Search

To solve the issues of low resource utilization and performance bottleneck in current server-centric data center networks (DCNs), we propose and experimentally demonstrate a disaggregated application-centric optical network (DACON) for data center infrastructures based on hybrid optical switches. DACON achieves flexible provision and dynamic reconfiguration of hardware nodes exploiting the softwared-define networking (SDN) orchestration plane. A four-node SDN-enabled disaggregated prototype is implemented with a field-programmable-gate-array-based controller of hardware nodes and nanosecond optical switches, performing a minimal node-to-node network latency of 378.6 ns and zero packet loss. Based on the unmodified Linux kernel and two different applications (distributed computing and a Memcached database), the application runtime of the disaggregated prototype is investigated and compared with the server-centric architecture. Experimental results show that the disaggregated prototype performs better with Memcached database applications, achieving a 1.46 × faster runtime than the server-centric network at a memory node access ratio of 0.9. Based on the customized control plane orchestrator and dynamic resource reallocation, the node-to-node latency is reduced by 21% when CPU nodes access memory nodes. The scalability of DACON is then numerically assessed based on experimentally measured parameters. Results show that the intra-rack node-to-node latency is less than 404.8 ns with a 6144-node network and memory node access ratio of 0.9. Finally, the cost and power consumption are also studied and compared with current DCN architectures. Results indicate DACON saves 13.4% of the cost of an interconnect network compared with current disaggregated architecture and consumes up to 31.1% less power with respect to server-centric DCN architectures.

We investigate the scalability and performance of a novel optical wireless data center network (OW-DCN) based on passive diffractive optics and fast tunable transmitters. The proposed OW-DCN architecture consists of two parallel inter- and intra- cluster networks to implement a flat communication between the top of racks within the DCN with at most two hops. The proposed DCN operation, the transmission performance and the scalability have been experimentally demonstrated and assessed for different modulation formats (NRZ-OOK and PAM4) with different sizes of DCN. A DCN with $8\times 8$ racks shows 20Gbit/s OOK error free transmission with a power penalty of only 1dB compared to the B2B performance. For a $16\times 16$ racks DCN, experiment indicates 16Gbit/s PAM4 transmission at FEC-limit of $\text {BER} . Moreover, scalability investigations of the proposed DCN architecture with optimized passive optics indicate that a $32\times 32$ racks DCN with spectral band around 1 nm per link is feasible.

Due to the lack of optical buffer, high packer loss caused by packet contention is one of the main challenges for the optical switching data centers (DCs). Flow control (FC) protocol employing electrical buffers at the top of racks (ToRs) and exploiting packet retransmission mechanism in case of contention has been extensively investigated to decrease the packet loss in optical DCs. However, the packet retransmission and the head-of-line (HOL) blocking in electrical buffers at substantial load traffic introduce extra latency and buffer-overflow that still cause packet loss. To overcome these issues, a novel contention resolution technique based on a software-defined networking (SDN) enabled optical polling flow control is proposed and experimentally assessed in this article. Experimental assessments show that the proposed contention resolution scheme achieves zero packet loss and 7.4 μs latency performance at the load of 0.4. In addition, we numerically modelled and investigated the scalability of the proposed contention resolution technique in a large scale DCN based on the experimental parameters. Results prove the excellent scalability performance of OPFC scheme, in which the packet loss increases from 2.1E-3 to 9.02E-3 and the average latency increases 5.17 μs at the load of 0.5 as the OPFC based network scales from 4 to 40960 servers.

