File Segments Research Articles

Automatic Segmentation of the Jaws Used in Guided Insertion of Orthodontic Mini Implants to Improve Their Stability and Precision.

Background and Objectives: With the goal of identifying regions with bicortical bone and avoiding root contact, the present study proposes an innovative technique for the simulation of the insertion of mini orthodontic implants using automatic jaw segmentation. The simulation of mini implants takes place in 3D rendering visualization instead of Multi-Planar Reconstruction (MPR) sections. Materials and Methods: The procedure involves utilizing software that automatically segments the jaw, teeth, and implants, ensuring their visibility in 3D rendering images. These segmented files are utilized as study models to determine the optimum location for simulating orthodontic implants, in particular locations characterized by limited distances between the implant and the roots, as well as locations where the bicortical structures are present. Results: By using this method, we were able to simulate the insertion of mini implants in the maxilla by applying two cumulative requirements: the implant tip needs to be positioned in a bicortical area, and it needs to be situated more than 0.6 mm away from the neighboring teeth's roots along all of their axes. Additionally, it is possible to replicate the positioning of the mini implant in order to distalize the molars in the mandible while avoiding the mandibular canal and the path of molar migration. Conclusions: The utilization of automated segmentation and visualization techniques in 3D rendering enhances safety measures during the simulation and insertion of orthodontic mini implants, increasing the insertion precision and providing an advantage in the identification of bicortical structures, increasing their stability.

“Love thy neighbor, but don’t pull down your hedge”: The influence of interdependence with major customers on supplier productivity

This research examines how mutual and relative dependence on major customers impact supplier productivity. We investigate two research questions: (1) Does the extent of mutual dependence between a supplier and its major customer base exhibit a non-linear impact on supplier productivity? (2) How does the combined influence of the supplier’s relative dependence and mutual dependence on major customers affect supplier productivity? Using major customer relationship data from Compustat segment files, we construct a longitudinal dataset in the manufacturing industry for the period 1998–2017. We estimate supplier productivity accounting for selection bias and employ the fixed effect regression approach to examine the effects of mutual dependence and relative dependence on supplier productivity. Moreover, we address potential endogeneity in the model with the control function approach. Our results indicate an inverted curvilinear relationship between a supplier’s mutual dependence on its major customers and its productivity: While increased mutual dependence initially enhances productivity, the marginal returns diminish. Interestingly, the curvilinear effect becomes more prominent when the supplier’s relative dependence increases. Our study’s findings offer three best-practice strategies including (1) spread and anchor risk, (2) reduce relative dependence, and (3) deepen customer understanding, all of which help suppliers manage the interdependence in major customer relationships.

Major Customers and the Speed of Leverage Adjustments

Purpose: This study investigates how major customers influence the speed of adjustment (SOA) in a firm's capital structure, specifically examining the effects of governmental versus non-governmental major customers.Study design/methodology/approach: The study employs Blundell and Bond’s (1998) GMM model to estimate SOA. Robustness methods include fixed effects, two-stage least squares (2SLS), propensity score matching (PSM), and the augmented doubly censored Tobit (DPF estimator).Sample and data: The sample consists of 186,240 firm-year observations from Compustat North America annual files and Compustat historical segment files, spanning 1978 to 2013. Firms in the utility and financial sectors are excluded.Results: Findings indicate that major customers significantly influence SOA by reducing market uncertainty through certifications. Firms with major customers have a higher SOA, with governmental customers exhibiting an even higher SOA than non-governmental ones. Results are consistent across various models and settings.Originality/value: This study uniquely identifies major customers as critical determinants of SOA, highlighting the heterogeneities in firms' SOA. It stresses the importance of considering customer type in financial strategies, differentiating between governmental and non-governmental impacts. These findings offer valuable insights for scholars, policymakers, and practitioners.Research limitations/implications: Focusing on publicly listed U.S. firms limits generalization. Future research should extend the analysis to other regions and recent periods. The findings provide insights for supplier firms in strategizing financial decisions and managing relationships with major customers. We emphasize the significant role of major customers in financial strategies, advocating for more detailed disclosure policies in the U.S. and globally to enhance transparency and market efficiency.

