The analysis of multimedia application traces can reveal important information to enhance program execution comprehension. However typical size of traces can be in gigabytes, which hinders their effective exploitation by application developers. In this paper, we study the problem of finding a set of sequences of events that allows a reduced-size rewriting of the original trace. These sequences of events, that we call blocks, can simplify the exploration of large execution traces by allowing application developers to see an abstraction instead of low-level events.

In pervasive environments, services are fastly developing and are being deployed everywhere. In this article, we introduce a Servicebook, a new social network of services, where services create and join group of service profile providing to users better access to all the services in their vicinity. We propose a novel technique to realize this Servicebook, the user-excentric service composition. This user-excentric composition relies on two service relations: the compatible relation and the composition relation. We developed and evaluated an OSGi-prototype as a proof-of-concept.

One of the challenges of ambient systems lies in providing all the available services of the environment to the ambient devices, even if they do not physically host those services. Although this challenge has come to find a solution through cloud computing, there are still few devices and operating systems that enable applications execution by only uploading the required components into the runtime environment. The ROCS (Remote OSGi Caching Service) framework is a novel proposal which relies on a heavy-weighted standard Java/OSGi stack. It is distributed between class servers and ambient devices to provide full functionalities to resource-constrained environments. The ROCS framework provides improvements in two areas. First, it defines a minimal bootstrap environment that runs a standard Java/OSGi stack. Secondly, it provides an architecture for loading any necessary missing class from remote servers into memory at runtime. Our first results show similar performances when classes are either remotely downloaded into the main memory from a local network or from a flash drive. These results suggest a way to design minimalistic middleware that dynamically obtain their applications from the network as a first step towards cloud-aware operating systems.

In this paper, we present a receiver design for Bluetooth transmission based on Laurent's decomposition of the Bluetooth transmit signal. The main features of this receiver are (a) its low complexity compared to alternative solutions, (b) its excellent performance close to the theoretical limit, and (c) its high robustness against frequency offsets, phase noise, and modulation index variations, which are characteristic for low-cost Bluetooth devices. In particular, we show that the devised noncoherent decision-feedback equalization receiver achieves a similar performance as a recently proposed 2-state noncoherent sequence detector, while it is advantageous in terms of complexity. The new receiver design is therefore highly attractive for a practical implementation.