Main Conference
Track 3 – Wireless Networks
Session T3-S1
Learning 1
Conference
11:00 AM
—
12:30 PM KST
Local
May 25 Mon, 7:00 PM
—
8:30 PM PDT
Robust Federated Learning Under Worst-case Model
Fan Ang (University of Science and Technology of China, China); Chen Li (University of Science And Technology of China, China); Weidong Wang (University of Science and Technology of China, China)
0
Federated learning provides a communication-efficient training process via alternating between local training and averaging updated local model. Nevertheless, it requires perfectly acquisition of the model which is hard to achieve in wireless communication practically, and the noise will cause serious effect on federated learning. To tackle this challenge, we propose a robust design for federated learning to decline the effect of noise. Considering the noise in communication steps, we first formulate the problem as the parallel optimization for each node under worst-case model. We utilize the samplingbased successive convex approximation algorithm to develop a feasible training scheme, due to the unavailable maxima noise condition and non-convex issue of the objective function. In addition, the convergence rate of proposed design are analyzed from a theoretical point of view. Finally, the prediction accuracy improvement and loss function value reduction of the proposed design are demonstrated via simulation.
SADroid: A Deep Classification Model for Android Malware Detection Based on Semantic Analysis
Dali Zhu (Institute of Information Engineering, Chinese Academy of Sciences, China); Tong Xi (Institute of Information Engineering, Chinese Academy of Sciences & School of Cyber Security, University of Chinese Academy of Sciences, China); Pengfei Jing (Institute of Information Engineering, Chinese Academy of Sciences, China); Qing Xia (Institute of Software, Chinese Academy of Sciences, China); Di Wu and Yiming Zhang (Institute of Information Engineering, Chinese Academy of Sciences, China)
0
Previous works have designed many deep learning models for Android malware detection using various features (e.g. permissions, APIs et.) to achieve better classification performance. However, these methods usually input each feature into the classifier independently and completely (using One- Hot Encoding) so that features are orthogonal to each other. This discrete representation is difficult to preserve the semantic information of features. In this paper, we design two feature segmentation methods to enhance the semantics of the features in preprocessing. Besides that, we propose a malware detection model that consists of a distributed representation process for Android features and an optimized convolutional neural network for classification, named Semantic Analysis Detection (SADroid). In SADroid, the distance between features with similar semantics is closer in vector space. It provides the semantic information of features to the classifier to improve the classification performance. In the evaluation, SADroid outperforms the advanced models in detection accuracy on a data set of 19,600 applications, while maintaining a low computational cost.
Semi-Federated Learning
Zhikun Chen, Daofeng Li, Ming Zhao, Sihai Zhang and Jinkang Zhu (University of Science and Technology of China, China)
1
Federated learning(FL) enables massive distributed Information and Communication Technology (ICT) devices to learn a global consensus model without any participants revealing their own data to the central server. However, the practicality, communication expense and non-independent and identical distribution (Non-IID) data challenges in FL still need to be concerned. In this work, we propose the Semi-Federated Learning (Semi-FL) which differs from the FL in two aspects, local clients clustering and in-cluster training. A sequential training manner is designed for our in-cluster training in this paper which enables the neighboring clients to share their learning models. The proposed Semi-FL can be easily applied to future mobile communication networks and require less uplink transmission bandwidth. Numerical experiments validate the feasibility, learning performance and the robustness to Non-IID data of the proposed Semi-FL. The Semi-FL extends the existing potentials of FL.
Optimization-driven Hierarchical Deep Reinforcement Learning for Hybrid Relaying Communications
Yuze Zou (Huazhong University of Science and Technology, China); Yutong Xie (Shenzhen Institutes of Advanced Technology, China); Canhui Zhang and Shimin Gong (Sun Yat-sen University, China); Hoang Thai Dinh (University of Technology Sydney (UTS), Australia); Dusit Niyato (Nanyang Technological University, Singapore)
2
In this paper, we employ multiple wireless-powered user devices as wireless relays to assist information transmission from a multi-antenna access point to a single-antenna receiver. To improve energy efficiency, we design a hybrid relaying communication strategy in which wireless relays are allowed to operate in either the passive mode via backscatter communications or the active mode via RF communications, depending on their channel conditions and energy states. We aim to maximize the overall SNR by jointly optimizing the access point's beamforming strategy as well as individual relays' radio modes and operating parameters. Due to the non-convex and combinatorial structure of the SNR maximization problem, we develop a deep reinforcement learning approach that adapts the beamforming and relaying strategies dynamically. In particular, we propose a novel optimization-driven hierarchical deep deterministic policy gradient (H-DDPG) approach that integrates the model-based optimization into the framework of conventional DDPG approach. It decomposes the discrete relay mode selection into the outer-loop by using deep Q-network (DQN) algorithm and then optimizes the continuous beamforming and relays' operating parameters by using the inner-loop DDPG algorithm. Simulation results reveal that the H-DDPG is robust to the hyper parameters and can speed up the learning process compared to the conventional DDPG approach.
Auction based Incentive Design for Efficient Federated Learning in Cellular Wireless Networks
Huong Tra Le (Kyung Hee University, Korea (South)); Nguyen H. Tran (The University of Sydney, Australia); Yan Kyaw Tun (Kyung Hee University, Korea (South)); Zhu Han (University of Houston, USA); Choong Seon Hong (Kyung Hee University, Korea (South))
0
Federated learning is an prominent machine learning technique that model is trained distributively by using local data of mobile users, which can preserve the privacy of users and still guarantee high learning performance. In this paper, we deal with the problem of incentive mechanism design for motivating users to participate in training. In this paper, we employ the randomized auction framework for incentive mechanism design in which the base station is a seller and mobile users are buyers. Concerning the energy cost incurred due to join the training, the users need to decide how many uplink subchannels, transmission power and CPU cycle frequency and then claim them in submitted bids to the base station. After receiving the submitted bids, the base station needs algorithms to select winners and determine the corresponding rewards so that the social cost is minimized. The proposed mechanism can guarantee three economic properties, i.e., truthfulness, individual rationality and efficiency. Finally, numerical results are provided to demonstrate the effectiveness, and efficiency of our scheme.
Session Chair
Jeongyeup Paek (Chung-Ang University, Korea (South))
Session T3-S2
Sensor Network and IoT 1
Conference
11:00 AM
—
12:30 PM KST
Local
May 25 Mon, 7:00 PM
—
8:30 PM PDT
Impact of Packet Routing Scheme on Post-Failure Industrial Wireless Sensor Networks
Rajith M Bandarage, Geoffrey G. Messier and Abraham O Fapojuwo (University of Calgary, Canada)
2
In most industrial applications, failed nodes will be repaired, but these repairs take time. For critical system applications, network operators need to understand how their networks function immediately after a failure but before a repair is possible. In this paper, we introduce an extension to the frame-level optimized routing/ scheduling algorithm to improve reliability and energy efficiency and compare it with the other industrial routing algorithms considering the oil refinery wireless sensor networks. We present a structured performance evaluation approach for studying the impact of the routing/scheduling algorithm on the post-failure/pre-repair regime. Simulation results show that the proposed algorithm exerts a positive impact on network performance: highly-reliable, low-latency, energy-efficient, and fitting with most industrial applications.