Fast and high capacity optical switching techniques have the potential to enable low latency and high throughput optical data center networks (DCNs) to afford the rapid increasing traffic boosted by multiple applications. Flexibility of the DCN is of key importance to provide adaptive and dynamic bandwidth and capacity to handle the variable traffic patterns of heterogeneous applications. Aiming at improving the network performance and the system flexibility of optical DCNs, we propose and investigate a novel optical DCN architecture named ROTOS based on reconfigurable optical top of rack (ToR) and fast optical switches. In the proposed DCN architecture, the novel optical flexible ToRs employing multiple transceivers (TRXs) and a wavelength selective switch (WSS) are reconfigured by the software-defined networking (SDN) control plane. The bandwidth can be dynamically allocated to the dedicated optical links on-demand according to the desired oversubscription (OV) and intra-/inter-cluster traffic matrix. Numerical investigations of the novel architecture under realistic traffic model indicate that dynamically allocating the TRXs and elastically controlling the WSS, a packet loss below 1E-5 and a server-to-server latency lower than 3 μs can be guaranteed for different traffic patterns at load of 0.4. With respect to the DCN with static interconnections, the average packet loss of ROTOS decreases two orders of magnitude and the average server-to-server latency performance improves by 21.5%. Scalability investigation to a large number of servers shows limited (11%) performance degradation as the network scale from 2560 to 40960 servers. Additionally, the dynamic bandwidth allocation of the DCN is experimentally validated. Network performance results show a packet loss of 0.05 and 5.85 μs end-to-end latency at the load of 0.8. Finally, investigations on the cost and power consumption confirm that the ROTOS DCN architecture has 28.4% lower cost and 35.0% better improvement for power efficiency with respect to the electrical switch based DCNs.

Optical switching techniques featuring the fast and large capacity have the potential to enable low latency and high throughput optical data center networks (DCN) to afford the rapid increasing of traffic-boosted applications. Flexibility of the DCN is of key importance to provide adaptive and dynamic bandwidth to handle the variable traffic patterns generated by the heterogeneous applications while optimizing the network resources. Aiming at providing the flexible bandwidth for optical DCNs, we propose and experimentally investigate a software-defined networking (SDN) enabled reconfigurable optical DCN architecture based on novel optical top of rack (OToR) switch exploiting photonic-integrated wavelength selective switch. Experimental results show that the optical bandwidth per link can be automatically reallocated under the management of the deployed SDN control plane according to the variable traffic patterns. With respect to the network with inflexible interconnections, the average packet loss of the reconfigurable DCN decreases 1 order of magnitude and the server-to-server latency performance improves of 42.2%. Scalability investigation illustrates limited (11.7%) performance degradation as the reconfigurable network scale from 2560 to 40960 servers. Both the numerical and experimental assessments validate the proposed DCN with reconfigurable bandwidth feature and lower latency variations with respect to the inflexible DCNs.

© 2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works Optical switching techniques have the potential to enable the optical data center network (DCN) interconnections providing high capacity and fast switching capabilities, overcoming thus the bandwidth and latency bottleneck of present electrical switch-based multi-tiered DCNs. The rapid growth of multi-tenant applications with heterogeneous traffic require specialized quality of service (QoS) in terms of packet loss and latency to the DCN infrastructure. Slicing the DCNs into dedicated pieces according to the deployed applications, differentiated QoS, and high resource utilization can be provided. However, slicing the optical DCNs still needs to be investigated because the Software-defined Networking (SDN) technique is developed for the electrical networks, not fully supporting the properties of the optical network. Additionally, Network Slices (NS) need to be automatically provisioned and reconfigured, to provide flexible slice interconnections in support of the multi-tenant applications to be deployed. In this article, we propose and experimentally assess the automatic and flexible NSs configurations of optical OPSquare DCN controlled and orchestrated by an extended SDN control plane for multi-tenant applications with differentiated QoS provisioning. Optical Flow Control (OFC) protocol has been developed to prevent packet losses at switch sides caused by packet contentions. The extended OpenFlow (OF) protocol of SDN is deployed as well in support of the optical switching characteristics. Based on the collected resource topology of data plane, the optical network slices can be dynamically provisioned and automatically reconfigured by the SDN control plane. Meanwhile, experimental results validate that the priority assignment of application flows supplies dynamic QoS performance to various slices running applications with specific requirements in terms of packet loss and transmission latency. In addition, the capability of exposing traffic statistics information of data plane to SDN control plane enables the implementation of load balancing algorithms further improving the network performance with high QoS. No packet loss and less than 4.8 µs server-to-server latency can be guaranteed for the sliced network with highest priority at a load of 0.5. This work is supported by Olympics project (ESTAR17207) and Metro-Haul project (G.A. 761727).

