Optimal Edge Research Articles

Dependent Function Embedding for Distributed Serverless Edge Computing

Edge computing is booming as a promising paradigm to extend service provisioning from the centralized cloud to the network edge. Benefit from the development of serverless computing, an edge server can be configured as a carrier of limited serverless functions, in the way of deploying Docker runtime and Kubernetes engine. Meanwhile, an application generally takes the form of directed acyclic graphs (DAGs), where vertices represent dependent functions and edges represent data traffic. The status quo of minimizing the completion time (a.k.a. makespan) of the application motivates the study on optimal function placement. However, current approaches lose sight of proactively splitting and mapping the traffic to the logical data paths between the heterogeneous edge servers, which could affect the makespan significantly. To remedy that, we propose an algorithm, termed as Dependent Function Embedding (DPE), to get the optimal edge server for each function to execute and the moment it starts executing. DPE finds the best segmentation of each data traffic by exquisitely solving several infinity norm minimization problems. DPE is theoretically verified to achieve the global optimality. Extensive experiments on Alibaba cluster trace show that DPE significantly outperforms two baseline algorithms in makespan by 43.19% and 40.71%, respectively.

Resource Allocation in DT-Assisted Internet of Vehicles via Edge Intelligent Cooperation

Applications in the Internet of Vehicles (IoV) are usually accompanied by ultralow network response latency requirement. A promising approach to meet this demand is combining the IoV with mobile edge computing and enabling edge devices to share their communication, computation, and caching (3C) resources via edge intelligent cooperation. However, the allocation of 3C resources supported by artificial intelligence (AI) demands a huge number of training data and strong computing ability which is impossible to achieve on resource-limited on board unit (OBU) or road side unit (RSU). In this article, we propose a digital twin (DT) supported edge intelligent cooperation scheme, which empowers the optimal 3C resource allocation and edge intelligent cooperation possible. We focus on the response delay minimization in order to meet the requirement of latency-sensitive applications in the IoV. Specifically, mathematical expressions of the network response time are formulated according to modeling the workflow of the edge server as an M/M/1/N/FCFS queuing process. Especially, we conduct a detailed analysis of the deviations in 3C resource between the physical world and the DT space, based on which we further discuss the impact of these deviations in offloading decision. Furthermore, a mathematical optimization model aiming at minimizing the latency is formulated. In view of its complexity, we apply a deep deterministic policy gradient algorithm to solve it by modeling the cooperation process between edge nodes as a Markov decision process. Finally, we carry out simulations to demonstrate that our algorithm outperforms the existing schemes in terms of network response latency.

Application of Virtual Network Mapping Algorithm Based on Optimal Subnet in Enterprise Cost Accounting Platform

The price of the product has a great impact on the market competitiveness of the product. The manager can adjust the operation strategy according to the cost information of the product, cut costs, adjust the market decision and make the product more competitive in the market. Therefore, the research of product costing method has a very important significance. Considering that the physical resources in the processing of the virtual network mapping (VNM) fragment embedded in the request are rejected, and the problem of physical resource usage is weakened, a VNM algorithm based on the optimal subnet edge matching algorithm is proposed. After the VNM request is coarsened to the subnet, the breadth-first search algorithm is used to create the physical subnet candidate set, and the constraints of the virtual node and the coarsening network topology are complied with. The simulation results show that the proposed algorithm reduces the number of link mapping hops and can improve the acceptance rate and cost-benefit ratio of virtual networks. First of all, this paper analyzes the research background of manufacturing cost calculation and the research situation at home and abroad, discusses the advantages and disadvantages of standard cost calculation and studies and analyzes the principles, advantages and disadvantages of standard cost calculation in daily business activities. Second, design a new mathematical accounting model based on the mathematical model of standard cost calculation in daily business activities. This combines the advantages of standard costing and time-based activity costing, and provides theoretical support for manufacturing companies to use this accounting method. Finally, the research results of this paper are summarized, and the shortcomings of this paper are analyzed. Cost accounting method needs to be further studied.

Deployment and Allocation Strategy for MEC Nodes in Complex Multi-Terminal Scenarios.