Wireless Backhaul Strategies for Real-Time High-Density Seismic Acquisition
Varun Amar Reddy and Gordon Stüber (Georgia Institute of Technology, USA); Suhail Al-Dharrab, Ali H Muqaibel and Wessam Mesbah (King Fahd University of Petroleum and Minerals, Saudi Arabia)
0
Modern geophysics methods for oil and gas exploration are able to generate subsurface images of superior quality and depth, albeit making real-time data acquisition a more challenging task. While current literature addresses data transfer issues primarily between the geophones and the gateway nodes, a communication scheme for the transfer of data from the gateway nodes to the sink largely remains unsolved. A novel wireless geophone network architecture for seismic data acquisition is described, with the objective of eliminating cable-based systems and providing Gigabit rates in order to support real-time acquisition. A performance analysis is conducted for various mesh networks employing IEEE 802.11ac, IEEE 802.11ad, and free space optical communication, from the geophysics perspective. Our findings suggest that the bottleneck links can be shifted from the top to the bottom of existing architectures, and a scalable approach requiring minimal number of gateway devices can be designed for high-density seismic surveys.
Asymmetric Wake-up Scheduling based on Block Designs in Wireless Sensor Networks
Teuk-Seob Song (Mokwon University, Korea (South)); Woosik Lee (SSIS, Korea (South)); Jong-Hoon Youn (University of Nebraska - Omaha, USA)
1
In wireless sensor networks (WSNs) with symmetric duty cycles, a block design technique produces an optimal solution for neighbor discovery in terms of the worst-case discovery latency. However, block design-based neighbor discovery methods may not be applicable to WSNs with asymmetric duty operations. Thus, to address this lack of support of asymmetric WSNs, we propose a new neighbor discovery protocol (NDP) that combines two block designs for generating a set of discovery schedules. We prove that the discovery schedule generated by the proposed NDP includes at least one common active slot with any neighboring nodes within a single cycle. We also conduct a simulation study and show that the proposed NDP is better than representative NDPs such as U-Connect, Disco, SearchLight, Hedis, and Todis in terms of discovery latency and energy efficiency.
On the Coverage Performance of Boolean-Poisson Cluster Models for Wireless Sensor Networks
Kaushlendra Pandey (Indian Institute of Technology, Kanpur, India); Abhishek K Gupta (Indian Institute of Technology Kanpur, India)
0
In this paper, we consider wireless sensor networks (WSNs) with sensor nodes exhibiting clustering in their deployment. We model the coverage region of such WSNs by Boolean Poisson cluster models (BPCM) where sensors nodes' location is according to a Poisson cluster process (PCP) and each sensor has an independent sensing range around it. We consider two variants of PCP, in particular Matern and Thomas cluster process to form Boolean Matern and Thomas cluster models. We first derive the capacity functional of these models. Using the derived expressions, we compute the sensing probability of an event and compare it with sensing probability of a WSN modeled by a Boolean Poisson model where sensors are deployed according to a Poisson point process. We also derive the power required for each cluster to collect data from all of its sensors for the three considered WSNs. We show that a BPCM WSN has less power requirement in comparison to the Boolean Poisson WSN, but it suffers from lower coverage, leading to a trade-off between per-cluster power requirement and the sensing performance. A cluster process with desired clustering may provide better coverage while maintaining low power requirements.
FMUCR: Fuzzy-based Multi-hop Unequal Cluster Routing for WSN
Lu Sixu, Wu Muqing and Min Zhao (Beijing University of Posts and Telecommunications, China)
1
Recently, wireless sensor networks play an important role in our life. Cluster routing has gained more attention in wireless sensor networks. However, hotpot problem always exists. One way to solve this problem is the unequal cluster routing. In most of the unequal routing protocols, nodes which closer to base station have the smaller cluster size than others. It will reduce the relay pressure of the node which near to the base station. In this paper, we propose the fuzzy-based multi-hop unequal cluster routing. In the cluster head election phase, relative inter cluster cost and relative intra cluster cost are proposed innovatively. Fuzzy system is used for unequal clustering which reduces the energy consumption for cluster members. In the cluster formation phase, a novel probability mechanism is proposed to let the cluster members decide which cluster to join. In the multi-hop routing phase, relative relay cluster cost is proposed innovatively for inter cluster routing. Two factors are considered for multi-hop routing which reduces the energy consumption for cluster heads. Self-adaptive rotation mechanism is proposed in the data transmission phase. It reduces the frequency for re-clustering self-adaptively which reduces the control overhead of the entire network. According to the simulation results, the proposed protocol balances and reduces the energy consumption as well as extends the network lifetime of the whole network.
Session Chair
Wei Liu (Chongqing University of Technology, P.R. China)
Session T3-S3
Fog Computing and Networking
Conference
11:00 AM
—
12:30 PM KST
Local
May 25 Mon, 7:00 PM
—
8:30 PM PDT
A Collaborative Task Offloading Scheme in D2D-Assisted Fog Computing Networks
Nanxin Fan and Xiaoxiang Wang (Beijing University of Posts and Telecommunications, China); Dongyu Wang (Beijing University of Posts and Telecommunications & Key Laboratory of Universal Wireless Communications, Ministry of Education, China); Yanwen Lan and Junxu Hou (Beijing University of Posts and Telecommunications, China)
2
Fog computing is a promising way to deal with computation-extensive tasks by deploying distributed nodes at the network edge. However, the resources of fog nodes may not be enough when many users need to be served simultaneously. Therefore, task offloading between a device- to-device (D2D) pair is introduced in existing works, which ignores the computing resources of other ubiquitous users. In this paper, we study the partial task offloading in D2D-assisted fog computing networks, where the task is collaboratively executed by multiple D2D users or a Fog-Access Point (F-AP). Firstly, the social relationship based on task offloading is introduced for user incentives. Secondly, aiming to maximize the total revenue, an optimization problem is formulated with the joint consideration of offloading strategy, computation resource allocation and D2D user selection. Next, to solve this non-linear and non-convex problem, a potential game is formulated and the existence of a Nash equilibrium (NE) is proved. Then, we propose a joint optimization algorithm to obtain the NE, where computation resource allocation problem is solved by Lagrange Multiplier Method while D2D users are selected through greedy strategy. Finally, simulations show the effectiveness of the proposed scheme.
An Energy-Efficient Mixed-Task Paradigm in Resource Allocation for Fog Computing
Xincheng Chen (Xi'an Jiaotong University, China); Yuchen Zhou (Xidian University, China); Long Yang (Xidian University, China & University of Alberta, Canada); Lu Lv (Xidian University, China)
3
We study the energy efficiency problem for a fog computing system with multiple users and fog nodes. In order to handle various kinds of tasks at the same time, a mixed- task paradigm is proposed to combine both binary offloading and partial offloading. Then, a mixed-task resource allocation problem with the consideration of computation and communication resources is formulated as a mixed integer nonlinear programming problem (MINLP), which cannot be handled by the traditional relaxation algorithm due to the joint design of binary and partial offloading. To solve this problem efficiently, we first adopt a replacement-based method to transform the problem, and then design an augmented Lagrange method (ALM)-based resource allocation scheme. To further accelerate the solution procedure, a novel optimization technique, AMSGrad, is applied to the designed scheme. The performance of the proposed scheme is demonstrated by simulation results.
