With rapid development of Graphics Processing Units (GPU) technologies, GPUs are strongly motivated to be adopted in many real-time applications. However, it is still a challenging work to efficiently integrate multiple GPUs into multicore/multiprocessor real-time systems, due to many real world constraints caused by the non-real-time closed-source GPU drivers. To avoid timing violations, k-exclusive locking protocols are developed to arbitrate exclusive access to each of the multiple GPUs. In this paper, a novel k-exclusion real-time locking protocol is proposed to handle multi-GPU sharing under partitioned fixed priority (P-FP) scheduling. The proposed protocol improves the prior work, the Clustered k-exclusion O(m) Locking Protocol (CK-OMLP) from two aspects: first, it allows multiple task on each CPU processor to make use of GPUs simultaneously, which improves the flexibility and increases GPU utilization in average cases, second, a suspension-aware analysis is presented to improve the schedulability, where task acquisition delays and GPU executions are modeled as self-suspensions. Schedulability experiments indicate that the proposed protocol outperforms the CK-OMLP in most considered scenarios.
{"title":"PK-OMLP: An OMLP Based k-Exclusion Real-Time Locking Protocol for Multi-GPU Sharing under Partitioned Scheduling","authors":"Maolin Yang, Hang Lei, Yong Liao, Furkan Rabee","doi":"10.1109/DASC.2013.63","DOIUrl":"https://doi.org/10.1109/DASC.2013.63","url":null,"abstract":"With rapid development of Graphics Processing Units (GPU) technologies, GPUs are strongly motivated to be adopted in many real-time applications. However, it is still a challenging work to efficiently integrate multiple GPUs into multicore/multiprocessor real-time systems, due to many real world constraints caused by the non-real-time closed-source GPU drivers. To avoid timing violations, k-exclusive locking protocols are developed to arbitrate exclusive access to each of the multiple GPUs. In this paper, a novel k-exclusion real-time locking protocol is proposed to handle multi-GPU sharing under partitioned fixed priority (P-FP) scheduling. The proposed protocol improves the prior work, the Clustered k-exclusion O(m) Locking Protocol (CK-OMLP) from two aspects: first, it allows multiple task on each CPU processor to make use of GPUs simultaneously, which improves the flexibility and increases GPU utilization in average cases, second, a suspension-aware analysis is presented to improve the schedulability, where task acquisition delays and GPU executions are modeled as self-suspensions. Schedulability experiments indicate that the proposed protocol outperforms the CK-OMLP in most considered scenarios.","PeriodicalId":179557,"journal":{"name":"2013 IEEE 11th International Conference on Dependable, Autonomic and Secure Computing","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133097046","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In real applications, there are two problems for the periodic patterns mining task: finding the frequent pattern(s) and determining their periodicity. In this paper, we propose a new method to investigate the periodic patterns form common frequent patterns. First, all the candidates patterns are generated by general frequent pattern mining algorithm. Then, for each pattern, all the time (order) attributes are extracted form its support records. Finally, all these time (order) attributes are partitioned into suitable n periods to obtain the feasible periodicity. To this end, two new parameters of per and fea are introduced to measure the periodicity and feasibility of the candidate patterns. The experiment results show that the method can be used to explore feasible periodic patterns efficiently and find some interesting patterns in business database.
{"title":"On Discovering Feasible Periodic Patterns in Large Database","authors":"Xiao Luo, Hua Yuan, Qian Luo","doi":"10.1109/DASC.2013.87","DOIUrl":"https://doi.org/10.1109/DASC.2013.87","url":null,"abstract":"In real applications, there are two problems for the periodic patterns mining task: finding the frequent pattern(s) and determining their periodicity. In this paper, we propose a new method to investigate the periodic patterns form common frequent patterns. First, all the candidates patterns are generated by general frequent pattern mining algorithm. Then, for each pattern, all the time (order) attributes are extracted form its support records. Finally, all these time (order) attributes are partitioned into suitable n periods to obtain the feasible periodicity. To this end, two new parameters of per and fea are introduced to measure the periodicity and feasibility of the candidate patterns. The experiment results show that the method can be used to explore feasible periodic patterns efficiently and find some interesting patterns in business database.","PeriodicalId":179557,"journal":{"name":"2013 IEEE 11th International Conference on Dependable, Autonomic and Secure Computing","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133150966","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper, a new system for license plate characters recognition is proposed, on the basis of a novel characters recognition algorithm called sliding window search. This system includes license plate location, license plate characters pre-segmentation and characters recognition based on sliding window search. The Gaussian mixture model is used for background modeling and a Bayesian classifier based on the target tracking technology is used to monitor all vehicles within the scene. The brightness difference between the license plate characters and background is considered. Based on the priori information about the characters width, the license plate characters pre-segmentation is easily acquired. Finally, sliding window search is conducted on the license plate characters. Compared with other license plate characters recognition systems, this system has a high recognition accuracy and has good robustness even in a complex environment.
