最新刊期
卷 40 ， 期 2 ， 2012
- 摘要：This paper describes a new automatic instruction-se textension approach with bitwise operaton analysis and trans- formations.It introduces a new intermediate bit-manipulation operation to directly describe bitwise accesses.It performs bit-level analysis and optimization,and builds an extended data-flow graph with the new bit-manipulation operations.Experimental results show that extending instruction-set selection based on the new data-flow graph can effectively improve the performance for bitwise computation-intensive applications.关键词：instruction-set extension;automatic selection;bitwise operation transformatio3|2|2更新时间：2025-07-16
- 摘要：A customized reconfigurable embedded Block RAM module for FPGA chip is proposed.The number of BRAM is eight and each module is a synchronous dual-port memory cell with the capacity of 18Kbits.The BRAM can be configured as six bit width modes including 16K×1bit、8K×2bits、4K×4bits、2K×9bits、1K×18bits、512×36bits;two write modes such as write-first、no-change.Several BRAM can be cascaded through interconnect circuit to achieve the expansion in depth or width.This paper focuses on the circuit that can realize the reconfiguration function.SMIC 0.13um 8-layer metal CMOS process is applied to produce the complete layout of FDP-II chip and then tape-out.The chip area is about 4.5mm×4.4mm and tested by cooperating with its software system which applies March C+ algorithm based MBIST test method.The test results show that BRAM in FDP-II has no fault and the function of reconfiguration work correctly,the design theory of such memory are verified.关键词：embedded memory;reconfigurable;FPGA;interconnect circuit;sense amplifier2|2|2更新时间：2025-07-16
- 摘要：Checkpointing is an effective technique to improve the reliability of large scale parallel computing systems.Data compression is a promising technique to reduce the size of data to be saved in the files in the storage subsystem and the amount of data to go through the communication subsystem.However,compression causes a huge amount of time overhead.The time overhead is the main technical barrier of its practical usability.In this paper,we propose a parallel compressed checkpointing technique to reduce the time overhead of compression in heterogenous architectures.It integrates a number of optimization techniques,which include transmitting checkpointing data between host and GPU in buffered pipelines,aggregating file write operations,employing a pipelined parallel compression algorithm,and delegating compression operations to GPU,etc.The paper reports an implementation of the technique in the TH-1 system and the evaluation experiments with the system.The experiment data show that the technique is efficient and practically useable.关键词：heterogenous architecture;checkpoint;data compression;pipeline;graphic processing unit (GPU)2|2|1更新时间：2025-07-16
- 摘要：We made Amdahl's Law and Rent's Rule understood and proposed one novel model to describe the multicore CPU speedup based on evaluating the known speedup models.Our methodology starts with traditional Amdahl's Law,in the steps of fixed-size,fixed-time,memory-bounded,and interconnection-bounded.Also the NoC-bandwidth and max-temperature of multicore features are discussed.The results showed,fixed-time model and memory-bounded model grasp speedups' up-limits,our model touches the middle gap between formers and Amdahl's Law,at high parallel ratio,and NoC bandwidth and Max temperature show that homogeneous multicore architectures long for higher ratio of parallelization.关键词：multicore CPU;Amdahl's law;speedup;Rent's rule;first principle;bandwidth;temperature2|3|7更新时间：2025-07-16
- 摘要：The fault probability of the elementary gate P for the gate-level circuits' reliability estimation had been given based on expert experience generally.Considering the layout structure information,the size and defect characteristics of elementary gates,analyzing the defect size distribution under different defect models.The parameters c of the probability density distribution function of particle size under the corresponding models was calculated.Then the P-expression was derived.Theoretical analysis and experimental results with ISCAS85 benchmark and 74-series circuits show that the defect piecewise linear interpolation model can well describe the lower level real defect of circuit reliability model.The reliability results of ISCAS85 benchmark circuits obtained by the probabilistic transfer matrix method based on the proposed P model and by the reliability calculation method recommended by MIL-HDBK-217 standard were compared.It shows that the proposed P model is reasonable.关键词：defect model;probability distribution of defect size;layout structure information;fault probability of elementary gate;evaluation of gate-level circuit reliability2|2|6更新时间：2025-07-16