Mobile edge computing (MEC) has become an effective solution for insufficient computing and communication problems for the Internet of Things (IoT) applications due to its rich computing resources on the edge side. In multi-terminal scenarios, the deployment scheme of edge nodes has an important impact on system performance and has become an essential issue in end–edge–cloud architecture. In this article, we consider specific factors, such as spatial location, power supply, and urgency requirements of terminals, with respect to building an evaluation model to solve the allocation problem. An evaluation model based on reward, energy consumption, and cost factors is proposed. The genetic algorithm is applied to determine the optimal edge node deployment and allocation strategies. Moreover, we compare the proposed method with the k-means and ant colony algorithms. The results show that the obtained strategies achieve good evaluation results under problem constraints. Furthermore, we conduct comparison tests with different attributes to further test the performance of the proposed method.

Blockchain-Based Incentive Energy-Knowledge Trading in IoT: Joint Power Transfer and AI Design

Recently, edge artificial intelligence techniques (e.g., federated edge learning) are emerged to unleash the potential of big data from Internet of Things (IoT). By learning knowledge on local devices, data privacy preserving and Quality of Service (QoS) are guaranteed. Nevertheless, the dilemma between the limited on-device battery capacities and the high energy demands in learning is not resolved. When the on-device battery is exhausted, the edge learning process will have to be interrupted. In this article, we propose a novel wirelessly powered edge intelligence (WPEG) framework, which aims to achieve a stable, robust, and sustainable edge intelligence by energy harvesting (EH) methods. First, we build a permissioned edge blockchain to secure the peer-to-peer (P2P) energy and knowledge sharing in our framework. To maximize edge intelligence efficiency, we then investigate the wirelessly powered multiagent edge learning model and design the optimal edge learning strategy. Moreover, by constructing a two-stage Stackelberg game, the underlying energy-knowledge trading incentive mechanisms are also proposed with the optimal economic incentives and power transmission strategies. Finally, simulation results show that our incentive strategies could optimize the utilities of both parties compared with classic schemes, and our optimal learning design could realize the optimal learning efficiency.

GW Quasiparticle Energies and Bandgaps of Two-Dimensional Materials Immersed in Water.

Computational simulations have become of major interest to screen potential photocatalysts for optimal band edge positions which straddle the redox potentials. Unfortunately, these methods suffer from a difficulty in resolving the dynamic solvent response on the band edge positions. We have developed a computational method based on the GW approximation coupled with an implicit solvation model that solves a generalized Poisson equation (GPE), that is, GW-GPE. Using GW-GPE, we have investigated the band edge locations of (quasi) 2D materials immersed in water and found a good agreement with experimental data. We identify two contributions of the solvent effect, termed a "polarization-field effect" and an "environmental screening effect", which are found to be highly sensitive to the atomic and charge distribution of the 2D materials. We believe that the GW-GPE scheme can pave the way to predict band edge positions in solvents, enabling design of 2D material-based photocatalysts and energy systems.

Study on Plant Crushing and Soil Throwing Performance of Bionic Rotary Blades in Cyperus esculentus Harvesting

Soil blocking and plant entanglement are the main challenges in Cyperus esculentus harvesting and excavating. The structure of the rotary blade is one of the critical factors affecting plant crushing and soil throwing performance. Inspired by the composite motion of longitudinal soil cutting and lateral soil throwing of the oriental mole cricket, a bionic rotary blade was designed with the contour curve of the excavation edge and excavation surface of its forefoot claw toe. The bionic rotary blade’s mechanical and kinematic analysis revealed its cutting mechanism. A flexible plant soil, bionic, rotary blade discrete element model was developed to simulate the Cyperus esculentus digging process. The optimal excavation edge and excavation surface were selected by a single factor experiment, and the optimal operating parameters of the bionic rotary blade were obtained by quadratic regression orthogonal rotational combination design. The results showed that the bionic rotary blade, based on the excavation edge and excavation surface of mole cricket first claw toe, had the longest throwing distance and the largest number of broken bonds. The best combination of operating parameters of the bionic rotary blade was 11.16 mm for blade spacing, 0.66 m/s for forward speed, and 300 rpm for shaft speed. The field experiment was carried out according to the best parameters. The results showed that the bionic rotary blade’s average soil throwing distance and plant crushing rate were 632.30 mm and 81.55%, respectively; thereby, not only meeting the requirements of Cyperus esculentus harvesting, but proving superior to the operation performance of the Chinese standard rotary blade IT245 and rotary blade with optimized cutting edge (IT245P). The results of this study can provide bionic design ideas and methods for the design of soil-cutting-based tillage machinery’s soil-engaging components, such as the rotary blade and returning blade.

