Networked estimation under contention-based medium access

Recent surveys conducted in the context of industrial automation have outlined that reliability concerns represent today one of the major barriers to the diffusion of wireless communications for sensing and control applications: this limits the potential of wireless sensor networks and slows down the adoption of this new technology. Overcoming these limitations requires that awareness on the causes of unreliability and on the possible solutions to this problem is created. With this respect, the main factor responsible for the perceived unreliability is radio interference: low-power communications of sensor nodes are in fact very sensitive to bad channel conditions and can be easily corrupted by transmissions of other co-located devices. In this thesis we investigate different techniques that can be exploited to avoid interference or mitigate its effects.We first consider interference avoidance through dynamic spectrum access: more specifically we focus on the idea of channel surfing and design algorithms that allow sensor nodes to identify interfered channels, discover their neighbors and maintain a connected topology in multi-channel environments. Our investigation shows that detecting and thus avoiding interference is a feasible task that can be performed by complexity and power constrained devices. In the context of spectrum sharing, we further consider the case of networked estimation and aim at quantifying the effects of intranetwork interference, induced by contention-based medium access, over the performance of an estimation system. We show that by choosing in an opportune manner their probability of transmitting, sensors belonging to a networked control system can minimize the average distortion of state estimates.In the second part of this thesis we focus on frequency hopping techniques and propose a new adaptive hopping algorithm. This implements a new approach for frequency hopping: in particular rather than aiming at removing bad channels from the adopted hopset our algorithm uses all the available frequencies but with probabilities that depend on the experienced channel conditions. Our performance evaluation shows that this approach outperforms traditional frequency hopping schemes as well as the adaptive implementation included in the IEEE 802.15.1 radio standard leading to a lower packet error rate.Finally, we consider the problem of sensor networks reprogramming and propose a way for ingineering a coding solution based on fountain codes and suitable for this challenging task. Using an original genetic approach we optimize the degree distribution of the used codes so as to achieve both low overhead and low decoding complexity. We further engineer the implementation of fountain codes in order to allow the recovery of corrupted information through overhearing and improve the resilience of the considered reprogramming protocol to channel errors.