- 摘要：Traditional TAP (Task Assignment Problem) is generally to minimize total execution cost and inter-node communication cost.This paper investigates New TAP (NTAP) considering additive conflict cost in emerging multi-core cluster systems.We analyze the complexity of the NTAP with network flow method and conclude that the conflict cost is a key to the complexity of the NTAP,and demonstrate the relationships between the complexity and the two costs including conflict cost and inter-node communication cost.Moreover,the solutions to the NTAP,and the applicability and effectiveness of the theorems and conclusions are also discussed.关键词：task assignment;complexity analysis;minimum cost flow;conflict cost;multi-core clusters2|2|2更新时间：2025-07-16
- 摘要：Due to the shortcoming of coarse partitioning granularity,utility-based single layer scheduling model fails to achieve good fairness among services and to maximize the whole utility.A two-layer hierarchical scheduling model named U-S as well as the corresponding algorithm USMS to solve the model is proposed.In order to improve the accuracy of the solution,we further design an improved algorithm named USIM.The evaluation experiments show that both USMS and USIM can achieve high solving effectiveness and efficiency and also have higher average whole utility than the single-layer model.关键词：utility;utility class;service;scheduling2|2|1更新时间：2025-07-16
- 摘要：How to determine the number of clusters is always a difficult problem in data cluster analysis.Therefore,a novel dynamic genetic clustering algorithm (DGCA) is proposed in this paper.The DGCA adopts a maximum attribute range partition method to overcome the sensitiveness to initial values of cluster centers for clustering algorithms.Furthermore,the two-stage dynamic selection and mutation operations are used in the DGCA to make selection probability and mutation probability vary with the consistency of the number of clusters in the population.Firstly the parallel search in different numbers of clusters is carried out.Then the optimal search for the best cluster centers is conducted.Numerical experiments on seven data sets show that the proposed DGCA can realize the global search for the best partition and find the optimal values for both the number of clusters and the cluster centers.关键词：cluster analysis;genetic algorithm;dynamic selection;mutation2|2|14更新时间：2025-07-16
- 摘要：Traditional kernelised classification methods could not perform well sometimes because of using a single and fixed kernel,especially on some complicated data sets.In this paper,a novel optimal double-kernel combination (ODKC) method is proposed for complicated classification tasks.Firstly,data are mapped by two basic kernels into different feature spaces respectively,and then three kinds of optimal composite kernels are constructed by integrating information of the two feature spaces.Comparative experiments demonstrate the effectiveness of our methods.关键词：kernel methods;double-kernel combination;classification2|2|6更新时间：2025-07-16
- 摘要：A data assured deletion approach adapted for cloud storage is proposed to protect data confidentiality of cloud storage system.We utilize a key derivation tree to organize and manage keys which are pushed to DHT network after partitioned by secret sharing scheme.The dynamic property of DHT network makes keys be deleted periodically causing ciphertext can not be decrypted or accessed when authorized time expires.And data assured deletion is implemented.The experiment results show that this method can delete keys effectively and the performance is low.It suffices for the requirements of assured deletion of expired data or expired data in the cloud storage system.关键词：cloud storage;data confidentiality;data deletion;key management2|2|27更新时间：2025-07-16
- 摘要：Improving accuracy,robustness and understandability is the objective of classification modeling.Regarding instability and performance limitation of existing rule learning techniques,we introduce an ensemble classifier based on randomized neighborhood reduction and neighborhood covering reduction.A set of reducts are obtained with randomized attribute reduction.A collection of rule sets are derived from the reducts based on neighborhood covering reduction.And then the classification result is output by combining the classification decision of different rule sets.The experiment result shows that the proposed technique is better than or equal to other classifiers,and is more stable when deals with noisy data.关键词：neighborhood;randomized reduction;ensemble learning;rule extraction;classifier2|2|13更新时间：2025-07-16
- 摘要：The decision functions of various kernelized classification methods can be expressed as a combination of Support Vectors (SVs),i.e.SVM,which contain the individual privacy information,so this information will be released during detecting unknown samples.Meanwhile,the amount of SVs limits classification speed,i.e.the computational time complexity of SVM is O(|SVs|).For overcoming the above drawbacks,a fast classification approach of SVM with privacy preservation is proposed,which is based on the agent preimage of the center of minimum enclosing ball (MEB),and two preimage-finding methods are presented in this paper,called QP-based solution and direct solution respectively.Experimental results on UCI and PIE face image demonstrate that two drawbacks as above can not only be solved,but also the obtained effectiveness of the proposed method is competitive.关键词：classification;support vector machine;fast classification;minimum enclosing ball (MEB);the agent of sphere center;preimage2|2|8更新时间：2025-07-16
