Mobile users can open program in new tab for better viewing.

Open program in new tab

Day 1 07/07/2020
Room #1

Organizing Committee Welcome 08:30 - 08:45

Start at 8:30am

Conference Manager's Message 08:45 - 09:00

Welcome message from Conference Manager and EAI

Keynote 1: Prof. Andreani Odysseos 09:00 - 10:00

Coffee Break #1 10:00 - 10:10

10 minutes

Main track #1 10:10 - 11:30

10:10 - 10:30
Clock Synchronization for Mobile Molecular Communication in Nanonetworks

Molecular communication is an emerging communication method using molecules or particles as a signal carrier, which enables nanodevices to exchange information at the nano- or micro-nano scale for information exchange and collaboration. Clock synchronization between nanomachines plays an important role in collaboration. The current research on the synchronization between nanodevices mainly focused on fixed molecular communication systems. However, the movement of nanodevices is widespread in molecular communication systems. This paper investigates the synchronization problem in mobile molecular communication systems based on diffusion. A simple but effective scheme for clock synchronization between mobile nanodevices is proposed. Based on the waveform of the molecular signal, the clock offset between mobile nanodevices is estimated by least square method. By using different types of molecules, the challenge for the practical varying molecule synthesis time duration is overcome. The proposed algorithm shows good performance by simulations.
Authors: Li Huang, Lin Lin (Tongji University, China), Fuqiang Liu (Tongji University), Hao Yan (Shanghai Jiao Tong University),
Hide Authors & Abstract

Show Authors & Abstract
10:30 - 10:50
A Cooperative Molecular Communication for targeted drug delivery

The lack of actively targeted nanoparticles and a low drug concentration in lesions are two of the main problems in drug delivery. This paper proposes a cooperative molecular communication system for drug delivery, electromagnetic control in the lead with bacteria followers. The leading particle is consisted by two function: could be controlled by electromagnetic field, and release attractant molecules. This cooperative scheme provides actively targeted ability by electromagnetic control, furthermore it expands the impact range of chemotactic substances to improve the chemotactic efficiency. To approach the specific position, this paper proposes electromagnetic field to control the nanoparticles, while bacteria could search the larger concentration positions and get closer to the leading particles. This paper develops mathematical modelling for the proposed model, as well as the self-adapted concentration gradient field searching algorithm. Finally, this paper performs biologically realistic simulation experiments to evaluate the performance of the proposed model.
Authors: Yutao Hsiang (Chengdu university of technology), Yue Sun (Chengdu University of Technology), Yifan Chen (University of Electronic Science and Technology of China), Yu Zhou (Beijing Institute of Collaborative Innovation),
Hide Authors & Abstract

Show Authors & Abstract
10:50 - 11:10
Performance of Diffusion-based MIMO Molecular Communications and Dual Threshold Algorithm

As the nanotechnology becomes more and more mature, the concept of molecular communication emerged and attracted lots of researchers鈥?attention. The most widespread model for a molecular communication channel is the diffusion-based channel, where the information-carrying molecules propagate randomly in the medium based on Brownian motion. As for Multi-Input Multi-Output (MIMO) transmissions, there are not only Inter-Symbol Interference (ISI), some molecules may arrive at the receiver after their intended time-slot, as interference. Another source of interference is the inter-link interference (ILI), which emerges when receiver receive other transmitters鈥?molecules. In this paper, we study the bit error rate (BER) performance of a molecular communications system having two transmitters and two receiver with two receptors by considering ISI and ILI. Last, dual threshold algorithm is proposed to optimize the system BER.
Authors: Zhiqiang Lu (University of Electronic Science and Technology of China), Qiang Liu (University of Electronic Science and Technology of China), Kun Yang (University of Essex), Yuming Mao (Key Lab of Optical Fiber Sensing and Communications, School of Communication & Information Engineering, University of Electronic Science and Technology of China, Chengdu, 611731 China),
Hide Authors & Abstract

Show Authors & Abstract
11:10 - 11:30
Binary Concentration Shift Keying with Multiple Measurements of Molecule Concentration in Mobile Molecular Communication