Euler-angle norms for tooth rotation, torque and tip.

The aim of the current study was to develop and describe a new measuring system for the orientation of a tooth in a digitalized cast of a jaw and provide new angular values for the rotation, torque and tip of maxillary and mandibular teeth. This retrospective cross-sectional study involved the utilization of a sub-group of extrinsic Euler-angles to derive optimal norm values per tooth in three different planes of orientation ('rotation', 'torque' and 'tip') by evaluating the digital representations of the teeth derived from a database containing over 17,500 patients. The process involved the entry of the .stl files of the jaw pairs into a fully automated software system (Smyl:Ai, Ulm, Germany) whereupon jaw alignment, teeth segmentation, landmark identification and visual validation of input files was conducted prior to calculation of the norm values for the three different planes of orientation. The digital scans in stereolithography (.STL)-file format of the upper and lower dentitions of 1914 individuals with optimal occlusion were chosen and evaluated. New mean (standard deviation) angular values were determined for the rotation, torque and tip of maxillary and mandibular teeth. The findings facilitate the reappraisal of rotation, torque and tip values currently acceptable as ideal. They will inform anthropologists and dental researchers about occlusion and alignment in orthodontic and non-orthodontic patients and provide baseline data for future studies. The methodology will also enable the evaluation of large numbers of data in relatively short timeframes.

Research on Parallel Reading and Drawing Techniques for Chemical Mechanical Polishing Simulation Data Based on Multi-Thread

In advanced integrated circuit manufacturing, the quality of chemical mechanical polishing (CMP) is a key factor affecting chip performance and yield. Designers need to use CMP simulation tools to locate and analyze the defects in the layout after the CMP process. However, the advancement of process nodes and the increase in data volume presents a great challenge to the speed of graphical display of CMP simulation data. To solve this issue, we propose a solution that uses multi-threading technology to optimize both data reading and drawing. In the process of data reading, we employ OpenMP and memory mapping (Mmap) technology to achieve parallel reading of file segmentation and propose a fast-string conversion algorithm based on the properties of simulation data. In the process of data drawing, we propose an adaptive downsampling method for data graphical display that combines multi-threading and double buffering technology to enable the parallel drawing of layouts. The effectiveness of this method is verified by testing CMP simulation data of various scales. Compared to traditional methods, this approach improves reading efficiency by over 8 times and drawing efficiency by more than 10 times. Furthermore, it enhances the smoothness of interaction with the CMP simulation tool.

Pediatric Unilateral Vocal Fold Paralysis

Purpose: The aim of this case study was to provide a review of a pediatric patient diagnosed with unilateral vocal fold paralysis. Method: The case history, voice samples, laryngeal examinations, and diagnostic probes were reviewed in a 10-year-old girl with the chief complaint of persistent dysphonia, characterized by use of both a breathy and a typical voice. Perceptual ratings were completed using both Consensus Auditory-Perceptual Evaluation of Voice and Grade, Roughness, Breathiness, Asthenia, and Strain scales. Cepstral peak prominence was determined and compared for the same file segment using both Praat and Analysis of Dysphonia in Speech and Voice analysis programs. Treatment probes during the initial preoperative visit are described. Results: The patient was able to compensate for short phonation tasks, but longer conversations were judged to have increased dysphonia and increased vocal fatigue. Because noted compensation was unable to be stabilized or sustained, both a reinnervation of the ansa cervicalis to the recurrent laryngeal nerve and voice therapy techniques were recommended as part of the multidisciplinary team plan. Conclusion: Pediatric patients with unilateral vocal fold paralysis may present with complex phonatory behaviors. This case study paired with supplemental acoustic and videostroboscopy files provides an opportunity to practice both an auditory-perceptual and acoustic assessment and to rate a laryngeal examination. Supplemental Material: https://doi.org/10.23641/asha.24163926

Deriving efficiency from the major customer network: the role of network connectedness and centeredness

