Search

In recent years, both the scientific community and the industry have focused on moving computational resources with remote data centres from the centralized cloud to decentralised computing, making them closer to the source or the so called "edge" of the network. This is due to the fact that the cloud system alone cannot sufficiently support the huge demands of future networks with the massive growth of new, time-critical applications such as self-driving vehicles, Augmented Reality/Virtual Reality techniques, advanced robotics and critical remote control of smart Internet-of-Things applications. While decentralised edge computing will form the backbone of future heterogeneous networks, it still remains at its infancy stage. Currently, there is no comprehensive platform. In this article, we propose a novel decentralised edge architecture, a solution called OMNIBUS, which enables a continuous distribution of computational capacity for end-devices in different localities by exploiting moving vehicles as storage and computation resources. Scalability and adaptability are the main features that differentiate the proposed solution from existing edge computing models. The proposed solution has the potential to scale infinitely, which will lead to a significant increase in network speed. The OMNIBUS solution rests on developing two predictive models: (i) to learn timing and direction of vehicular movements to ascertain computational capacity for a given locale, and (ii) to introduce a theoretical framework for sequential to parallel conversion in learning, optimisation and caching under contingent circumstances due to vehicles in motion., Universiti Teknologi Malaysia; Research University Grant Scheme Tier 2; Ambeent Inc.

Many of the machine learning tasks rely on centralized learning (CL), which requires the transmission of local datasets from the clients to a parameter server (PS) entailing huge communication overhead. To overcome this, federated learning (FL) has been suggested as a promising tool, wherein the clients send only the model updates to the PS instead of the whole dataset. However, FL demands powerful computational resources from the clients. In practice, not all the clients have sufficient computational resources to participate in training. To address this common scenario, we propose a more efficient approach called hybrid federated and centralized learning (HFCL), wherein only the clients with sufficient resources employ FL, while the remaining ones send their datasets to the PS, which computes the model on behalf of them. Then, the model parameters are aggregated at the PS. To improve the efficiency of dataset transmission, we propose two different techniques: i) increased computation-per-client and ii) sequential data transmission. Notably, the HFCL frameworks outperform FL with up to 20% improvement in the learning accuracy when only half of the clients perform FL while having 50% less communication overhead than CL since all the clients collaborate on the learning process with their datasets., European Union (EU); Horizon 2020; European Research Council (ERC); Project AGNOSTIC; CHIST-ERA; Scientific and Technological Council of Turkey (TÜBİTAK)

Proper determination of the transmission rate in ultra-reliable low latency communication (URLLC) needs to incorporate a confidence interval (CI) for the estimated parameters due to the large amount of data required for their accurate estimation. In this paper, we propose a framework based on the extreme value theory (EVT) for determining the transmission rate along with its corresponding CI for an ultra-reliable communication system. This framework consists of characterizing the statistics of extreme events by fitting the generalized Pareto distribution (GPD) to the channel tail, deriving the GPD parameters and their associated CIs, and obtaining the transmission rate within a confidence interval. Based on the data collected within the engine compartment of Fiat Linea, we demonstrate the accuracy of the estimated rate obtained through the EVT-based framework considering the confidence interval for the GPD parameters. Additionally, we show that proper estimation of the transmission rate based on the proposed framework requires a lower number of samples compared to the traditional extrapolation-based approaches., NA

Vehicular visible light communications (V-VLC) is a promising intelligent transportation systems (ITS) technology for vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communications with the utilization of light emitting diodes (LEDs). The main degrading factor for the performance of V-VLC systems is noise. Unlike traditional radio frequency (RF) based systems, V-VLC systems include many noise sources: solar radiation, background lighting from vehicle, street, parking garage and tunnel lights. Traditional V-VLC system noise modeling is based on the additive white Gaussian noise assumption in the form of shot and thermal noise. In this paper, to investigate both time correlated and white noise components of the V-VLC channel, we propose a noise analysis based on Allan variance (AVAR), which provides a time-series analysis method to identify noise from the data. We also propose a generalized Wiener process based V-VLC channel noise synthesis methodology to generate different noise components. We further propose convolutional autoencoder (CAE) based denoising scheme to reduce V-VLC signal noise, which achieves reconstruction root mean square error (RMSE) of 0.0442 and 0.0474 for indoor and outdoor channels, respectively., European Union (EU); Horizon2020; CHIST-ERA; Scientific and Technological Research Council of Turkey (TÜBİTAK); Ford Otosan

