Microglia, the brain's resident macrophages, actively contribute to the homeostasis of cerebral parenchyma by sensing neuronal activity and supporting synaptic remodeling and plasticity. While several studies demonstrated different roles for astrocytes in sleep, the contribution of microglia in the regulation of sleep/wake cycle and in the modulation of synaptic activity in the different day phases has not been deeply investigated. Using light as a zeitgeber cue, we studied the effects of microglial depletion with the colony stimulating factor‐1 receptor antagonist PLX5622 on the sleep/wake cycle and on hippocampal synaptic transmission in male mice. Our data demonstrate that almost complete microglial depletion increases the duration of NREM sleep and reduces the hippocampal excitatory neurotransmission. The fractalkine receptor CX3CR1 plays a relevant role in these effects, because cx3cr1 GFP/GFP mice recapitulate what found in PLX5622‐treated mice. Furthermore, during the light phase, microglia express lower levels of cx3cr1 and a reduction of cx3cr1 expression is also observed when cultured microglial cells are stimulated by ATP, a purinergic molecule released during sleep. Our findings suggest that microglia participate in the regulation of sleep, adapting their cx3cr1 expression in response to the light/dark phase, and modulating synaptic activity in a phase‐dependent manner.

Mouse strains differ markedly in all behaviors, independently of their genetic background. We undertook this study to disentangle the diurnal activity and feature key aspects of three non-genetically altered mouse strains widely used in research, C57BL/6NCrl (inbred), BALB/cAnNCrl (inbred) and CRL:CD1(ICR) (outbred). With this aim, we conducted a longitudinal analysis of the spontaneous locomotor activity of the mice during a 24-h period for 2 months, in two different periods of the year to reduce the seasonality effect. Mice (males and females) were group-housed in Digital Ventilated Cages (Tecniplast), mimicking standard housing conditions in research settings and avoiding the potential bias provided in terms of locomotor activity by single housing. The recorded locomotor activity was analyzed by relying on different and commonly used circadian metrics (i.e., day and night activity, diurnal activity, responses to lights-on and lights-off phases, acrophase and activity onset and regularity disruption index) to capture key behavioral responses for each strain. Our results clearly demonstrate significant differences in the circadian activity of the three selected strains, when comparing inbred versus outbred as well as inbred strains (C57BL/6NCrl versus BALB/cAnNCrl). Conversely, males and females of the same strain displayed similar motor phenotypes; significant differences were recorded only for C57BL/6NCrl and CRL:CD1(ICR) females, which displayed higher average locomotor activity from prepuberty to adulthood. All strain-specific differences were further confirmed by an unsupervised machine learning approach. Altogether, our data corroborate the concept that each strain behaves under characteristic patterns, which needs to be taken into consideration in the study design to ensure experimental reproducibility and comply with essential animal welfare principles.

Amyotrophic Lateral Sclerosis (ALS) is a devastating neurodegenerative disease that affects both central and peripheral nervous system, leading to the degeneration of motor neurons, which eventually results in muscle atrophy, paralysis, and death. Sleep disturbances are common in patients with ALS, leading to even further deteriorated quality of life. Investigating methods to potentially assess sleep and rest disturbances in animal models of ALS is thus of crucial interest. We used an automated home cage monitoring system (DVC®) to capture irregular activity patterns that can potentially be associated with sleep and rest disturbances and thus to the progression of ALS in the SOD1G93A mouse model. DVC® enables non-intrusive 24/7 long term animal activity monitoring, which we assessed together with body weight decline and neuromuscular function deterioration measured by grid hanging and grip strength tests in male and female mice from 7 until 24 weeks of age. We show that as the ALS progresses over time in SOD1G93A mice, activity patterns start becoming irregular, especially during day time, with frequent activity bouts that are neither observed in control mice nor in SOD1G93A at a younger age. The increasing irregularities of activity pattern are quantitatively captured by designing a novel digital biomarker, referred to as Regularity Disruption Index (RDI). We show that RDI is a robust measure capable of detecting home cage activity patterns that could be related to rest/sleep-related disturbances during the disease progression. Moreover, the RDI rise during the early symptomatic stage parallels grid hanging and body weight decline. The non-intrusive long-term continuous monitoring of animal activity enabled by DVC® has been instrumental in discovering novel activity patterns potentially correlated, once validated, with sleep and rest disturbances in the SOD1G93A mouse model of the ALS disease.