- 摘要：Test schemes with multiple scan chains can speed up test schedule and are more fitting for testing VLSI,and hence a test data compression scheme is proposed and applied to multiple scan chain testing.Treatment model of cyclic shift introduced can dynamically adjust reference vectors,retain don't care bits in vectors and increase extensions of test vectors,and thus compatibility and inverse compatibility between vectors can be heightened.At the same time,an efficient technique for replacement of reference vectors can be exploited to further heighten correlations between vectors and decrease number of code words.Moreover,existing shift registers can be utilized to lower unnecessary hardware overhead.The experiment results demonstrate that the proposed scheme can further improve test compression ratios,meet deterministic fault test and hybrid built-in self test under maintaining advantages of multiple scan chain testing.关键词：test compression;test data;multiple scan chains;cyclic shift2|2|2更新时间：2025-07-16
- 摘要：In order to provide scientific bases for developing rational benign worm propagation policy in P2P networks,the propagation characteristics of benign worms in countering against malicious worms should be detailed researched.In this paper,a modeling and analysis method for propagation of worm-anti-worm in P2P network is proposed based on stochastic process algebra (SPA).Through analyzing the interactions between benign worms and malicious worms and the network node's state transition process,two SPA models are built by using PEPA syntax to describe the malicious worms' initial propagation phase and the worm-anti-worm phase respectively.To analysis these models,the paper makes use of continuous state-space approximation to transform the models into a set of ODEs.Simulation results show that the benign worms can effectively contain the spread of malicious worms,but it should apply scientific propagation policy according to the current network conditions,so as to ease the pressure of the network leaded by the propagation of benign worm.关键词：P2P network;benign worm;propagation model;stochastic process algebra2|3|4更新时间：2025-07-16
- 摘要：In this paper,propagation pattern of reactive worms is studied.The propagation model and the immunization model of reactive worms are proposed,in which dynamic factors of P2P networks are taken into account.Further,the sufficient condition of worms not attaining an endemic state is deduced from the model of propagation of reactive worms in applying Epidemiology.Large scale simulation experiments validate the models and the sufficient condition.All the simulations also show that among all P2P-related parameters,the downloading rate is the crucial factor to the propagation of reactive worms,the basic reproduction number of worms is the index of propagation capability of worms,and the sufficient condition is helpful to early warn the presence of an epidemic.In addition,by using data from the Gnutella network,the propagation capability,propagation speed and the risk of reactive worms are predicted and evaluated based on the propagation model.Prediction and evaluation show that it is time that reactive worms should be taken into account.关键词：P2P networks;reactive worms;dynamics;propagation modeling;basic reproduction number;simulation2|3|5更新时间：2025-07-16
- 摘要：This paper proposes a simple estimate-and-forward (EF) scheme for two-way relay networks.The relay node firstly calculates log-likelihood ratio (LLR) of bits from source nodes,then physical network encoding is performed in LLR domain instead of binary bits.The resulting new LLR is employed to generate the transmitted symbols at relay.Both upper and lower bounds are derived on the achievable sum rate of the proposed EF scheme.Performance evaluations confirm that EF scheme can improve the achievable sum rate and the bit error rate compared to conventional amplify-and-forward (AF) and decode-and-forward (DF) schemes.关键词：two-way relay;network coding;estimate-and-forward2|2|4更新时间：2025-07-16
- 摘要：Based on multi-permutation matrix,a constructing method for structure LDPC codes suitable for cooperative relaying joint with network coding is proposed.Multi-permutation is defined firstly,and then some important characteristic is summarized and proved.Detail process to construct full rank LDPC codes with column-weight-3 and girth-6 is given.It is analyzed that the generation matrix of proposed LDPC codes with structure and spare feature is suitable for joint encoding and iterative decoding in cooperative relaying with network coding.Compared to array LDPC codes,LDPC codes with dual diagonal and triple diagonal structure,which are the same 2/3 code rate,code length and quasi-cyclic matrix Y structure,simulation results show favorable decoding performance of LDPC codes proposed in the paper.关键词：network coding;structure LDPC codes;multi-permutation;full rank3|2|1更新时间：2025-07-16
