Prediction-based Scheme Research Articles

Federated Visualization: A Privacy-preserving Strategy for Aggregated Visual Query.

We present a novel privacy preservation strategy for aggregated visual query of decentralized data. The key idea is to imitate the flowchart of the federated learning framework, and reformulate the visualization process within a federated infrastructure. The federation of visualization is fulfilled by leveraging a shared global module that composes the encrypted externalizations of transformed visual features of data pieces in local modules. We design two implementations of federated visualization: a prediction-based scheme, and a query-based scheme. We demonstrate the effectiveness of our approach with a set of visual forms, and verify its robustness with evaluations. We report the value of federated visualization in real scenarios with an expert review.

Deep Transfer Learning-Enabled Energy Management Strategy for Smart Home Sensor Networks

The applications of wireless sensor networks are extensively used to detect and control home residents’ activities in smart homes. However, the sensors are battery-powered, so keeping them in active mode consumes tremendous energy. In this regard, we propose a solution to activate the smart home sensors based on detecting the upcoming activities using a Deep Long-Short Term Memory (DLSTM) model. The pre-trained model is then transferred to the same and different Target Domains (TDs) to reduce the time for training. The proposed system applies to preprocess and feature mapping steps to both the source and target data to make grounds for efficient transfer. Further, applying the trained model to the TD may miss the essential activities. Therefore, a reinforcement learning model is applied in the TD. To handle unusual activities in real-time, guard sensors are appointed among the idle sensors. The performance evaluation shows that the proposed scheme detects the activities with an accuracy of 96.1%. Additionally, the proposed scheme outperforms the sentry and prediction-based schemes in energy consumption of the sensors and network lifetime.

Can We Improve the Information Freshness With Prediction for Cognitive IoT?

Timely status updates are significant for some critical Internet of Things applications. However, due to the transmission delay of status packets, the status information arriving at the destination is not fresh enough. Especially, when the transmission delay is large, the destination cannot know the current status of the target well. In this article, the prediction technology is introduced into status update systems to solve this problem. First, we study a prediction-based status update scheme for machine-type communications (MTCs) and use the recently proposed metric, Age of Information (AoI), to characterize the status update performance. With the scarce spectrum resources, an MTC device (MTCD), as an unlicensed user, adopts the overlay mode to reuse the spectrum resource of the primary user and sends status packets to the central controller (CC). Then, we consider prediction errors and packet decoding errors to characterize the reliability of MTC, analyze the status update process in the prediction-based scheme, and derive the closed-form expression of the average AoI. Finally, in the proposed prediction-based scheme, we analyze the tradeoff between the status update performance and energy consumption and design an optimization algorithm to improve the status update performance by adjusting the transmit power and prediction horizon of the MTCD. Simulation results verify the correctness of the theoretical analysis and show that the proposed scheme can help the MTCD effectively improve the status update performance.

Modeling of Intelligent Sensor Duty Cycling for Smart Home Automation

The advancement of wireless sensor networks (WSNs) improves various smart home automation services and home users’ living standards. However, efficiently collecting data and automating smart home services require the extensive deployment of the sensors. Thus, one of the crucial and challenging tasks is to minimize the sensors’ energy consumption for monitoring and automating various activities in a smart home. In this article, we present a solution to control the excessive energy consumption of sensors used to detect various activities of daily living (ADL) of a smart home resident. The sensors within a smart home network are divided into various groups employing the recurrent neural network (RNN) and dynamic time warping (DTW) techniques to predict the activities with high accuracy and less energy consumption. The smart home users’ future activities are forecast with bidirectional long short-term memory (BLSTM) RNN model to select those sensors that are likely to predict the upcoming activities. Similarly, to predict the home users’ unusual activities, a guard sensor is elected among sensors with high similarities with each other using DTW. The sensor’s role is evenly switched between different modes to maintain a fair tradeoff between energy and accuracy. An extensive set of simulations is performed to validate the proposed scheme’s work integrating datasets from authentic sources. Finally, the proposed system significantly reduces the sensors’ energy consumption and prolongs the battery lifetime to approximately 137 days. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —This article presented an energy-efficient duty-cycling scheme for automating sensors’ operations in a smart home scenario. The traditional duty-cycling schemes mainly provide solutions based on appointing sentries or predicting smart home users’ next activity using models, such as Bayesian networks. We design a system that integrates the advantages of both sentry and prediction-based schemes to reduce the amount of energy required by sensors to detect and automate smart home users’ activities with high accuracy and precision. The active sensors are appointed using a bidirectional long short-term memory recurrent neural network. Similarly, the guard sensors are assigned to detect unusual activities using the similarities among idle sensors. This study could be used to automate the smart home sensors for detecting home user’s activities with less energy, which ultimately prolongs the battery lifetime of the sensors.