Automated home cage monitoring represents a key technology to collect animal activity information directly from the home cage. The availability of 24/7 cage data enables extensive and quantitative assessment of mouse behavior and activity over long periods of time than possible otherwise. When home cage monitoring is performed directly at the home cage rack, it is possible to leverage additional advantages, including, e.g., partial (or total) reduction of animal handling, no need for setting up external data collection system as well as not requiring dedicated labs and personnel to perform tests. In this work we introduce a home cage-home rack monitoring system that is capable of continuously detecting spontaneous animal activity occurring in the home cage directly from the home cage rack. The proposed system is based on an electrical capacitance sensing technology that enables non-intrusive and continuous home cage monitoring. We then present a few animal activity metrics that are validated via comparison against a video camera-based tracking system. The results show that the proposed home-cage monitoring system can provide animal activity metrics that are comparable to the ones derived via a conventional video tracking system, with the advantage of system scalability, limited amount of both data generated and computational capabilities required to derive metrics.

Consider a discrete-time system in which a centralized controller (CC) is tasked with assigning at each time interval (or slot) K resources (or servers) to K out of M ≥ K nodes. A node can execute a task when assigned to a server . The tasks are independently generated at each node by stochastically symmetric and memoryless random processes and stored in a finite-capacity task queue. The tasks are time-sensitive since there is a non-zero probability, within each slot, that a task expires before being scheduled. The scheduling problem is tackled with the aim of maximizing the number of tasks completed over time (or the task-throughput) under the assumption that the CC has no direct access to the state of the task queues. The scheduling decisions at the CC are based on the outcomes of previous scheduling commands, and on the known statistical properties of the task generation and expiration processes.

Consider a single-hop wireless sensor network, where a central node (or fusion center, FC) collects data from a set of M energy harvesting (EH)-capable sensors (or nodes). In each time-slot only a subset of K ≤ M nodes can be scheduled by the FC for transmission over K orthogonal communication resources (e.g., frequencies). The scheduling problem is tackled by assuming that the FC has no direct access to the instantaneous states of the nodes' batteries, but it only knows the outcomes of previous transmissions attempts and the statistical properties of the energy harvesting/discharging processes. Based on a simple Markovian modeling of the EH and battery leakage processes, the FC's scheduling problem is formulated as partially observable Markov decision processes (POMDPs) and then cast into a restless multi-armed bandit (RMAB) framework. It is shown that in some special cases, a myopic (or greedy) scheduling policy is optimal, and that such a policy coincides with the so called Whittle index policy.

A novel random access protocol for data collection from a set of energy harvesting (EH) capable wireless nodes is proposed. The scheme is a variant of the dynamic framed ALOHA (DFA) protocol, tailored to EH networks. The proposed scheme, referred to as energy group-DFA (EG-DFA), is based on the observation that, when DFA is operated with EH-capable nodes, the optimal number of slots in a frame (i.e., the frame size) must balance two conflicting performance requirements. First, increasing the data collection rate (throughput) is well known to require a frame size equal to the backlog, namely the number of transmitting nodes. Second, since each node can store and harvest a finite energy, the number of (re)transmissions attempts that each node can perform during the channel contention process is limited. Thus, decreasing the number of uncollected data packets due to energy shortages (referred to as delivery error rate, DER) calls for a larger frame so as to avoid energy-wasting collisions. Moreover, the optimal frame size depends on both the residual energy at the nodes and the harvesting rate. Leveraging these insights, EG-DFA creates groups of nodes according to their energy availability and runs optimized and separated instances of DFA for each group. Simulation results show the advantages of EG-DFA in terms of throughput for a given DER, especially in the low-DER regime.