Wireless powered communication networks (WPCNs) will act as a major enabler of massive machine type communications (MTCs), which is a major service domain for 5G and beyond systems. The MTC networks will be deployed by using low-power transceivers with finite discrete configurations. This paper considers minimum length scheduling problem for full-duplex WPCNs, where users transmit information to a hybrid access point at a rate chosen from a finite set of discrete-rate levels. The optimization problem considers energy causality, data and maximum transmit power constraints, and is proven to be NP-hard. As a solution strategy, we define the minimum length scheduling (MLS) slot, which is slot of minimum transmission completion time while starting transmission at anytime after the decision time. We solve the problem optimally for a given transmission order based on the optimality analysis of MLS slot. For the general problem, we categorize the problem based on whether the MLS slots of users overlap over time. We propose optimal algorithm for non-overlapping scenario by allocating the MLS slots, and a polynomial-time heuristic algorithm for overlapping scenario by allocating the transmission slot to the user with earliest MLS slot. Through simulations, we demonstrate significant gains of scheduling and discrete rate allocation., Scientific and Technological Research Council of Turkey (TÜBİTAK)

A key building block in the design of ultra-reliable communication systems is a wireless channel model that captures the statistics of rare events occurring due to the significant fading. In this paper, we propose a novel methodology based on extreme value theory (EVT) to statistically model the behavior of extreme events in a wireless channel for ultra-reliable communication. This methodology includes techniques for fitting the lower tail distribution of the received power to the generalized Pareto distribution (GPD), determining the optimum threshold over which the tail statistics are derived, ascertaining the optimum stopping condition on the number of samples required to estimate the tail statistics by using GPD, and finally, assessing the validity of the derived Pareto model. Based on the data collected within the engine compartment of Fiat Linea under various engine vibrations and driving scenarios, we demonstrate that the proposed methodology provides the best fit to the collected data, significantly outperforming the conventional extrapolation-based methods. Moreover, the usage of the EVT in the proposed method decreases the required number of samples for estimating the tail statistics significantly., Ford Otoson

A proper channel modeling methodology that characterizes the statistics of extreme events is key in the design of a system at an ultra-reliable regime of operation. The strict constraint of ultra-reliability corresponds to the packet error rate (PER) in the range of 10(-9)-10(-5) within the acceptable latency on the order of milliseconds. Extreme value theory (EVT) is a robust framework for modeling the statistical behavior of extreme events in the channel data. In this paper, we propose a methodology based on EVT to model the extreme events of a non-stationary wireless channel for the ultra-reliable regime of operation. This methodology includes techniques for splitting the channel data sequence into multiple groups concerning the environmental factors causing non-stationarity, and fitting the lower tail distribution of the received power in each group to the generalized Pareto distribution (GPD). The proposed approach also consists of optimally determining the time-varying threshold over which the tail statistics are derived as a function of time, and assessing the validity of the derived Pareto model. Finally, the proposed approach chooses the best model with minimum complexity that represents the time variation behavior of the non-stationary channel data sequence. Based on the data collected within the engine compartment of Fiat Linea under various engine vibrations and driving scenarios, we demonstrate the capability of the proposed methodology in providing the best fit to the extremes of the non-stationary data. The proposed approach significantly outperforms the channel modeling approach using the stationary channel assumption in characterizing the extreme events.

ARINC 664 is an Ethernet based deterministic networking standard providing data transmission with bounded delays among avionics sub-systems. This paper presents a Genetic Algorithm (GA) based ARINC 664 network delay optimization using the network calculus (NC), where the GA is used to effectively search the mapping of Virtual Links (VLs) to priority levels using the extended priority scheme. While there are only two priority levels in the ARINC 664 standard, the extended priority concept increases the number of priority levels to improve the schedulability of VLs. For each possible assignment of the VLs to the priority levels, the NC analysis provides the worst-case delay results for all VLs. We define three different fitness functions aiming to minimize the maximum, the average, and the standard deviation of the worst-case VL delays, respectively. The results demonstrate that the extended priority concept improves the schedulability of VLs and the GA optimization approach can successfully achieve the desired objectives for the VL delays if the appropriate cost function is selected.