A flexible optical switch and control system with dynamic bandwidth allocation is experimentally assessed to enable reconfigurable DCNs. Experiments validate the flexible system achieves 3.15 µs end-to-end latency (improving 48.61%) and decreases one order magnitude of packet loss, compared with the scheme with fixed connections.

We experimentally demonstrate a fast optical wireless datacenter network architecture using nanoseconds SOA-based wavelength selector and Arrayed Waveguide-Grating Router for optical switching. 4x4 prototype experiments show error-free 50Gb/s OOK per link with power penalty

A 4-node prototype of SDN-controlled disaggregated data-center network is experimentally demonstrated based on the nanoseconds optical switch, enabling flexible hardware resource provisioning and dynamic resource reallocation. Experimental results show that, based on monitoring statistics, real-time reconfiguration reduces the memory node access latency by 21%.

The rapid increasing traffic in data centers (DCs) puts tremendous pressure to the present multi-tier network architectures and electrical switching techniques. Switching traffic in the optical domain featuring ultra-high bandwidth, therefore, has been intensively investigated to build the high capacity data center networks (DCNs). To handle the variable traffic pattern in DCs, the network reconfigurability with adaptable optical bandwidth allocation is of key importance to flexibly assign the optical bandwidth. To this end, we propose and experimentally evaluate a software-defined networking enabled reconfigurable optical DCN with dynamic bandwidth allocation in this work, based on novel optical top of racks exploiting a wavelength selective switch. Experimental assessments show that the proposed solution can automatically reallocate the optical bandwidth in real-time to adapt the dynamic traffic pattern. Compared with the conventional optical DCN with static bandwidth provision, the end-to-end latency performance of the reconfigurable scheme with adaptable bandwidth allocation improves of 58.3% and the average packet loss decreases one order of magnitude. Moreover, the reconfigurable optical DCN features deterministic latency performance, with much lower time variations of packets delivery completion. Based on the experimental parameters, the simulation platform is also built to validate the good scalability of the proposed reconfigurable DCN. Numerical results illustrate the negligible performance degradation (11%) as the network scales from 2560 to 40 960 servers.

Optical networks based on fast optical switches (FOSes) could potentially solve the latency, bandwidth, cost and power consumption challenges in current electrical switches (ESes) based high performance computing (HPC) networks. In this work, we present a novel HPC network which employs distributed FOS interconnecting by removing ES (Firefly). In Firefly, Dragonfly topology is adopted for the inter-group connection of blades, while the intra-group connection of blades is implemented by FOS with fast optical flow control. The Firefly exploits the wavelength, space, and time switching domain with nanoseconds reconfiguration time of the FOS to achieve efficient statistical multiplexing operation. We numerically investigate the Firefly performance with real HPC traffic traces collected by running multiple computing applications in MareNostrum 3 HPC infrastructure with Leaf-Spine architecture. Compared with Leaf Spine architecture, results show that Firefly performs 62.4%, 54%, 68.6%, and 71.8% less latency for the applications Conjugate Gradient (CG), Multi-Grid (MG), Multiple Instruction Lattice Computation (MILC), and Miniature Molecular Dynamics (MINI_MD), respectively. Moreover, Firefly can save 56.4% cost and 65.7% power consumption, respectively, with respect to Leaf-Spine when both support around 10,000 blades.

This paper presents a discussion on theoretical solution for realizing accurate spatial channel model extension (SCME) in reverberation chamber (RC) and highlights the practical challenges on its implementations. For SCME to be correctly implemented in RC, the theoretical solution considers a reverse engineering approach, which requires an additional channel to be convoluted with the known RC channel. This additional channel could be divided into two parts, namely, the reference part and the delay-and-inverse part. This paper also presents the potential solutions in the realization.

A fast switch and control system is developed for the practical deployment of optical switches in DCNs. Implemented label control and OFC technique experimentally achieve zero packet-loss and 5µs server-to-server latency at 0.5 traffic load.

A disaggregated DCN prototype based on nanoseconds optical switch is experimentally assessed. Results show a network latency of 122.3ns, no packet loss, and error-free operation with power penalty of 0.5dB at BER of 1E-9.