- 摘要：Mobile ad hoc network relies on multi-point to complete routing tasks in collaboration,relationship of trust need to be established in the trusted routing between two nodes,but most of trust routing model only pursue the confidence of route whereas neglecting robustness.Basing on ANTHOCNET algorithm,the TrANTHOCNET algorithm is designed while taking into account both routing confidence and routing robustness.We introduce Fuzzy Petri Net formalized reasoning process algorithm to deal with the uncertain relationship among nodes,and use location information to update pheromone in real time which enhances routing robustness.The experiment results show that the TrANTHOCNET behaves better than ANTHOCNET,AODV and T-AODV when facing malicious nodes,it also has been proved effectively in routing performance.关键词：mobile Ad Hoc network;fuzzy petri net;ant colony algorithm;trusted routing2|3|1更新时间：2025-07-16
- 摘要：An in-depth study was carried out on the genetic algorithm for MFO(Multi-modal function optimization) in noise environment.The effect of noise on MFO was theoretically analyzed.The probability of searching global area and the precision of global convergence were proposed to analyze the global convergence of genetic algorithm for MFO.It was found that the complexity of Multi-modal function and the strength of noise have influence on the performance of genetic algorithm for MFO.The result shows resampling method could lower the effect of noise,and the performance of MPGA and DCGA was better than that of EGA.关键词：genetic algorithm;Multi-modal function optimization;noise environment2|3|5更新时间：2025-07-16
- 摘要：Enabling the parity-check matrices to have a large girth is an efficient method to improve the decoding performance for many binary and non-binary QC-LDPC codes.A new explicit framework to construct QC-LDPC codes with girth at least eight is proposed,without any computer search procedures.The QC-LDPC codes constructed by the framework not only have a girth at least eight,but also possess an advantage of consecutive cyclic permutation matrix (CPM) sizes.The framework includes two steps:first,a parity-check matrix is explicitly constructed with a girth at least eight under an infinite CPM size;second,the tight lower bound of consecutive CPM sizes is precisely calculated directly from the corresponding shift matrix of the parity-check matrix,by employing a new girth property discovered by this paper.关键词：low-density parity-check (LDPC) codes;quasi-cyclic (QC);girth;explicit construction2|3|5更新时间：2025-07-16
- 摘要：Test application time of IP is based on the maximum scan in/out chains.A well balance wrapper scan chains design is an important approach to minimize the test application time and test cost.Best Fit Decrease (BFD) is the most classical scan chains balance method,which is easy to realize,but the result is not the best.This paper describes a wrapper scan chains balance algorithm base on twice-assigned method by the chains difference.The main idea of the proposed solution is that select a standard chain,with its length L,find the chains which length longer than L from all internal scan chains of the IP core,compute these chains the difference di with L,regard these chains as L in the first-assigned.Sort the length shorter than L and di in descending order of length,assign to the enable shortest wrapper scan chain in the second-assigned.The proposed method is also easy to use,experimental results for the ITC'02 SOC Test Benchmarks show that the method in this paper can get better balanced result when compared to the existing solutions.关键词：scan chains balance;best fit decrease(BFD);chains difference;twice-assigned2|4|2更新时间：2025-07-16
- 摘要：A novel non-contact vital signal detection method based on Principal Component Analysis (PCA) and Empirical Mode Decomposition (EMD) is presented for the impulse ultra wideband (UWB) vital signal module.The clutter is reduced by PCA on the basis of the characteristics of the UWB clutter signal and the objective vital signal.The distance information of the vital object is estimated by the delay time,which is extracted from the appropriate principal component eigenvector peak value.The sequence of the target echo is decomposed into several IMFs by EMD,and the flatness curve of the vital sign is reconstructed in time domain.The respiration and heartbeat signals can be separated with the high signal-to-noise ratio condition.The experiment demonstrated that this signal processing algorithm is simple but feasible.The frequency spectrum characteristics,the waveform in time domain of the vital signal and the distance between the object and the antenna can be obtained simultaneously.The reconstructed vital signal is complied with the characteristics of the time variety and non-stationary for the actual signal comparatively.关键词：vital signal;ultra wideband (UWB);principal component analysis (PCA);empirical mode decomposition (EMD);singular value decomposition (SVD)2|2|8更新时间：2025-07-16