SCMA-Based Multiaccess Edge Computing in IoT Systems: An Energy-Efficiency and Latency Tradeoff

Sparse code multiple access (SCMA) is a kind of code-domain non-orthogonal multiple access (NOMA) scheme, which can support the increasing requirements for high spectral efficiency and massive connections. Meanwhile, multi-access edge computing (MEC) is a promising technology for providing resource-constrained users with computing resources. In this paper, we propose a novel optimization scheme in SCMA-based MEC network from the perspective of energy and latency for the Internet of Things (IoT) devices. Specifically, a system utility is first used to calculate the weighted energy consumption and task execution latency. The initial utility minimization problem is non-convex and then can be subdivided into two tractable subproblems by fixing task offloading decisions, namely, optimal local computing via CPU frequency scheduling and optimal edge computing via SCMA codebook assignment, subcarrier power allocation, and MEC server computing resources distribution. Primarily, a joint SCMA codebook assignment based on the bidirectional matching principle and optimal power allocation algorithm is proposed. Moreover, we come up with CPU frequency scheduling strategies utilizing convex optimization to optimize the computing resources allocation of local devices and the MEC server. Finally, a low-complexity task offloading policy based on simulated annealing is presented. Numerical results show that our proposed joint optimization algorithm for resource allocation and task offloading can achieve a good compromise between time delay and energy consumption for IoT devices. It is demonstrated that the proposed strategy has a remarkable advantage compared to previous SCMA-MEC schemes.

Multi-Armed Bandits Learning for Task Offloading in Maritime Edge Intelligence Networks

In the context of complex and dynamic marine environment, the offloading of computing tasks for ships of Internet of Things (IoT) users is a very challenging problem considering the different quality of service (QoS) requirements of maritime applications. Mobile edge computing driven by powerful computing capability and edge intelligence is taken as a promising solution, especially for the resource-constrained and delay-sensitive maritime IoT users. In this paper, we study the optimal edge server selection problem for ship IoT users to jointly minimize the latency and energy consumption for task offloading. Specifically, we first propose a novel space-air-ground-edge (SAGE) integrated maritime network architecture to offload computation-intensive IoT services at sea. Then, the latency and energy consumption of data transmission and processing during offloading are modelled. Based on the models, the edge server selection problem is formulated into a Multi-Armed Bandits learning problem, with considering the task latency requirement and energy budget. To achieve the optimal solution, a novel algorithm, referred to as UCB1-ESSS, is developed, which links the latency, energy consumption, and network constraints by introducing both reward and cost. The simulation results show that the proposed algorithm can achieve considerably lower offloading latency and weighted latency-energy cost compared with the traditional algorithms under different QoS requirements, which proves the efficacy of theproposed algorithm.

COME-UP: Computation Offloading in Mobile Edge Computing with LSTM Based User Direction Prediction

In mobile edge computing (MEC), mobile devices limited to computation and memory resources offload compute-intensive tasks to nearby edge servers. User movement causes frequent handovers in 5G urban networks. The resultant delays in task execution due to unknown user position and base station lead to increased energy consumption and resource wastage. The current MEC offloading solutions separate computation offloading from user mobility. For task offloading, techniques that predict the user’s future location do not consider user direction. We propose a framework termed COME-UP Computation Offloading in mobile edge computing with Long-short term memory (LSTM) based user direction prediction. The nature of the mobility data is nonlinear and leads to a time series prediction problem. The LSTM considers the previous mobility features, such as location, velocity, and direction, as input to a feed-forward mechanism to train the learning model and predict the next location. The proposed architecture also uses a fitness function to calculate priority weights for selecting an optimum edge server for task offloading based on latency, energy, and server load. The simulation results show that the latency and energy consumption of COME-UP are lower than the baseline techniques, while the edge server utilization is enhanced.

