Speaker verification can be viewed as a task of modeling and testing two hypotheses: the null hypothesis and the alternative hypothesis. Since the alternative hypothesis involves unknown impostors, it is usually hard to characterizeapriori. In this paper, we propose improving the characterization of the alternative hypothesis by designing two decision functions based, respectively, on a weighted arithmetic combination and a weighted geometric combination of discriminative information derived from a set of pretrained background models. The parameters associated with the combinations are then optimized using two kernel discriminant analysis techniques, namely, the kernel Fisher discriminant (KFD) and support vector machine (SVM). The proposed approaches have two advantages over existing methods. The first is that they embed a trainable mechanism in the decision functions. The second is that they convert variable-length utterances into fixed-dimension characteristic vectors, which are easily processed by kernel discriminant analysis. The results of speaker-verification experiments conducted on two speech corpora show that the proposed methods outperform conventional likelihood ratio-based approaches.

Speaker verification based on the log-likelihood ratio (LLR) is essentially a task of modeling and testing two hypotheses: the null hypothesis and the alternative hypothesis. Since the alternative hypothesis involves unknown imposters, it is usually hard to characterize a priori. In this paper, we propose a framework to better characterize the alternative hypothesis with the goal of optimally separating client speakers from imposters. The proposed framework is built on either a weighted arithmetic combination or a weighted geometric combination of useful information extracted from a set of pre-trained anti-speaker models. The parameters associated with the combinations are then optimized using minimum verification error training such that both the false acceptance probability and the false rejection probability are minimized. Our experiment results show that the proposed framework outperforms conventional LLR-based approaches.

It is usually difficult to characterize the alternative hypothesis precisely in a log-likelihood ratio (LLR)-based speaker verification system. In a previous work, we proposed using a weighted arithmetic combination (WAC) or a weighted geometric combination (WGC) of the likelihoods of the background models instead of heuristic combinations, such as the arithmetic mean and the geometric mean, to better characterize the alternative hypothesis. In this paper, we further propose learning the parameters associated with WAC or WGC via an evolutionary minimum verification error (MVE) training method, such that both the false acceptance probability and the false rejection probability can be minimized. Our experiment results show that the proposed methods outperform conventional LLR-based approaches. Index Terms: genetic algorithm, log-likelihood ratio, minimum verification error training, speaker verification

The performance of a likelihood ratio-based speaker verification system is highly dependent on modeling of the target speaker’s voice (the null hypothesis) and characterization of non-target speakers ’ voices (the alternative hypothesis). To better characterize the ill-defined alternative hypothesis, this study proposes a new likelihood ratio measure based on a composite-structure Gaussian mixture model, the so-called GMM2. Motivated by the combined use of a variety of background models to represent the alternative hypothesis, GMM2 is designed with an inner set of mixture weights connected to the significance of each individual Gaussian density, and an outer set of mixture weights connected to the significance of each individual background model. Through the use of kernel discriminant analysis namely, Kernel Fisher Discriminant (KFD) or Support Vector Machine (SVM), GMM2 is trained in such a manner that the utterances of the null hypothesis can be optimally separated from those of the alternative hypothesis. Index Terms: speaker verification, likelihood ratio, kernel Fisher discriminant, support vector machine

Clearly, the linear discriminant classifier is not robust enough to cope with most real-world data classification problems. Kernel Fisher discriminant analysis (KFDA) tries to increase the expressiveness of the discriminant based on the high order statistics of the data set. In this paper, we propose the GMM-based KFDA with the Bhattacharyya kernel to obtain a transformation, called a speaker eigenspace, based on which the transformed MFCC features are more discriminative for speaker recognition. In our approach, the eigenspace is directly constructed from the complete GMM parameter set, rather than the supervectors considering mean vectors only as in the eigenvoice approach. Moreover, FDA, which is believed to be more appropriate for classification accuracy than principal component analysis (PCA), is applied for eigenspace construction. The speaker identification experiments show that the new features outperform the MFCC features, in particular when the amount of enrollment data for each speaker is very small.