Latency Minimization with Optimum Workload Distribution and Power Control for Fog Computing
Saman Atapattu (University of Melbourne, Australia); Chathuranga Weeraddana (University of Moratuwa, Sri Lanka); Minhua Ding (Sri Lanka Institute of Information Technology, Sri Lanka); Hazer Inaltekin (Macquarie University, Australia); Jamie S Evans (University of Melbourne, Australia)
3
This paper investigates a three-layer IoT-fog-cloud computing system to determine the optimum workload and power allocation at each layer. The objective is to minimize maximum per-layer latency (including both data processing and transmission delays) with individual power constraints. The resulting optimum resource allocation problem is a mixed-integer optimization problem with exponential complexity. Hence, the problem is first relaxed under appropriate modeling assumptions, and then an efficient iterative method is proposed to solve the relaxed but still non-convex problem. The proposed algorithm is based on an alternating optimization approach, which yields close-to-optimum results with significantly reduced complexity. Numerical results are provided to illustrate the performance of the proposed algorithm compared to the exhaustive search method. The latency gain of three-layer distributed IoT-fog-cloud computing is quantified with respect to fog-only and cloud-only computing systems.
Mobility Prediction-Based Joint Task Assignment and Resource Allocation in Vehicular Fog Computing
Xianjing Wu, Shengjie Zhao and Rongqing Zhang (Tongji University, China); Liuqing Yang (Colorado State University, USA)
1
Most recently, vehicular fog computing (VFC) has been regarded as a novel and promising architecture to effectively reduce the computation time of various vehicular application tasks in Internet of vehicles (IoV). However, the high mobility of vehicles makes the topology of vehicular networks change fast, and thus it is a big challenge to coordinate vehicles for VFC in such a highly mobile scenario. In this paper, we investigate the joint task assignment and resource allocation optimization problem by taking the mobility effect into consideration in vehicular fog computing. Specifically, we formulate the joint optimization problem from a Min-Max perspective in order to reduce the overall task latency. Then we decompose the non- convex problem into two sub-problems, i.e., one to one matching and bandwidth resource allocation, respectively. In addition, considering the relatively stable moving patterns of a vehicle in a short period, we further introduce the mobility prediction to design a mobility prediction-based scheme to obtain a better solution. Simulation results verify the efficiency of our proposed mobility prediction-based scheme in reducing the overall task completion latency in VFC.
Distributed V2V Computation Offloading Based on Dynamic Pricing Using Deep Reinforcement Learning
Jinming Shi (Tsinghua University, China); Jun Du (Tsinghua University, Beijing, China); Jian Wang and Jian Yuan (Tsinghua University, China)
1
Vehicular computation offloading is a promising paradigm that improves the computing capability of vehicles to support autonomous driving and various on-board infotainment services. Comparing with accessing the remote cloud, distributed vehicle-to-vehicle (V2V) computation offloading is more efficient and suitable for delay-sensitive tasks by taking advantage of vehicular idle computing resources. Due to the high dynamic vehicular environment and the variation of available vehicular computing resources, it is a great challenge to design an effective task offloading mechanism to efficiently utilize vehicular computing resources. In this paper, we investigate the computation task allocation among vehicles, and propose a distributed V2V computation offloading framework, in which wireless channel states and variation of idle computing resources are both considered. Specially, we formulate the task allocation problem as a sequential decision making problem, which can be solved by using deep reinforcement learning. Considering that vehicles with idle computing resources may not share their computing resources voluntarily, we thus propose a dynamic pricing scheme that motivates vehicles to contribute their computing resources according to the price they receive. The performance of designed task allocation mechanism is validated by simulation results which reveal the effectiveness of our mechanism compared to the other algorithms.
Session Chair
Byung-Seo Kim (Hongik University, Korea)
Session T3-S4
Localization
Conference
2:00 PM
—
3:30 PM KST
Local
May 25 Mon, 10:00 PM
—
11:30 PM PDT
Wimage: Crowd Sensing based Heterogeneous Information Fusion for Indoor Localization
Fangmin Li (Shenzhen Institutes of Advanced Technology Chinese Academy of Science, China); Yubin Zhao (Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, China); Xiaofan Li (The State Radio Monitoring Center and Testing Center, China); Cheng-Zhong Xu (University of Macau, China)
0
Crowd sensing is an efficient way to collect heterogeneous information in the complicated infrastructures for fingerprinting based indoor localization. However, the information related to the dynamic trajectory are difficult to fuse due to the reliability issues from different devices and user moving habits. In this paper, we proposed a crowd sensing based indoor localization system with heterogeneous information fusion, which is called Wimage. Wimage can efficiently fuse multiple information sources related to location information, e.g., visual image, WiFi and geomagnetic data, even if the targets are moving with different and variable speeds. Then we design image-base subregion matching algorithm to locate the initial position and segmented weighted K-nearest neighbor algorithm to attain the matched trajectories in the database. A dynamic temporal warping algorithm is proposed for further calibrating the estimations. The experimental results indicate that with the helps from different kinds of information, the root mean square error is only below 0.4m, which is highly accurate for locating a target in a large scale of indoor environment.
Flight Path Optimization for UAVs to Provide Location Service to Ground Targets
Youpeng Wang and Xiaojun Zhu (Nanjing University of Aeronautics and Astronautics, China); Lijie Xu (Nanjing University of Posts and Telecommunications, China)
1
We consider using UAVs to provide location service to ground targets. UAVs fly over the target area and broadcast their locations periodically via wireless signals. When a UAV flies along a straight line, a target will observe the strongest signal when the UAV is at the closest location, based on which it can infer its location. The problem is to minimize the flight length subject to the constraint that the transmission range of a UAV is limited and all targets in the area should be located. We formulate the problem and give an optimal flight path to cover a square with side length no greater than a certain value. In the general case when the given area cannot be covered by the square, we propose to divide UAVs into two groups, whose flight paths are either parallel or orthogonal. We conduct simulations to compare our approach with existing approaches. Results verify the superiority of our approach in terms of flight length and localization error.
Cooperative Localization in Wireless Sensor Networks with AOA Ranging Measurements
Xianbo Jiang (Beijing University of Posts and Telecommunications, China); Shengchu Wang (Beijing University Of Posts And Telecommunications, China)
0
This paper researches the cooperative localization in wireless sensor networks (WSNs) with 2π/π -periodic angle-of-arrival (AOA) ranging measurements. When the orientation angles of the antenna arrays at WSN nodes are known, a ranging link loss is defined based on the tan-relationships between AOA observations and æy-minus coordinates of two neighboring nodes. Subsequently, the positioning problem under 2π -periodic AOAs is converted as a convex optimization problem about minimizing total ranging-link loss through optimizing agent positions, which is resolved by the gradient-descent (GD) method. Under π-periodic AOAs, additional 0/1 integers are introduced to indicate the front-or-back impinging directions. By relaxing 0/1 integers as continuous variables within [0,1], the positioning problem is relaxed as a nonconvex optimization one about minimizing total link loss over the agent positions and indicating variables, which is solved by the projected GD (PGD) method. Finally, Type-I least-square (LS) localizer is developed for WSNs with both 2π and π-periodic AOAs. When the orientation angles are unknown, Type-II LS localizer is developed by combining Type-I LS localizer with a maximum-likelihood (ML) orientation estimator, which alternatively updates agent positions and orientation angles. Simulation results validate that the proposed LS-type localizers outperform existing localizers.