Binary concentration shift keying in molecular communication is a modulation technique that transforms binary information onto the concentration of molecules. Transmitter releases a pre-specified number of molecules into the environment according to the information it wishes to transmit to receiver. In this paper, we consider binary concentration shift keying in mobile molecular communication where a mobile receiver performs multiple measurements of molecule concentration to demodulate information that the transmitter transmits. Numerical experiments are conducted to evaluate the performance of mobile molecular communication where the binary concentration shift keying with multiple measurements of molecule concentration is employed.
Authors: Yutaka Okaie (Osaka University Institute for Datability Science), Tadashi Nakano (Osaka University Institute for Datability Science),
Hide Authors & Abstract

Show Authors & Abstract

Lunch 11:30 - 12:00

30 minutes

Special Session: Data Driven Intelligent Modeling, Application and Optimization 12:00 - 14:00

12:00 - 12:20
Causal Network Analysis and Fault Root Point Detection Based on Symbolic Transfer Entropy

Transfer entropy (TE) is a model-free method based on data-driven information theory. It can obtain causal relationships between variables. It has been used for modeling, monitoring and fault diagnosis of complex industrial processes. It can detect the causal relationship between variables without the need to assume any underlying model, but its calculation process is complicated and the calculation time is long. In order to overcome this limitation, symbol transfer entropy is pro-posed. The symbol transfer entropy is robust and fast to calculate. It can also quantify the dominant direction of information flow between time series with identical and non-identical coupling systems, thereby improving the accuracy of causal paths. Sex. Through the symbolic transfer of entropy, a causal network di-agram can be obtained, and the root cause of the fault can be found. The effec-tiveness and accuracy of the method are verified by simulation and actual indus-trial cases (Tennessee-Eastman process).
Authors: Jian-Guo Wang (School of Mechatronical Engineering and Automation, Shanghai University), Xiang-Yun Ye (School of Mechatronical Engineering and Automation, Shanghai University, Shanghai Key Lab of Power Station Automation Technology), Yuan Yao (Department of Chemical Engineering, National Tsing-Hua University),
Hide Authors & Abstract

Show Authors & Abstract
12:20 - 12:40
Personalized EEG feature extraction method based on filter bank and elastic network

In the practical application of the Brain Computer Interface (BCI) system, be-cause of the diversity between the individuals in the electroencephalogram (EEG) system, the manifestation of Brain signals of each individual is different, so it is necessary to conduct personalized screening for different individuals to obtain in-formation that is conducive to the classification of the EEG signals of the move-ment imagination. Because the EEG signal manifestation and corresponding rhythm range of different individuals are different, and the EEG characteristics corresponding to different frequency bands are also different, this paper proposes a personalized feature extraction method based on filter bank and elastic network. Based on several commonly used feature extraction and classification algorithms in the current BCI system, the analysis and research are carried out. The best combination method to obtain higher calculation rate and recognition accuracy pro-vides some theoretical reference for the practical application of BCI system. Thus, the shortcomings of the CSP algorithm with better feature extraction effect are improved, and the proposed method can eliminate the individual differences of EEG signals, realize automatic feature selection, and improve classification ac-curacy.
Authors: Jian-Guo Wang (School of Mechatronical Engineering and Automation, Shanghai University), Zeng Chen (School of Mechatronical Engineering and Automation, Shanghai University, Shanghai Key Lab of Power Station Automation Technology), Yuan Yao (Department of Chemical Engineering, National Tsing-Hua University),
Hide Authors & Abstract

Show Authors & Abstract
12:40 - 13:00
Release rate optimization based on M/M/c/c queue in local nanomachine-based targeted drug delivery

As the basis for the modern medical therapeutics, targeted drug delivery is one of the most important topics in nanomedicine. In nanomachine-based targeted drug delivery, it should be taken into consideration that nanomachines have limited resources and drug molecules are expensive and lost molecules may cause undesired side effect. This paper aims to optimize drug release rate of nanomachine and is expected to pave a way for designing a drug delivery system. To this end, we proposed a method to calculate the optimized drug release rate producing a full drug response in local targeted drug delivery. In the method: first, a drug reception model based on M/M/c/c queue to simulate the interactions between ligands and receptors is established; second, the least effective concentration of drug molecules is derived from the least ratio of receptors occupied by drug molecules to produce full drug response according to the drug response theory named occupancy theory; finally, the optimized release rate is derived from the least concentration of drug molecules according to molecular diffusion law. Simulations reflecting diffusion of drug molecules and occupancy of receptors are established. The obtained simulation results match well with the results derived from the proposed analytical method.
Authors: Qingying Zhao (Changshu Institute of Technology), Min Li (Shanghai University),
Hide Authors & Abstract