This paper proposes the development of a dynamic-reconfigurable protocol stack, which allows the programmer to create, to remove, and to replace protocol modules during their operation. Moreover, this protocol stack also aims to preserve the module state, such as the data structures that manage the existing connections. To achieve these goals, a Java-based component framework is developed so that the programmers are able to implement their components under the proposed framework. This framework can dynamically reconfigure the components at a safe period and can help the components transfer their states, and the dynamic reconfiguration is transparent to the user application running on top of the stack. To demonstrate the component framework, a TCP component is implemented. While maintaining active connections for the user application, the TCP component is able to be dynamically replaced by another version. Copyright © 2005 John Wiley & Sons, Ltd.

Previous real-time disk scheduling algorithms assume that each disk request incurs a physical disk mechanical operation and only consider how to move the disk head under real-time constraints. However, with the increased capacity of on-disk cache, modern disk drives read-ahead data aggressively. Thus, the on-disk cache may service many disk requests without incurring physical disk access. By exploring the design methodology of on-disk cache, in this paper, we propose cache-aware real-time disk scheduling algorithms that take the on-disk cache into consideration during scheduling. Therefore, the scheduling algorithm can help the cache replacement scheme to minimize the cache miss ratio. Besides, the service timing estimation is more accurate in schedulability analysis since the cache effect is considered during scheduling. A simulation-based evaluation shows the proposed scheduling algorithms to be highly successful as compared with the classical real-time disk scheduling algorithms. For example, under sequential workload with 10 sequential streams, the data throughput of our scheme is 1.1 times that of DM-SCAN.

In this paper, we design and implement a Bluetooth system with mobile LAN access capability. First, by adding a virtual network interface card (NIC) driver between IP and Bluetooth protocol stack, Bluetooth users can use their Bluetooth devices to access the LAN resources via the Bluetooth access points. Besides, to accommodate with the mobility characteristics in Bluetooth network, we also propose a new handoff scheme. Instead of passively waiting for a handoff indication by the expiration of link supervision timer, in this paper, we detect a handoff actively and promptly by measuring the RSSI (receive signal strength indication) value. In addition, our proposed handoff scheme eliminates the inquiry procedure and only requires paging procedure during a handoff. We have implemented a prototype of Bluetooth system with mobile LAN access capability by our proposed scheme. Experimented results show that the handoff duration is significant reduced than the normal scheme.

The objective of software radio is reconfigurability - the ability to dynamically configure software and hardware functionalities of mobile terminals. To achieve this, the concept of software download is emerging. With software download, a mobile terminal can be reconfigured using downloadable modules instead of saving all the software in its non-volatile memory. This paper presents the design of a configurable Java architecture. With this architecture, we may dynamically build secure and flexible execution environments for different software requirements. The implementation of the configurable Java virtual architecture is based on K virtual machine (KVM) and the constructed execution environment supports user application, man-machine interface (MMI), and protocol stack reconfigurability.

As network-enabled embedded devices and Java grow in their popularity, embedded system researchers start seeking ways to make these devices Java-enabled. However, it is a challenge to apply Java technology to these devices due to their shortage of resources. In this paper, we propose EJVM (Economic Java Virtual Machine), an economic way to run Java programs on network-enabled and resource-limited embedded devices. Espousing the architecture proposed by distributed JVM, we store all Java codes on the server to reduce the storage needs of the client devices. In addition, we use two novel techniques to reduce the client-side memory footprints: server-side class representation conversion and on-demand bytecode loading. Finally, we maintain client-side caches and provide performance evaluation on different caching policies. We implement EJVM by modifying a freely available JVM implementation, Kaffe. From the experiment results, we show that EJVM can reduce Java heap requirements by about 20–50% and achieve 90% of the original performance. Copyright © 2001 John Wiley & Sons, Ltd.