{"title":"A New Method for License Plate Characters Recognition Based on Sliding Window Search","authors":"Ying-Jia Bu, M. Xie","doi":"10.1109/DASC.2013.79","DOIUrl":"https://doi.org/10.1109/DASC.2013.79","url":null,"abstract":"In this paper, a new system for license plate characters recognition is proposed, on the basis of a novel characters recognition algorithm called sliding window search. This system includes license plate location, license plate characters pre-segmentation and characters recognition based on sliding window search. The Gaussian mixture model is used for background modeling and a Bayesian classifier based on the target tracking technology is used to monitor all vehicles within the scene. The brightness difference between the license plate characters and background is considered. Based on the priori information about the characters width, the license plate characters pre-segmentation is easily acquired. Finally, sliding window search is conducted on the license plate characters. Compared with other license plate characters recognition systems, this system has a high recognition accuracy and has good robustness even in a complex environment.","PeriodicalId":179557,"journal":{"name":"2013 IEEE 11th International Conference on Dependable, Autonomic and Secure Computing","volume":"326 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122440799","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Video caching has gained significant attention due to its cost efficiency. In this paper, we propose an optimized cache replacement scheme based on the user request arrival pattern. The users are grouped among all possible request intervals, and the user density is calculated in each interval. The groups with highest user density are cached. Whenever a group with high user density arrives, the groups with lower density will be replaced. The simulation results show that the proposed schemes can double the hit ratio compared with the Least recently Used (LRU) scheme.
{"title":"Optimized Cache Replacement Scheme for Video on Demand Service","authors":"Xiaocui Sun, Zhijun Wang","doi":"10.1109/DASC.2013.61","DOIUrl":"https://doi.org/10.1109/DASC.2013.61","url":null,"abstract":"Video caching has gained significant attention due to its cost efficiency. In this paper, we propose an optimized cache replacement scheme based on the user request arrival pattern. The users are grouped among all possible request intervals, and the user density is calculated in each interval. The groups with highest user density are cached. Whenever a group with high user density arrives, the groups with lower density will be replaced. The simulation results show that the proposed schemes can double the hit ratio compared with the Least recently Used (LRU) scheme.","PeriodicalId":179557,"journal":{"name":"2013 IEEE 11th International Conference on Dependable, Autonomic and Secure Computing","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114825123","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Mingjun Du, Xiaoxiang Wang, Dongyu Wang, Yulong Wang
Device-to-Device (D2D) communications make high-speed multicast services possible since the multicast receivers with poor downlink channel conditions can be retransmitted by devices nearby via D2D links. In this paper, we consider how to efficiently use D2D communications to help enhance the quality of wireless multicast services in cellular networks. To achieve this, a dynamic D2D retransmission scheme with maximized utility is proposed, which can adaptively select the retransmission algorithm according to the state of the network load. Through both analysis and simulations, we show that our algorithms achieve a significant gain in terms of utility, and reduce the burden of the base station (BS).
{"title":"Device-to-Device Dynamic Clustering Algorithm in Multicast Communication","authors":"Mingjun Du, Xiaoxiang Wang, Dongyu Wang, Yulong Wang","doi":"10.1109/DASC.2013.129","DOIUrl":"https://doi.org/10.1109/DASC.2013.129","url":null,"abstract":"Device-to-Device (D2D) communications make high-speed multicast services possible since the multicast receivers with poor downlink channel conditions can be retransmitted by devices nearby via D2D links. In this paper, we consider how to efficiently use D2D communications to help enhance the quality of wireless multicast services in cellular networks. To achieve this, a dynamic D2D retransmission scheme with maximized utility is proposed, which can adaptively select the retransmission algorithm according to the state of the network load. Through both analysis and simulations, we show that our algorithms achieve a significant gain in terms of utility, and reduce the burden of the base station (BS).","PeriodicalId":179557,"journal":{"name":"2013 IEEE 11th International Conference on Dependable, Autonomic and Secure Computing","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123646063","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Fast Fourier Transform (FFT) algorithm has an important role in the image processing and scientific computing, and it's a highly parallel divide-and-conquer algorithm. In this paper, we exploited the Compute Unified Device Architecture CUDA technology and contemporary graphics processing units (GPUs) to achieve higher performance. We focused on two aspects to optimize the ordinary FFT algorithm, multi-threaded parallelism and memory hierarchy. We also proposed parallelism optimization strategies when the data volume occurs and predicted the possible situation when the amount of data increased further.it can be seen from the results that Parallel FFT algorithm is more efficient than the ordinary FFT algorithm.