PurposePrior studies on major customer relationships (i.e. embedded ties) focus mostly on the ties between a focal firm and its immediate customers, hindering the understanding of the influence of indirect ties (both upstream and downstream) on a focal firm's operational performance. In this study, the authors analyze how a focal firm's upstream and downstream connectedness and network location affect its productive efficiency.Design/methodology/approachUtilizing Compustat segment files, the authors constructed large-scale major customer networks covering the period 2007–2013. The authors applied a fixed-effect panel stochastic frontier model to conduct estimation. Moreover, the authors applied an endogenous panel stochastic frontier model to ensure the robustness of the main analysis.FindingsThe authors found that a focal firm's upstream and downstream connectedness both have a positive influence on a firm's productive efficiency, whereas a focal firm's centeredness in the major customer network has a negative influence on productive efficiency. Moreover, it was found that centeredness lessens the positive influences of upstream and downstream connectedness on productive efficiency. The post hoc analysis further confirmed that a focal firm's indirect ties, both upstream and downstream, positively influence a focal firm's productive efficiency.Originality/valueThis study contributes to the literature by evaluating the relative effectiveness of a focal firm's direct and indirect major customer ties, both upstream and downstream. More importantly, this study suggests potential exploitation–exploration trade-offs (i.e. productive efficiency vs. innovation) triggered by a firm's network location.

Replicating File Segments between Multi-Cloud Nodes in a Smart City: A Machine Learning Approach

The design and management of smart cities and the IoT is a multidimensional problem. One of those dimensions is cloud and edge computing management. Due to the complexity of the problem, resource sharing is one of the vital and major components that when enhanced, the performance of the whole system is enhanced. Research in data access and storage in multi-clouds and edge servers can broadly be classified to data centers and computational centers. The main aim of data centers is to provide services for accessing, sharing and modifying large databases. On the other hand, the aim of computational centers is to provide services for sharing resources. Present and future distributed applications need to deal with very large multi-petabyte datasets and increasing numbers of associated users and resources. The emergence of IoT-based, multi-cloud systems as a potential solution for large computational and data management problems has initiated significant research activity in the area. Due to the considerable increase in data production and data sharing within scientific communities, the need for improvements in data access and data availability cannot be overlooked. It can be argued that the current approaches of large dataset management do not solve all problems associated with big data and large datasets. The heterogeneity and veracity of big data require careful management. One of the issues for managing big data in a multi-cloud system is the scalability and expendability of the system under consideration. Data replication ensures server load balancing, data availability and improved data access time. The proposed model minimises the cost of data services through minimising a cost function that takes storage cost, host access cost and communication cost into consideration. The relative weights between different components is learned through history and it is different from a cloud to another. The model ensures that data are replicated in a way that increases availability while at the same time decreasing the overall cost of data storage and access time. Using the proposed model avoids the overheads of the traditional full replication techniques. The proposed model is mathematically proven to be sound and valid.

Multi-material polyjet printing for more lifelike endodontic practice teeth.