We demonstrate an optical wireless DCN architecture employing fast switches and reconfiguration with FPGA-based ToRs and SOA-based wavelength switch. Experimental investigation indicates 13.3ns switching time and 1E-5 packet loss by SDN-assisted reconfiguration with automatic reallocation.

The upcoming 5G and beyond 5G heterogeneous applications with different quality of service (QoS) will impose strict latency, bandwidth, and flexibility requirements on optical metro access networks. Conventional cloud computing is gradually unable to fulfill the application requirements, especially on latency due to the distance causing propagation and networking delay. Therefore, the edge computing that distributed in metro access networks is promising to serve the applications with the requirements of ultra-low latency. As the resources of edge computing nodes are restricted and light compared with cloud data centers (DC), it is significant to manage across multiple edge computing nodes to enable joint allocation of the distributed resources. To address this issue, the optical metro network infrastructure should be flexible on the data plane and able to interact with the control and orchestration plane to automatically adapt to the communication requirements of multiple edge computing nodes. Related works have been focused on the simulation and numerical study. In this paper, an experimental testbed of a flexible optical metro access network including hardware and software components is built, and the performance is validated with real server traffic. The presented network system is based on the field-programmable gate array (FPGA), and hardware adapted open source network management and telemetry tools. Different from the commercial electrical switches, FPGA is fully programmable making it able to flexibly forward and monitor the traffic, in the meantime, to dynamically control the optical devices according to the feedback from the control plane. By exploiting dynamic software defined networking (SDN) control and network service orchestration, the proposed network is able to establish capacity adapted network slices for edge computing connections. Successful telemetry-assisted dynamic network service chain (NSC) generation, automatic bandwidth resources assignment, and QoS protection are demonstrated.

Nanoseconds WDM optical switch is exploited to implement a rack-scale memory disaggregated datacenter network (DCN). Numerical and experimental results validate the disaggregated DCN architecture prototype with 122.3ns network latency, zero packet loss, and error-free operation.

We demonstrate an automatic optical metro network based on programmable traffic interface and SOA ROADM. Automatic telemetry assisted network service chains deployment has been experimentally investigated and optimized based on the user request and application requirements. Results shows less than 200us latency for critical application.

We propose a novel optical wireless data center network based on passive diffractive optics and fast tunable transmitters. Transmission performance of the proposed network with 8times 8 racks is assessed with 20-Gb/s NRZ-OOK. Results show a maximum power penalty of 1dB.

We investigate operational performance of a novel rack-scale disaggregated network. Results show that the disaggregated network achieves 30.6% higher workloads acceptance rate, 12.9% higher resource utilization, and 33% more power saving compared with the server-centric.

Quantitative imaging biomarkers are increasingly used in oncology clinical trials to assist the evaluation of tumor responses to novel therapies. To identify these biomarkers and ensure smooth clinical translation once they have been validated, it is critical to develop a reliable workflow-efficient imaging platform for integration in clinical settings. Here we will present a web-based volumetric response-assessment system that we developed based on an open-source image viewing platform (WEASIS) and a DICOM image archive (DCM4CHEE). Our web-based response-assessment system offers a DICOM imaging archiving function, standard imaging viewing and manipulation functions, efficient tumor segmentation and quantification algorithms, and a reliable database containing tumor segmentation and measurement results. The prototype system is currently used in our research lab to foster the development and validation of new quantitative imaging biomarkers, including the volumetric computed tomography technique, as a more accurate and early assessment method of solid tumor responses to targeted and immunotherapies.

—We investigated the measurement techniques of thenonlinear vector network analyzer (NVNA) test bench, to fullycharacterize the nonlinear behavior of radio frequency (RF)power amplifiers (PAs) driven by complex modulated signals.In order to extend the baseband measurements, two kinds ofNVNA phase reference approaches are developed as alternativesolutions, so that the baseband phase measurements can beachieved with those of modulated components at multiharmonicRF bands. In the first approach, a modulated baseband signalis combined with another modulated RF one to become thedesired NVNA phase reference. While as an alternative solution,the second phase reference design is based on a “stepped”multisine, which is stepped through the baseband and eachharmonic following the NVNA swept measurements, to achievestable phase measurements. To validate these proposed NVNAtest bench designs, an RF PA, driven by a large-signal long-termevolution-like multisine, was tested and compared with digitalreal-time oscilloscope measurements.