Minimization of schedule length is key in ensuring the delay performance of wireless powered communication networks (WPCNs) demanding strict timing and reliability guarantees. Previous solution methodologies proposed for these wireless networks suffer from high run-time complexity, making it very difficult to solve the problem in real time. This paper considers a run-time efficient deep learning based approach for solving minimum length scheduling problem in a full-duplex WPCN. Leveraging upon the universal approximation capability of neural networks, a multi-output feed forward deep neural network based framework is proposed where inputs are the channel coefficients and outputs are the optimal power, transmission length and schedule of users. Simulation results indicate that the proposed deep learning based approach can very well approximate the true outputs with a percentage error below 1% for different network configurations while maintaining a very low run-time complexity., Scientific and Technological Research Council of Turkey (TÜBİTAK); Ford Otosan

Many of the machine learning tasks are focused on centralized learning (CL), which requires the transmission of local datasets from the clients to a parameter server (PS) leading to a huge communication overhead. Federated learning (FL) overcomes this issue by allowing the clients to send only the model updates to the PS instead of the whole dataset. In this way, FL brings the learning to edge level, wherein powerful computational resources are required on the client side. This requirement may not always be satisfied because of diverse computational capabilities of edge devices. We address this through a novel hybrid federated and centralized learning (HFCL) framework to effectively train a learning model by exploiting the computational capability of the clients. In HFCL, only the clients who have sufficient resources employ FL; the remaining clients resort to CL by transmitting their local dataset to PS. This allows all the clients to collaborate on the learning process regardless of their computational resources. We also propose a sequential data transmission approach with HFCL (HFCL-SDT) to reduce the training duration. The proposed HFCL frameworks outperform previously proposed non-hybrid FL (CL) based schemes in terms of learning accuracy (communication overhead) since all the clients collaborate on the learning process with their datasets regardless of their computational resources., European Union (EU); CHIST-ERA; Scientific and Technological Research Council of Turkey (TÜBİTAK)

The high cost and power consumption of digital beamforming, as a result of the high number of RF chains, has overshadowed its performance on multi-antenna wireless powered communication networks (WPCNs). This setback forced researchers to low-cost alternatives such as hybrid beamforming, which decreases the number of expensive RF chains by utilizing cheaper phase shifters. This cost-cutting, however, comes with reduced control over beamforming weights and compromise performance. To circumvent this deficiency, scheduling of energy harvesting (SEH), utilizing the degree of freedom in the time domain, has been proposed. In SEH, the downlink slot is subdivided into multiple variable-length subslots with different beamforming weights. In this paper, we examine the effect of SEH on the optimization of minimum length scheduling for space division multiple access (SDMA) uplink transmission compared to time division multiple access (TDMA) uplink transmission. Via simulations, we demonstrate that SDMA benefits more from the additional degree of freedom provided by the usage of SEH for any number of nodes. However, SDMA yields inferior delay performance compared to TDMA as the number of nodes increases, which restricts the application of SDMA with SEH, making it impractical., Scientific and Technological Research Council of Turkey (TÜBİTAK)

Millimeter wave multiple-input multiple-output (mmWave-MIMO) systems with small number of radio-frequency (RF) chains have limited multiplexing gain. Spatial path index modulation (SPIM) is helpful in improving this gain by utilizing additional signal bits modulated by the indices of spatial paths. In this paper, we introduce model-based and model-free frameworks for beamformer design in multi-user SPIM-MIMO systems. We first design the beamformers via model-based manifold optimization algorithm. Then, we leverage federated learning (FL) with dropout learning (DL) to train a learning model on the local dataset of users, who estimate the beamformers by feeding the model with their channel data. The DL randomly selects different set of model parameters during training, thereby further reducing the transmission overhead compared to conventional FL. Numerical experiments show that the proposed framework exhibits higher spectral efficiency than the state-of-the-art SPIM-MIMO methods and mmWave-MIMO, which relies on the strongest propagation path. Furthermore, the proposed FL approach provides at least 10 times lower transmission overhead than the centralized learning techniques., Scientific and Technological Research Council of Turkey (TÜBİTAK); European Union (EU); CHIST-ERA