- 摘要：With the development of sequencing technology,Next-Generation Sequencing brought opportunities and challenges to the bioinformatics.As the importance of alternative splicing in gene expression and protein diversity in eukaryotes,alternative splicing identification has been the focus of bioinformatics research.The computation methods for alternative splicing need to be improved since the appearance of Next-Generation Sequencing.We introduce the past and current research of AS firstly;then conclude and compare the research methods,software and databases of AS based on RNA-seq data;finally try to discuss the development and prospect of the computational methods on AS.关键词：next-generation sequencing;RNA-seq;alternative splicing;splice sites;read mapping;bioinformatics2|2|0更新时间：2025-07-16
- 摘要：A new method of discriminant feature extraction,called Class Preserving Projection (CPP),is proposed from the point view of class relation of pairwise samples.Compared to LDA,CPP has the following two advantages.One is that the optimal subspace dimension is not restricted to the number of categories of samples,and the other is that computational complexity is lower.Experiments are performed on gene expression data for sample classification,and the results confirm the effectiveness of the method.Kernel CPP (KCPP) is presented by generalizing CPP to nonlinear space to solve the problem of nonlinear feature extraction,and the experiments on gene expression data verify the feasibility of the method.关键词：feature extraction;Fisher's linear discriminant analysis(LDA);small sample size;gene expression data2|2|2更新时间：2025-07-16
- 摘要：The applications for large-scale data distribution under timing constraint have been widely deployed on Internet.Most existing systems organize the involved clients with mesh structure and control the data distribution through the scheduling algorithms running on requesting peers.However,requesting peers request data only according to their own needs,regardless of the overall service quality of the system.Therefore,we take aim at improving the overall service quality of the system and propose a scheduling algorithm running on serving peers:DSF (Deadline Sensitive Fast distribution algorithm),which transfers urgently needed data to the peers that have strong follow-up service ability with high priority,to reduce the ratio of late data chunks and improve the service capacity of the whole system.The experimental results show that DSF has better performance with respect to transmission quality,distribution rate and load balance than the frequently used scheduling algorithms.关键词：data distribution;timing constraint;scheduling2|2|0更新时间：2025-07-16
- 摘要：The possibilistic fuzzy clustering algorithm overcomes the problem of sensitivity to noises and coincident clusters,but it assumes the contribution of each sample is equal,which leads to strong impact from outliers or noises and too many iterations.For this reason,this paper proposes a novel faster possibilistic fuzzy clustering algorithm based on the sample-weighted idea.The results of the experiments on standard data sets and synthetic data sets show that the sample-weighted algorithm is more robust against noises and outliers and reduces the time complexity effectively,and can obtain good clustering accuracy at the same time.关键词：sample-weighted;possibilistic c-means clustering;possibilistic fuzzy clustering2|3|13更新时间：2025-07-16
- 摘要：As removable storage medias are passive devices,their security policies depend on mechanisms in connected terminal systems,which will reduce the availability while providing security.This paper presents TRSF,a framework of removable storage based on trust virtual domain to implement active protection.TRSF solidifies a smart card and an isolation mechanism into the storage device and builds trust data channels from the device to the isolated usage environment in terminal system.So TRSF is able to provide trust virtual environment for data access and usage of removable storage even in untrust terminal systems by untrust processes.We implement an intelligent USB disk based on TRSF called UTrustDisk to evaluate the framework.The average overhead on read and write caused by trust chain mechanism is 7.5% and 11.5%.关键词：trusted virtual domains (TVDs);active protection;trusted storage;trust chain;isolation;chip operation system (COS)2|2|1更新时间：2025-07-16