Show Authors & Abstract
13:00 - 13:20
Research on Course Control of Unmanned Surface Vehicle

The unmanned surface vehicle (USV) system has the characteristics of large inertia, long time delay and under drive, the traditional linear control method is not robust enough to achieve course control. A course controller of USV has been proposed in this paper based upon the fuzzy control theory. The mathematical model of USV is established and the model parame-ter has been identified by least square method in different speed. A hardware platform of USV also are obtained. The course controller of USV has been verified and validated through com-puter simulation and ocean experiment and has been proven to work effectively.
Authors: Xinming Hu (Shanghai University), Huaichun Fu (Shanghai University), Qixing Cheng (Shanghai University),
Hide Authors & Abstract

Show Authors & Abstract
13:20 - 13:40
Design and Experiment of a Double-layer Vertical Axis Wind Turbine

This paper introduces the design and experiment of a double-layer vertical axis wind turbine. This system is mainly oriented to the polar environment as a supplement to mobile robot energy. After consulting a large amount of lit-erature and combining theoretical calculations, the system uses NACA4412 blades to form a double-layer wind turbine, and uses a permanent magnet di-rect-drive synchronous generator to convert mechanical energy into electrical energy, and charges the battery through a circuit. In this paper, ANSYS soft-ware is used to model and verify the structure of the main components of the wind turbine. Finally, the prototype is developed and tested. The maximum conversion efficiency of wind energy can reach 24.66%
Authors: Qixing Cheng (Shanghai University), Xinming Hu (Shanghai University),
Hide Authors & Abstract

Show Authors & Abstract
13:40 - 14:00
Real-Time Obstacle Detection Based on Monocular Vision for Unmanned Surface Vehicles

The reliable obstacle detection is a challenging task in autonomous navigation of unmanned surface vehicles (USVs). In this paper, we present a novel real-time obstacles detection based on monocular vision which can effectively tell apart ob-stacles on the sea surface from complex background. The main innovation is to propose an edge detection algorithm based on semantic segmentation and random sample con-sensus(RANSAC) line fitting algorithm, and a simple and effective significance detection method based on background prior and contrast prior to de-tect obstacles under the edge.
Authors: Rui Zhang (Shanghai University), Jingyi Liu (Shanghai University), Hengyu Li (Shanghai University), Qixing Cheng (Shanghai University),
Hide Authors & Abstract

Show Authors & Abstract

Coffee break #2 14:00 - 14:10

10 minutes

Special Session: Intelligent Internet of Things and Network Applications 14:10 - 15:30

14:10 - 14:30
A Method of Data Integrity Check and Repair in Big Data Storage Platform

In the big data storage platform, in order to ensure the security of user data, it is necessary to perform cyclic verification on the stored data and repair the damaged data in time. Considering the problems of low verification efficiency, low check frequency and low calibration accuracy of HDFS data integrity check, this paper proposed a new HDFS storage platform security check and repair scheme. The process can effectively reduce the amount of calculation and communication overhead, and can support the dynamic operation of the data. Experiments show that this method has certain advantages in security, scalability and flexibility.
Authors: Jiaxin Li (School of Electronics and Information Engineering, Beijing Jiaotong University), Yun Liu (School of Electronics and Information Engineering, Beijing Jiaotong University), Zhenjiang Zhang (School of Electronics and Information Engineering, Beijing Jiaotong University), Han-Chieh Chao (Department of Electrical Engineering, National Dong Hwa University),
Hide Authors & Abstract

Show Authors & Abstract
14:30 - 14:50
A Study of Image Recognition for Standard Convolution and Depthwise Separable Convolution

Artificial intelligence and deep learning techniques are all around our life. Image recognition and natural language processing are the two major topics. Through using TensorFlow-GPU as backend in convolutional neural network (CNN) and deep learning network, image recognition has been an extreme breakthrough in recent years. However, more and more model parameters result in overfitting problem and computation overhead. In the paper, the performance of image recognition between standard CNN and depthwise separable CNN is experimented and investigated. In addition, data augmentation technique is applied to both standard and depthwise separable CNNs to improve the image recognition accuracy. The experiments are implemented by an open source API called Keras with using CIFAR-10 dataset. Experimental results showed that the depthwise separable CNN improves validation accuracy compared with the standard CNN. Moreover, schemes with data augmentation achieve higher validation accuracy but training accuracy.
Authors: Fan-Hsun Tseng (National Taiwan Normal University), Fan-Yi Kao (National Taiwan Normal University),
Hide Authors & Abstract

