks to offload to each cloudlet at each observed system state such that the user's utility obtained by task execution is maximized while minimizing the energy consumption, task processing delay, task loss probability and required payment. Then, we use a deep reinforcement learning scheme, called deep Q-network (DQN) to learn an efficient solution for the proposed MDP-based offloading problem. Extensive simulations were performed to evaluate the performance of the proposed offloading scheme. The simulation results validate the effectiveness of the offloading policies obtained by the proposed scheme.

Core networks have to contend with a huge amount of traffic, making the networks heavy, congested or even fail, specifically at peak times. However, not all traffic has to be processed immediately. Background (BG) traffic can be delayed to yield network resources for foreground (FG) traffic when needed. This paper proposes an efficient approach to BG and FG traffic control leveraging software defined network (SDN) techniques. We devised an effective data structure, namely a flow tree, for flow aggregation, to reduce the inherent communication costs (between the Controller and Openflow switches) and the storage cost in the switch's expensive memory. Experimental results from both simulated and real datasets reveal the effectiveness and efficiency of the proposed scheme, confirming the viability of the proposed SDN-based BG and FG traffic control approach.

An ad-hoc mobile cloud allows a mobile user to access a potential cloud resource with a short offloading latency and a low bandwidth consumption from nearby mobile devices, namely mobile cloudlets via short-range communications. However, due to the user and cloudlet mobility, the time-varying wireless channel properties, and the cloudlet's limited computation resource, designing an effective offloading algorithm involves challenges in workload distribution, cost estimation and energy minimization. In this paper, to address these challenges, we develop an optimization-based offloading algorithm that enables the mobile user to make an optimal offloading decision. The proposed algorithm takes into account effects of the user's workload, the diverse connectivity of cloudlets, and the wireless environment on the offloading action. More specifically, we formulate and solve a Markov decision process (MDP) scheme to achieve an optimal offloading policy for the mobile user with the objective of maximizing the user's utility while minimizing the offloading cost. Extensive simulations were performed to evaluate the performance of the proposed MDP scheme. The simulation results show that the proposed scheme outperforms baseline schemes.

Due to its advantages of providing a large geographical coverage and having no strict limits on energy and sensing and processing capabilities, a vehicular sensor network (VSN) has recently emerged as a promising paradigm for air quality monitoring in an urban area. However, designing an efficient VSN-based air monitoring system has challenges due to the vehicles' heterogeneous temporal and spatial coverage and the relatively expensive communication cost over cellular networks. In this paper, we propose a machine learning (ML)-based Air quality Monitoring (MLAirM) system which aims at reducing communication and sensing costs by allowing vehicles to process the collected data in a distributed fashion. More specifically, in MLAirM, vehicles are first assigned to take measurements at sets of locations in a sensing area. The vehicle then utilizes a distributed machine learning algorithm to learn a local model of air quality based on its collected data. Finally, the vehicle sends parameters of its model to a monitoring center which combines multiple local models to build a global air quality map. Furthermore, assigning the sensing locations to vehicles can be viewed as a successful measurement probability aware location assignment problem. An integer linear optimization problem is formulated and a heuristic algorithm is proposed to find the solution. Simulations based on realistic vehicular traces are performed to compare the proposed MLAirM system with other approaches. The simulations results show that the MLAirM can achieve a similar accuracy of building the global air quality map with a significant reduction in communication and sensing costs compared to other approaches.

This paper proposes an approach to optimize service placement on Fog landscape in the context of the Internet of Things (IoT). A multi-tier fog computing architecture that supports IoT service provision is devised. Based on this architecture, a novel service placement mechanism that optimizes service decentralization on Fog landscape leveraging context-aware information such as location, time, quality of services (QoS) has been proposed. An experiment has been conducted to evaluate the proposed approach with several simulations applying to smart grid applications. The results reveal the effectiveness of the proposed approach in terms of reducing latency, energy consumption, and network load in comparison with the conventional cloud computing model.