This paper proposes a novel sequential identification method for enhancing the anti-jamming performance and for accurate recognition rate of the emitters’ individual identification in the complicated environment. The proposed method integrates the D-S evidence theory and features extraction that can get the utmost out of features of information systems and decrease the influence of uncertain factors in the signal processing. Firstly, selected features are extracted from intercepted signals. Then, the proposed self-adaptive fusing rule based on the decision vector is utilized to fuse the evidences that are transformed by features and the previous fusing information. Finally, recognition results can be obtained by judgment rules. The simulation analysis demonstrates that self-adaptive fusing rule can achieve a great balance between computational efficiency and accurate identifying rate. While comparing with other identifying methods, the proposed sequential identifying method can provide more accurate and stable recognition results, which makes the utmost care and use of existing information.

An SDN-controlled optical DCN enabling automatic QoS-driven network-slice provisioning, reconfiguration and flow-priority updating is demonstrated for the first-time. Network-slice reconfigurations with guaranteed QoS based on monitored statistics are achieved within 125 ms with 3.485 µs server-to-server latency and zero packet-loss at 0.7 load for high-priority. The authors would like to thank the Olympics (ESTAR17207) and QAMeleon (G.A. 761727) project for partially supporting this work.

We propose a novel scalable and modular low-radix fast optical switch based DCN with sphere topology (FOSphere). Numerical analyses on 10880-server indicates that FOSphere achieves 4.1 μs server-to-server latency and 2.6E-3 packet loss at load 0.4.

Network virtualization in today's data center (DC) creates new heterogeneous traffic patterns different from what so far have been reported in literatures. To properly evaluate the DC network performance, a virtualized DC traffic model is developed from the real application traffics. In this paper, we propose a novel lowest path buffer (LPB) algorithm and evaluate the network performance of LPB in OPSquare DC under the developed traffic model. Compared with Round- Robin and Localflow, LPB could achieve 23.7% and 32.1% less latency, respectively. Besides, LPB provides lower packet loss with respect to Round-Robin, Drill and Localflow.

An SDN-controlled DCN based on photonic integrated switches is experimentally assessed. Network-slice can be reconfigured within 150 ms and deployed photonic integrated switches perform error-free 10 Gb/s switching with < 2dB penalty and zero packet-loss.

We propose a novel multi-parallel layer DCN architecture based on low radix FOS. Results show cost and power consumption reduction of 4 parallel-layer architecture is 60% and 26.7% with respect to the FatTree.

We investigate a novel disaggregated architecture based on nanoseconds optical switches. Results show that under data-traffic from real applications the disaggregated architecture with 0.2 local memory rate and 40Gb/s optical links outperforms the server-centric architecture.

Maintaining high-performance operation under dynamic and nonuniform network traffic has been a technical challenge in current data center networks (DCNs). With the aim to provide better quality of service (QoS) for diverse applications, this work presents a dynamic and adaptive DCN reconfiguration framework based on deep reinforcement learning (DCR2L). The proposed framework is integrated into the SDN control plane of the DCN, implementing real-time and automatic DCN reconfiguration. Performance of the DCR2L framework is experimentally demonstrated in our DCN lab, including 4 racks and 16 servers. Experimental results show a network latency improvement of 6.9% based on the DCR2L at an average network bandwidth of 2.3 Gb/s. Based on measured traffic and physical parameters in the experiment, performance of the DCR2L framework is numerically assessed with the realistic traffic of diverse QoS requirements for both electrical and optical DCNs and for different data center scales. Leaf–spine electrical and OPSquare optical networks are set up in an OMNeT++ simulator. For a data plane network consisting of 16 racks and 320 servers, results indicate that the DCR2L framework improves the network latency of up to 16.4% under leaf–spine and up to 24.6% under OPSquare for the overall traffic with respect to the classical heuristic method. For a DCN scale of 10,240 servers, the DCR2L framework provides up to 12.5% lower latency for the leaf–spine electrical network and up to 17.8% latency improvement for the OPSquare optical network.