Root canal therapy is a highly technical procedure that has several unique challenges. Extracted human teeth have traditionally been used to train dental students which have come under increased scrutiny in recent years.1 Multiple published studies have demonstrated the value of three-dimensional (3D) printed teeth for dental education.2-5 A limitation common to these publications has been the inability to 3D print using multiple materials simultaneously.3 One strategy for creating a multi-material tooth was to create a tooth with a hollow pulp chamber and cavity preparation that was later filled with impression material and resin composite, respectively.2 Polyjet printing that allows for the use of six materials with different colors and physical properties was used to create endodontic practice teeth that had a white enamel layer, an A2 shaded dentin and root structure as well as a hollow, red pulp chamber full of a water-soluble, gelatinous support material (J5 Dentajet; Stratasys). A cone beam computed tomography (CBCT) radiograph of an extracted maxillary first premolar was made (75 μm, 90 kV, and 10.0 mA). The pulp chamber and canals were segmented from the reconstruction and Standard Tesselation Language files of the pulp chamber and tooth were exported. A 3D modeling program (Meshmixer, Autodesk) was used to further segment the file into an enamel layer, a dentin and root layer, and a hollow pulp chamber (Figure 1). These three files were brought into the printing software (GrabCAD; Stratasys) as an assembly which allows for the selection of different materials to be printed simultaneously in a single object. The enamel layer was printed using a white resin, DEN847 VeroDent PureWhite, an A2 shade resin for the root layer, MED620 VeroGlaze, the hollow pulp layer was printed using a red resin, RGD 852 VeroMagenta-V, and the hollow cavity of the pulp layer was printed with SUP711. Figure 2 shows the three layers assembled into a single item as well as disassembled into its three constituent parts. PolyJet printing was capable of fabricating endodontic practice teeth with multiple textures throughout the printed item. Figure 3 shows the printed teeth in cross-section and whole demonstrating the three solid materials and the gelatinous material filling the pulp chamber and canals. The canals can be filed with endodontic files and the gelatinous support material removed in a similar fashion to the pulp tissue. The initial expense of 3D printing technology can be high as identified by several authors.3 The use of the described technology in this article was able to produce 100 of the practice teeth in an overnight printing cycle for a per-tooth cost of less than half of one US dollar. With the ease of producing segmented files from a CBCT image of a tooth and the low cost of printing, students can be given greater access to standardized practice teeth for endodontic education. This same process has been applied by the author to design and create practice teeth with soft, excavatable carious lesions for a similarly low cost per tooth.

Cyclic fatigue resistance of EdgeTaper Platinum, Protaper Gold, and TruNatomy Prime rotary files before and after autoclave sterilization.

This in vitro investigation aimed to determine the influence of multiple autoclave cycles on the cyclic fatigue resistance of three heat-treated nickel-titanium (NiTi) files: EdgeTaper Platinum (ETP), ProTaper Gold (PTG), and TruNatomy Prime (TN). Sixty NiTi files, twenty of each NiTi file type: ETP 25/.06, PTG 25/.08, and TN 26/.04 were randomly divided into four equal subgroups (n=5). The files for the control group were left un-autoclaved. Different autoclave sterilization cycles (one, five, and ten) were used for the other three groups. The files were then placed in a metal canal block and rotated according to the manufacturer's instructions until fracture. The length of the broken segment and the time taken for fracture were measured. The fractured surfaces were subsequently subjected to SEM imaging. The Kruskal-Wallis test was used to analyze the data, followed by Dunn-Bonferroni pairwise comparisons. Statistical significance was set at p<0.05. ETP showed significantly greater resistance to cyclic fatigue than TN in all autoclave groups and PTG after five autoclave cycles (p=0.014). Fatigue resistance was not affected by the number of autoclaving cycles, except for ETP. After the first and tenth autoclaving cycles, they required significantly more rotations to failure than the non-sterilized files (p=0.039 and p=0.021, respectively). The fractured segments of the ETP files in these two groups were also longer than those in the control group (p=0.010). The cyclic fatigue resistance of ETP was greater than that of TN in all tested conditions. Repeated autoclave cycles of sterilization improved the cyclic fatigue resistance of the ETP files only and did not affect the cyclic fatigue resistance of TN and PTG. However, the ETP files separated at a longer distance from the tip with increased autoclaving cycles.

Speech emotion recognition using 2D-convolutional neural network

&lt;span lang="EN-US"&gt;This research proposes a speech emotion recognition model to predict human emotions using the convolutional neural network (CNN) by learning segmented audio of specific emotions. Speech emotion recognition utilizes the extracted features of audio waves to learn speech emotion characteristics; one of them is mel frequency cepstral coefficient (MFCC). Dataset takes a vital role to obtain valuable results in model learning. Hence this research provides the leverage of dataset combination implementation. The model learns a combined dataset with audio segmentation and zero padding using 2D-CNN. Audio segmentation and zero padding equalize the extracted audio features to learn the characteristics. The model results in 83.69% accuracy to predict seven emotions: neutral, happy, sad, angry, fear, disgust, and surprise from the combined dataset with the segmentation of the audio files.&lt;/span&gt;

Amharic Speech Search Using Text Word Query Based on Automatic Sentence-like Segmentation