Show Authors & Abstract
14:50 - 15:10
A Novel Genetic Algorithm-based DES Key Generation Scheme

The data encryption is closely related to our daily lives, especially after the concept of e-commerce has become popular. The main reason is that although various network service platforms facilitate our lives, they also bring many potential problems such that there are many ways that people with bad intentions can tamper or steal data via Internet and further decrypt it. In order to maintain user privacy, encryption methods are adopted to prevent stealers from easily reading data. Data encryption standard (DES) is currently the most widely used encryption method in commercial financial units. However, since DES has a short password length and the generated key is quite linear, weak keys may be generated so that the security is still insufficient. In this paper, a genetic algorithm (GA) method is used to generate high-variability keys to solve the problem of a high probability that DES generates weak keys.
Authors: Min-Yan Tsai (National Ilan University), Hsin-Hung Cho Cho (National Ilan University), Chi-Yuan Chen (National Ilan University), Wei-Min Chen (National Dong Hwa University),
Hide Authors & Abstract

Show Authors & Abstract
15:10 - 15:30
Developing an Intelligent Agricultural System based on Long Short-Term Memory

There were many undeveloped countries upgraded to emerging countries in re-cent years; as a result, the farmland has been transferred to commercial or indus-trial lands that significantly reduce the areas of farmland, lowers down the agri-cultural labor force due to the population aging and further decreases agricultural output. Additionally, many of the farmland are outdoor farms, which are limited by water resources and electricity. The study develops an intelligent agricultural system based on Long Short-Term Memory (LSTM), through utilizing solar power to monitor crop environments. The key features presented in this study are 1. reducing the electrical wiring cost by using solar power; 2. adding weather forecast information to initiate the equipment and avoid the waste of electricity; 3. using the environmental monitor to check whether the crop is at a suitable envi-ronment and the system will alarm if the environment is not suitable. Through LSTM to monitor environments and lower the initiating power for avoiding elec-tricity waste. From the experiments of the research, the method is proved to be feasible and is usable without the need for additional power-supply equipment.
Authors: Hsin-Te Wu (Department of Computer Science and Information Engineering, National Ilan University, Taiwan), Jun-Wei Zhan (Department of Computer Science and Information Engineering, National Penghu University of Science and Technology, Taiwan), Fan-Hsun Tseng (Department of Technology Application and Human Resource Development, National Taiwan Normal University, Taiwan),
Hide Authors & Abstract

Show Authors & Abstract

Special Session: Intelligent Sensor Networks 15:30 - 16:30

15:30 - 15:50
Detection of atherosclerotic lesions based on molecular communication

Atherosclerotic plaques in the human circulatory system are a major cause of diseases in the blood vessels and the heart. These plaques can grow and block blood vessels, preventing blood from being supplied to the distal end. Mild to moderate stenosis does not cause a significant reduction in blood flow, and clinical signs do not appear unless the lesion has progressed to an advanced stage, and there is no reliable way to detect the lesion at early stages. Digital subtraction angiography is a commonly used method to detect atherosclerosis in medical clinics. DAS is considered to be the "gold standard" for the diagnosis of vascular diseases. This article will analyze, model and evaluate the indicators of atherosclerosis development from the perspective of molecular communication and angiography. The main idea is to use the propagation of contrast agents as a function of the cross-sectional area of the blood vessel. Its specific implementation can be expressed by the propagation index of the contrast agent obtained after DAS processing. DAS processing is easily achieved in medicine. This article has practical significance for detecting similar vascular diseases.
Authors: Meiling Liu (Chengdu University of Technology), Yue Sun (Chengdu University of Technology), Yifan Chen (University of Electronic Science and Technology of China),
Hide Authors & Abstract

Show Authors & Abstract
15:50 - 16:10
Design for Detecting Red Blood Cell Deformation at Different Flow Velocities in Blood Vessel