To cope with the dynamic datacenter (DC) traffic matrix generated by multiple applications, load balancing algorithms can be deployed to improve the datacenter network (DCN) performance. OPSquare, as a promising optical DCN architecture, adopts fast optical switches to provide high bandwidth, low latency, large scalability, and cost and energy efficiency. Current load balancing algorithms are not optimal for improving the network performance of an OPSquare DCN. To optimize the network performance of OPSquare, a dedicated load balancing algorithm considering the peculiarity of OPSquare should be addressed. Moreover, network virtualization in today’s data DC creates new heterogeneous traffic patterns different from what so far have been reported in the literature. A synthetic virtualized DC traffic model is needed to properly evaluate the performance of the load balancing algorithms. In this paper, we provide a synthetic traffic model by capturing the traffic traces of the virtualized DC applications running in a DC. The captured traffic model is used in the simulator to quantitatively evaluate the network performance of an OPSquare DCN. Then we propose and evaluate the lowest path buffer (LPB) algorithm to optimize the OPSquare DCN performance. LPB considers the buffer occupation in the whole path (from the source to the destination) rather than the local buffer occupation. The network performance improvements of LPB are demonstrated under different central controller update periods, traffic destination distributions, and network sizes. Compared with the round-robin, DRILL, and LocalFlow load balancing algorithms, LPB could achieve 23.7%, 46%, and 32.1% less latency, respectively. Besides, LPB provides lower packet loss in comparison with round-robin, DRILL, and LocalFlow.

Aiming to solve the low utilization and high operational cost in current data centers and the diversified resource requirements for different application types, we propose and investigate what we believe to be a novel rack-scale disaggregated optical data center network (RDON). Different types of hardware resources (i.e., CPU, memory, and storage) are deployed in standalone resource pools of the RDON. The RDON architecture encompasses nanosecond distributed fast optical switches for the low latency and high bandwidth interconnects between the computing and memory nodes, while electrical packet switches are utilized to support the traffic of latency insensitivity and low bandwidth between memory nodes and storage nodes. By replacing the onboard data bus with a network interconnection, RDON architecture can provide a fine granularity hardware resource and relieve the constraints of server-centric architecture. To assess the performance of the RDON under realistic scenarios, statistics of diverse applications are captured from the data center lab. The RDON is modeled based on the OMNeT++ discrete event simulator and compared with the server-centric architecture. Numerical results show that the performance of the RDON is closely coupled with the type of deployed application. Compared to the server-centric architecture, the RDON can achieve up to 2.44 × faster running time when running a single application, whereas at most 2.25 × is the fastest running time under coexisting diverse types of applications. Finally, the analysis of costs and power consumption show that, with respect to server-centric architecture, the RDON requires 21.5% extra capital cost but 46.7% less power consumption.

Purpose Automatic and accurate three-dimensional (3D) segmentation of liver with severe diseases from computed tomography (CT) images is a challenging task. Fully convolutional networks (FCNs) have emerged as powerful tools for automatic semantic segmentation, with multiple potential applications in medical imaging. However, the use of a large receptive field and multiple pooling layers in the network leads to poor localization around object boundaries. The network usually makes pixel-wise prediction independently, making it difficult to respect local label consistency and enforce the smoothness of the object boundary. Methods We have developed an automatic liver segmentation method based on a novel framework that integrates fully convolutional network predictions into active contour models (ACM). We use only a single network architecture to generate a pixel label map containing spatial regional information (foreground and background) as well as layered boundary information. We exploit the structured network outcome to define an external constraint force of active contour models. A unique property of the designed force is that both its strength and direction are adaptive to its position and relative distance to the object boundary. The resulting integrated active contour models have the advantages of incorporating both high-level and low-level image information simultaneously, while enforcing the smoothness of the contour. Because the external constraint force can push the evolving contour to the liver boundary and exists everywhere in the image domain, it allows us to place the initial contour far away from the liver boundary. It potentially allows us to control the evolution of the contour in order to preserve the topology of the liver. Results We have trained and evaluated our model on 73 liver CT scans from a clinic study. The integrated ACM model yields mean dice coefficients (DICE) 95.8 ± 1.4 (%). Without further fine-tuning the network weights for two independent datasets, it yields mean DICE 96.2 ± 0.9 (%) for the SLIVER07 training dataset, and mean DICE 94.3 ± 2.7 (%) for the LiTS training dataset. In comparison with FCN alone model, the integrated ACM model yields improvements in terms of surface distance and DICE values for almost all the cases. Furthermore, the initialization of the active contour can be very far away from the liver boundary. Conclusions Experimental results for segmenting livers (with severe diseases on CT images resulting in shape and density abnormalities) have revealed that our proposed model improves segmentation results in comparison with FCN alone. Without further fine-tuning the network weights for two independent datasets, the model is capable of handling image variations from different datasets due to its inherent deformable nature. It is relatively easy to integrate more advanced (either existing or future) FCN architecture into our framework to further improve the segmentation performance.