A proper channel modeling methodology that characterizes the statistics of extreme events is key in the design of a system at an ultra-reliable regime of operation. The strict constraint of ultra-reliability corresponds to the packet error rate (PER) in the range of 10(-9)-10(-5) within the acceptable latency on the order of milliseconds. Extreme value theory (EVT) is a robust framework for modeling the statistical behavior of extreme events in the channel data. In this paper, we propose a methodology based on EVT to model the extreme events of a non-stationary wireless channel for the ultra-reliable regime of operation. This methodology includes techniques for splitting the channel data sequence into multiple groups concerning the environmental factors causing non-stationarity, and fitting the lower tail distribution of the received power in each group to the generalized Pareto distribution (GPD). The proposed approach also consists of optimally determining the time-varying threshold over which the tail statistics are derived as a function of time, and assessing the validity of the derived Pareto model. Finally, the proposed approach chooses the best model with minimum complexity that represents the time variation behavior of the non-stationary channel data sequence. Based on the data collected within the engine compartment of Fiat Linea under various engine vibrations and driving scenarios, we demonstrate the capability of the proposed methodology in providing the best fit to the extremes of the non-stationary data. The proposed approach significantly outperforms the channel modeling approach using the stationary channel assumption in characterizing the extreme events., Ford Otosan

ARINC 664, which is a heavily used protocol for modern avionics networks, is preferred due to its simplicity although its mixed-criticality support is limited. Time Triggered Ethernet (TTEthernet), and IEEE Time Sensitive Networking (TSN), which utilize time synchronized schedule, are more suitable for supporting mixed-criticality applications; however, both require a fault tolerant time synchronization that makes the certification process more challenging. In this paper, we propose a novel dynamic priority assignment (DPA) concept together with the burst limiting shaper (BLS) from the IEEE TSN standard to enhance the schedulability and the mixed-criticality support of ARINC 664. The decision of flow re-assignment to a new priority class is done by calculating the high priority (HP) and low priority (LP) class worst-case delays using the network calculus framework. The numerical results show that the class utilization rates can be significantly increased by using the DPA concept with and without the BLS while the deadline constraints for all classes are satisfied. Thus, the DPA can improve the schedulability and mixed-criticality of ARINC 664 without using any time synchronization mechanism., NA

ARINC 664 is an Ethernet based deterministic networking standard providing data transmission with bounded delays among avionics sub-systems. This paper presents a Genetic Algorithm (GA) based ARINC 664 network delay optimization using the network calculus (NC), where the GA is used to effectively search the mapping of Virtual Links (VLs) to priority levels using the extended priority scheme. While there are only two priority levels in the ARINC 664 standard, the extended priority concept increases the number of priority levels to improve the schedulability of VLs. For each possible assignment of the VLs to the priority levels, the NC analysis provides the worst-case delay results for all VLs. We define three different fitness functions aiming to minimize the maximum, the average, and the standard deviation of the worst-case VL delays, respectively. The results demonstrate that the extended priority concept improves the schedulability of VLs and the GA optimization approach can successfully achieve the desired objectives for the VL delays if the appropriate cost function is selected., NA

Vehicular Visible Light Communication (VVLC) is preferred as a vehicle-to-everything (V2X) communications scheme due to its highly secure, low complexity and radio frequency (RF) interference free characteristics, exploiting the line-of-sight (LoS) propagation of visible light and usage of already existing vehicle light emitting diodes (LEDs). Current VVLC channel models based on deterministic and stochastic methods provide limited accuracy for path loss prediction since deterministic methods heavily depend on site-specific geometry and stochastic models average out the model parameters without considering environmental effects. Moreover, there exists no wireless channel model that can be adopted for channel frequency response (CFR) prediction. In this paper, we propose novel framework for the machine learning (ML) based channel modeling of the VVLC with the goal of improving the model accuracy for path loss and building the CFR model through the consideration of multiple input variables related to vehicle mobility and environmental effects. The proposed framework incorporates multiple measurable input variables, e.g., distance, ambient light, receiver inclination angle, and optical turbulence, with the exploitation of feed forward neural network based multilayer perceptron neural network (MLP-NN), radial basis function neural network (RBF-NN) and decision tree based Random Forest learning methods. The framework also includes data pre-processing step, with synthetic minority over-sampling technique (SMOTE) data balancing, and hyper-parameter tuning based on iterative grid search, to further improve the accuracy. The accuracy of the proposed ML based channel modeling is demonstrated on the real-world VVLC vehicle-to-vehicle (V2V) communication channel data. The proposed MLP-NN, RBF-NN, and Random Forest based channel models generate highly accurate path loss predictions with 3.53 dB, 3.81 dB, 3.95 dB root mean square error(RMSE), whereas the best performing stochasti, European Union (EU); Horizon 2020; CHIST-ERA; Scientific and Technological Research Council of Turkey (TÜBİTAK); Ford Otosan