Molecular communication (MC) holds considerable promise as the next generation of design for drug delivery that allows for targeted therapy with minimal toxicity. Most current studies on flow-based MC driven drug delivery application consider a Newtonian fluid and laminar flow. However, blood is a complex biological fluid composed of deformable cells especially red blood cells, proteins, platelets, and plasma. For blood flow in capillaries, arterioles and venules, the particulate nature of the blood needs to be considered in the delivery process. The ability to change shape is essential for the proper functioning of red blood cells in microvessels. The different shapes of red blood cells have a great impact on the performance characteristics of whole blood (blood and plasma). Changes in the properties and shape of RBC substances are often associated with different blood diseases and diseases, such as sickle cell anemia, diabetes, and malaria. Based on the state of the red blood cells in the microtubules at different flow rates, this paper proposes a design for detecting the ability of the cells to deform. Based on the difference in the concentration of the nanoparticles at the receiving end at different flow rates, the ability of the red blood cells to deform is determined, and the blood state is determined. Further, the related blood diseases can be initially predicted.
Authors: RuiZi Zhang, Yue Sun (Chengdu University of Technology), Yifan Chen (University of Electronic Science and Technology of China),
Hide Authors & Abstract

Show Authors & Abstract
16:10 - 16:30
Intelligent Power Controller of Wireless Body Area Networks based on Deep Reinforcement Learning

Wireless Body Area networks allow groups of tiny sensors to communicate for purpose of medical applications. With the progress of sensor manufacture and artificial intelligence, abundant wearing devices are produced and applied with powerful intelligence functionalities. In wireless body area networks, battery energy capacity and inter-network interference are two serious threats to restrict the raise of performance. In this work, we focus on the power controlling theme in wireless body area networks. First, we introduce the primer overview of the deep-Q- Network algorithm, which is the method utilized in this work. Second, we present our communication system which is composed of two interfered WBANs. Third, we show how to design the power controller based on the deep-Q-network algorithm. The results reveal that our proposed power controller significantly decreases energy consumption by sacrificing little throughput performance.
Authors: Peng He (Chongqing University of Posts and Telecommunications), Zhenli Liu (Chongqing University of Posts and Telecommunications), Lei Fu (Chongqing University of Posts and Telecommunications), Zhongyuan Tao (Chongqing University of Posts and Telecommunications), Jia Liu (Chongqing University of Posts and Telecommunications), Tong Tang (Chongqing University of Posts and Telecommunications), Zhidu Li (Chongqing University of Posts and Telecommunications),
Hide Authors & Abstract

Show Authors & Abstract
Day 2 08/07/2020
Room #1

Keynote 2: Prof. Pedro Antonio Valdes-sosa 09:00 - 10:00

Coffee Break #1 10:00 - 10:10

10 minutes

Main track #2 10:10 - 11:30

10:10 - 10:30
Real-Time Seven Segment Display Detection and Recognition Online System using CNN

Normally, the machinery shows the information via the seven-segment display. The user must manually access the information. In the actual situation, users can-not check the machine at all times. For real-time and automatic tracking of the ma-chine display, the real-time detection and recognition seven-segment display online system is designed. The camera module is used to take pictures of the ma-chine display. The images are processed and displayed online in real-time as numbers. However, the image processing process has often encountered prob-lems, such as noise, contrast, brightness, and rotation of the image. To cope with these challenges, This paper proposed an Interpretation Seven-Segment display (ISS) framework that can interpret the image to numerical data. The experiment result shows that the systems can track the display of the machine consistently. The proposed framework (ISS) has an accuracy of interpreting up to 97%.
Authors: Autanan Wannachai (Chiang Mai University), Wanarut Boonyung (Chiang Mai University), Paskorn Champrasert (Chiang Mai University),
Hide Authors & Abstract

Show Authors & Abstract
10:30 - 10:50
A novel method for extracting high-quality RR intervals from noisy single-lead ECG signals

