Setting up connections to hosts behind Network Address Translation (NAT) equipment has last been the subject of research debates half a decade ago when NAT technology was still immature. This paper fills this gap and provides a solid comparison of two essential TCP hole punching approaches: sequential and parallel TCP hole punching. The comparison features current conditions and thoroughly compares setup delay, implementation complexity, resource usage, and effectuality of the two approaches. The result is a list of recommendations and a portable, effectual, and open-source Java implementation.

Seamless roaming in heterogeneous circuit-switched and IP networks is key for successful migration towards all-IP. The Unified Mobility Manager (UMM) keeps track of users' locations whatever network they are roaming, and that routes calls from any originating to any terminating network. In a UMM enabled network, calls are addressed to users and dynamically routed to the active devices. This notion of user mobility breaks up the traditional coupling of addresses and devices, and replaces it with rules for the dynamic mapping from logical to device addresses. We present the UMM multi-protocol subscriber base that is built upon the concept of Incoming Call Handling (ICH) profiles that are the addressable entities in the UMM. An ICH profile contains all address level service data as well as the routing rules to the actual terminating devices. We also present our implementation of the UMM subscriber base in detail, and compare the UMM approach with two other approaches - a federated approach and an approach that combines ANSI-41 user mobility with GAIT interworking.

Session-based Internet protocol (IP) applications, such as Internet telephony, are an important component of the emerging mobile Internet. The ubiquitous availability of these services is critical to the success of the mobile Internet. Because all-IP networks will be deployed in phases and current mobile telecommunication systems will be in operation for decades to come, the interworking and migration between current network services and all-IP services is a key problem. In this paper, we address seamless roaming for session initiation protocol-based services across current cellular telecommunication networks and emerging all-IP wireless networks, such as those using third-generation and WiFi networks. We present an abstract mobility model, and map this model to three basic approaches for supporting seamless mobility: a master-slave approach, a federated system, and a unified approach. We discuss the challenges and implementation of an instance of the unified mobility management approach, called the Unified Mobility Manager, and then compare the tradeoffs of the three systems using a comparative performance analysis. We conclude that unified mobility management is most efficient if a great deal of interworking is required, and as more users invoke IP-based services; the federated approach is efficient when a single network technology is dominant and data access is limited, but requires sharing of data across networks; the master-slave approach is the least efficient, but is easy to introduce if the number of network types is small.

Query processing on partly inaccessible databases generally does not yield exact, but vague result sets. A good notion of vague sets fulfills two aims: It keeps the degree of vagueness of the query result as small as possible, and it clarifies the degree of and the reasons for the vagueness to the end user. The first goal requires a good internal representation, while the second goal requires a good external representation of a vague set. In this paper, we present a novel calculus for expressive vague sets that meets both requirements. This is the first approach that is well suited for both internal and external representation of vagueness induced by partial inaccessibility. It consists of a data representation that is capable of holding all the necessary information. Complementary, we have accordingly adapted the usual query language operations. These adaptations are independent of a concrete query language, to make them applicable to most existing query languages. The adapted operations minimize the vagueness of the result, propagate the reasons of uncertainty of the individual vague candidates, and compute an expressive description of the missing elements.

An important characteristic of distributed object management systems is that due to network or machine failure, the environment may become partitioned into subenvironments that cannot communicate with each other. In some application scenarios, it is important that the subenvironments remain operable even in this case. In particular, queries should be processed in an appropriate way. To this end, the final and all intermediate results of a query in a distributed object management system must be regarded as potentially vague. We propose a hybrid representation for vague sets and vague multisets designed for this application context. The representation consists of an enumerating part, which contains the elements we could access during query processing, and a descriptive part, which describes the relevant elements we could not access. We introduce propagation rules which can be used to minimize the vagueness of a query result represented in this hybrid way. The main advantage of our approach is that the descriptive part of the representation can be used to improve the enumerating part during query processing.

An interesting research area for distributed database systems is inaccessibility.Even if part of the database is inaccessible, a query should be evaluated reasonably.This means, that appropriate rules for propagation and minimization of errors induced by inaccessibility are needed.Furthermore, the result of a query in a distributed database system must be regarded to be potentially vague.In this paper we present a hybrid representation for vague sets, that consists of(1) an enumerating part, which contains the elements we could access during query processing, and (2) a descriptive part, which describes the relevant elements we could not access.Further we introduce propagation rules, which can be used to minimize the vagueness of a query result represented in this hybrid way.Thereby, we propose different levels of accuracy for the descriptive part.As far as we know this is the first approach using a descriptive part to qualify the missing elements of a result.An example for the usefulness of the descriptive part arises, if we have to calculate the intersection of two subqueries.Here, the descriptive part of the first operand can be used to check for each element of the second operand, whether it should be part of the first operand, and vice versa.