• Distance Measures for QOS Performance Management in Mixed Networks

      Dean, Richard; Astatke, Yacob; Morgan State University (International Foundation for Telemetering, 2008-10)
      The integrated Network Enhanced Telemetry effort (iNET) was launched to create a telemetry network that will enhance the traditional point-to-point telemetry link from test articles (TAs) to ground stations (GS). Two of the critical needs identified by the Central Test and Evaluation Investment Program (CTEIP) are, "the need to be able to provide reliable coverage in potentially high capacity environments, even in Over-The-Horizon (OTH) settings", and "the need to make more efficient use of spectrum resources through dynamic sharing of said resources, based on instantaneous demand thereof". Research conducted at Morgan State University (MSU) has focused on providing solutions for both critical problems. The Mixed Network architecture developed by MSU has shown that a hybrid network can be used to provide coverage for TAs that are beyond the coverage area of the GS. The mixed network uses clustering techniques to partition the aggregate network into clusters or sub-networks based on properties of each TA, which currently include signal strengths, and location. The paper starts with a detailed analysis of two parameters that affect the performance of each sub-network: contention between the TAs in the mobile ad-hoc network, and queuing at the Gateway TAs that serve as the link between the mobile ad-hoc and the Cellular networks. Contention and queuing will be used to evaluate two performance (distance) measures for each sub-network: throughput and delay. We define a new distance measure known as "power", which is equal to the ratio of throughput over delay, and is used as a measure of performance of the mixed network for Quality of Service (QOS). This paper describes the analytical foundation used to prove that the "power" performance measure is an excellent tool for optimizing the clustering of a mixed network to provide QOS.
    • The Optimization of Node Configuration in a Mixed Network for Quality of Service

      Dean, Richard; St. Ange, Laurie; Morgan State University (International Foundation for Telemetering, 2008-10)
      Morgan State University's iNET project is the development of a complete network architecture to enhance telemetry performance. Existing telemetry standards employ an RF link standard which allows only unidirectional data transfer from a test article (TA) to a ground station over a dedicated channel per TA. This limits the capacity of the system as resources are not efficiently utilized. Additionally, this standard does not allow data exchange between Over the Horizon TAs and the ground station. Their data must be recorded for subsequent retrieval by test engineers. The growth of telemetry demands that spectrum resources are maximized to increase transfer capacity. Real time data exchange capability is essential to success in telemetry. Bidirectional streaming of data to both near and over the horizon TAs is desired with no compromise of Quality of Service for real time applications. The network approach addresses these demands by allowing telemetry data to be multiplexed over a shared radio channel to and from the TAs. A mixed cellular and mobile adhoc network (mixed network) has been shown to satisfy the capacity and coverage elements of iNET requirements. This paper begins with an introduction to the iNET objective, continues with some background information on the major technical elements applied in this endeavor and finally shows that an optimum node configuration can be achieved and maintained in this mixed network using clustering techniques to include the joint optimization of the Quality of Service, Signal-to-Noise ratio, and Interference requirements as outlined by the iNET Needs Discernment.