In previous studies, plenty of high-accuracy R-peak detection methods were per-formed on electrocardiogram (ECG) signal analysis. However, these excellent results were usually obtained from some standard and common databases. When applying these detectors on ECG signals collected in daily life and ordinary experiments, or acquired from wearable single-lead ECG devices, the R peak detection accuracies were usually unsatisfying. Due to the influence of data-acquiring environment and devices, the collected ECG signals were often noisy. In this study, we proposed a method combining seven R-peak detection methods to get high-quality RR Intervals (RRIs) from noisy ECGs. This new method included two steps, 1) obtain preliminary R-peak annotations through combining seven R-peak detection methods, and 2) calculate the quality score of each R-peak annotations detected in 1) according to the ECG waveform features including kurtosis, skewness and the frequency band power ratio, then exclude the wrong annotations based on the quality scores. The proposed method was evaluated on two databases: MIT-BIH Arrhythmia database and the CPSC2019 training set. The R peak detection average accuracies on these two da-tabases were 98.89% and 55.47% respectively. The results showed that the method proposed in this paper performed better than the seven common R-peak detection methods, especially in noisy ECG signals.
Authors: Shan Xue (School of Biological Science & Medical Engineering, Southeast University), Leirong Tian (School of Biological Science & Medical Engineering, Southeast University), Zhilin Gao (School of Biological Science & Medical Engineering, Southeast University), Xingran Cui (School of Biological Science & Medical Engineering, Southeast University),
Hide Authors & Abstract

Show Authors & Abstract
10:50 - 11:10
Leak-resistant design of DNA strand displacement systems

Although a number of dynamically-controlled nanostructures and programmable behaviors have been constructed using DNA strand displacement, predictability and scalability of these DNA-based systems remain limited due to leakages introduced by spuriously triggered dis- placement events. We present a systematic design method for implement- ing leak-resistant DNA strand displacement systems in which each legit- imate displacement event requires signal species to bind cooperatively at the two designated teohold binding sites in the protected fuel com- plexes, and thus prevents spurious displacement events. To demonstrate the potential of the leak-resistant design approach for the construction of arbitrary complex digital circuits and systems with analog behaviors, we present domain-level designs and displacement pathways of the basic building blocks of the DNA strand displacement cascades, e.g. OR, AND gates, and an elementary bimolecular reaction.
Authors: Vinay Gautam (Aalto University, Finland),
Hide Authors & Abstract

Show Authors & Abstract
11:10 - 11:30
Chessboard EEG Images Classification for BCI Systems Using Deep Neural Network

Classification of electroencephalography (EEG) signals is a fundamental issue of Brain Computer Interface (BCI) systems, and deep learning techniques are still under investigation although they are dominant in other fields like computer vision. In this paper, we applied two approaches to transform the EEG signals into images in order to be classified using a deep learning model, the first is the azimuthal equidistant projection with Clough-Tocher interpolation algorithm and the second is our proposed chessboard image transformation approach. The Physionet dataset for EEG motor movement/imagery tasks was used which consists of 109 subjects and the Motor Imagery (MI) signals for two frequency bands (Mu [8-13 Hz] and Beta [13-30 Hz]) were transformed into 2-channel images (one channel for each band). The network model consists of Deep Convolutional Neural Network (DCNN) to extract the spatial and frequency features followed by Long Short Term Memory (LSTM) to extract temporal features and then finally to be classified into 5 different classes (4 motor imagery tasks and one rest). The results were promising (68.13% average accuracy for the azimuthal projection approach and 68.72% for the chessboard approach) compared to 64.64% average accuracy for the Support Vector Machine (SVM).
Authors: Ward Fadel (Pazmany Peter Catholic University), Moutz Wahdow (Pazmany Peter Catholic University), Csaba Kollod (Pazmany Peter Catholic University), Gergely Marton (Pazmany Peter Catholic University), Istvan Ulbert (Pazmany Peter Catholic University),
Hide Authors & Abstract

Show Authors & Abstract

Special Session: Internet of Everything 11:30 - 12:30

11:30 - 11:50
Target Tracking Based on DDPG in Wireless Sensor Network

For target tracking in mission critical sensors and sensor networks (MC-SSN), the contribution of the measured value of each sensor node to the data fusion center is different, so better weighted node fusion and scheduling node participation in tracking can obtain better tracking performance.In this paper, to address this problem and fully utilize the network transmission capability, we proposed a collaborative perception and intelligent scheduling to jointly optimize system responding latency and tracking accuracy while guaranteeing low energy consumption. Based on the unreliable historical tracking data, we formulate the joint optimization problem as the infinite horizon Markov Decision Process (MDP), we propose an intelligent collaboration scheme based on the deep deterministic policy gradient (DDPG) approach to perform the optimal tracking with low energy consumption and high tracking accuracy.
Authors: Yinhua Liao (University of Electronic Science and Technology of China), Qiang Liu (University of Electronic Science and Technology of China),
Hide Authors & Abstract

