Samoadaptivno upravljanje vezom za usluge digitalnih sadržaja

scientific master's theses

• Title (english): Self-adaptive connection control for digital content services
• Author: Krešimir Čunko
• Mentor: Dragan Jevtić (mentor)
• Committee member: Marijan Kunštić (predsjednik povjerenstva)
• Committee member: Dragan Jevtić (član povjerenstva)
• Committee member: Antun Carić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; (Department of Telecommunications); Zagreb
• Defense date and country: 2007-11-30, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Electrical Engineering; Telecommunications and Informatics
• Universal decimal classification (UDC): 654; APPLIED SCIENCES. MEDICINE. TECHNOLOGY; Telecommunication and telecontrol
• Universal decimal classification (UDC): 004; GENERALLY; Computer science and technology. Computing. Data processing

Abstract

Usmjeravanje prometa u telekomunikacijskoj mreži uobičajeno se obavlja determinističkim metodama za koje je karakteristična slaba prilagodljivost na promjene izazvane prometnim opterećenjima. Uvođenjem samoadaptivnih svojstava u takvu mrežu može se pojačati učinkovitost komunikacije ponajprije kroz poboljšanu prilagodljivost pristupa mrežnim resursima. Za očekivati je pojavu pojačanih potreba takvih svojstava kod isporuke usluga digitalnih sadržaja zbog promjenjivih prijenosnih kapaciteta, a uz istovremenu prisutnost više spojnih puteva između korisnika i poslužitelja. Zadaća ovog magistarskog rada bila je istražiti modele inteligentnih agenata pogodnih za podršku pristupa digitalnim sadržajima kroz paketski orijentirane mreže. Cilj je usmjeren ka maksimiziranju prijenosnog kapaciteta tijekom realizacije usluge između korisnika i poslužitelja. Samoadaptivnost je postignuta primjenom paradigme inteligentnog agenta fiksno pozicioniranog izvan mrežnih čvorova. Za tu svrhu dizajnirani su inteligentni agenti bazirani na Q i SARSA algoritmima, čiji je rad uspoređen sa klasičnom metodom usmjeravanja Round Robin. U svrhu simulacije rada agenata i verifikacije rezultata razvijen je laboratorijski model implementiran u programskom jeziku C++ koristeći razvojnu okolinu Microsoft Visual Studio .NET 2003. Na temelju dobivenih rezultata provedenih simulacija i njihove analize, može se zaključiti da je korištenje Q i SARSA agenta na opisani način učinkovit pristup rješavanja problema transporta digitalnih sadržaja u uvjetima nestabilnog mrežnog okruženja.

Abstract (english)

In telecommunication networks traffic routing algorithms are usually based on deterministic methods which are not adaptive enough for rapid changes in traffic density. By introducing self-adaptive characteristics to such network it is possible to increase communication performance through improved access to network resources. Increased requirements for such characteristics can be expected especially in distribution of digital content due to constant changes in network transmission capacity and existence of several interconnection routes between client and server. The objective for this master thesis was to design intelligent agents which could be used for accessing digital content in packet-oriented networks. The goal was to maximize transmission capacity during realization of service. Self-adaptiveness was realized by using paradigm of intelligent agent positioned outside network nodes. For this purpose intelligent agents based on Q and SARSA algorithms were designed and compared against Round Robin routing method. For behavior analysis of designed agents a laboratory model was implemented in C++ language using Microsoft Visual Studio .NET 2003. Based on simulation results it can be concluded that the usage of intelligent Q and SARSA agents in a way described in this thesis is an effective way of solving digital content transport problems in conditions of unstable network surroundings.

• Keywords: samoadaptivno usmjeravanje; inteligentni agent; Q učenje; SARSA učenje; digitalni sadržaj
• Keywords (english): self-adaptive routing; intelligent agent; Q learning; SARSA learning; digital content
• Language: croatian
• URN:NBN: urn:nbn:hr:168:083934
• Study programme: Title: Postgraduate master programme in electrical engineering; Study programme type: university; Study level: postgraduate; Academic / professional title: Magistar znanosti elektrotehnike (Magistar znanosti elektrotehnike)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2020-04-23 09:20:45