Low-Energy Clustering Protocol for Query-Based Wireless Sensor Networks

Nowadays, clustering protocols have been widely used to reduce energy consumption in wireless sensor networks (WSNs). However, when existing clustering protocols are used for query-based WSNs, the energy efficiency becomes limited because in such networks, clusters are formed without considering the positional distribution of query targets, and the procedures of node clustering, cluster head selection and inter-cluster routing are performed in a distributed way, which is energy-consuming. In this paper, a centralized clustering protocol, called Low-Energy Dynamic Clustering (LEDC), is developed to enhance the energy efficiency in query-based WSNs. The main contributions of LEDC include: (1) a novel prediction-based scheme is raised to make the centralized network information maintenance energy-efficient and scalable; (2) centralized optimal dynamic cluster formation algorithms are proposed to cluster the query targets in a centralized way, which not only helps the distribution of clusters match with that of query targets but also saves the high cost incurred in the distributed procedures of node clustering, cluster head selection and inter-cluster routing; (3) an energy-aware forwarding strategy is suggested to improve the energy efficiency and balance the energy load among nodes in inter-cluster communication; (4) a parallel processing algorithm is raised to increase the network throughput and reduce the latency of queries. The performance of LEDC is then compared to the latest clustering protocol. Simulation results demonstrate that LEDC considerably saves the energy of nodes and prolongs the network lifetime.

An adaptive routing protocol in flying ad hoc networks

Nowadays, ad hoc networks play a vital role in wireless communication. It is a temporary network to relay the data between nodes. Ad hoc network is divided into different types such as MANET, VANET & FANET. FANET is a combination of MANET and VANET. In FANET, routing is a big challenging issue due to unique characteristics like dynamic topology, frequent changes of link quality, mobility of UAV nodes, etc. Reliability of routes is also a real challenge due to the very high mobility in FANET. In this research work, we have introduced an adaptive routing protocol, PF-WGTR - A predicted future weight-based routing scheme for FANETs that considers the node’s existing and upcoming values of assured parameters to determine the reliable routing path. This proposed routing protocol assigns weight to every node in the network by calculating the node’s future. Depending on the predicted future weight of every node in a network, the communicating nodes can establish a reliable path that persists for a long time. This adaptive, future prediction-based routing scheme ensures better data delivery with minimum overhead and optimized energy consumption in all conditions compared with the existing routing protocols. The NS-2 simulator is used to compare the proposed routing protocol with previous protocols in terms of FANET parameters. Finally, the simulation results show better performance than the existing works.

Task-Offloading Strategy Based on Performance Prediction in Vehicular Edge Computing

In vehicular edge computing, network performance and computing resources dynamically change, and vehicles should find the optimal strategy for offloading their tasks to servers to achieve a rapid computing service. In this paper, we address the multi-layered vehicle edge-computing framework, where each vehicle can choose one of three strategies for task offloading. For the best offloading performance, we propose a prediction-based task-offloading scheme for the vehicles, in which a deep-learning model is designed to predict the task-offloading result (success/failure) and service delay, and then the predicted strategy with successful task offloading and minimum service delay is chosen as the final offloading strategy. In the proposed model, an automatic feature-generation model based on CNN is proposed to capture the intersection of features to generate new features, avoiding the performance instability caused by manually designed features. The simulation results demonstrate that each part of the proposed model has an important impact on the prediction accuracy, and the proposed scheme has the higher Area Under Curve (AUC) than other methods. Compared with SVM- and MLP-based methods, the proposed scheme has the average failure rate decreased by 21.2% and 6.3%, respectively. It can be seen that our prediction-based scheme can effectively deal with dynamic changes in network performance and computing resources.