We demonstrate a dynamic network reconfiguration method (ACRO) based on deep reinforcement learning that can autonomously learn the scheme to reconfigure the application traffic with different QoS requirements. Numerical results based on real application traffics indicate that the ACRO provides up to 24.6% network latency improvement.

We investigate a novel scalable low-latency HPC network for blade-to-blade interconnection exploiting multiple-parallel recursive architecture with fast optical switches under realistic HPC application traffics. Compared with Leaf-Spine HPC interconnecting 4096-blade, results show that HFOS4 reduces up to 75.9% latency, 64.2% cost and 30.9% power consumption.

We propose a novel scalable and low-latency DCN architecture (FOScube) based on multiple parallel interconnected networks employing distributed low-radix FOS. Numerical analyses on 92160-server indicates that FOScube can interconnect with 3.0us end-to-end latency and 2E-4 packet loss at load 0.3.

This paper investigates the bandwidth extension of phase reference and standard, so that a group of band-limited delay-unknown multisine signals can be used to realize an equivalently time-synchronized wideband phase reference. To minimize the PAPR, the multisines are designed with Schroeder-like phase spectra. The phases of overlapping components are made identical for alignment purpose. By adjusting the carrier phase of a multisine, the time delay is effectively compensated according to the VNA-based measurements. As a result, an equivalent wideband phase reference signal is obtained based on sweeping multisines, the bandwidth of which can be arbitrarily extended in practice.

Considering that the tradition al emitter threat assessment methods are difficult to intuitively reflect the degree of target threaten and the deficiency of real-time and complexity, on the basis of radar chart method(RCM), an algorithm of emitter combined threat assessment based on ICW-RCM (improved combination weighting method, ICW) is proposed. The coarse sorting is integrated with fine sorting in emitter combined threat assessment, sequencing the emitter threat level roughly accordance to radar operation mode, and reducing task priority of the low-threat emitter; On the basis of ICW-RCM, sequencing the same radar operation mode emitter roughly, finally, obtain the results of emitter threat assessment through coarse and fine sorting. Simulation analyses show the correctness and effectiveness of this algorithm. Comparing with classical method of emitter threat assessment based on CW-RCM, the algorithm is visual in image and can work quickly with lower complexity.

Introduction The National Cancer Institute Quantitative Research Network (QIN) is a collaborative research network whose goal is to share data, algorithms and research tools to accelerate quantitative imaging research. A challenge is the variability in tools and analysis platforms used in quantitative imaging. Our goal was to understand the extent of this variation and to develop an approach to enable sharing data and to promote reuse of quantitative imaging data in the community. Methods We performed a survey of the current tools in use by the QIN member sites for representation and storage of their QIN research data including images, image meta-data and clinical data. We identified existing systems and standards for data sharing and their gaps for the QIN use case. We then proposed a system architecture to enable data sharing and collaborative experimentation within the QIN. Results There are a variety of tools currently used by each QIN institution. We developed a general information system architecture to support the QIN goals. We also describe the remaining architecture gaps we are developing to enable members to share research images and image meta-data across the network. Conclusions As a research network, the QIN will stimulate quantitative imaging research by pooling data, algorithms and research tools. However, there are gaps in current functional requirements that will need to be met by future informatics development. Special attention must be given to the technical requirements needed to translate these methods into the clinical research workflow to enable validation and qualification of these novel imaging biomarkers.