Effect of cutting edge radius on end milling Ti–6Al–4V under minimum quantity cooling lubrication – Chip morphology and surface integrity study

The effect of cutting edge radius (30–55 μm) during milling Ti–6Al–4V under minimum quantity cooling lubrication (MQCL) conditions is studied. A trial study is conducted to compare the performance of dry, wet, and MQCL using unprepared tools with edge radius of 30 μm alone. Subsequently, inserts having initial radius of 30 μm are prepared to 40–55 μm radius using drag finishing and milling of Ti–6Al–4V under MQCL is studied for different edge radii. Rounded edge tools are found to produce thinner and more curvilinear chips resulting in easier chip removal from the machining zone. The chip thickness and curl radius were up to 54% and 47% lower for the larger edge radius tools as compared to the 30 μm edge radius tool. This improves the penetration of the MQCL aerosol jet into the cutting zone thus producing better tribological results. The better penetration of MQCL aerosol into the cutting zone for larger edge radius tools results in lowering the resultant process forces (up to 16%), tool-chip interface temperature (15%), tool wear (40%) and surface roughness (30%) for the 48 μm radius tool as compared to the unprepared (30 μm) tool. There is deterioration in process forces, tool wear and surface integrity for the 55 μm tool possibly suggesting the existence of an optimum edge radius closer to 48 μm.

Optimal Edge Nodes Deployment With Multi Association for Smart Health

The further evolution and integration of internet of things (IoT) and artificial intelligence (AI) technologies would transform traditional health care processes into smart health care services, which leverages various wireless sensors, cameras, and controllers to collect multi-dimensional medical information and analyze them intelligently. In case of this, the massive amount of data generated by smart health devices need to be transferred, processed and stored quickly, which poses great challenges to traditional cloud-based communications with single association mechanism. To address this problem, this paper proposes an innovative multi-access edge computing based access network with multi-association mechanism, where a single smart device could associate and communicate with multiple edge nodes. The deployment of edge nodes for proposed access network is formulated as an extended knapsack problem, which considers not only signal to interference plus noise rate (SINR) requirement of devices, but also power constraint and load balance of edge nodes and it is solved by the simulated annealing (SA) algorithm. Numerical results via SA algorithm show that the proposed multi-association mechanism can improve the SINR of devices significantly, while achieving better load balancing among edge nodes. All these together provides good reference to build an efficient access network for future smart healthcare system.

Upgrading edges in the maximal covering location problem

We study the upgrading version of the maximal covering location problem with edge length modifications on networks. This problem aims at locating p facilities on the vertices (of the network) so as to maximise coverage, considering that the length of the edges can be reduced at a cost, subject to a given budget. Hence, we have to decide on: the optimal location of p facilities and the optimal edge length reductions.This problem is NP-hard on general graphs. To solve it, we propose three different mixed-integer formulations and a preprocessing phase for fixing variables and removing some of the constraints. Moreover, we strengthen the proposed formulations including valid inequalities. Finally, we compare the three formulations and their corresponding improvements by testing their performance over different datasets.

Design of noble metal-free NiTiO3/ZnIn2S4 heterojunction photocatalyst for efficient visible-light-assisted production of H2 and selective synthesis of 2,5-Bis(hydroxymethyl)furan

In this work, the development of noble metal-free NiTiO3/ZnIn2S4 (1:0.25 (S1), 1:0.5 (S2), 1:1 (S3), and 1:2 (S4)) heterojunction photocatalysts possessing optimal band edge positions suitable for efficient production of H2 from water and in situ reduction of biomass derivative, 5-hydroxymethylfurfural (HMF) to value-added 2,5-Bis(hydroxymethyl)furan (BHMF) in the absence of any external reducing agent is presented. The electron microscopy analysis of these heterojunctions revealed that ZnIn2S4 nanosheets are decorated uniformly over the surface of NiTiO3 microrods. Interestingly, heterojunction, S3 having NiTiO3/ZnIn2S4 (1:1) showed the best photocatalytic activity with a high H2 generation rate of 4.43 mmol g−1h−1 which is about eight times higher than that of pure ZnIn2S4. Further, the photocatalytic H2 evolution activity of S3 was coupled with in situ reduction of biomass derivative, HMF to obtain value-added chemical, BHMF with > 99% yield along with 100% selectivity. This high photocatalytic activity of S3 is aided by the Z-scheme heterojunction between NiTiO3 and ZnIn2S4. Moreover, photocatalyst, S3, showed excellent photostability and retained the catalytic activity for several cycles of reuse. Overall, this work represents a unique demonstration of H2 generation and high yield production of an important commodity chemical, BHMF from biomass-derivative and provides a greener path for harvesting solar energy and its conversion to chemical energy.