Depthwise Separable Convolution based Passive Indoor Localization using CSI Fingerprint
Wenjing Xun, Lijuan Sun, Chong Han, Zhaoxiao Lin and Jian Guo (Nanjing University of Posts and Telecommunications, China)
0
Wi-Fi-based indoor localization has received extensive attention in the academic community. However, most Wi-Fi-based indoor localization systems have complex models and high localization delays, which limit the universality of these localization methods. To solve these problems, we propose a depthwise separable convolution based passive indoor localization system (DSCP) using Wi-Fi channel state information (CSI). DSCP is a fingerprint-based localization system, which includes an offline training phase and an online localization phase. In the offline training phase, the indoor scenario is first divided into different areas to set training locations for collecting CSI. Then amplitude differences of these CSI subcarriers are extracted for constructing location fingerprints, thereby training the CNN. In the online localization phase, CSI data is first collected at the test locations, then the location fingerprint is extracted and finally fed to the trained network to obtain the predicted location. The experimental results show that DSCP has a short training time and a low localization delay. DSCP achieves a high localization accuracy, upper than 97%, and a small median localization distance error of 0.98 m in open indoor scenarios.
Packet Corruption Tolerant Localization for Underwater Acoustic Sensor Networks
Keyong Hu, Xianglin Song and Zhongwei Sun (Qingdao University of Technology, China); Hanjiang Luo (Shandong University of Science and Technology, China); Zhongwen Guo (Ocean University of China, China)
0
Existing range-based localization schemes for underwater acoustic sensor networks (UASNs) rely on sufficient and accurate distance measurements. However, in practice, ranging packets are inevitably corrupted due to packet collisions and signal noises, resulting in missing and noisy distance measurements and further degrading localization performance significantly. In this paper, we propose a packet corruption tolerant localization algorithm to address this challenge. First, we design an energy-efficient mechanism to gather inter-node distance measurements and form partially observed Square Distance Matrix (SDM). Then, leveraging the intrinsic low-rank structure of SDM, the reconstruction of true SDM is formulated as a Frobenius-norm regularized matrix factorization problem and an improved Newton-Raphson method is designed to solve this problem. Finally, we apply Multi-Dimension Scaling technique to localize all the nodes based on the reconstructed SDM. Simulation results demonstrate that, our proposed algorithm outperforms the benchmark approaches in terms of localization accuracy, coverage and stability.
On the Localization of Unmanned Aerial Vehicles with Cellular Networks
Irshad A. Meer, Mustafa Ozger and Cicek Cavdar (KTH Royal Institute of Technology, Sweden)
1
Localization plays a key role for safe operation of UAVs enabling beyond visual line of sight applications. Compared to GPS based localization, cellular networks can reduce the positioning error and cost since cellular connectivity is becoming a prominent solution as a communication system for UAVs. As a first step towards localization, UAV needs to receive sufficient number of localization signals each having a signal to interference plus noise ratio (SINR) greater than a threshold. On the other hand, three-dimensional mobility of UAVs, altitude dependent channel characteristics between base stations (BSs) and UAVs, its line of sight and non-line of sight conditions, and resulting interference from the neighboring BSs pose challenges to receive usable signals from the required number of BSs. In this paper, we utilize a tractable approach to calculate localizability probability, which is defined as the probability of successfully receiving usable signals from at least a certain number of BSs. Localizability has an impact on overall localization performance regardless of the localization technique to be used. In our simulation study, we investigate the relation between the localizability probability with respect to the number of participating BSs, post-processing SINR requirement, air-to-ground channel characteristics, and network coordination, which are shown to be the most important factors for the localizability performance of UAVs. We observe the localizability performance is better at higher altitudes which indicates that localizability with cellular networks for UAVs is more favorable than for terrestrial users.
Session Chair
Jiguang He (University of Oulu, Finland)
Session T3-S5
Energy Efficiency
Conference
2:00 PM
—
3:30 PM KST
Local
May 25 Mon, 10:00 PM
—
11:30 PM PDT
Energy and Spectral Efficiency Tradeoff in OFDMA Networks via Antenna Selection Strategy
Ata Khalili (Amirkabir University of Technology, Iran); Derrick Wing Kwan Ng (University of New South Wales, Australia)
1
In this paper, we investigate the joint resource allocation and antenna selection algorithm design for uplink orthogonal frequency division multiple access (OFDMA) communication system. We propose a multi-objective optimization framework to strike a balance between spectral efficiency (SE) and energy efficiency (EE). The resource allocation design is formulated as a multi-objective optimization problem (MOOP), where the conflicting objective functions are linearly combined into a single objective function employing the weighted sum method. In order to develop an efficient solution, the majorization minimization (MM) approach is proposed where a surrogate function serves as a lower bound of the objective function. Then an iterative suboptimal algorithm is proposed to maximize the approximate objective function. Numerical results unveil an interesting tradeoff between the considered conflicting system design objectives and reveal the improved EE and SE facilitated by the proposed transmit antenna selection in OFDMA systems.
Heterogeneity-based Energy-efficient Transmission in Dense Small Cell Networks
Shie Wu (Yantai University, P.R. China); Rui Yin (School of Information and Electrical Engineering, Zhejiang University City College, Hangzhou, China); Ningfei Dong (Yantai University, China); Xia Liu (Yantai University, P.R. China)
0
Green communications in dense small cell network (DSCNs) has attracted much attention. Energy saving (ES) and energy efficiency (EE) are two main goals and they are usually optimized separately. In this paper, taking into account the heterogeneity and cooperation opportunities among small cells, EE and ES are jointly optimized through subframe configuration and power allocation in the DSCNs. To quantize the heterogeneity, we define an EE preference function. Accordingly, a multi-objective optimization problem is formulated while considering the EE and ES optimization simultaneously. Due to the coupling of EE and ES, obtaining the solution is non-trivial. A heterogeneity-based ES and EE (HESEE) optimization algorithm is proposed, where the sleep mechanism is adopted via the subframe configuration. Particularly, the concave-convex procedure (CCCP) method is applied to solve the non-concave sum-of-ratios optimization for system EE. Simulation results show that the proposed HESEE algorithm can optimize the EE of small cell base stations (SBSs) according to their EE preference weights. In addition, compared with the base scheme, the HESEE algorithm can save energy by over 35.2% while improving system EE by up to 23%.