The design of Medium Access Control (MAC) protocols for wireless sensor networks (WSNs) has been conventionally tackled by assuming battery-powered devices and by adopting the network lifetime as the main performance criterion. While WSNs operated by energy-harvesting (EH) devices are not limited by network lifetime, they pose new design challenges due to the uncertain amount of energy that can be harvested from the environment. Novel design criteria are thus required to capture the trade-offs between the potentially infinite network lifetime and the uncertain energy availability.This paper addresses the analysis and design of WSNs with EH devices by focusing on conventional MAC protocols, namely TDMA, framed-ALOHA (FA) and dynamic-FA (DFA), and by accounting for the performance trade-offs and design issues arising due to EH. A novel metric, referred to as delivery probability, is introduced to measure the capability of a MAC protocol to deliver the measurement of any sensor in the network to the intended destination (or fusion center, FC). The interplay between delivery efficiency and time efficiency (i.e., the data collection rate at the FC), is investigated analytically using Markov models. Numerical results validate the analysis and emphasize the critical importance of accounting for both delivery probability and time efficiency in the design of EH-WSNs.

The Dynamic Framed-ALOHA (DFA) protocol is studied for wireless sensor networks with energy limitations and energy-harvesting capability. The performance of DFA in this scenario is evaluated in terms of the time efficiency (or throughput), which is routinely used to evaluate medium access protocols, and by introducing a new metric, referred to as detection efficiency, which is tailored to scenarios with energy constraints. Specifically, detection efficiency measures the ability of a multiple access protocol to collect data from nodes without depleting their energy reserves. Analysis is first performed by assuming that DFA is operated with a perfect backlog (i.e., number of sensors left to be interrogated) knowledge. Then, a low-complexity backlog estimation algorithm is presented, which is shown by numerical results to perform close to the ideal case of perfect backlog knowledge.

A critical performance criterion in backscatter modulation-based RFID sensor networks is the distance at which a RFID reader can reliably communicate with passive RFID sensors (or tags). This paper proposes to introduce a power amplifier (PA) and an energy storage device (such as a capacitor or a battery), in the hardware architecture of conventional passive RFID tags, with the aim of allowing amplification of the backscatter signal to increase the read range. This new tag architecture, referred to as Amplified Backscattering via Energy Harvesting (ABEH), can still be considered as passive, since the energy storage device is charged exclusively by harvesting energy from the RF-signal transmitted by the reader and received by the tag during idle periods. The harvested and stored energy is then used by the tags to opportunistically amplify the backscatter signal. It is noted that this architecture is significantly different from active RFID tags where the battery, charged at the time of installation, is used to supply a complete onboard transceiver so that no backscatter modulation is employed. Energy scheduling strategies, based on the trade-off between energy harvesting rate and successful transmission probability, are proposed. Performance analysis of tags with the proposed ABEH architecture is carried out over quasi-static fading channels by framing the design problem as a Markov Decision Process. Numerical results show remarkable improvement of the ABEH approach with respect to conventional passive RFID tags and provide insight into the effect of system parameters on the energy scheduling.

One and two-way communication strategies are studied in a two-hop model in which the intermediate node (relay) piggybacks data packets intended for the end-users. Assuming quasi-static fading channels, memoryless processing at all the nodes and no latency constraints, the region of simultaneously achievable throughputs for the two end-users and the relay, measured in reliably transmitted bits per second per Hz, is investigated. For both one-way and two-way transmission, outer and inner bounds to the throughput region are derived. The considered achievable strategies involve type-I HARQ, decode-and-forward and joint or superposition encoding at the relay. The trade-off between the user and relay throughputs is analyzed and studied via numerical simulations, yielding insight into optimal design choices.

A Two-Way Relay Channels (TWRC) is studied over quasi-static fading channels by focusing on the throughput of Type-I HARQ strategies. An upper bound is evaluated, along with a number of achievable results obtained by proposing different protocols combining Type-I HARQ with standard amplify or decode-and-forward techniques or more sophisticated lattice code-based strategies. Performance comparison among the different protocols shows a trade-off between the achievable throughput and the system overhead required to obtain partial channel state information at the nodes.