Dynamic topology analysis for spatial patterns of multifocal lesions on MRI.

Quantitatively analysing the spatial patterns of multifocal lesions on clinical MRI is an important step towards a better understanding of the disease and for precision medicine, which is yet to be properly explored by feature engineering and deep learning methods. Network science addresses this issue by explicitly modeling the inter-lesion topology. However, the construction of the informative graph with optimal edge sparsity and quantification of community graph structures are the current challenges in network science. In this paper, we address these challenges with a novel Dynamic Topology Analysis framework on the basis of persistent homology, aiming to investigate the predictive values of global geometry and local clusters of multifocal lesions. Firstly, Dynamic Hierarchical Network is proposed to construct informative global and community-level topology over multi-scale networks from sparse to dense. Multi-scale global topology is constructed with a nested sequence of Rips complexes, from which a new K-simplex Filtration is designed to generate a higher-level topological abstraction for community identification based on the connectivity of k-simplices in the Rips Complex. Secondly, to quantify multi-scale community structures, we design a new Decomposed Community Persistence algorithm to track the dynamic evolution of communities, and then summarise the evolutionary communities incorporated with a customisable descriptor. The quantified community features are encapsulated with global geometric invariants for topological pattern analysis. The proposed framework was evaluated on both diagnostic differentiation and prognostic prediction for multiple sclerosis that is a typical multifocal disease, and achieved ROC_AUC 0.875 and 0.767, respectively, outperforming seven state-of-the-art persistent homology methods and the reported performance of six feature engineering and deep learning methods.

High-quality AlGaN epitaxy on lattice-engineerable AlN template for high-power UVC light-emitting diodes

AlGaN-based UVC light-emitting diodes (LED) were fabricated on high-quality AlN templates with an engineerable in-plane lattice constant. The controllability of the in-plane strain originated from the vacancy formation in Si-doped AlN (AlN:Si) and their interaction with edge dislocations. The strain state of the Si:AlN top interface could be well depicted by a dislocation-tilt model depending on the buffer strain state, threading dislocation density (TDD), and regrown Si:AlN thickness. The validity of the model was verified by cross-sectional TEM analysis. With a gradually widened lattice constant of regrown Si:AlN layer, strain-induced defects of subsequently grown n-AlGaN was suppressed. Therefore, growing a current spreading layer which possesses a moderate Al content (<65%), decent thickness (>1.5 µm), and a low TDD (<1.0 × 109 cm−2) simultaneously becomes possible. Additionally, the idea of an optimal edge TDD (ρe,opt) in the AlN buffer was revealed for growing high-quality n-AlGaN layers with a targeted thickness. After a deliberate strain-TDD engineering for Si:AlN and n-AlGaN, high-power UVC LEDs (λ = 275 nm, P > 200 mW) with a low forward voltage (Vf = 5.7 volt) were demonstrated at I = 1.35 A. The low forward voltage under high current injection density was attributed to the success in preparation of a low series resistance and high-quality n-AlGaN current spreading layer.

Next-Generation Edge Computing Assisted Autonomous Driving Based Artificial Intelligence Algorithms

Edge Computing and Network Function Virtualization (NFV) concepts can improve network processing and multi-resources allocation when intelligent optimization algorithms are deployed. Multiservice offloading and allocation approaches pose interesting challenges in the current and next-generation vehicle networks. The state-of-the-art optimization approaches still formulate exact algorithms, and tune approximation methods to get sufficient solutions. These approaches are data-centric that aim to use heterogeneous data inputs to find the near optimal solutions. In the context of connected and autonomous vehicles (CAVs), these techniques show an exponential computational time and deal only with small and medium scale networks. Therefore, we are motivated by using recent Deep Reinforcement Learning (DRL) techniques to learn the behavior of exact optimization algorithms while enhancing the Quality of Service (QoS) of network operators and satisfying the requirements of the next-generation Autonomous Vehicles (AVs). DRL algorithms can improve AVs service offloading and optimize edge resources. An Optimal Virtual Edge Autopilot Placement (OVEAP) algorithm is proposed using Integer Linear Programming (ILP). Moreover, an autopilot placement protocol is presented to support the algorithm. Optimal allocation and Virtual Network Function (VNF) placement and chaining of the autopilot, based on several new constraints such as computing and networking loads, network edge infrastructure, and placement cost, are designed. Further, a DRL approach is formulated to deal with dense Internet of Autonomous Vehicle (IoAV) networks. Extensive simulations and evaluations are carried out. Results show that the proposed allocation strategies outperform the state-of-the-art solutions and give better performance in terms of Total Edge Servers Utilization, Total Edge Servers Allocation Time, and Successfully Allocated autopilots.