Dynamic Load Adjustments for Small Cells in Heterogeneous Ultra-dense Networks
Qi Zhang (Beijing University of Posts and Telecommunications, China); Xiaodong Xu (Beijing University of Posts and Telecommunications & Wireless Technology Innovation Institute, China); Jingxuan Zhang and Xiaofeng Tao (Beijing University of Posts and Telecommunications, China); Cong Liu (China Mobile Research Institute, China)
1
The ultra-dense deployment of small cells has been applied to the 5th-generation (5G) mobile networks. A large number of base stations (BSs) will lead to a dramatic increase in energy consumption, and network resources will be more difficult to fully utilize. In this paper, we propose the dynamic load adjustments (DLA) algorithm for small cells in heterogeneous ultra-dense networks. The proposed algorithm applies Q-learning to learn effective offloading policies which could combine the energy-saving function and the load balancing function. Based on the DLA algorithm, the heterogeneous ultra-dense networks could adjust the traffic load to turn off some redundant BSs or balance the load between heavily loaded BSs and lightly loaded BSs. The simulation results show that the algorithm not only improves the network energy efficiency when the average load of the networks is light, but also improves the network throughput when the average load of the networks is heavy.
Energy Efficient Joint Resource Allocation and Clustering Algorithm for M2M Communication Systems
Changzhu Liu, Rong Chai and Ahmad Zubair (Chongqing University of Posts and Telecommunications, China); Qianbin Chen (Chongqing University of Posts and Telecommunication, China)
1
In recent years, machine-to-machine (M2M) communications have attracted great attentions from both academia and industry. In M2M communication systems, machine type communication devices (MTCDs) are capable of communicating with each other intelligently under highly reduced human interventions. Although diverse types of services are expected to be supported for MTCDs, various quality of service (QoS) requirements and network states pose difficulties and challenges to the resource allocation and clustering schemes of M2M communication systems. In this paper, we address the joint resource allocation and clustering problem in M2M communication systems. To achieve the efficient resource management of the MTCDs, we propose a joint resource management architecture, and design a joint resource allocation and clustering algorithm. More specifically, by defining system energy efficiency as the sum of the energy efficiency of the MTCDs, the joint resource allocation and clustering problem is formulated as an energy efficiency maximization problem. As the original optimization problem is a nonlinear fractional programming problem, which cannot be solved conveniently, we transform the optimization problem into power allocation subproblem and clustering subproblem. Applying iterative method-based energy efficiency maximization algorithm, we first obtain the optimal power allocation strategy based on which, we then propose a modified K-means algorithm to obtain the clustering strategy. Numerical results demonstrate the effectiveness of the proposed algorithm.
Energy Efficient Bidirectional Relaying Network Coded HARQ Transmission Scheme for S-IoT
Zilin Ni (Harbin Institute of Technology, China); Jian Jiao (Harbin Institute of Technology - Shenzhen, China); Shiqi Liu (Harbin Institute of Technology (Shenzhen), China); Shaohua Wu (Harbin Institute of Technology, China); Qinyu Zhang (Shenzhen Graduate School, Harbin Institute of Technology, China)
0
Recently, with the development of the next generation of high throughput satellites, deploying satellite-based Internet of Things (S-IoT) is suggested to solve the increasing demand for ubiquitous broadband access capability terrestrial communications. Under the current situation that the number of communication devices and the hardware capabilities of devices continue to increase, network coding becomes an effective way to further improve the throughput and efficiency in S-IoTs. In this paper, a Network Coded Hybrid Automatic Repeat Request (NCed HARQ) transmission scheme is proposed based on typical bidirectional relaying scenarios of S-IoT, and a general process of the NCed HARQ is presented. The corresponding detailed transmission process is given, and the theoretical performance index is derived and verified by simulations, which emphasizes the benefit of network coding. Besides, we adopt matrix exponential distribution in the calculation to make formulations more concise and unified.
Session Chair
Rong Chai (Chongqing University of Posts and Telecommunications, China)
Session T3-S6
Routing
Conference
2:00 PM
—
3:30 PM KST
Local
May 25 Mon, 10:00 PM
—
11:30 PM PDT
A Three-level Routing Hierarchy in improved SDN-MEC-VANET Architecture
Xuefeng Ji, Wenquan Xu, Chuwen Zhang and Bin Liu (Tsinghua University, China)
1
Existing routing algorithms that based on traditional Vehicular ad-hoc NETwork (VANET) architectures cannot provide fast and diverse routing services due to dynamic and unstable environment. To address this issue, we propose a three- level routing hierarchy in improved Software-Defined VANET architecture based on Mobile Edge Computing (SDN-MEC- VANET) to improve routing performance and enrich the data transmission mode for the VANET. Moreover, it can be applied to almost all VANET protocols, enabling protocol-independent forwarding. Besides, this improved architecture can coordinate different edge devices to timely adjust the service delivery strategy under the predictive correction from controllers, providing high-bandwidth and low-delay transmission for Internet of Vehicles (IoV). Meanwhile, MEC technology is introduced to perform local control, leveraging the storage and computing capabilities of edge devices to reduce the processing pressure of the controller. Simulation results show that our routing algorithm in improved network architecture can achieve a higher packet delivery ratio within a reasonable delay than other approaches under different scenarios of network scale, communication frequency and vehicular velocity.
NIHR: Name/ID Hybrid Routing in Information-Centric VANET
Wenquan Xu, Xuefeng Ji, Chuwen Zhang and Bin Liu (Tsinghua University, China)
0
Vehicular Ad hoc network (VANET) has received great attention in recent research, but many challenges still lie in innovating efficient routing protocols to support the highly dynamic environment. Existing ID-based routing protocols cannot fundamentally tackle the dynamic topology problem in VANET. The recent emerging Information-Centric Networking (ICN) makes routing decisions based on data itself instead of a particular host, seeming to have the potential to handle the dynamic topology, but problems (e.g., severe flooding overhead) still remain. Therefore, inspired by the idea of ICN, we propose a name/ID hybrid routing (NIHR) protocol that combines the data-name- based routing and host-ID-based routing to address the above two issues simultaneously. In particular, we develop an announce strategy to improve the efficiency of the in-network cache, and we design a bloom filter based structure to achieve fast content lookup. Simulation results show NIHR's high performance in terms of Packet Delivery Ratio (PDR), Roundtrip time (RTT) and roundtrip hop count. Especially, to verify NIHR's performance in real-world scenarios, we have implemented a vehicular real-time video conference system based on MK5 OBU [1] (On-Board Unit).
FNTAR: A Future Network Topology-aware Routing Protocol in UAV Networks
Jianfei Peng (Nanjing University of Aeronautics and Astronautics, China); Hang Gao and Liang Liu (Colleage of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, China); Yuting Wu (Nanjing University of Aeronautics and Astronautics, China); Xiangyu Xu (Nanjing University Of Aeronautics And Astronautics, China)
0
Unmanned aerial vehicles (UAVs) can gather data in the air and transmit the data to the ground station. Multi- UAV systems have been used in an increasing number of mission scenarios and routing protocols play a critical role in UAV network communications. It is now well established that unstable link quality and frequently changing network topology pose significant challenges for messages forwarding in UAV networks. Hence, traditional mobile ad-hoc network routing protocols do not fit well in UAV networks. In many UAV applications, the flight paths of UAVs are planned in advance before performing missions. The positions and motion information of UAVs are available through Global Positioning System (GPS) and inertial sensors, which can be utilized to calculate the future positions of UAVs. Therefore, the future topology of the UAV network is also available. However, existing work does not take advantage of this information. Based on the trajectory, location and motion information of the UAVs, this paper proposes a future network topology-aware routing (FNTAR) protocol, which uses future location information to make superior routing decisions. Moreover, to mitigate data loss problems caused by unstable links and highly dynamic topology, FNTAR can forward messages to multiple excellent next-hop UAVs based on future network topology, and these UAVs can deliver messages to destinations faster. We implement FNTAR in the simulation experiment, the simulation results demonstrate that FNTAR can achieve lower latency and higher delivery ratio than DTNgeo protocol.
