Albert M. K. Cheng

Assigning real-time tasks to heterogeneous processors by applying ant colony optimization

The problem of determining whether a set of periodic tasks can be assigned to a set of heterogeneous processors without deadline violations has been shown, in general, to be NP-hard. This paper presents a new algorithm based on ant colony optimization (ACO) metaheuristic for solving this problem. A local search heuristic that can be used by various metaheuristics to improve the assignment solution is proposed and its time and space complexity is analyzed. In addition to being able to search for a feasible assignment solution, our extended ACO algorithm can optimize the solution by lowering its energy consumption. Experimental results show that both the prototype and the extended version of our ACO algorithm outperform major existing methods; furthermore, the extended version achieves an average of 15.8% energy saving over its prototype.

Analysis of real-time rule-based systems with behavioral constraint assertions specified in Estella

Rule-based expert systems are increasingly used to monitor and control the operations of complex real-time systems which require intensive knowledge-decision processing and human expertise. These embedded AI systems must respond to events in the rapidly changing external environment so that the results of the expert systems computation in each monitor-respond cycle are valid in safely operating the real-time system. Determining how fast an expert system can respond under all possible situations is a difficult problem. We have developed an efficient analysis methodology for a large class of rule-based EQL programs to determine whether a program in this class has bounded response time. In particular, we have identified several sets of primitive behavioral constraint assertions: an EQL program which satisfies all constraints in one of these sets of assertions is guaranteed to have bounded response time. Here, we enhance the applicability of our analysis technique by introducing a facility with which the rule-based programmer can specify application-specific knowledge that is too difficult to be mechanically detected in the new language Estella in order to determine the performance of an even wider range of programs. We also describe efficient algorithms for implementing the analysis tools. >

Applying Ant Colony Optimization to the partitioned scheduling problem for heterogeneous multiprocessors

The problem of determining whether a set of periodic tasks can be assigned to a set of heterogeneous processors in such a way that all timing constraints are met has been shown, in general, to be NP-hard. This paper presents a new algorithm based on Ant Colony Optimization (ACO) metaheuristic for solving this problem. Experimental results show that our ACO approach can outperform the major existing methods. In addition to being able to search for a feasible assignment solution, our ACO approach can further optimize the solution to reduce its energy consumption.The problem of determining whether a set of periodic tasks can be assigned to a set of heterogeneous processors in such a way that all timing constraints are met has been shown, in general, to be NP-hard. This paper presents a new algorithm based on Ant Colony Optimization (ACO) metaheuristic for solving this problem. Experimental results show that our ACO approach can outperform the major existing methods. In addition to being able to search for a feasible assignment solution, our ACO approach can further optimize the solution to reduce its energy consumption.

A digital map/GPS based routing and addressing scheme for wireless ad-hoc networks

As intelligent transportation systems move towards advanced communication systems, it is fair to assume that every vehicle will be equipped with a digital map database, GPS receiver and 802.11 based radio in the near future. Ad-hoc networks by definition are infrastructure-less and are formed on the fly. Mobile nodes in the form of vehicles can effectively form ad-hoc networks. In this paper we present a routing algorithm and an addressing scheme for such a class of ad-hoc networks, which utilizes digital maps and the geographical position of the nodes. It can be shown that, for this class of ad-hoc networks the proposed scheme will outperform traditional MANET protocols.

ANDES: an Anomaly Detection System for Wireless Sensor Networks

In this paper, we propose ANDES, a framework for detecting and finding the root causes of anomalies in operational wireless sensor networks (WSNs). The key novelty of ANDES is that it correlates information from two sources: one in the data plane as a result of regular data collection in WSNs, the other in the management plane implemented via a separate routing protocol, making it resilient to routing anomaly in the data plane. Evaluation using a 32-node sensor testbed shows that ANDES is effective in detecting fail-stop failures and most routing anomalies with negligible computing and storage overhead.

An imprecise algorithm for real-time compressed image and video transmission

One major requirement for multimedia systems is the efficient transmission of multimedia information over a communication or computer network. Several commercial products heave been developed to apply compression techniques to images and videos in order to reduce their transmission time. Sufficient available time and bandwidth lead to satisfactory results, but image/video loss occurs if these resources are insufficient. Using imprecise computation theory, we can achieve a good tradeoff between the quality of the transmitted image/video and the available resources such as the time for transmission. However, for a compressed image file, this very flexible technique cannot be easily applied since a compressed file cannot be uncompressed if it is not transmitted completely. This paper proposes an imprecise compressed image/video transmission technique which combines the advantages of both compression and imprecise computation for the first time. Our experimental results show that the proposed algorithm has better results compared to simply using compression or imprecise transmission techniques, and has a good potential for commercial use.

Response time analysis of OPS5 production systems

The paper focuses on the problem of determining a priori the maximal response time of rule based programs. The response time analysis problem is an important problem, especially for real time systems. We study this problem in the context of OPS5 production systems. Two aspects of the response time of a program are investigated, the maximal number of rule firings and the maximal number of basic comparisons made by the Rete network during the execution of the program. The response time analysis problem is in general undecidable. However, a program terminates in a finite time if the rule triggering pattern of this program satisfies certain conditions. We present four such termination conditions for OPS5 production systems. An algorithm for computing an upper bound on the number of rule firings is then given. To have a better idea of the time required during execution, we present an algorithm that computes the maximal time required during the match phase in terms of the number of comparisons made by the Rete network. This measurement is sufficient since the match phase consumes about 90 percent of the execution time.

MRL: a real-time rule-based production system

The response time analysis of rule-based expert systems is discussed. The rule-based production system MRL (macro-rule-based language) is introduced. MRL has been designed to facilitate more accurate analysis of the response times of programs while maintaining the flexibility and expressiveness of traditional production systems such as OPS5. Research on modular analysis of rule-based systems is described. Several timing analysis algorithms based on this approach have been developed. One of them, a fixed-point detection algorithm is discussed to show that efficient and effective analysis of MRL programs can be achieved. In particular, a general technique called the transfer principle is introduced for exploiting analysis algorithms which are simpler to analyze. The design of the match algorithm Rhyme, an algorithm uniquely suited for more accurate analysis of the performance of real-time expert systems, is presented.<<ETX>>

Multisite co-allocation algorithms for computational grid

Efficient multisite job scheduling facilitates the cooperation of multi-domain massively parallel processor systems in a computing grid environment. However, co-allocation, heterogeneity, adaptability, and scalability emerge as tough challenges for the design of multisite job scheduling models and algorithms. This paper presents a new multisite job scheduling schema based on the multisite job scheduling model and the performance model for a heterogeneous grid environment. There are three key components: resource selection, reservation, and backfilling. The optimal and greedy-heuristic adaptive resource selection strategies are introduced. The conservative and easy backfilling are incorporated into the backfilling procedure. Experiments indicate that the scheduler and the algorithm are effective and perform better than a non-adaptive algorithm

Shortening matching time in OPS5 production systems

A rule-based system must satisfy stringent timing constraints when applied to a real-time environment. As the scale of rule-based expert systems increases, the efficiency of systems becomes a pressing concern. The most critical performance factor in the implementation of a production system is the condition-testing algorithm. We propose a new method based on the widely used RETE match algorithm. We show an approach designed to reduce the response time of rule-based expert systems by reducing the matching time. There are two steps in the method we propose: The first makes an index structure of the tokens to reduce the /spl alpha/-node-level join candidates. The second chooses the highest time tag for certain /spl beta/-nodes to reduce the amount of combinatorial match that is problematical in a real-time production system application. For this purpose, a simple compiler is implemented in C and the response time of test programs is measured.