Showing papers by "Mihail L. Sichitiu published in 2017"

Biobotic motion and behavior analysis in response to directional neurostimulation

Abstract: This paper presents preliminary results for motion behavior analysis of Madagascar hissing cockroach biobots subject to stochastic and periodic neurostimulation pulses corresponding to randomly applied right and left turn, and move forward commands. We present our experimental setup and propose an unguided search strategy based stimulation profile designed for exploration of unknown environments. We study a probabilistic motion model fitted to the trajectories of biobots, perturbed from their natural motion by the stimulation pulses. Furthermore, we provide a statistical assessment of the biobotic directional response to turn commands and its correlation with stimuli profile over time. This study paves the way towards reliable control for more realistic under-rubble search and rescue applications.

On the Accuracy of Pairwise Time Synchronization

Abstract: Time synchronization is a fundamental problem in any distributed system. In particular, wireless sensor networks require scalable time synchronization for implementing distributed tasks on multiple sensor nodes. Different synchronization schemes and hardware systems are able to achieve widely different levels of synchronization accuracy. Since synchronization accuracy is very sensitive to random delays and clock rate fluctuations, we model the pairwise synchronization error as a function of random delays and clock drifts to quantitatively model synchronization performance as a function of the relevant parameters, such as the beacon interval, the number of beacons per synchronization period, and the measurement point. The synchronization error is analyzed for different state-of-the-art pairwise synchronization schemes, based on one-way sender–receiver, two-way sender–receiver, and receiver–receiver approaches, using application layer or medium access control layer timestamping. The models are validated through numerical and experimental results.

Temperature compensated Kalman distributed clock synchronization

Abstract: Time synchronization is a fundamental problem in any distributed system. In particular, wireless sensor networks (WSNs) require scalable time synchronization for implementing distributed tasks on multiple sensor nodes. We propose a temperature-compensated Kalman based distributed synchronization protocol (TKDS) using a two-way sender-receiver synchronization scheme, to achieve high synchronization accuracy while modelling the clock skew change based on its physical characteristics. By asynchronously combining estimates from neighbours, TKDS achieves better performance than the spanning tree based protocols in a fully-distributed fashion. The synchronization performance is evaluated numerically and compared with that of other well-known synchronization protocols.

Towards acoustic localization for biobotic sensor networks

Abstract: A cyberphysical network of insect biobots or biological robots could aid first-responders for search-and-rescue applications in uncertain disaster environments. In such networks, the nodes are insect biobots equipped with miniature backpacks utilizing a system-on-chip. These applications demand fine-grained, real-time localization of the sensor nodes. In this paper, we use a combination of radio-frequency and acoustic signal transmission to estimate the distance between two nodes with an aim of localizing the biobotic sensor nodes. We describe how digital processing, peak detection, and noise mitigation can be applied to meet the severe resource constraints of the specific application, yet yielding average localization errors of 12 cm within the effective range.

The Effect of Increasing the Number of Transceivers in an Anti-jamming Channel-Hopping Scheme

Abstract: —In this paper, we analyze a channel-hopping scheme under a smart jamming attack in wireless military communication systems. In particular, we focus on the effect of using multiple transceivers in channel-hopping schemes and compare them with the case of a single transceiver. To evaluate the performance of the multiple transceiver system, we present an analytical model that can calculate the average jammed time when a channel is detected by a smart jammer; consequently network throughput can also be computed. The numerical results show that the multiple transceiver system achieves a significantly higher throughput than the single transceiver system. It is also shown that the effect of increasing the number of transceivers in anti-jamming channel-hopping scheme improves network throughput.

Evaluating the Accuracy of Vehicular Channel Models in a Large-Scale DSRC Test

Abstract: Vehicle-to-vehicle channel characteristics differ significantly from those of conventional cellular channels, especially in terms of fading statistics due to varying environmental conditions, link types, vehicle types, and objects that result in complex propagation effects. Accurate modeling of the vehicular channel remains a complex challenge. In this paper, existing, common vehicular channel models (VCMs) are evaluated and compared to measurement data from the UMTRI large-scale Dedicated Short-Range Communications (DSRC) testbed involving nearly 3000 vehicles and conducted over several months around Ann Arbor, MI. While many VCMs can predict reasonably well the frequent packet error rates observed near Ann Arbor, they over-estimate inter-packet gap and under-estimate the likelihood of runs of consecutively and successfully received packets. The testbed environment shows significant fading (i.e., sub-Rayleigh) and/or shadowing effects that challenge the accuracy of traditional VCMs. Furthermore, despite 60% of the inter-vehicle paths being obstructed by items available using OpenStreetMap (OSM) geodata, a deterministic obstacle shadowing model that makes use of such geodata does not account for all shadowing effects. Evaluating vehicular channel models in terms of real-world, large-scale experiments helps researchers better understand actual behaviors and allows for the development of new and/or improved models that more accurately reflect reality.