Exploiting Mobile Contact Patterns for Message Forwarding in Mobile Opportunistic Networks
Mohd Yaseen Mir and Chih-Lin Hu (National Central University, Taiwan)
1
When sparse and challenged networks are characterized by small node population, non-randomized node distribution, and low node density, it is hard to maintain end-to- end connectivity among mobile nodes. The mobile opportunistic networking technology provides data dissemination services by means of transient inter-node communications during node movements in a network. Recent studies intend to exploit node mobility to store, carry and forward data upon opportunistic contacts among nodes. To address the uncertainty nature of future contacts, traditional ways replicate message copies to increase data delivery. However, most replication techniques forward message copies either in a greedy manner or select the next forwarding node by considering only contact history. In this paper, we exploit contact patterns among mobile nodes and propose a Regular and Sporadic Contact-Based Routing (RSCR) scheme where regular and sporadic contact patterns are defined to distinguish contacts of which either periodically or occasionally appear in a time scale, to enhance delivery rate in a cost- effective manner. We conduct synthetic simulation to examine the RSCR scheme under the SLAW and Infocom'05 mobility traces. Performance results show its efficiency on the successful delivery rate with lower message overhead as compared with the typical Epidemic and PRoPHETv2 schemes.
Reliable and Power Confined Routing in Large and Densely Deployed 6TiSCH Mesh Networks
Yichao Jin (Toshiba Research Europe Ltd, United Kingdom (Great Britain)); Michael Baddeley (Toshiba Research Europe Ltd., United Kingdom (Great Britain)); Usman Raza (Toshiba Research Europe Limited, United Kingdom (Great Britain)); Aleksandar Stanoev (Toshiba Research Europe Ltd, United Kingdom (Great Britain)); Mahesh Sooriyabandara (Toshiba Research Europe Limited, United Kingdom (Great Britain))
0
RPL, the de-facto standard for low power and lossy networks, forms multi-hop routing structures between network nodes and a single root. It intrinsically minimises the number of hops across the mesh, however this can result in large distances between adjacent forwarding nodes. Consequently, during forwarding operations, the received signal strength can be close to the receiver sensitivity threshold and result in frequent packet losses due to link fluctuations. RECLAIM, our proposed approach, overcomes this challenge by transmitting routebuilding messages at a reduced power level at first while switching back to the normal transmission power level during the actual data communication phase. This results in more reliable links per hop and ensures link budget above the receiver sensitivity threshold. This however creates more hops and a longer routing path in order to reach destination. Hence, RECLAIM further applies efficient and non-conflicting 6TiSCH scheduling method to coordinate those communication events in a non conflicting manner which do not collide or cause interference to each other, resolving the issue that has traditionally prevented this approach. Our detailed simulation shows that RECLAIM drops 60 times less packets than standard RPL, and achieves 99.9999% packet delivery ratio.
Session Chair
Yichao Jin (Toshiba Research Europe Ltd, United Kingdom)
Session T3-S7
Learning 2
Conference
4:00 PM
—
5:30 PM KST
Local
May 26 Tue, 12:00 AM
—
1:30 AM PDT
End-Edge Coordinated Inference for Real-Time BYOD Malware Detection using Deep Learning
Xinrui Tan (State Key Laboratory of Information Security & Institute of Information Engineering, China); Hongjia Li, Liming Wang and Zhen Xu (Chinese Academy of Sciences, China)
0
Bring-Your-Own-Device (BYOD) has been widely viewed as a definite trend among enterprises in which employees bring and use their personal smartphones for work. Despite the perceived opportunities of increasing productivity and reducing costs, BYOD raises severe security and privacy concerns: the corporate networks and data are directly exposed to malware apps running on the personal smartphones. This highlights the necessity for performing real-time mobile malware detection in BYOD environments. Deep learning seems to be a natural choice to handle such detection, due to its state-of-the-art detection effectiveness. However, deep learning inference is usually too computationally complex for resource-constrained smartphones, and the communication overhead of cloud-based inference may be unacceptable. As a result, it is hard to seek the tradeoff between the real-time demand and optimality of detection accuracy. In this paper, we tackle this issue by proposing an end- edge coordinated inference approach that can support highly- accurate and average latency guaranteed malware detection. Our proposed approach integrates the early-exit and model partitioning methods to allow fast, correct and smartphone- localized inference to occur frequently. Extensive evaluations are carried out, demonstrating that our proposed approach offers a good compromise between detection accuracy and efficiency.
MSDF: A Deep Reinforcement Learning Framework for Service Function Chain Migration
Ruoyun Chen and Hancheng Lu (University of Science and Technology of China, China); Yujiao Lu (USTC, China); Jinxue Liu (University of Science and Technology of China, China)
0
Under dynamic traffic, service function chain (SFC) migration is considered as an effective way to improve resource utilization. However, the lack of future network information leads to non-optimal solutions, which motivates us to study reinforcement learning based SFC migration from a long-term perspective. In this paper, we formulate the SFC migration problem as a minimization problem with the objective of total network operation cost under constraints of users' quality of service. We firstly design a deep Q-network based algorithm to solve single SFC migration problem, which can adjust migration strategy online without knowing future information. Further, a novel multi-agent cooperative framework, called MSDF, is proposed to address the challenge of considering multiple SFC migration on the basis of single SFC migration. MSDF reduces the complexity thus accelerates the convergence speed, especially in large scale networks. Experimental results demonstrate that MSDF outperforms typical heuristic algorithms under various scenarios.
L-FGADMM: Layer-Wise Federated Group ADMM for Communication Efficient Decentralized Deep Learning
Anis Elgabli (University of Oulu, Finland); Jihong Park (Deakin University, Australia); Sabbir Ahmed (University of Oulu, Finland); Mehdi Bennis (Centre of Wireless Communications, University of Oulu, Finland)
0
This article proposes a communication-efficient decentralized deep learning algorithm, coined layer-wise federated group ADMM (L-FGADMM). To minimize an empirical risk, every worker in L-FGADMM periodically communicates with two neighbors, in which the periods are separately adjusted for different layers of its deep neural network. A constrained optimization problem for this setting is formulated and solved using the stochastic version of GADMM proposed in our prior work. Numerical evaluations show that by less frequently exchanging the largest layer, L-FGADMM can significantly reduce the communication cost, without compromising the convergence speed. Surprisingly, despite less exchanged information and decentralized operations, intermittently skipping the largest layer consensus in L-FGADMM creates a regularizing effect, thereby achieving the test accuracy as high as federated learning (FL), a baseline method with the entire layer consensus by the aid of a central entity.
