In this paper we present a vehicle detection system using convolutional neural networks on 2d image data. Since realtime capabilities are crucial for object detection systems running in real-traffic situations, we will show how the calculation time of our algorithm can be significantly reduced by taking advantage of depth information from lidar sensors. One part of this work focusses on useful network topologies and network parameters to increase the classification precision. We will test the presented algorithm on an autonomous car in different real-traffic scenarios with regards to detection accuracy and calculation time and show experimental results.

Planing the topology of wireless sensor networks (WSN) for a specific application is a complex task.Each application defines requirements to its WSN.Some of these requirements have to been fulfilled by the wireless technology, e.g., energy consumption and throughput, and some by the network topology, e.g., redundancy and latency.Topology makes restrictions to the wireless technology and the wireless technology makes restrictions to the network topology.In this paper we present an algorithm to select a network topology and a wireless technology depending on application's requirements automatically.The algorithm is part of the Sens4U approach, which aims to simplify and possibly automate the process of building WSN applications and support applications development done by non-WSN-experts.

The wireless Bluetooth standard, generally used for short range point-to-point communication between mobile devices, can in principle connect any number of stations into a multi-hop network. For this purpose a "scatternet" must be built up, which is a collection of small overlapping "piconets". Bluetooth technology is basically well-suited to some types of wireless sensor networks, but such deployment has been been limited due to the difficulty in setting up and maintaining a scatternet. We demonstrate the SFX algorithm, an extension of the earlier SHAPER procedure. The SFX algorithm builds up a scatternet by a decentralized procedure with the properties of self-organization, self-optimization, and self-healing. The procedure has been verified by deployment in a photovoltaic power plant.

Wireless sensor and actor networks based on Bluetooth technology require an algorithm to build up and maintain a Bluetooth scatternet. Although much research has been done, actual deployment is rare because a self-organizing and self-healing procedure suitable for actual Bluetooth nodes has not yet been developed. We present the SFX algorithm to build a scatternet tree with the required properties. The method was verified by simulations for up to 400 nodes and by field tests.