The paper proposes an improved underdetermined blind source separation (UBSS) method of nonlinear mixture, which can be applied to the communication signals' separation during data link transmission in the complicated electromagnetic environment. The proposed method combines the advantages of wavelet analysis and kernel canonical correlation analysis (KCCA), and can enhance the signals' detection performance to the data link communication via separating the communication signals from the aliasing signals effectively. The simulation results show that the wavelet-KCCA BSS method can pick up the objective communicative signals even in low SNR condition, which has proper feasibility and effectiveness.

In this paper we report the set-up and results of the Multimodal Brain Tumor Image Segmentation Benchmark (BRATS) organized in conjunction with the MICCAI 2012 and 2013 conferences. Twenty state-of-the-art tumor segmentation algorithms were applied to a set of 65 multi-contrast MR scans of low- and high-grade glioma patients - manually annotated by up to four raters - and to 65 comparable scans generated using tumor image simulation software. Quantitative evaluations revealed considerable disagreement between the human raters in segmenting various tumor sub-regions (Dice scores in the range 74%-85%), illustrating the difficulty of this task. We found that different algorithms worked best for different sub-regions (reaching performance comparable to human inter-rater variability), but that no single algorithm ranked in the top for all sub-regions simultaneously. Fusing several good algorithms using a hierarchical majority vote yielded segmentations that consistently ranked above all individual algorithms, indicating remaining opportunities for further methodological improvements. The BRATS image data and manual annotations continue to be publicly available through an online evaluation system as an ongoing benchmarking resource., National Center for Research Resources (U.S.) (P41-RR14075), National Institute of Biomedical Imaging and Bioengineering (U.S.) (R01EB013565), National Institute of Biomedical Imaging and Bioengineering (U.S.) (U54-EB005149), National Institutes of Health (U.S.) (P41-RR13218), National Institute of Biomedical Imaging and Bioengineering (U.S.) (P41-EB-015902), National Cancer Institute (U.S.) (R15CA115464)

Given the ease of alteration of digital data, integrity verification and tamper detection for medical images are becoming ever more important. In this paper, instead of using the conventional irreversible block-based watermarking approach to achieve tamper localization, we propose to incorporate such functionality into the region-based lossless watermarking scheme. This is achieved by partitioning an image into certain non-overlapping regions and appending the associated local authentication information directly into the watermark payload. A region of authentication, which can be flexibly specified by the user, is partitioned into small regions in a multilevel hierarchical manner. Such hierarchical structure allows the user to easily adjust the localization accuracy, and makes the tamper detection efficient. Experimental results demonstrate the effectiveness of tamper localization.

This paper presents a lossless watermarking scheme in the sense that the original image can be exactly recovered from the watermarked one, with the purpose of verifying the integrity and authenticity of medical images. In addition, the scheme has the capability of not introducing any embedding-induced distortion in the region of interest (ROI) of a medical image. Difference expansion of adjacent pixel values is employed to embed several bits. A region of embedding, which is represented by a polygon, is chosen intentionally to prevent introducing embedding distortion in the ROI. Only the vertex information of a polygon is transmitted to the decoder for reconstructing the embedding region, which improves the embedding capacity considerably. The digital signature of the whole image is embedded for verifying the integrity of the image. An identifier presented in electronic patient record (EPR) is embedded for verifying the authenticity by simultaneously processing the watermarked image and the EPR. Combining with fingerprint system, patient’s fingerprint information is embedded into several image slices and then extracted for verifying the authenticity.

We propose a lossless watermarking scheme, in a sense that the original image can be exactly recovered from the watermarked one, with the purpose of verifying the integrity and authenticity of medical images in PACS. Our embedding method includes imposing invertible modulations that induce low distortion level, as well as losslessly compressing the status of the original image and carrying such information in the watermarked image. Digital signature of the whole image is embedded for verifying the integrity of the image. An identifier also presented in electronic patient record (EPR) is embedded for verifying the authenticity by simultaneously processing of the watermarked image and the EPR. Combining with fingerprint system, patient’s fingerprint information is further embedded into several image slices and then extracted for verifying the authenticity. No visual quality degradation is detected in the watermarked image.