Root Cause Analysis of Noisy Neighbors in a Virtualized Infrastructure
Hedi Bouattour (ORANGE, France); Yosra Ben Slimen (Orange Labs Belfort, France); Marouen Mechtri and Hanane Biallach (Orange Labs Networks, France)
0
This paper proposes a model to identify the noise source in a virtualized infrastructure. This phenomenon appears when network functions running under virtual machines that are deployed on the same physical server compete for physical resources. First, an anomaly detection model is proposed to identify the machines that are in an abnormal state in the infrastructure by performing an unsupervised learning. An investigation of the root cause is later achieved by searching how anomalies are propagated in the system. To do this, a supervised learning of the anomaly propagation paths is proposed. A propagation graph is automatically created with a score assigned to its components. With a testbed created using Openstack, an experimentation study with real data is held giving promising results.
Federated Learning under Channel Uncertainty: Joint Client Scheduling and Resource Allocation
Madhusanka Dinesh Weeraratne Manimel Wadu and Sumudu Samarakoon (University of Oulu, Finland); Mehdi Bennis (Centre of Wireless Communications, University of Oulu, Finland)
1
In this work, we propose a novel joint client scheduling and resource block (RB) allocation policy to minimize the loss of accuracy in federated learning (FL) over wireless compared to a centralized training-based solution, under imperfect channel state information (CSI). First, the problem is cast as a stochastic optimization problem over a predefined training duration and solved using the Lyapunov optimization framework. In order to learn and track the wireless channel, a Gaussian process regression (GPR)-based channel prediction method is leveraged and incorporated into the scheduling decision. The proposed scheduling policies are evaluated via numerical simulations, under both perfect and imperfect CSI. Results show that the proposed method reduces the loss of accuracy up to 25.8% compared to state-of-the-art client scheduling and RB allocation methods.
Session Chair
Anis Elgabli (University of Oulu, Finland)
Session T3-S8
Sensor Network and IoT 2
Conference
4:00 PM
—
5:30 PM KST
Local
May 26 Tue, 12:00 AM
—
1:30 AM PDT
On the Age of Information for Multicast Transmission with Hard Deadlines in IoT Systems
Jie Li and Yong Zhou (ShanghaiTech University, China); He Chen (The Chinese University of Hong Kong, Hong Kong)
0
We consider the multicast transmission of a real-time Internet of Things (IoT) system, where a server transmits time-stamped status updates to multiple IoT devices. We apply a recently proposed metric, named age of information (AoI), to capture the timeliness of the information delivery. The AoI is defined as the time elapsed since the generation of the most recently received status update. Different from the existing studies that considered either multicast transmission without hard deadlines or unicast transmission with hard deadlines, we enforce a hard deadline for the service time of multicast transmission. This is important for many emerging multicast IoT applications, where the outdated status updates are useless for IoT devices. Specifically, the transmission of a status update is terminated when either the hard deadline expires or a sufficient number of IoT devices successfully receive the status update. We first calculate the distributions of the service time for all possible reception outcomes at IoT devices, and then derive a closed-form expression of the average AoI. Simulations validate the performance analysis, which reveals that: 1) the multicast transmission with hard deadlines achieves a lower average AoI than that without hard deadlines; and 2) there exists an optimal value of the hard deadline that minimizes the average AoI.
Exact algorithms for barrier coverage with line-based deployed rotatable directional sensors
Zijing Ma, Shuangjuan Li and Dong Huang (South China Agricultural University, China)
0
Barrier coverage is an important coverage model for intrusion detection, which requires a chain of sensors across the deployment area with the adjacent sensors' sensing areas overlapping. Directional sensors are often dispersed from an airplane following a predetermined line. However, barrier coverage cannot be guaranteed after initial sensor deployment due to the sensors' random offsets and random orientations. Fortunately, directional sensors can rotate to mend the barrier gaps using this line-based sensor deployment model. Existing work proposed a greedy heuristic approach to mend the gaps by rotating sensors, but it cannot answer whether there exists a barrier. We fill in this gap by presenting an exact algorithm which can determine whether there exists a barrier. We first introduce the notion of feasible orientation range and then try to calculate each sensor's feasible orientation starting from the leftmost sensor. We also propose a fast algorithm of choosing the sensors' orientations from their feasible orientation ranges to form a barrier if there exists a barrier, or form a set of sub-barriers if there does not exist a barrier. Simulation results show that our algorithm outperforms the distributed algorithm in the existing work.
Genetic Algorithm-based Periodic Charging Scheme for Energy Depletion Avoidance in WRSNs
Huong Thi Tran (Ha Noi University of Science, Vietnam); Phi Le Nguyen (Hanoi University of Science and Technology, Vietnam); Huynh Thi Thanh Binh (HUST, Vietnam); Kien Nguyen (Chiba University, Japan); Minh Hai Ngo (Hanoi University of Science and Technology, Vietnam); Vinh Le (Vietnam National University, Vietnam)
0
Thanks to the advancements in wireless power transfer technologies, a new paradigm of wireless sensor network (WSNs) called wireless rechargeable sensor networks (WRSNs) has recently emerged. For a WRSN, designing an efficient charging schedule is a challenging issue due to the inherent constraints of WSNs. Although there have been many efforts to optimize the charging schedule, the existing works suffer from several critical problems. Firstly, they rarely tackle the dead node minimization problem, which is the ultimate objective of wireless charging. Secondly, most of the existing works assume impractical conditions, which include the unlimited battery capacity of the charger, and a fully charging scheme at the sensors. In this paper, aiming at minimizing the number of dead nodes, we propose a novel charging scheme based on the genetic algorithm. Our scheme works when the mobile charger has only limited capacity, and the sensors are charged partially at each charging round. The experiment results show that our proposed algorithm reduces the number of dead nodes significantly compared to other existing studies.
Simplified Theoretical Model based Self-adaptive Packet Reception Rate Estimation in Sensor Networks
Wei Liu, Yu Xia and Jian Xie (Chongqing University of Technology, China); Ming Xu (Nanjing University of Aeronautics and Astronautics, China); Rong Luo (Tsinghua University, China); Shunren Hu (Chongqing University of Technology, China); Xiaoyu Dang and Daqing Huang (Nanjing University of Aeronautics and Astronautics, China)
0
Real-time and accurate packet reception rate estimation is crucial for wireless sensor networks. However, existing approaches usually rely on offline data collection and training, which limits their generality. Specifically, models fitted by the test data acquired under specific conditions cannot be used in all environments and for arbitrary packet sizes. In this paper, a simplification method for the theoretical bit error rate model of IEEE 802.15.4 2.4 GHz physical layer is proposed, which is about 18 to 38 times faster than the original one. Then, with the simplified model, a lightweight packet reception rate estimation approach is designed, which is self-adaptive to different environments and arbitrary packet sizes. With the proposed approach, offline data collection and training are no longer needed, which will reduce deployment cost effectively. Compared with state-of-the-art approaches, estimate error of the proposed one is reduced by 2.46%~74.97% in different environments, and by 2.46%~62.00% for different packet sizes.