Show Authors & Abstract
11:50 - 12:10
A fuzzy tree system based on cuckoo search algorithm for target tracking in Wireless Sensor Network

Wireless Sensor Network (WSN) consists of sensors with small vol-ume and limited power. These sensors can communicate with each other and fuse data to make different decisions. Target tracking is an important application in wireless sensor network. How to schedule nodes for tracking the moving target and how to improve the tracking accuracy are the problems that we face. In this paper, we introduce a fuzzy tree system in target tracking. The fuzzy tree system is composed of two layers, in which the first one is to decide which nodes to move and the second one is to decide the distance and angle. All the parameters are tuned by the Cuckoo Search algorithm (CS). We performed a large number of simulations in choosing different numbers of the moving nodes. The results of my experimental data show that the adaptive fuzzy system has a good effect on target tracking, and the Cuckoo Search algorithm outperforms the algorithms widely used now in tuning the parameters.
Authors: Qing Xia (University of Electronic Science and Technology of China), Junjun Lin (University of Electronic Science and Technology of China), Qiang Liu (University of Electronic Science and Technology of China), Supeng Leng (University of Electronic Science and Technology of China),
Hide Authors & Abstract

Show Authors & Abstract
12:10 - 12:30
Sensor scheme for target tracking in Mobile Sensor Networks

Wireless sensor networks is an important component of Internet of everything, and can be deployed in many applications, such as search and rescue, border pa-trols, environmental monitoring, and combat scenarios. In these applications, tar-get tracking is a crucial difficulty. Compared with the traditional static wireless sensor networks (WSN), the mobile sensor networks (MSN) has the advantages of strong robustness, flexibility, energy saving, etc., and has been widely de-ployed. For target tracking applications in mobile wireless sensor networks, this paper investigates an extended Kalman filter(EKF) algorithm in a dynamic sce-nario, and proposes a low-power, high-accuracy sensor scheduling strategy based on the extend Kalman filter algorithm. The properly sensors selection and path planning at each sample time of target tracking can make the EKF algorithm in dynamic scenarios complete target trajectory prediction more efficiently. Simu-lation results show that the proposed sensor scheduling strategies have better per-formances in power consumption and tracking accuracy, compared with the static network extend Kalman filter algorithm.
Authors: Qiang Liu (University of Electronic Science and Technology of China), Hao Dong (University of Electronic Science and Technology of China),
Hide Authors & Abstract

Show Authors & Abstract

Lunch 12:30 - 13:00

30 minutes

Workshop on Applications, Testbeds, and Simulation Design for Molecular Communication 13:00 - 14:20

13:00 - 13:20
Molecular MIMO Communications Platform with BTSK for In-Vessel Network Systems

In this paper, we propose a molecular multiple-input multipleout (MIMO) communication platform using binary time shift keying (BTSK) modulation to model in-vessel network systems. A notable prior work introduced a vessel-like communication testbed, yet leaving a challenge to achieve a higher data rate. We suggest an improved version of testbed adding MIMO configurations with modulation techniques. The flow-assist channel model for MIMO systems has been limitedly investigated yet, the feasibility of MIMO systems with timing-based modulation is shown in this paper. The platform uses acid and base molecules as information carriers, and the received output is a set of pH values varying over time. The MIMO platform yields a higher data rate than the singleinput single-output (SISO) systems. Furthermore, the system is flexible to any desired configurations, which can illustrate actual blood vessel environments.
Authors: Changmin Lee (Yonsei University), Bon-Hong Koo (Yonsei University), Chan-Byoung Chae (Yonsei University),
Hide Authors & Abstract

Show Authors & Abstract
13:20 - 13:40
Preliminary Studies on Flow Assisted Propagation of Fluorescent Microbeads in Microfluidic Channels for Molecular Communication Systems