In a multimedia system, disk I/O subsystem is the most important component due to its relatively limited throughput and large delay. Previously, by applying SCAN to reschedule tasks having the same deadline, SCAN-EDF tries to improve disk throughput while real time constraints can be satisfied. In DM-SCAN, groups of tasks that can be successfully rescheduled by SCAN under specified real time requirements are identified. They are called MSGs (maximum-scannable-groups). An enlarged-MSG (E-MSG) is proposed to further expand the MSG concept and thus to obtain more improvement in disk throughput. By removing some excess constraints on MSG, E-MSG merges several MSGs as a new scannable group. Experimental results show that the E-MSG scheme is better than both SCAN-EDF and MSG in the disk throughput obtained.

Real-time disk scheduling is one of the most important problems in designing a multimedia system. It has been proved to be NP-complete. Recently, various approaches have been proposed to improve disk throughput under guaranteed real-time requirements. SCAN-EDF, which scans the disk surface to retrieve the task data block under the disk head in order to re-schedule tasks in a real-time EDF (earliest deadline first) schedule, is one of the best-known real-time disk scheduling methods. Since tasks rescheduled in SCAN-EDF should have the same deadline, its efficiency depends on the number of tasks with the same deadline. If all tasks have different deadlines, the scheduling results of SCAN-EDF would be the same as EDF. In this paper, we improve SCAN-EDF by applying different hybrid local-merging and global-inserting schemes. As opposed to SCAN-EDF, in our method tasks rescheduled by SCAN may have different deadlines. Its efficiency is not limited by the number of tasks that have the same deadlines. Experiments show that the proposed method is significantly better than SCAN-EDF. In terms of disk throughput, the improvement obtained is 24% greater than the best-known SCAN-EDF method.

Both numerical and subdivision methods are widely used approaches for ray tracing parametric surfaces. However, the expense of finding the ray–surface intersection points is a major drawback. Thus, simpler and less memory-intensive strategies are needed to improve these methods without further complicating them. This work presents an efficient algorithm for enhancing the performance of both numerical and subdivision methods. The proposed technique can be extended to most applications based on these two methods. The computational time of both approaches is improved by 16–40%. Copyright © 2000 John Wiley & Sons, Ltd.

Flash memory offers attractive features for storage of data, such as non-volatility, shock resistance, fast access speed, and low power consumption. However, it requires erasing before it can be overwritten. The erase operations are slow and consume comparatively a great deal of power. Furthermore, flash memory can only be erased a limited number of times. To overcome hardware limitations, we use the non-in-place update mechanism that requires a cleaner to reclaim space occupied by obsolete data. To improve cleaning performance and prolong flash memory lifetime, we propose a new data reorganization method. By this method, data in flash memory are dynamically classified and clustered together according to their accessing frequencies. Experimental results show that this clustering technique significantly improved the cleaning performance for a variety of cleaning algorithms. The number of erase operations performed is greatly reduced and flash memory lifetime is prolonged. Even wearing is ensured as well. Copyright © 1999 John Wiley & Sons, Ltd.

Process checkpointing is a procedure which periodically saves the process states into stable storage. Most checkpointing facilities select hard disks for archiving. However, the disk seek time is limited by the speed of the read-write heads, thus checkpointing process into a local disk requires extensive disk bandwidth. In this paper, we propose an approach that exploits the memory on idle workstations as a faster storage for checkpointing. In our scheme, autonomous machines which submit jobs to the computation server offer their physical memory to the server for job checkpointing. Eight applications are used to measure the remote memory performance in four checkpointing policies. Experimental results show that remote memory reduces at least 34.5 per cent of the overhead for sequential checkpointing and 32.1 per cent for incremental checkpointing. Additionally, to checkpoint a running process into a remote memory requires only 60 per cent of the local disk checkpoint latency time. Copyright © 1999 John Wiley & Sons, Ltd.