A named data networking prediction-based mobility solution in space–air–terrestrial networks

With the rapid development of satellite constellations and mobile devices, multimedia-content transfer in space–air–terrestrial networks (SATNs) has become increasingly popular. Governed by the current internet protocol-based architecture and the distance between the space and the ground, real-time transmissions of multimedia encounter an enormous challenge. Recently, named data networking has been considered to be a promising approach for solving this problem in SATN scenarios. However, a new issue, producer and consumer mobility management, arises in this approach. In this paper, we propose a location prediction-based scheme to reduce the round-trip time during the handover process and further enhance the system throughput, as well as to enable the producer and consumer to simultaneously move from an old access point (AP) to a new AP. Moreover, the proposed scheme can be applied in two kinds of SATN scenarios. Finally, the effectiveness of the proposed scheme is verified by simulations.

State estimation via prediction-based scheme for linear time-varying uncertain networks with communication transmission delays and stochastic coupling

In this paper, the design problem of recursive state estimation algorithm based on delay-prediction compensation is considered for a class of linear time-varying uncertain dynamical networks with network-induced communication transmission delays and stochastic coupling phenomenon. The random variables obeying Bernoulli distribution under uncertain occurrence probabilities are selected to depict the stochastic coupling phenomenon. Moreover, the modelling uncertainties and communication transmission delays are taken into account. Afterwards, a feasible prediction equation is constructed to compensate the communication transmission delays, and an updated time-varying state estimator is designed. Subsequently, the upper bound expression of the state estimation error covariance matrix is given by using the stochastic analysis technique and matrix theory, and the gain matrix of time-varying state estimator is obtained via minimizing the trace of such an upper bound. Furthermore, we present the mathematical proof which illustrates the monotonic relationship between the occurrence probabilities of stochastic coupling and the upper bound of error covariance. Finally, a simulation example is utilized to demonstrate the effectiveness of the proposed estimation strategy based on the delay prediction idea.

Prediction-based control for a class of unstable time-delayed processes by using a modified sequential predictor

This work presents a prediction-based scheme to control a class of unstable delayed (bio)-chemical processes. The main characteristic of the control scheme is that it is based on a predicted future state that is estimated using a prediction-observation protocol that allows compensating large time-delays. The convergence of the prediction error is shown through standard stability arguments. The stability of the closed-loop system is shown by invoking the separation principle between the prediction error and the control based on estimated variables feedback. Four numerical examples are used to illustrate our results, including two time-delayed benchmark case studies taken from the control process literature.

Bidirectional Charging in V2G Systems: An In-Cell Variation Analysis of Vehicle Batteries

Vehicle-to-grid (V2G) technology enables bidirectional charging of electric vehicle (EV) and facilitates power grid ancillary services. However, battery pack in EV may develop in-cell dynamic variations over time. This is due to the structural complexity and electrochemical operations in the battery pack. These variations may arise in V2G systems due to: first, additional charging and discharging cycles to power grid; second, external shocks; and third, long exposures to high temperatures. A particular source of these variations is due to faulty sensors. Therefore, it can be argued that the battery packs in EV are highly reliant on the monitoring of these in-cell variations and their impact of propagation with each involved component. In this article, a prediction-based scheme to monitor the health of variation induced sensors is proposed. First, a propagation model is developed to predict the in-cell variations of a battery pack by calculating the covariance using a median-based expectation. Second, a hypothesis model is developed to detect and isolate each variation. This is obtained by deriving a conditional probability-based density function for the measurements. The proposed monitoring framework is evaluated using experimental measurements collected from Li-ion battery pack in EVs. The in-cell variation profiles have been verified using D-SAT Chroma 8000ATS hardware platform. The performance results of the proposed scheme show accurate analysis of these emerged variations.

