With the rapid advancement of wearable devices, sensor-based human activity recognition has emerged as a fundamental research area with broad applications in various domains. While significant progress has been made in this research field, energy consumption remains a critical aspect that deserves special attention. Recognizing human activities while optimizing energy consumption is essential for prolonging device battery life, reducing charging frequency, and ensuring uninterrupted monitoring and functionality. The primary objective of this survey paper is to provide a comprehensive review of energy-aware wearable human activity recognition techniques based on wearable sensors without considering vision-based systems. In particular, it aims to explore the state-of-the-art approaches and methodologies that integrate activity recognition with energy management strategies. Finally, by surveying the existing literature, this paper aims to shed light on the challenges, opportunities and potential solutions for energy-aware human activity recognition.

The design of IoT systems supporting deep learning capabilities is mainly based today on data transmission to the cloud back-end. Recently, edge computing solutions, which keep most computing and communication as close as possible to user devices have emerged as possible alternatives to reduce energy consumption, limit latency, and safeguard privacy. Early-exit models have been proposed as a way to combine models with different depths into a single architecture. The aim of this article is to investigate the energy expenditure of a distributed IoT system based on early exit architectures, by taking human activity recognition as a case study. We propose a simulation study based on an analytical model and hardware characterization to estimate the trade-off between the accuracy and energy of early exit-based configurations. Experimental results highlight nontrivial relationships between architectures, computing platforms, and communication link. For instance, we found that early-exit strategies do not ensure energy reductions with respect to a cloud-based solution if the same accuracy levels are kept; nonetheless, by tolerating a 1.5% decrease in accuracy, it is possible to achieve a reduction of around 40% of the total energy consumption.

Deep learning is nowadays considered state-of-the-art technology in many applications thanks to huge performance capabilities. However, the accuracy levels that can be obtained with these models entail computationally demanding resources. This results in a challenging task when such systems have to be deployed on edge devices with tight computing, memory, and communication requirements and when energy expenditure and inference delays have to be kept under control. Early exit is a design methodology aimed at reducing the burden of neural networks on computational resources, trading off accuracy for latency. In this work, we aim at exploring the use of early exit for human activity recognition tasks. In particular, we propose an experimental assessment of the accuracy–latency trade-off on different deep network architectures across various publicly available datasets. We also evaluate the impact of early exiting in distributed environments by taking into account communication technologies. Experimental results provide evidence of the significant gain provided by early exits in terms of latency (up to 35×), without a reduction in accuracy (in most cases), confirming the viability of an adaptive approach. In a distributed environment, early exit results are not beneficial in all situations. In particular, it is not convenient for models that are very fast (with inference latency lower than, or as equal as, that of communication) and for models that are forced to make extensive use of far exit points to satisfy the accuracy requirements. Therefore, communication delays in a distributed environment shape performance in an architecture-dependent way.

Abstract The capability of effectively quantifying the uncertainty associated to a given prediction is an important task in many applications that range from drug design to autonomous driving, providing valuable information to many downstream decision-making processes. The increasing capacity of novel machine learning models, and the growing amount of data on which these systems are trained poses however significant issues to be addressed. Recent research advocated the need for evaluating learning systems not only according to traditional accuracy metrics but also according to the computational complexity required to design them, toward a perspective of sustainability and inclusivity. In this work, we present an empirical investigation aimed at assessing the impact of uniform sampling on the reduction in computational requirements, the quality of regression, and on its uncertainty quantification. We performed several experiments with recent state-of-the-art methods characterized by statistical guarantees whose performances have been measured according to different metrics for evaluating uncertainty quantification (i.e., coverage and length of prediction intervals) and regression (i.e., errors measures and correlation). Experimental results highlight possible interesting trade-offs between computation time, regression and uncertainty evaluation quality, thus confirming the viability of sampling-based approaches to overcome computational bottlenecks without significantly affecting the quality of predictions.