We consider a wireless powered, harvest-then-transmit communication network, which consists of multiple, single antenna, energy and information access points (APs) and multiple, single antenna users with energy harvesting capabilities and rechargeable batteries, and allows simultaneous information transmission. We formulate the joint power control and scheduling problem with the objective of minimizing the total schedule length, subject to the constraints on the minimum amount of data to be sent by the users to the APs, and the maximum transmit power for the information transmission. This problem is a nonlinear and non-convex, mixed integer programming problem for which there is no known polynomial time algorithm. The proposed heuristic algorithm is based on, first, finding the solution for a fixed energy harvesting time and then searching for the optimal energy harvesting time that minimizes the total schedule length. For the former, a scheduling problem is formulated as an integer programming problem, which we solve with Branch and Price based methods upon solving the power control problem separately. Simulation results demonstrate that the proposed algorithm outperforms previously proposed time minimization algorithms that do not consider simultaneous transmission scenarios up to 3:5% for larger AP power, 25:4% for tighter maximum transmit power limit, and 6:5% for greater number of users per AP.

Multi-connectivity (MC) is proposed in Fifth-Generation mobile communications systems (5G) to mitigate the deterioration of Quality-of-Service owing to line-of-sight blockage and lack of communication resources. The main idea of MC is to associate single user equipment with multiple network layers and multiple radio access technologies, simultaneously. In previous studies, MC is demonstrated to increase capacity, provide high reliability and decrease outage probability. However, for the first time in literature, in this letter, we incorporate MC into the optimization of the total power consumption in 5G Heterogeneous Cloud Radio Access Networks ( HCRAN )s. Upon formulation of the problem as a binary integer linear programming problem, proving its NP-hardness, a heuristic algorithm is proposed consisting of linear programming relaxation and rounding and generalized assignment problem heuristic, which takes the outputs of the linear programming relaxation and rounding as inputs and checks if it further minimizes the power consumption. We analyze power consumption, time complexity and achievable rate of the proposed algorithm and verify its superiority over existing methods by simulations.

Hybrid beamforming is a low-cost alternative to digital beamforming with its capability to operate with fewer radio frequency (RF) chains than antennas. However, the diminishing degree of freedom from utilizing fewer RF chains results in considerable performance degradation. To circumvent this setback, scheduling of energy harvesting (SEH) has been recently proposed to provide an additional degree of freedom. In this letter, we study the optimization of SEH together with the beamforming weights, energy and data transfer intervals, and uplink transmit power. The objective is minimizing the total duration of uplink and downlink transmission, whereas the constraints include minimum data transfer, maximum allowed transmit power and hybrid beamforming requirements. We formulate the non-convex optimization problem and convert it to the equivalent rank constrained semidefinite programming problem. We then propose efficient solution methodologies based on iteratively moving the rank constraints to the objective function as a penalty. Extensive simulations demonstrate that SEH is an effective circumvention to the RF chain scarcity, culminating in up to 26% delay performance gain.

Multi-connectivity (MC) is proposed in Fifth-Generation mobile communications systems (5G) to mitigate the deterioration of Quality-of-Service owing to line-of-sight blockage and lack of communication resources. The main idea of MC is to associate single user equipment with multiple network layers and multiple radio access technologies, simultaneously. In previous studies, MC is demonstrated to increase capacity, provide high reliability and decrease outage probability. However, for the first time in literature, in this letter, we incorporate MC into the optimization of the total power consumption in 5G Heterogeneous Cloud Radio Access Networks ( HCRAN )s. Upon formulation of the problem as a binary integer linear programming problem, proving its NP-hardness, a heuristic algorithm is proposed consisting of linear programming relaxation and rounding and generalized assignment problem heuristic, which takes the outputs of the linear programming relaxation and rounding as inputs and checks if it further minimizes the power consumption. We analyze power consumption, time complexity and achievable rate of the proposed algorithm and verify its superiority over existing methods by simulations., NA