More than 7000 languages are spoken in the world today. Amharic is one of the languages spoken in the East African country Ethiopia. A lot of speech data is being made every day in different languages as machines are getting better at processing and have improved storing capacity. However, searching for a particular word with its respective time frame inside a given audio file is a challenge. Since Amharic has its own distinguishing characteristics, such as glottal, palatal, and labialized consonants, it is not possible to directly use models that are developed for other languages. A popular approach in developing systems for searching particular information in speech involves using an automatic speech recognition (ASR) module that generates the text version of the speech where the word or phrase is searched based on text query. However, it is not possible to transcribe a long audio file without segmentation, which in turn affects the performance of the ASR module. In this paper, we are reporting our investigation on the effects of manual and automatic speech segmentation of Amharic audio files in a spiritual domain. We have used manual segmentation as a baseline for our investigation and found out that sentence-like automatic segmentation resulted in a word error rate (WER) closer to the WER achieved on the manually segmented test speech. Based on the experimental results, we propose Amharic speech search using text word query (ASSTWQ) based on automatic sentence-like segmentation. Since we have achieved lower WER using the previously developed speech corpus, which is in a broadcast news domain, together with the in-domain speech corpus, we recommend using both in- and out-domain speech corpora to develop the Amharic ASR module. The performance of the proposed ASR is a WER of 53% that needs further improvement. Combining two language models (LMs) developed using training text from the two different domains (spiritual and broadcast news) allowed a WER reduction from 53% to 46%. Therefore, we have developed two ASSTWQ systems using the two ASR modules with WERs of 53% and 46%.

The diagnostic utility of DNA copy number analysis of core needle biopsies from soft tissue and bone tumors

AbstractMorphologic and immunohistochemical analysis of preoperative core needle biopsies (CNB) is important in the management of patients with soft tissue and bone tumors (STBTs). Most SBTB subtypes have more or less extensive DNA copy number aberrations (CNA), potentially providing useful diagnostic information. To evaluate the technical feasibility of single nucleotide polymorphism (SNP) array analysis and the diagnostic usefulness of the copy number profiles, we studied CNBs from 171 patients with suspected STBTs. SNP array analysis could be performed on 168 (98%) of the samples. The CNA profile was compatible with the CNB diagnosis in 87% of the cases. Discrepant cases were dominated by false-negative results due to nonrepresentative material or contamination with normal cells. 70 genomic profiles were indicative of specific histopathologic tumor entities and in agreement with the corresponding CNB diagnoses in 83%. In 96 of the cases with aberrant CNA profiles, the SNP profiles were of sufficient quality for segmentation, allowing clustering analysis on the basis of the Jaccard similarity index. The analysis of these segment files showed three major CNA clusters, based on the complexity levels and the predominance of gains versus losses. For 43 of these CNB samples, we had SNP array data also from their corresponding surgical samples. In 33 of these pairs, the two corresponding samples clustered next to each other, with Jaccard scores ranging from 0.61 to 0.99 (median 0.96). Also, for those tumor pairs that did not cluster together, the Jaccard scores were relatively high (median 0.9). 10 cases showed discrepant results, mainly due to varying degrees of normal cell contamination or technical issues. Thus, the copy number profile seen in a CNB is typically highly representative of the major cell population in the tumor.

Performing forced alignment with Wav2vec 2.0

Forced alignment, the task of aligning segmentation of audio speech files with an orthographic or phonetic transcript, is fundamental to many types of speech research. Yet the available toolkits for forced alignment are mostly based on the classic HMM/GMM systems, which are outperformed by neural network-based speech recognition models, especially the large-scale speech pre-trained models in recent years. We propose a method of forced alignment utilizing the pre-trained transformer-based model, Wav2vec 2.0. This model has been pre-trained on massive audio datasets, and is subsequently fine-tuned on 360 h of speech data guided by connectionist temporal classification (CTC) loss. The model has learned to jointly recognize phonemes and perform segmentation with or without orthographic transcriptions. During inference, the hidden states are converted to frame-level alignments through post-processing. Our preliminary analysis shows that the model performs competitively when segmenting the TIMIT benchmark, even without orthographic transcription provided.