The development of large-scale systems of networked embedded devices with sensing capabilities relies on the availability of low-cost and resource-constrained components However, the reduced precision and accuracy of low-cost sensors on board of low-power platforms risks to impair the overall reliability of these systems, thus preventing their potential diffusion, especially in deployments with several (e.g., hundreds or more) nodes. Hence, ensuring the required quality of measurements along the lifetime of a sensor network represents a key challenge, which is often tackled also by means of calibration techniques. In this article, we propose a novel approach to multihop calibration, targeting the derivation of a spanning tree that encompasses the optimization of a biobjective problem. Indeed, since minimum spanning trees can be related to the energy budget of a network and shortest path trees can be used as a model for the minimization of the cumulative calibration errors, the search for a spanning tree that simultaneously optimizes the two metrics represents a useful direction toward the design of effective and efficient calibration strategies. To this aim, we introduce a method based on the Prim-Dijkstra algorithm, which represents an effective heuristics for effective search of solutions that could represent a tradeoff between the accuracy of multihop calibration and the energy expenditure needed to calibrate sensors. The proposed approach allows fast derivation of different solutions by means of a single parameter, thus enabling the efficient exploration of the design space even in large-scale scenarios as confirmed by numerical results obtained for validation.

Reliability is nowadays considered a key requirement in wireless sensor networks for their increasing diffusion in various Internet of Things applications. However, radio interference from various sources may heavily affect the performance of wirelessly connected embedded devices, resulting into increased packet collisions and congestions. There is therefore a widely recognized need of both theoretical and practical investigations capable of shedding light on the factors affecting the operativeness of sensor networks subject to interference. In this paper, we investigate the role of packet length into the reliability and energy efficiency of low-power medium access protocols. Specifically, we propose a mathematical model to explore the functional dependence of the reliability of a pair of sensor nodes under interference from packet length. We also present a wide range of experimental evaluations aimed at validating the model and at providing novel insights on dependability issues. In particular, we assess the performance of Contiki's default medium access control layer (together with that of an always listening receiver, as a baseline) in terms of packet loss rate and energy efficiency for varying payload lengths. Experimental results highlight the interplay between packet size and interference and in particular the tradeoff between the robustness against interference and the overhead imposed to communication as a function of the length of data packets. The Pareto curve describing the energy efficiency as a function of the packet loss rate, demonstrates the existence of intermediate packet size representing an optimal choice for balancing energy consumption and communication reliability, enabling adequate system dimensioning at design-level.

No abstract available.

Out-of-band signaling provides a valuable support to the management of wireless sensor networks. Among other signals, sound has a propagation speed in air which is fast enough to be neglected in typical sensor network applications, and slow enough to be measured by means of low-cost embedded systems. An ultrasonic triggering mechanism has been recently developed for VirtualSense, an ultra-low-power sensor node featuring a Java runtime environment. Ultrasonic triggers provide on-demand wakeup capabilities that can be exploited by routing nodes to be switched on whenever there is a packet to be routed towards the sink, while spending all the idle time in an ultra low power inactive state where even the radio module is turned off. In addition, the same hardware components can be used to perform pairwise distance measurements that can be exploited for localization. In this demonstration we show the possible use of VirtualSense ultrasonic wake-up modules as distance estimators.

Predicting protein function at the proteomic-scale is a key task in computational systems biology. High-throughput experimental methods have recently made available many protein interaction networks that need to be analyzed in order to provide insight into the functional role of proteins in the organization of the cell. We propose here a new approach to computational function annotation of protein interaction maps based on a random walk algorithm. Our method exploits the whole topology of the network according to the basic principles of a ranking algorithm for link analysis. We apply the proposed algorithm to analyze the yeast protein interaction network and show that it represents a valid alternative to other annotation techniques based on network analysis by comparing it with the effective majority vote algorithm.

MOTIVATION: Matching a biological sequence against a probabilistic pattern (or profile) is a common task in computational biology. A probabilistic profile, represented as a scoring matrix, is more suitable than a deterministic pattern to retain the peculiarities of a given segment of a family of biological sequences. Brute-force algorithms take O(NP) to match a sequence of N characters against a profile of length P << N. RESULTS: In this work, we exploit string compression techniques to speedup brute-force profile matching. We present two algorithms, based on run-length and LZ78 encodings, that reduce computational complexity by the compression factor of the encoding.