- 摘要：Fault-tolerant system of Reconfigurable Hardware (RH) with centralized controller has the shortcomings of complex reconfiguration algorithm and long reconstruction time,and the system will be invalidation when the controller is in fail.To realize online distributed fault-tolerant in-chip,a new RH architecture of cell arrays was proposed,which has the ability to achieve in-chip self-repairing.The method of fault localization and self-repairing for interconnection circuit between electronic cells in RH are described in detail.The electronic cell circuit and a Fault-Tolerant Switch Block (FT-SB) in RH are designed.The fault-tolerant method of interconnection circuit includes two stages.Firstly,the MUXs in FT-SB and the connection lines between FT-SBs in the fault channels are detected,then the re-routing and line-shift methods in interconnection circuit are introduced to heal the fault MUXs and the broken lines respectively.The implementation and simulation experiment of a 4-bit serial-parallel multiplier are presented.The performance analysis of fault-tolerant time and hardware resources consumption show that the fault-tolerant performance of the interconnection circuits in new RH is improved greatly.关键词：reconfigurable hardware;in-chip fault-tolerant;distributed control;fault localization;self-repairing2|3|20更新时间：2025-07-16
- 摘要：In order to extent the lifetime of cognitive radio network based on clustered architecture,the role of cluster head need be changed for balance the energy consumption in cognitive radio network.A energy consumption model is proposed to describe the different between the cluster head nodes and common nodes,and then DTACRA (Double Thresholds Adaptive Cluster Rotation Algorithm) is presented,which double rotation energy thresholds are estimated using cluster head real-time energy load.The simulation results show that comparing with LEACH and EDAC,DTACRA maximizes the utilization of the nodes energy and prolongs the network lifetime.关键词：cognitive radio network;clustering;double thresholds rotation;energy-driven2|2|0更新时间：2025-07-16
- 摘要：The template-based classification ,which is considered as a conventional synthetic-aperture radar(SAR) automatic target recognition(ATR) approach,needs a stored set of templates of the targets at a reasonable number of different orientations.However,it results in the great consumption of computation and storage.Depending on the term,View-aspect,in Computer Vision and thus an approach is proposed to construct simplified prototype templates to replace the conventional library.The prototype templates are extracted by the Gustafson-Kessel(GK) algorithm with target key features according to the rules of optimizing recognition rate.Experiment was runned on simulating SAR data sets of typical ship targets and the results demonstrate that the approach can not only simplify the templates,but also improve the recognition performance effectively.关键词：synthetic aperture radar (SAR);automatic target recognition;view-aspect;fuzzy clustering2|2|5更新时间：2025-07-16
- 摘要：This paper proposes a new CABAC pipeline driving by syntax element instructions and a CABAC-based parallel arithmetic entropy coder for on-chip large-scale parallel H.264 video coders is also presented in this paper.Furthermore,the hardware architecture and transistor cost of the new parallel entropy coder are evaluated.This new parallel entropy coder can cooperate with the traditional symbol-level parallel algorithms and it suits the manycore platform well.Compared with the traditional CABAC hardware,the new parallel entropy coder can double its throughput to >1Gbins/s by a cost of 55% transistors.关键词：parallel video coder;arithmetic coder;VLSI implementation2|2|1更新时间：2025-07-16
- 摘要：In this paper,a medical image encryption algorithm is proposed.The algorithm is based on the design principle of AES,and combined with Skew tent map.The standard procedure of AES is modified so that it can be suitable for the DICOM medical digital images.At first,the block size 4*4 in AddRoundKey is replaced by the size of an image.Then the basic information of patient is protected by our proposed algorithm.In addition,the AES's MixColumns and key schedule are improved.The theoretical analysis and numerical simulations show that the improved algorithm not only has better confusion diffusion effects with less computational overhead,but also can maintain the compatibility of DICOM file format.关键词：medical image encryption;AES;digital imaging commucation in medicine (DICOM);chaotic system2|2|9更新时间：2025-07-16
- 摘要：A novel timing and frequency synchronization method by superimposed sequence with lower power was proposed for OFDM systems.The training sequence,which was constructed and had low/zero autocorrelation and weak energy performance,was superimposed with total energy accounted for only a transmitter total power 4.1 percent on the information symbol.The novel frequency offset estimation range using guard interval is enlarged from 0.5 to 1,while the total frequency offset estimation range is 1/4 bandwidth.Theory and simulation show that the performances of the proposed algorithm are much better than those of the conventional schemes in accuracy and the mean square error.关键词：orthogonal frequency division multiplexing (OFDM);synchronization;superimposed weak power sequence2|2|3更新时间：2025-07-16
- Technical support is provided by Beijing Founder electronics co., LTD 京ICP备09064830号-19
京公网安备11010802024621 - It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
- Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.
0