The “Digital race — Driverless car” was a student competition where each team was given a model car with fixed specification in both mechanical and electronics. The mission was to program the car to run automatically in the condition of road having obstacles, missing lane lines and bridges with completely out lane lines in the field of view. This paper aims to address the issues facing in the competition as well as a sustainable solution to make car run autonomously and rapidly in the condition of real time processing and limited computing resource. The vision-based solution consists of modules to infer the center of lane from images in normal flat road as well as road having obstacles and bridge. It also features a fuzzy controller and a PID controller to control the car accordingly. The proposed solution has been proven in practice by reaching the first position of the competition.

Recently, the vehicular sensor network (VSN) is emerging as an efficient solution for executing different sensing tasks in urban environments. However, due to the heterogeneity of vehicles in sensing capability and uncontrollable movement trajectory, it is a challenge to best provide the required quality of information (QoI) of the sensing task in VSNs. In this paper, we introduce a VSN architecture, in which multiple vehicles cooperatively sense a particular urban area of interest, and process the sensed data to achieve the QoI requirements while considering incentives for environment sensing, data processing and communication. Furthermore, we formulate and solve an optimization problem for determining the optimal sampling rates for vehicles with the objective of minimizing the total incentive under the constraints related to QoI requirements. Various numerical results based on realistic vehicular traces are presented to justify the effectiveness of proposed approach in the vehicles' QoI-aware cooperative sensing operations.

Aspect extraction is one of most challenging tasks in opinion mining. Many researches have attempted to solve this problem for English text. For less popular languages such as Vietnamese, their complex structure causes difficulties in management or semantic analysis tasks. In this paper, we propose an approach to extracting and classifying aspect-terms for Vietnamese language. The semi-supervised learning GK-LDA is proved to have better performance than the traditional topic modeling LDA. In the aspect inference, we use dictionary-based method which can extract noun-phrases for obtaining better performance than just extract word seeds or use a complete sentence to infer aspects. Our experimental results show that our proposed method can effectively perform the aspect extraction and classification task. Even though our approach is initially proposed for handling Vietnamese text, we believe that it is also applicable to other languages.

The task of monitoring a moving phenomenon in an unbounded area using a mobile sensor network (MSN) brings out several challenges due to the high movement speed of phenomenon, the limited sensing/communication capabilities of mobile sensor nodes. To address the challenges and achieve a high weighted sensing coverage, in this paper we propose a monitoring algorithm, namely VirFID-MP (Virtual Force (VF)-based Interest-Driven moving phenomenon monitoring with Mobility Prediction). In VirFID-MP, the movement of phenomenon is first predicted based on its previous movements. Then, the predicted information is used to determine a global virtual force, which is utilized to speed up the MSN toward the moving phenomenon. Simulation results show that VirFID-MP outperforms original VirFID in terms of weighted coverage efficiency, when the MSN monitors a moving phenomenon.

NiSi pipes are formed on SiGe during conventional (He based) RTA anneal processes in NiSi module due to Ni diffusion. NiSi pipes increase leakage and reduce device Vt's. The control of Nickel diffusion is essential for device control. In this paper we use a laser anneal tool at NiSi RTA2 to control Ni diffusion. The formation of NiSi-pipes is studied as a response of process factors, such as laser and chuck temperature, wafer rotation, and dwell time. The high process temperature in the range of 700–950°C can melt NiSi. The onset of NiSi melting as a function of process parameters and layout locations are studied.

This paper addresses the design of instrumented single joint and multi-joints active orthoses for musculoskeletal and neuromuscular disorder exploration and functional rehabilitation. The use of these multi-contact exoskeletons, operating a selective limb joint mobilization, allows the measurement and the control of physiological parameters (instantaneous helical axis (IHA) motion, joint motion and torque) of each individual anatomical joint. Passive joints have to be added into the kinematic chain of the exoskeleton in order to eliminate residual stresses in limb/orthose interaction due to misalignment between the device principal axis of rotation and the IHA of the anatomical joint. The kinematic design and the control of spatial orthotic mechanisms to obtain an adaptation and a best-fit to the joints kinematics are discussed afterward. Examples of active orthosis design for knee, elbow joint and lower limbs are also given.