Hardware-based security approach for secure distributed cloud data storage and retrieval

The East African Community (EAC) recognizes the fundamental role of science and technology in economic development. This led the Partner States to promote cooperation in the development and application of science and technology within the Community. The adoption of cloud computing in East Africa Community member states may induce a number of paths for Internet-based service provisioning to satisfy various demands in the community. However, data privacy and security has turned into crucial issues that resist cloud adoption for the past decades in the area of computerized systems. The dominating issue regarding security and privacy is pointed out on the data accessibility whereby cloud service providers and consumers simply gain access to secret information (data). The issue resulted in the cloud service consumer’s fear and slowed down the adoption speed of cloud computing from diverse areas of applications, i.e., governmental, medical and financial institutions; therefore, in this article we address this issue and propose a robust Hardware-Based Security (HBS) approach, whereby the cloud service provider and consumers are limited to simply access any stored data in the cloud data centers. This approach halves data files into data segments and distributes encrypted file segments to different cloud data centers. Our approach is built on Secure Data Distribution (SDD) and Data Retrieval (DR) Algorithms described in a Secure Efficient Data Distribution Model (SEDD). The analysis shows robust data security and less computation power which can resist unauthorized data accessibility with less latency. Therefore, once the East African Community member states adopt a secure cloud services, the development and application of science and technology within the community will boost the sustainable economic development.

Computer-aided design-CAM-guided endodontic microsurgical localization and retrieval of two separated instruments from the periapical area of a mandibular second molar

Three-dimensional (3D) imaging devices used for the visualization of anatomic structures not only allow for a more accurate diagnosis but also facilitate precise planning of surgical treatments, such as in guided surgery for instrument retrieval using templates. The innovative template-based treatment methods could also play a crucial role in future modern surgical endodontic treatments, enabling exact preplanning and precision-guided surgical interventions, and resulting in greater accuracy and success rates. We present here a case of a patient with a periapical lesion below a right mandibular second molar, with a fractured file segment extruding beyond the apex in the distal canal, and another instrument half-way extruded from the mesial canal, with radiolucencies evident at the mesial and distal root apices. The patient was treated by employing a 3D-guided microsurgical approach. First, 3D optical scans were imported into a guided surgery program. Periapical lesions and extruded fractured instruments were marked within the software. The osteotomy size, apical resection level, and bevel angle were defined pretreatment. This case introduces a novel-guided microsurgical endodontic technique, incorporating recommended guidelines of modern surgical endodontic treatment with the aid of 3D-printed surgical templates, by which minimally invasive surgical treatment was ensured with preservation of tooth, bone, and surrounding anatomical structures.

DASH Live Broadcast Traffic Model: A Time-Bound Delay Model for IP-Based Digital Terrestrial Broadcasting Systems

This paper proposes a live broadcast traffic model for an internet protocol (IP)-based terrestrial digital broadcasting system to transmit dynamic adaptive streaming over hypertext transfer protocol (DASH) media. The IP-based terrestrial digital broadcasting systems such as Advanced Television Systems Committee (ATSC) 3.0 transmit media content (e.g., full high definition and ultra-high definition) in units of DASH segment files. Although the DASH segment file has the same quality and playback time, the size of each DASH segment file can vary according to the media composition. The transmission resource of the terrestrial broadcasting system has increased the transmission capacity of broadcasting with new technologies. However, the transmission capacity is still limited and fixed compared to wired broadcasting networks. Therefore, a problem occurs with the efficiency of broadcasting resources and transmission delay when transmitting a variable segment file to a terrestrial digital broadcasting network. In this paper, the resource efficiency and transmission delay results that occur when transmitting the actual DASH segment file are simulated through the live broadcast traffic model, and the maximum delay time that a viewer accessing the terrestrial broadcast can experience is presented.