{"title":"Design and Implementation of Parallel FFT on CUDA","authors":"Xueqin Zhang, K. Shen, Cheng-Hai Xu, K. Wang","doi":"10.1109/DASC.2013.130","DOIUrl":"https://doi.org/10.1109/DASC.2013.130","url":null,"abstract":"Fast Fourier Transform (FFT) algorithm has an important role in the image processing and scientific computing, and it's a highly parallel divide-and-conquer algorithm. In this paper, we exploited the Compute Unified Device Architecture CUDA technology and contemporary graphics processing units (GPUs) to achieve higher performance. We focused on two aspects to optimize the ordinary FFT algorithm, multi-threaded parallelism and memory hierarchy. We also proposed parallelism optimization strategies when the data volume occurs and predicted the possible situation when the amount of data increased further.it can be seen from the results that Parallel FFT algorithm is more efficient than the ordinary FFT algorithm.","PeriodicalId":179557,"journal":{"name":"2013 IEEE 11th International Conference on Dependable, Autonomic and Secure Computing","volume":"126 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131911614","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The serial concatenated convolutional code (SCCC) structure of Turbo code shows excellent decoding performance in the region of moderate-to-high signal-noise-ratio (SNR), which has been widely applied to many fields, such as deep space communication and so on. So, the recognition of SCCC turbo code is very important in the cognitive radio and information interception. We firstly estimated the parameters of outer encoder by Gaussian elimination and Euclidean algorithm. Then, the proposed method exploits the invariance property between the original data and interleaved data. Parameters of interleaver can be estimated easily by calculating the correlation between the check bits and the data. The computer simulations show that the proposed method has good performance when errors exist in the interception data.
{"title":"A Method of Parameters Estimation of SCCC Turbo Code","authors":"Dan Li, L. Gan","doi":"10.1109/DASC.2013.29","DOIUrl":"https://doi.org/10.1109/DASC.2013.29","url":null,"abstract":"The serial concatenated convolutional code (SCCC) structure of Turbo code shows excellent decoding performance in the region of moderate-to-high signal-noise-ratio (SNR), which has been widely applied to many fields, such as deep space communication and so on. So, the recognition of SCCC turbo code is very important in the cognitive radio and information interception. We firstly estimated the parameters of outer encoder by Gaussian elimination and Euclidean algorithm. Then, the proposed method exploits the invariance property between the original data and interleaved data. Parameters of interleaver can be estimated easily by calculating the correlation between the check bits and the data. The computer simulations show that the proposed method has good performance when errors exist in the interception data.","PeriodicalId":179557,"journal":{"name":"2013 IEEE 11th International Conference on Dependable, Autonomic and Secure Computing","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133409713","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yunong Zhang, Dechao Chen, Long Jin, Y. Wang, Feiheng Luo
This paper proposes a novel multi-input Bernoulli-polynomial neuronet (MIBPN) on the basis of function approximation theory. The MIBPN is trained by a weights-and-structure-determination (WASD) algorithm with twice pruning (TP). The WASD algorithm can obtain the optimal weights and structure for the MIBPN, and overcome the weaknesses of conventional BP (back-propagation) neuronets such as slow training speed and local minima. With the TP technique, the neurons of less importance in the MIBPN are pruned for less computational complexity. Furthermore, this MIBPN can be extended to a multiple input multiple output Bernoulli-polynomial neuronet (MIMOBPN), which can be applied as an important tool for classification. Numerical experiment results show that the MIBPN has outstanding performance in data approximation and generalization. Besides, experiment results based on the real-world classification data-sets substantiate the high accuracy and strong robustness of the MIMOBPN equipped with the proposed WASD algorithm for classification. Finally, the twice-pruning aided WASD neuronet of Bernoulli-polynomial type in the forms of MIBPN and MIMOBPN is established, together with the effective extension to robust classification.