Real-time disk scheduling is important to multimedia systems support for digital audio and video. In these years, various approaches are presented to use the seek-optimizing scheme to improve the disk throughput of a real-time guaranteed schedule. However, as these conventional approaches apply SCAN only to the requests with the same deadline or within the same constant-sized group, their improvements are limited. In this paper, we introduce the DM-SCAN (deadline-modification-SCAN) algorithm with an idea of MSG (maximum-scannable-group). The proposed DM-SCAN method can apply SCAN to MSG iteratively by modifying request deadlines. We have implemented the DM-SCAN algorithm on UnixWare 2.01. Experiments show that DM-SCAN is significantly better than that of the best-known SCAN-EDF method in both the obtained disk throughput and the number of supported disk requests.

Checkpointing is a basic mechanism for backward error-recovery in fault-tolerant systems. A checkpointed process stops execution and saves its states to files periodically. To reduce the file sizes, only data modified between two consecutive checkpoint times is saved. However, existing approaches do not consider operating system paging activities; which, if ignored may double the number of disk accesses required to checkpoint non-resident dirty pages. In this paper, we propose continuous checkpointing, which combines the checkpoint facility with virtual memory paging operations. Thus, checkpointing is continuous during the lifetime of a process without extra overhead. Checkpoint size is no longer proportional to application size, but rather is bounded by resident dirty pages. Experimental results show that disk accesses can be reduced by about 80% when checkpointing large applications. © 1997 John Wiley & Sons, Ltd.

Conventional operating systems employ a kernel-controlled caching strategy that cannot properly serve all access-pattern types used by applications. When running under these systems, many memory-intensive applications with mis-matching access patterns cause excessive page faults and page replacements that reduce the application's performance. This paper presents the hipec system, which allows applications to have their own caching strategies for managing page frames with negligible overhead. Since application designers know the access patterns of their applications, the specific caching strategies can be tuned to meet the needs of each application. Empirical results show that the hipec system significantly improves application performance and system throughput. © 1997 John Wiley & Sons, Ltd.

Abstract The texture of feather is one of the most fascinating, complicated, and beautiful texture patterns in nature. In this paper, we propose a new and effective texture generation approach that uses the traits of iteration behavior to synthesize the textures of Galliformes feathers realistically. We also propose an interactive feather modeling approach which provides a close connection between the user's intuition and the resulting branching pattern. In texturing the feather structure, we use an object‐space mapping technique. Experimental results are presented to show the effectiveness of our method.

This paper presents a parallel file system, called Pasda, that not only addresses the problem of data rate mismatches between the requirements of an application, storage devices, and the inter-connection medium but also supports parallel file accesses in general purpose distributed systems. The Pasda (PArallel Storage Devices Accelerator) parallel file system uses a high-speed interconnection medium to provide high data transfer rates by driving multiple slower storage devices in parallel. A good parallel file accessing interface is also supported by Pasda. The user of Pasda can control the physical layout and partitioning of the file data, and can access files in parallel. An FDDI-based prototype of Pasda is constructed. The prototype provides faster data rates than access to the local SCSI disk.

Abstract The dynamic constraints technique has been proposed for building geometrical models composed of rigid bodies, which are made to act naturally, according to Newtonian laws, by specifying constraints on their states. In computer animation, the dynamic constraints technique alleviates the work‐load of animators who formerly had to plan animated sequences in detail by intuition alone. Nevertheless, for some real‐world applications, it is desirable to have a mechanism that makes physically‐based elements move according to a given scenario by providing some control states. The control states can be represented by transient constraints that are to be met and then released immediately. In this paper, a technique called the sequential‐goal constraints technique is proposed to provide such a mechanism. With the sequential‐goal constraints technique, it is easy to specify transient constraints according to a given scenario and derive proper forces and torques to drive an element to meet each transient constraint exactly at a specified time so that the whole motion of the element is continuous and integral.