Continuous object tracking protocol with selective wakeup based on practical boundary prediction in wireless sensor networks

In wireless sensor networks, the selective wakeup approach to activate minimum sensor nodes for object tracking is considered as an attractive way for efficient energy consumption. Different from individual tracking, continuous object tracking is necessary to maintain only sensor nodes on the boundary of continuous objects in the active mode. Recently, a prediction-based scheme, named PRECO, is proposed to energy-efficiently track a continuous object using a selective wakeup scheme. It continuously predicts the next boundary position of a continuous object through mathematical computing and activates sleeping sensor nodes on the next boundary to detect the object in advance. However, PRECO has critical problems for practical implementation in terms of three domains: space, quantity, and time domains. We propose a new continuous object tracking protocol that practically predicts and accurately detects a continuous object at the right time by solving the problems in the domains. To avoid high complex prediction in the space domain, a virtual cell-based prediction scheme is applied to estimate next diffusing areas of continuous object. To reduce the number of nodes for the prediction participation in the quantity domain, our protocol requests only cell heads for estimating the diffusing area of the object and announcing the decision. To remove synchronous prediction in the time domain, each cell head asynchronously predicts the next diffusing area of the object without collaboration with other cell heads. Simulation results conducted in various environments verifies that our protocol is superior to PRECO in terms of energy efficiency and prediction accuracy.

Anomaly Detection in UASN Localization Based on Time Series Analysis and Fuzzy Logic

Underwater acoustic sensor network (UASN) offers a promising solution for exploring underwater resources remotely. For getting a better understanding of sensed data, accurate localization is essential. As the UASN acoustic channel is open and the environment is hostile, the risk of malicious activities is very high, particularly in time-critical military applications. Since the location estimation with false data ends up in wrong positioning, it is necessary to identify and ignore such data to ensure data integrity. Therefore, in this paper, we propose a novel anomaly detection system for UASN localization. To minimize computational power and storage, we designed separate anomaly detection schemes for sensor nodes and anchor nodes. We propose an auto-regressive prediction-based scheme for detecting anomalies at sensor nodes. For anchor nodes, a fuzzy inference system is designed to identify the presence of anomalous behavior. The detection schemes are implemented at every node for enabling identification of multiple and duplicate anomalies at its origin. We simulated the network, modeled anomalies and analyzed the performance of detection schemes at anchor nodes and sensor nodes. The results indicate that anomaly detection systems offer an acceptable accuracy with high true positive rate and F-Score.

Distinguishable brain networks relate disease susceptibility to symptom expression in schizophrenia.

Disease association studies have characterized altered resting-state functional connectivities describing schizophrenia, but failed to model symptom expression well. We developed a model that could account for symptom severity and meanwhile relate this to disease-related functional pathology. We correlated BOLD signal across brain regions and tested separately for associations with disease (disease edges) and with symptom severity (symptom edges) in a prediction-based scheme. We then integrated them in an "edge bi-color" graph, and adopted mediation analysis to test for causality between the disease and symptom networks and symptom scores. For first-episode schizophrenics (FES, 161 drug-naïve patients and 150 controls), the disease network (with inferior frontal gyrus being the hub) and the symptom-network (posterior occipital-parietal cortex being the hub) were found to overlap in the temporal lobe. For chronic schizophrenis (CS, 69 medicated patients and 62 controls), disease network was dominated by thalamocortical connectivities, and overlapped with symptom network in the middle frontal gyrus. We found that symptom network mediates the relationship between disease network and symptom scores in FEP, but was unable to define a relationship between them for the smaller CS population. Our results suggest that the disease network distinguishing core functional pathology in resting-state brain may be responsible for symptom expression in FES through a wider brain network associated with core symptoms. We hypothesize that top-down control from heteromodal prefrontal cortex to posterior transmodal cortex contributes to positive symptoms of schizophrenia. Our work also suggests differences in mechanisms of symptom expression between FES and CS, highlighting a need to distinguish between these groups.