{"title":"Twice-Pruning Aided WASD Neuronet of Bernoulli-Polynomial Type with Extension to Robust Classification","authors":"Yunong Zhang, Dechao Chen, Long Jin, Y. Wang, Feiheng Luo","doi":"10.1109/DASC.2013.85","DOIUrl":"https://doi.org/10.1109/DASC.2013.85","url":null,"abstract":"This paper proposes a novel multi-input Bernoulli-polynomial neuronet (MIBPN) on the basis of function approximation theory. The MIBPN is trained by a weights-and-structure-determination (WASD) algorithm with twice pruning (TP). The WASD algorithm can obtain the optimal weights and structure for the MIBPN, and overcome the weaknesses of conventional BP (back-propagation) neuronets such as slow training speed and local minima. With the TP technique, the neurons of less importance in the MIBPN are pruned for less computational complexity. Furthermore, this MIBPN can be extended to a multiple input multiple output Bernoulli-polynomial neuronet (MIMOBPN), which can be applied as an important tool for classification. Numerical experiment results show that the MIBPN has outstanding performance in data approximation and generalization. Besides, experiment results based on the real-world classification data-sets substantiate the high accuracy and strong robustness of the MIMOBPN equipped with the proposed WASD algorithm for classification. Finally, the twice-pruning aided WASD neuronet of Bernoulli-polynomial type in the forms of MIBPN and MIMOBPN is established, together with the effective extension to robust classification.","PeriodicalId":179557,"journal":{"name":"2013 IEEE 11th International Conference on Dependable, Autonomic and Secure Computing","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130274143","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Probability density function (PDF) plays a vital role in many applications involving stochastic process. A good approximation for real-time PDF conditioned on certain performance criterion could help to acquire unknown information about the system. With the help of this kind of information, which was not available earlier, many features of various models that describe the real system can be estimated effectively, especially for non-linear non-Gaussian stochastic system. In this paper, we elucidate some PDFs with only one parameter that have a definite physical meaning based on Tsallis entropy. The PDFs that we calculated here are all Gaussian-like, and Gaussian distribution is attained when the parameter of Tsallis entropy approaches zero. Based on these explicit form of Gaussian-like PDFs we calculated here, an extension of Gaussian particle filter (GPF) called Gaussian-like particle filter (GLPF) is proposed and the simulation results show that the GLPF is a more effective way to estimate the state of non-linear stochastic system compared with the GPF.
{"title":"A New Gaussian-Like Density Model and Its Application to Object-Tracking","authors":"Xifeng Li, Yongle Xie","doi":"10.1109/DASC.2013.124","DOIUrl":"https://doi.org/10.1109/DASC.2013.124","url":null,"abstract":"Probability density function (PDF) plays a vital role in many applications involving stochastic process. A good approximation for real-time PDF conditioned on certain performance criterion could help to acquire unknown information about the system. With the help of this kind of information, which was not available earlier, many features of various models that describe the real system can be estimated effectively, especially for non-linear non-Gaussian stochastic system. In this paper, we elucidate some PDFs with only one parameter that have a definite physical meaning based on Tsallis entropy. The PDFs that we calculated here are all Gaussian-like, and Gaussian distribution is attained when the parameter of Tsallis entropy approaches zero. Based on these explicit form of Gaussian-like PDFs we calculated here, an extension of Gaussian particle filter (GPF) called Gaussian-like particle filter (GLPF) is proposed and the simulation results show that the GLPF is a more effective way to estimate the state of non-linear stochastic system compared with the GPF.","PeriodicalId":179557,"journal":{"name":"2013 IEEE 11th International Conference on Dependable, Autonomic and Secure Computing","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128236454","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper, the fairness problem of opportunistic multicast scheduling (OMS) is investigated and a parity-based OMS is proposed. In conventional multicast system, the system throughput is limited by the users with poor channel state information (CSI). In the proposed OMS, to improve the system throughput performance, the data reception of each user is taken into consideration by the base station (BS). Each user is assigned two aspects of priority, one of them is based on the system throughput, and the other is based on the data reception of users. For the users with poor channel condition, fewer packets would be received than the users with fine channel conditions. Thus, by raising the priority level of the user with worse channel condition, the performance of fairness among users can be improvement. Simulation results show that the proposed OMS can improve the system throughput performance while ensuring the fairness performance among users.
{"title":"A Parity-Based Opportunistic Multicast Scheduling Scheme over Cellular Networks","authors":"Yulong Wang, Xiaoxiang Wang, Mingming Li, Jing Qu","doi":"10.1109/DASC.2013.126","DOIUrl":"https://doi.org/10.1109/DASC.2013.126","url":null,"abstract":"In this paper, the fairness problem of opportunistic multicast scheduling (OMS) is investigated and a parity-based OMS is proposed. In conventional multicast system, the system throughput is limited by the users with poor channel state information (CSI). In the proposed OMS, to improve the system throughput performance, the data reception of each user is taken into consideration by the base station (BS). Each user is assigned two aspects of priority, one of them is based on the system throughput, and the other is based on the data reception of users. For the users with poor channel condition, fewer packets would be received than the users with fine channel conditions. Thus, by raising the priority level of the user with worse channel condition, the performance of fairness among users can be improvement. Simulation results show that the proposed OMS can improve the system throughput performance while ensuring the fairness performance among users.","PeriodicalId":179557,"journal":{"name":"2013 IEEE 11th International Conference on Dependable, Autonomic and Secure Computing","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133960762","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：