Decision Triggered Data Transmission and Collection in Industrial Internet of Things
Jiguang He (University of Oulu, Finland); Long Kong (Interdisciplinary Centre for Security, Reliability and Trust (SnT) & University of Luxembourg, Luxembourg); Tero Frondelius (R&D and Engineering, Wärtsilä, Vaasa, Finland); Olli Silvén and Markku Juntti (University of Oulu, Finland)
0
We propose a decision triggered data transmission and collection (DTDTC) protocol for condition monitoring and anomaly detection in the industrial Internet of things (IIoT). In the IIoT, the collection, processing, encoding, and transmission of the sensor readings are usually not for the reconstruction of the original data but for decision making at the fusion center. By moving the decision making process to the local end devices, the amount of data transmission can be significantly reduced, especially when normal signals with positive decisions dominate in the whole life cycle and the fusion center is only interested in collecting the abnormal data. The proposed concept combines compressive sensing, machine learning, data transmission, and joint decision making. The sensor readings are encoded and transmitted to the fusion center only when abnormal signals with negative decisions are detected. All the abnormal signals from the end devices are gathered at the fusion center for a joint decision with feedback messages forwarded to the local actuators. The advantage of such an approach lies in that it can significantly reduce the volume of data to be transmitted through wireless links. Moreover, the introduction of compressive sensing can further reduce the dimension of data tremendously. An exemplary case, i.e., diesel engine condition monitoring, is provided to validate the effectiveness and efficiency of the proposed scheme compared to the conventional ones.
Session Chair
Wei Liu (Chongqing University of Technology, P.R. China)
Session T3-S9
Device-to-Device Communication
Conference
4:00 PM
—
5:30 PM KST
Local
May 26 Tue, 12:00 AM
—
1:30 AM PDT
Distributed Deep Learning Power Allocation for D2D network Based on Outdated Information
Jiaqi Shi, Qianqian Zhang, Ying-Chang Liang and Xiaojun Yuan (University of Electronic Science and Technology of China, China)
2
In the overlay D2D networks, multiple D2D pairs coexist with full frequency reuse resulting in complicated interference. Traditional centralized power control methods require instantaneous interference information and so are difficult to implement in a D2D network due to the backhaul delay and high computational requirements. To overcome this challenge, we propose a distributed power allocation algorithm called interference feature extractor aided recurrent neural network (IFE-RNN). The core idea of the scheme is described as follows. First, we design linear filters with various sizes termed IFEs to extract the local interference patterns from the outdated interference information. This feature extraction process enables our network to precisely learn the interference patterns around D2D links, and so as to provide more effective power allocation strategies. Then, we propose to predict the real-time interference pattern based on the outputs of the IFEs and further make power decision. The prediction and decision can be modelled as a Markov decision problem (MDP) and solved by using a recurrent neural network (RNN). The acquisition of the channel correlation can greatly improve the efficiency and the accuracy of our network according to our simulation results. It is worth noting that an input reduction process is also designed to reduce the space complexity from O(N2 ) to O(1) which speeds up the operation time and reduces the system overhead. Finally, extensive simulation results show that the proposed algorithm achieves an encouraging performance compared to the state- of-the-art power allocation algorithm.
Efficient Load Rearrangement of Small Cells with D2D Relay for Energy Saving and QoS Support
You-Chiun Wang (National Sun Yat-Sen University, Taiwan); Zong-Han Lin (National Sun Yat-sen University, Taiwan)
0
To satisfy the growing demand of wireless access, a mass of small cells are deployed in the service area to intensify signal quality and team up with macrocells. However, it spends lots of energy to keep the operation of small cells, which collides with the goal of green communications. In the paper, we propose an efficient load sharing (ELS) scheme to conquer this problem by transferring services of user equipments (UEs) among different cells. For each small cell in the off-peak period, its serving UEs will be adaptively taken over by other cells through handover or user-to-network relay (i.e., D2D relay). Thus, its base station can switch to the sleep mode and save energy. The above mechanism is also applied to the small cells whose base stations are overloaded for mitigating congestion. Simulation results show that ELS raises energy efficiency, curtails energy expense of the base stations in small cells, and provides better QoS support for UEs.
Intelligent Bluetooth Device to Device Connection Shift
Praneeth Juturu, Veerabhadrappa Chilakanti and Gowtham Babu (Samsung R&D Institute Bangalore, India)
0
Bluetooth Low Energy (BLE) is a wireless technology designed to provide reduced power consumption while maintaining similar communication range as compared to Classic Bluetooth [1]. Bluetooth (BT) and BLE are predominantly available in smartphones and are very useful for device-to-device communication. Most Bluetooth accessory (BTA) devices like BT headset, BT speaker and BT car-kit support only single connection. This behavior of BTA devices will hinder today's expectation of seamless continuity across multiple devices. User will encounter a connection failure, when user tries to connect a Smartphone to BTA device, which is already connected to other device. However, the reported reason for connection failure is "connection failed to be established" and is different from the exact reason, which is, "remote device is already connected". This is because BT device cannot differentiate between non-availability and the case where device is not responding. This will leave the user confused on which steps can ensure a successful connection. Device to device (D2D) communication can solve this connection issue and avoid confusion. To the best of our knowledge, this paper for the first time proposes a new concept that uses BLE's advertisement and scanning features to solve this connection issue by providing a connection shift from an already connected device, to the device that is trying to connect with the same BTA device. We have developed a prototype of our concept to present the results of connection shift and to provide analysis on difference in the time taken for connection shift during multiple attempts.
Contextual Multi-armed Bandit Based Pricing Scheme for Cooperative D2D Communications
Yue Meng and Zhaowei Wang (China Academy of Electronics and Information Technology (CAEIT), China)
0
Device-to-device (D2D) is a promising technology in 5G to improve network throughput, spectrum and energy efficiency, and transmission delay. It brings a new communication paradigm, cooperative D2D communication, which allows a user equipment (UE) to relay data for other UEs to the base station (BS). A relay UE consumes computing and communication resources, so that it should be compensated by setting price for the relay. In this paper, we investigate the pricing of relay UEs by modeling it with contextual multi-armed bandit (C-MAB). An online algorithm is designed to solve the C-MAB problem. Theoretical analysis shows that the algorithm converges to the optimal strategy asymptotically, and simulation results show that the C-MAB based pricing scheme results in both higher rewards for relay UEs and higher channel rate for the source UE.
D2D Assisted Beamforming for Coded Caching
Hamidreza Bakhshzad Mahmoodi (University of Oulu, Finland); Jarkko Kaleva (Solmu Technologies, Finland); Pooya Shariatpanahi (Institute for Research in Fundamental Sciences (IPM), Iran); Antti Tölli (University of Oulu, Finland)
0
Device-to-device (D2D) aided beamforming for coded caching is considered in a finite signal-to-noise ratio regime. A novel beamforming and resource allocation scheme is proposed where the local cache content exchange among nearby users is exploited. The transmission is split into two phases: local D2D content exchange and downlink transmission. In the D2D phase, users can autonomously share content with the adjacent users. The downlink phase utilizes multicast beamforming to simultaneously serve all users to fulfill the remaining content requests. A low complexity D2D-multicast mode selection algorithm is proposed with comparable performance to the optimal exhaustive search. We first explain the main procedure via one simple example and then present the general formulation. Furthermore, D2D transmission scenarios and conditions useful for minimizing the overall delivery time are identified. By exploiting the direct D2D exchange of file fragments, the common multicasting rate for delivering the remaining file fragments in the downlink phase is increased, providing greatly enhanced overall content delivery performance.
Session Chair
You-Chiun Wang (National Sun Yat-Sen University, Taiwan)