Institute for Systems Research Technical Reports

Permanent URI for this collectionhttp://hdl.handle.net/1903/4376

This archive contains a collection of reports generated by the faculty and students of the Institute for Systems Research (ISR), a permanent, interdisciplinary research unit in the A. James Clark School of Engineering at the University of Maryland. ISR-based projects are conducted through partnerships with industry and government, bringing together faculty and students from multiple academic departments and colleges across the university.

Browse

Subject: search.filters.subject.Performance

Search Results

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Simulation Studies of a Hybrid Network in Order to Enhance the Performance of Hybrid Internet Service
    (1998) Gaid, Ahmed Y.; Baras, John S.; ISR; CSHCN
    The work reported here addresses the modeling and simulation of a hybrid network that conforms to Hybrid Internet Service, which is an Internet access currently commercialized by Hughes Network Systems. The objective is to research and discover performance enhancement techniques for Hybrid Internet Service. Hybrid Internet Service intends to provide Internet end users with high bandwidth by delivering packets over a satellite channel. It takes advantage of the fact that the vast majority of Internet end-users don't send as much data as they receive. However, long delays experienced over the satellite channel (in the case of a GEO satellite) have catastrophic consequences on the overall throughput with Commercial-Off-The-Shelf TPC/IP stacks. This problem is addressed by the Asymmetric TCP/IP protocol implemented within a gateway, the hybrid gateway. The hybrid gateway acts as a go-between for the server and the end user. The hybrid gateway both acknowledges packets on behalf of the end user and advertises a larger window to the server. We have implemented the Asymmetric TCP/IP within the hybrid gateway with OPNET, which is an industrial-strength, popular network modeling and simulation software. A discussion of the current results of the simulation in the broader perspective of hybrid networks development is provided following the presentation of both Hybrid Internet Service and OPNET along with the various built-in OPNET models used in the course of the simulation. The methodology we have followed is also included in this report.
  • Thumbnail Image
    Item
    Markov Decision Models with Weighted Discounted Criteria
    (1991) Feinberg, Eugene A.; Shwartz, Adam; ISR
    We consider a discrete time Markov Decision Process with infinite horizon. The criterion to be maximized is the sum of a number of standard discounted rewards, each with a different discount factor. Situations in which such criteria arise include modeling investments, modeling projects of different durations and systems with different time-scales, and some axiomatic formulations of multi-attribute preference theory. We show that for this criterion for some positive e there need not exist an e - optimal (randomized) stationary strategy, even when the state and action sets are finite. However, e - optimal Markov (non-randomized) strategies and optimal Markov strategies exist under weak conditions. We exhibit e - optimal Markov strategies which are stationary from some time onward. When both state and action spaces are finite, there exists an optimal Markov strategy with this property. We provide an explicit algorithm for the computation of such strategies.