Hybrid beamforming is a low-cost alternative to digital beamforming with its capability to operate with fewer radio frequency (RF) chains than antennas. However, the diminishing degree of freedom from utilizing fewer RF chains results in considerable performance degradation. To circumvent this setback, scheduling of energy harvesting (SEH) has been recently proposed to provide an additional degree of freedom. In this letter, we study the optimization of SEH together with the beamforming weights, energy and data transfer intervals, and uplink transmit power. The objective is minimizing the total duration of uplink and downlink transmission, whereas the constraints include minimum data transfer, maximum allowed transmit power and hybrid beamforming requirements. We formulate the non-convex optimization problem and convert it to the equivalent rank constrained semidefinite programming problem. We then propose efficient solution methodologies based on iteratively moving the rank constraints to the objective function as a penalty. Extensive simulations demonstrate that SEH is an effective circumvention to the RF chain scarcity, culminating in up to 26% delay performance gain., Scientific and Technological Research Council of Turkey (TÜBİTAK)

We consider a wireless powered, harvest-then-transmit communication network, which consists of multiple, single antenna, energy and information access points (APs) and multiple, single antenna users with energy harvesting capabilities and rechargeable batteries, and allows simultaneous information transmission. We formulate the joint power control and scheduling problem with the objective of minimizing the total schedule length, subject to the constraints on the minimum amount of data to be sent by the users to the APs, and the maximum transmit power for the information transmission. This problem is a nonlinear and non-convex, mixed integer programming problem for which there is no known polynomial time algorithm. The proposed heuristic algorithm is based on, first, finding the solution for a fixed energy harvesting time and then searching for the optimal energy harvesting time that minimizes the total schedule length. For the former, a scheduling problem is formulated as an integer programming problem, which we solve with Branch and Price based methods upon solving the power control problem separately. Simulation results demonstrate that the proposed algorithm outperforms previously proposed time minimization algorithms that do not consider simultaneous transmission scenarios up to 3:5% for larger AP power, 25:4% for tighter maximum transmit power limit, and 6:5% for greater number of users per AP., Scientific and Technological Research Council of Turkey (TÜBİTAK)

Radio frequency (RF) energy harvesting is key in attaining perpetual lifetime for time-critical wireless powered communication networks (WPCNs) due to full control on energy transfer, far field region, small and low-cost circuitry. In this paper, we propose a novel minimum length scheduling problem to determine the optimal power control, time allocation and schedule subject to data, energy causality and maximum transmit power constraints in a full-duplex WPCN. We first formulate the problem as a mixed integer non-linear programming problem and conjecture that the problem is NP-hard. As a solution strategy, we demonstrate that the power control and time allocation, and the scheduling problems can be solved separately in the optimal solution. For the power control and time allocation problem, we derive the optimal solution by evaluating Karush-Kuhn-Tucker conditions. For the scheduling, we introduce a penalty function allowing reformulation of the problem as a sum penalty minimization problem. Upon derivation of the optimality conditions based on the characteristics of the penalty function, we propose two polynomial-time heuristic algorithms and a reduced-complexity exact algorithm employing smart pruning techniques. Via extensive simulations, we illustrate that the proposed heuristic schemes outperform the schemes for predetermined transmission order of users and achieve close-to-optimal solutions., Scientific and Technological Research Council of Turkey (TÜBİTAK)

Machine learning for hybrid beamforming has been extensively studied by using centralized machine learning (CML) techniques, which require the training of a global model with a large dataset collected from the users. However, the transmission of the whole dataset between the users and the base station (BS) is computationally prohibitive due to limited communication bandwidth and privacy concerns. In this work, we introduce a federated learning (FL) based framework for hybrid beamforming, where the model training is performed at the BS by collecting only the gradients from the users. We design a convolutional neural network, in which the input is the channel data, yielding the analog beamformers at the output. Via numerical simulations, FL is demonstrated to be more tolerant to the imperfections and corruptions in the channel data as well as having less transmission overhead than CML., Scientific and Technological Research Council of Turkey (TÜBİTAK); European Union (EU); Horizon 2020; CHIST-ERA Grant; Ford Otosan