High throughput microfluidic devices coupled with optical detection systems bring several advantages to study molecular communication (MC) by mimicking capillary vessels and arterioles. Motivated by this, we present an MC platform using fluorescence polystyrene (PS) beads as messenger molecules to transfer encoded information in microfluidic channels via flow induced diffusion. To this end, we couple multiple production and analysis techniques to construct and characterize our micro scale MC system. PS microbeads are introduced into microchannels via programmable syringe pumps serving as transmitters, while the received signal is recorded by inverted fluorescence microscope. Time lapsed images of microparticles are presented as they move across diffusion channels.
Authors: M. Gorkem Durmaz (Bogazici University, Computer Engineering Department, NETLAB, 34242, Turkey), Abdurrahman Dilmac (Bogazici University, Computer Engineering Department, NETLAB, 34242, Turkey), Berk Camli (Bogazici University, Electrical and Electronics Engineering Department, 34242, Turkey), Elif Gencturk (Bogazici University, Chemical Engineering Department, 34242, Turkey), Z. Cansu Canbek Ozdil (Bogazici University, Computer Engineering Department, NETLAB, 34242, Turkey), Ali Emre Pusane (Bogazici University, Electrical and Electronics Engineering Department, 34242, Turkey), Arda Deniz Yalcinkaya (Bogazici University, Electrical and Electronics Engineering Department, 34242, Turkey), Kutlu Ulgen (Bogazici University, Chemical Engineering Department, 34242, Turkey), Tuna Tugcu (Bogazici University, Computer Engineering Department, NETLAB, 34242, Turkey),
Hide Authors & Abstract

Show Authors & Abstract
13:40 - 14:00
Comparative Evaluation of a New Sensor for Superparamagnetic Iron-Oxide Nanoparticles in a Molecular Communication Setting

Testbeds are required to assess concepts and devices in the context of molecular communication. These allow the observation of real-life phenomena in a controlled environment and therefore present the basis of future work. A testbed using superparamagnetic iron-oxide nanoparticles (SPIONs) as information carriers was constructed in lieu of this context and requires a sensitive receiver for the detection of SPIONs. This paper focusses on the comparison between a newly presented device (inductance sensor), a previously constructed SPION sensor (resonance bridge), and a commercial susceptometer as reference. The new inductance sensor is intended to improve on a low sensitivity achieved with the previous device and restrictions regarding sample rate and measurement aperture encountered with the susceptometer. The signal-to-noise ratio (SNR) for each device is assessed at a variety of SPION concentrations. Furthermore, the sensors bit error rates (BER) for a random bit sequence are determined. The results show the device based on an inductance sensor to be the most promising for further investigation as values both for BER and SNR exceed those of the resonance bridge while providing a sufficiently high sample rate. On average the SNR of the new device is 13dB higher while the BER for the worst transmission scenario is 9% lower. The commercial susceptometer, although returning the highest SNR, lacks adaptability for the given use-case.
Authors: Max Bartunik (Institute for Electronis Engineering, Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), Germany), Harald Unterweger (Section of Experimental Oncology and Nanomedicine (SEON), University Hospital Erlangen, Germany), Christoph Alexiou (Section of Experimental Oncology and Nanomedicine (SEON), University Hospital Erlangen, Germany), Robert Schober (Institute for Digital Communications, Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), Germany), Maximilian Lübke (Institute for Electronics Engineering, Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), Germany), Georg Fischer (Institute for Electronis Engineering, Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), Germany), Jens Kirchner (Institute for Electronis Engineering, Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), Germany),
Hide Authors & Abstract

Show Authors & Abstract
14:00 - 14:20
Localization of a Passive Molecular Transmitter with a Sensor Network

Macroscale molecular communication (MC), which has a potential for practical applications, is a promising area for communication engineering. In a practical scenario such as monitoring air pollutants released from an unknown source, it is essential to estimate the location of the molecular transmitter (TX). This paper presents a novel Sensor Network-based Localization Algorithm (SNCLA) for passive transmission by using a novel experimental platform which mainly comprises a clustered sensor network (SN) with 24 sensor nodes and evaporating ethanol molecules as the passive TX. With the usage of the SN concept, novel methods can be developed for the problems in macroscale MC by utilizing the wide literature of sensor networks. In SNCLA, Gaussian plume model is employed to derive the location estimator. The parameters such as transmitted mass, wind velocity, detection time and actual concentration are calculated or estimated from the measured signals via the SN to be employed as the input for the location estimator. The numerical results show that the performance of SNCLA is better for stronger winds in the medium. Our findings show that evaporated molecules do not propagate homogeneously through the SN due to the presence of the wind. In addition, the estimation error of SNCLA decreases for higher detection threshold values.
Authors: Fatih Gulec (Izmir Institute of Technology, Turkey), Baris Atakan (Izmir Institute of Technology, Turkey),
Hide Authors & Abstract

Show Authors & Abstract

Best Paper announcement 14:20 - 14:25