Bubble Image Segmentation Based on a Novel Watershed Algorithm With an Optimized Mark and Edge Constraint

Bubble size contains important indication information that is closely related to flotation production conditions and process indicators. However, bubble images often have low contrast, noise, and many other shortcomings, making foam segmentation a difficult problem that the existing segmentation methods cannot solve. In this article, an improved watershed algorithm based on optimal labeling and edge constraints is proposed. Three algorithms are designed to obtain different initial tags, and then the extracted content of different tags is fused to obtain the combined foreground tag. To reduce the offset of the segmentation line, the edge operator is applied to extract the bubble boundary, and the boundary priori condition is used as a constraint to correct the segmentation line. Finally, the optimal segmentation line is obtained by fusing foreground markers and external constraints. Industrial experiments show that this method is effective and has a higher accuracy than the other methods. The average value and variance of rand index (RI) are 92.88% and 0.69, respectively.

Blockchain-Based SQKD and IDS in Edge Enabled Smart Grid Network

Smart Grid is a power grid that improves flexibility, reliability, and efficiency through smart meters. Due to extensive data exchange over the Internet, the smart grid faces many security challenges that have led to data loss, data compromise, and high power consumption. Moreover, the lack of hardware protection and physical attacks reduce the overall performance of the smart grid network. We proposed the BLIDSE model (Blockchain-based secure quantum key distribution and Intrusion Detection System in Edge Enables Smart Grid Network) to address these issues. The proposed model includes five phases: The first phase is blockchain-based secure user authentication, where all smart meters are first registered in the blockchain, and then the blockchain generates a secret key. The blockchain verifies the user ID and the secret key during authentication matches the one authorized to access the network. The secret key is shared during transmission through secure quantum key distribution (SQKD). The second phase is the lightweight data encryption, for which we use a lightweight symmetric encryption algorithm, named Camellia. The third phase is the multi-constraint-based edge selection; the data are transmitted to the control center through the edge server, which is also authenticated by blockchain to enhance the security during the data transmission. We proposed a perfect matching algorithm for selecting the optimal edge. The fourth phase is a dual intrusion detection system which acts as a firewall used to drop irrelevant packets, and data packets are classified into normal, physical errors and attacks, which is done by Double Deep Q Network (DDQN). The last phase is optimal user privacy management. In this phase, smart meter updates and revocations are done, for which we proposed Forensic based Investigation Optimization (FBI), which improves the security of the smart grid network. The simulation is performed using network simulator NS3.26, which evaluates the performance in terms of computational complexity, accuracy, false detection, and false alarm rate. The proposed BLIDSE model effectively mitigates cyber-attacks, thereby contributing to improved security in the network.

Pre-Demosaic Graph-Based Light Field Image Compression

An unfocused plenoptic light field (LF) camera places an array of microlenses in front of an image sensor in order to separately capture different directional rays arriving at an image pixel. Using a conventional Bayer pattern, data captured at each pixel is a single color component (R, G or B).The sensed data then undergoes demosaicking (interpolation of RGB components per pixel) and conversion to an array of sub-aperture images (SAIs). In this paper, we propose a new LF image coding scheme based on graph lifting transform (GLT), where the acquired sensor data are coded in the original captured form without pre-processing. Specifically, we directly map raw sensed color data to the SAIs, resulting in sparsely distributed color pixels on 2D grids, and perform demosaicking at the receiver after decoding. To exploit spatial correlation among the sparse pixels, we propose a novel intra-prediction scheme, where the prediction kernel is determined according to the local gradient estimated from already coded neighboring pixel blocks. We then connect the pixels by forming a graph, modeling the prediction residuals statistically as a Gaussian Markov Random Field (GMRF). The optimal edge weights are computed via a graph learning method using a set of training SAIs. The residual data is encoded via low-complexity GLT. Experiments show that at high PSNRs -- important for archiving and instant storage scenarios -- our method outperformed significantly a conventional light field image coding scheme with demosaicking followed by High Efficiency Video Coding (HEVC).