A prediction-based handoff scheme for QoS in WLAN systems

A prediction-based handoff scheme for QoS in WLAN systems

A prediction-based handoff scheme for QoS in WLAN systems

In this paper, we introduce a technique to reduce the authentication overhead for handoff process in wireless local area networks (WLAN). It is based on user mobility prediction to reduce the network scanning area during the handoff procedure. The central server facilitates an intelligent mobility prediction mechanism to identify the next possible mobile access points for mobile subscriber. The predicted information is managed by an event log table, and it helps to reduce the authorisation overheads for cellular networks. It follows the reserve channel scheme to maintain the QoS for users during the call and handoff procedure and the handover-call has higher priority over local initiated call. The analytical model shows the effectiveness and efficiency of proposed work, we apply the Markov random walk model and M/M/1 queuing model to formulate our work. It is compared with the tradition technique, and has 40% better result than legacy techniques.

Prediction-Based Control for Linear Systems with Input- and State- Delay – Robustness to Delay Mismatch

This paper addresses the design of a robust prediction-based controller for linear systems with both input and state delays. We extend the usual prediction-based scheme to state delay and prove its robustness to sufficiently small delay mismatches. Our approach is grounded on the linking of two recently proposed infinite-dimensional techniques: a Complete-Type Lyapunov functional, which enables state delay systems stability analysis, and tools from the field of Partial Differential Equations, reformulating the delays as transport equations and introducing a tailored backstepping transformation. We illustrate the merits of the proposed technique with simulations on a process dryer system.

Space-filling curves applied to compression of ultraspectral images

Space-filling curves (SFCs) have been widely used in image compression as a way to assist prediction-based schemes by providing a method to map an $$m$$ -dimensional space into a highly correlated unidimensional space. In this paper, we not only analyze the effect of different state-of-the-art curves but also propose a novel scanning procedure applied to the preprocessing stage of an ultraspectral lossy compression architecture. Specifically, we focus on Atmospheric infrared sounder images that are good compression candidates as they account for well over 25 megapixels of information per cube. In addition, we introduce a metric to estimate the efficiency of the different SFCs and present a mathematical model that, based on the Laplacian nature of the pixels, it is used to derive bounds. Finally, to verify the accuracy of this metric, we obtain rate-distortion ratios calculated by applying the corresponding curve to the compression architecture.

A Novel Prediction-Based Location Management Technique for Mobile Networks

In this paper, the authors have proposed a new prediction-based location management scheme for cellular networks. The current mobile switching centre predicts the previous mobile switching center of user, and then imports the data about the mobile terminal from that. For this purpose it uses a prediction table, which is based on the neighboring location areas. The analytical model and formulation presents the conceptual schema of proposed scheme. The performance comparisons and numerical analysis is helpful to find out the efficiency and effectiveness of the proposed system, with respect to existing location management techniques. It has been observed that, the registration cost, and the network overhead at mobile switching center have been reduced in the proposed prediction-based scheme.

Network Performance Optimization in Constrained Queueing Systems

Most of literature assumed infinite buffer for users, which, however, is not practical in real networks. In this paper, we investigate downlink resource scheduling with constrained queueing. We first formulate an optimization model with the objective of maximizing the system rate under a limited queue length. Then two scheduling methods are proposed to solve this problem. One scheme is based on a virtual alarming threshold; the other is a prediction-based scheme. However, they still suffer from unfairness among rate allocations. To improve performance with respect to rate fairness, the factor of rate fairness is introduced into this optimization formulation finally. Numerical results are presented to demonstrate the efficiency of the proposed scheduling methods in terms of average system rate, maximum queue length and rate fairness, compared to some existing methods.