Woojik Chun

Improvements on UIO Sequence Generation and Partial UIO Sequences

Abstract Two necessary conditions are derived for determining a shortest Unique Input Output (UIO) sequence. Using these conditions, Sabnani and Dahburas original algorithm [SD88] for UIO sequence generation in test case generation is made more efficient. For states having no UIO sequence, an algorithm for generating Partially Unique Input Output (PUIO) sequences is introduced and analyzed. While a UIO sequence distinguishes a state from all other states, a PUIO sequence distinguishes a state from only a subset of other states. A test case generation technique combining UIO and PUIO sequences is explained.

ID based communication in Domain-Insulated Autonomous Network Architecture (DIANA)

Although the Internet achieved tremendous success in the recent decades, several limitations have emerged from recent technical advancement and challenging requirements of diverse applications. For overcoming these limitations, the future Internet can be redesigned in the “clean slate” basis. This paper proposes a novel architecture that considers both horizontal and vertical separations called, Domain-Insulated Autonomous Network architecture (DIANA). In this architecture, a network is composed of a set of basic building blocks, called “domains”, which can be defined recursively by topological, administrative, or operational grouping of communicating entities. In this paper, the design principles of DIANA and its distinguishing features are described. Then the inter-domain routing mechanism is described along with some use cases of DIANA.

YANAIL: yet another definition on names, addresses, identifiers, and locators

This paper proposes a new definition on names, addresses, identifiers, and locators based on a different framework. Starting with observation of the patterns how those terms are used, a new definition for the terms and their relations are presented. First, name and address are defined based on assignment, where a name denotes an entity itself, and an address denotes a point to which the entity is attached. On the other hand, identifier is used for locating an entity within a given space. Next, the relationship among those four terms is presented. Finally, we show how communication is performed with respect to the new definition.

A flat name based routing scheme for information-centric networking

The paradigm shift from the current host-centric networking to information-centric networking (ICN) generates much interest in the research community. In ICN, the primary communication entity is contents and uniquely identifiable names are assigned directly to the contents, where the names are independent of their locations. Thus, content naming and routing are the key challenges in ICN. In this paper, we propose a flat name based routing scheme for ICN. For scalable routing, we exploit the structure of container which is a space where contents or information objects reside. We also utilize bloom filters as an aggregated form of the flat names and enable the deepest match routing.

Demonstrating A Scalable Name Resolution System for Information-Centric Networking

In implementing Information-Centric Networking (ICN), an efficient Name Resolution System (NRS) is required since names assigned directly to the named data objects (NDOs) are location independent. On designing NRS, the most important challenge is scalability on the ever increasing number of NDO, which becomes even more critical when flat names are used in ICN. In this demonstration, we present a scalable NRS with flat names for ICN, where bloom filters are utilized to aggregate flat names. Specifically, we show the feasibility of our NRS with flat names and highlight the benefits of utilizing bloom filters.