Search

We propose the novel RTPD/MACSC dynamic cache size tuning approach to shorten the information retrieval roundtrip time by a SFF (small-form-factor) client in a pervasive computing environment. By itself the MACSC (model for adaptive cache size control) does not compensate for the ill effects due to the Internet traffic. In the proposed approach, however, the MACSC is enhanced with the real-time traffic pattern detection (RTPD) capability. With this support the MACSC tuner strives to maintain the given hit ratio as the minimum and reduces the amount of cyber foraging. In the process the MACSC component self-tunes its core point-estimate (PE) calculation to neutralize any ill effects of the traffic pattern currently detected and identified by the supporting RTPD capability.

This paper proposes to use the novel Fuzzy Logic Controller (FLC), which is a dynamic buffer size tuning technique, to support more dependable location-aware information retrieval. Within a smart space supported by mobile/pervasive computing a user may request service from the surrogate through a portable SFF (small-for-factor) device (e.g. PDA). Therefore, the one-surrogate-to-many-SFF-clients relationship or asymmetric rendezvous can easily lead to surrogate reception buffer overflow during peak demand periods. In fact, the traffic of requests to a surrogate is tied in with the transient traveler traffic through the smart space, for example, an airport. Once overflow has happened the user cannot benefit from the cyber foraging supported by the infrastructure. The FLC adaptively tunes the surrogate buffer size so that the buffer length always covers the queue length by the given? safety margin. In this way the chance of surrogate reception buffer overflow is eliminated, and as a result the asymmetric rendezvous becomes more reliable and dependable, leading to shorter SFF-client/surrogate service roundtrip time.

Smart Sensor Networks are one of the most exciting research areas in information technology today; their potential for business applications is vast, but yet to be realised. In this paper we argue that intelligent sensor networks can only reach their potential for business applications if the network is grounded so as to support meaningful information sharing and knowledge generation as the basis for an effective business decision model. Our work is based on the idea that sensors extract information from their environment which, typically, must be fused with other sensor data and network knowledge so that effective business decision making is supported.

not available

Vol. 15, No. 5; Weekly edition. Items scanned from microfilm of varying quality--some editions are difficult to read, mostly from 1972-1974.

Vol. 15, No. 19; Weekly edition. Items scanned from microfilm of varying quality--some editions are difficult to read, mostly from 1972-1974.

Includes the beginning of 4 different songs on the final page., D, 1 low voice, 1 piano, This archival material has been provided for educational purposes. Ball State University Libraries recognizes that some historic items may include offensive content. Our statement regarding objectionable content is available at: https://dmr.bsu.edu/digital/about

Includes a list of Sacred Songs and Duetts published by Arthur P. Schmidt, Boston., D, 1 low voice, 1 piano, This archival material has been provided for educational purposes. Ball State University Libraries recognizes that some historic items may include offensive content. Our statement regarding objectionable content is available at: https://dmr.bsu.edu/digital/about