Modern object recognition algorithms have very high precision. At the same time, they require high computational power. Thus, widely used low-power IoT devices, which gather a substantial amount of data, cannot directly apply the corresponding machine learning algorithms to process it due to the lack of local computational resources. A method for fast detection and classification of moving objects for low-power single-board computers is shown in this paper. The developed algorithm uses geometric parameters of an object as well as scene-related parameters as features for classification. The extraction and classification of these features is a relatively simple process which can be executed by low-power IoT devices. The algorithm aims to recognize the most common objects in the street environment, e.g., pedestrians, cyclists, and cars. The algorithm can be applied in the dark environment by processing images from a near-infrared camera. The method has been tested on both synthetic virtual scenes and real-world data. The research showed that a low-performance computing system, such as a Raspberry Pi 3, is able to classify objects with acceptable frame rate and accuracy.

Sharing physical resources among virtual instances introduces time overhead in comparison with direct access to hardware. Such lag is not significant for most of the everyday tasks however it can influence much more in dealing with time-critical applications and especially in case of reliable network services. The purpose of this paper is to compare time overhead on timing operations such as time acquisition and sleep introduced by different virtualization environments: Xen, VMWare ESXi, QEMU. The current research focus to establish possibility of using such platforms in real-time applications with high resource utilization. In terms of present work, there are several load types to be used to simulate real conditions. Measurement performance includes different methods of time measurements and waiting operations. Considering results, the certain recommendations about timing mechanisms using different virtualization environment have been offered.

Available bandwidth parameter is a crucial characteristic in terms of networking and data transmission. The beforehand knowledge of its value and use of this parameter in various traffic engineering algorithms and QoS calculations is a key for high-efficient multigigabit data transport in nowadays networks. The challenge in available bandwidth estimations is not only in its accuracy and processing speed but also in the reduction of the amount of probe traffic injected into the network by keeping an adequate level of estimation accuracy. In this paper we extend existing active probing measurement algorithms for end-to-end available bandwidth estimation along with methods to reduce estimation times and amount of injected traffic while keeping measurement accuracy constant and even reducing the uncertainty of estimations. The main goal of this research was to detect a sufficient ratio of MTU, packet train size with the link capacity and available bandwidth (AvB) in up to 10 Gbps networks. In order to explore measurement accuracy under different conditions, a new tool for the AvB estimation named Kite2 has been developed and is presented in the paper. Comparative performance of AvB estimations using Kite2, Kite and Yaz is presented. Finally we calculate with statistical means dependency between the estimation error probability, measurement probing overhead and the measurement time.

The impact of cloud computing and sharing physical resources simultaneously between several virtual instances can be crucial for the designing real-time and time-critical applications. This especially applicable to cases when the target is to deliver reliable network services. In this paper, we primarily aim at investigating the timing behavior in the cloud environments where the resources are shared among up to 50 instances. Moreover, we extend the previously performed experiment with the simulation of various background tasks where the key use-case is the deployment of secure and reliable network applications in virtualized environments. The measurements have been performed on host-based model QEMU/KVM and include different method for time acquisitions and waiting operations. Providing detailed statistical data, we have been analyzed the performance of timing part for the general network application and found some interesting phenomena about the behavior of Linux standard sleep calls. Considering these results, corresponding recommendations about using timing mechanisms have been offered.

In this paper we present a measurement study to characterize the impact of hardware virtualization on basic software timing, as well as on precise sleep operations of an operating system. We investigated how timer hardware is shared among heavily CPU-, I/O- and Network-bound tasks on a virtual machine as well as on the host machine. VMware ESXi and QEMU/KVM have been chosen as commonly used examples of hypervisor- and host-based models. Based on statistical parameters of retrieved distributions, our results provide a very good estimation of timing behavior. It is essential for real-time and performance-critical applications such as image processing or real-time control.