Customer–Supplier Relationships and Abnormal Accruals

We investigate the relationship between customer and supplier firms’ abnormal accruals to examine whether the supply chain is an important transmission channel of abnormal accruals. We propose “earnings management” hypothesis and “customer demand shock” hypothesis. Empirically, we examine the relation between a supplier’s estimated abnormal accruals and those of its major customers using Compustat Business Segment Files over the period 1987–2015. To further explore the demand shock channel, we directly test the impact of the bullwhip effect (BWE) on the linkage in abnormal accruals along the supply chain. Following the literature in operation management, we construct the amplification ratio, measured as the coefficient of variation of a firm’s orders divided by the coefficient of variation of the firm’s demand. We find that customer firms’ demand shocks link customer and supplier abnormal accruals as they propagate along the supply chain, via the BWE. Our evidence supports “customer demand shock” hypothesis. Consistent with the view that improving predictions on orders from their customers would mitigate this BWE, we find that a customer’s abnormal accruals have a much smaller impact on those of its suppliers whose auditors have expertise in the customer’s industries. Overall, our results suggest that the supply chain is an important transmission channel of abnormal accruals, and auditor expertise serves to reduce information opaqueness during this process. Our paper contributes to the literature examining the impact of BWEs on firms’ financial performance and the role of auditors’ expertise in reducing information opaqueness in supply chain.

Optimization on data offloading ratio of designed caching in heterogeneous mobile wireless networks

Caching context at small base stations (SBSs) is a prospective approach to relieve backhaul pressure. Numerous prior arts focus on caching strategy by separately considering user mobility and wireless interference, which are two inherent characteristics of heterogeneous mobile wireless networks. Unlike these existing works, we consider the influences of both characteristics together on caching strategy by adopting data offloading ratio (DOR) as a performance metric, which reflects the utilization of resources of SBSs. Here, we will focus on two caching models, namely short length coded caching (SLC) and long length coded caching (LLC), to avoid the repetition of index of file segments in SLC. First, we derive explicit expressions of DOR under LLC and SLC in environments with both characteristics (Scenario-A). The maximization of DORs are then investigated and near optimal solutions are obtained in this NP-hard problem. Second, a degraded environment that only considers user mobility (Scenario-B) is investigated, where the near optimal solutions are obtained by submodular expressions with linear complexity. Finally, simulations firmly support the necessity of combining both characteristics. Specifically, the DORs of our caching strategies under LLC and SLC in Scenario-A are increased by 30% on average in comparison to the state of the art.

An offline framework for high-dimensional ensemble Kalman filters to reduce the time to solution

Abstract. The high computational resources and the time-consuming IO (input/output) are major issues in offline ensemble-based high-dimensional data assimilation systems. Bearing these in mind, this study proposes a sophisticated dynamically running job scheme as well as an innovative parallel IO algorithm to reduce the time to solution of an offline framework for high-dimensional ensemble Kalman filters. The dynamically running job scheme runs as many tasks as possible within a single job to reduce the queuing time and minimize the overhead of starting and/or ending a job. The parallel IO algorithm reads or writes non-overlapping segments of multiple files with an identical structure to reduce the IO times by minimizing the IO competitions and maximizing the overlapping of the MPI (Message Passing Interface) communications with the IO operations. Results based on sensitive experiments show that the proposed parallel IO algorithm can significantly reduce the IO times and have a very good scalability, too. Based on these two advanced techniques, the offline and online modes of ensemble Kalman filters are built based on PDAF (Parallel Data Assimilation Framework) to comprehensively assess their efficiencies. It can be seen from the comparisons between the offline and online modes that the IO time only accounts for a small fraction of the total time with the proposed parallel IO algorithm. The queuing time might be less than the running time in a low-loaded supercomputer such as in an operational context, but the offline mode can be nearly as fast as, if not faster than, the online mode in terms of time to solution. However, the queuing time is dominant and several times larger than the running time in a high-loaded supercomputer. Thus, the offline mode is substantially faster than the online mode in terms of time to solution, especially for large-scale assimilation problems. From this point of view, results suggest that an offline ensemble Kalman filter with an efficient implementation and a high-performance parallel file system should be preferred over its online counterpart for intermittent data assimilation in many situations.