In recent years, networking hardware development has primarily focused on speed rather than power efficiency. By contrast, computing hardware has received a lot more attention given its dominant power footprint, especially in machine-learning (ML) data centers. With faster networks, we spend less time communicating and get more useful work out of the (increasingly expensive) computing hardware. But, the faster the network, the more time it idles and the worse its energy efficiency, which is magnified by the notorious lack of power proportionality of networking equipment.

Reducing our society's energy demand is critical to address the sustainability challenge. While the Internet currently accounts for 1–-1.5% of global electricity consumption and continues to grow, the energy demands of one of its core components---routers---remain poorly understood. The available power data is limited and not fine-grained enough, offering little actionable insight into strategies for effectively reducing the Internet's energy consumption.

Generalizing machine learning (ML) models for network traffic dynamics tends to be considered a lost cause. Hence for every new task, we design new models and train them on model-specific datasets closely mimicking the deployment environments. Yet, an ML architecture called Transformer has enabled previously unimaginable generalization in other domains. Nowadays, one can download a model pre-trained on massive datasets and only fine-tune it for a specific task and context with comparatively little time and data. These fine-tuned models are now state-of-the-art for many benchmarks.

Low-power wireless communication is a central building block of Cyber-physical Systems and the Internet of Things. Conventional low-power wireless protocols make avoiding packet collisions a cornerstone design choice. The concept of synchronous transmissions challenges this view. As collisions are not necessarily destructive, under specific circumstances, commodity low-power wireless radios are often able to receive useful information even in the presence of superimposed signals from different transmitters. We survey the growing number of protocols that exploit synchronous transmissions for higher robustness and efficiency as well as unprecedented functionality and versatility compared to conventional designs. The illustration of protocols based on synchronous transmissions is cast in a conceptional framework we establish, with the goal of highlighting differences and similarities among the proposed solutions. We conclude the paper with a discussion on open research questions in this field.

Closing feedback loops fast and over long distances is key to emerging cyber-physical applications; for example, robot motion control and swarm coordination require update intervals of tens of milliseconds. Low-power wireless communication technology is preferred for its low cost, small form factor, and flexibility, especially if the devices support multi-hop communication. Thus far, however, feedback control over multi-hop low-power wireless networks has only been demonstrated for update intervals on the order of seconds. To fill this gap, this article presents a wireless embedded system that supports dynamic mode changes and tames imperfections impairing control performance (e.g., jitter and message loss), and a control design that exploits the essential properties of this system to provably guarantee closed-loop stability for physical processes with linear time-invariant dynamics in the presence of mode changes. Using experiments on a cyber-physical testbed with 20 wireless devices and multiple cart-pole systems, we are the first to demonstrate and evaluate feedback control and coordination with mode changes over multi-hop networks for update intervals of 20 to 50 milliseconds.

The Dual Processor Platform (DPP) is a novel architecture template for networked embedded systems based on a strictly asynchronous processor interconnect that allows to minimize interference with a proven predictable behavior [7]. Contrary to traditional platforms [1, 5] DPP tries to mitigate interference by isolating different tasks and mapping them onto dedicated hardware resources. Typically, two different task sets - (i) sensing/actuation/data processing and (ii) communication - are mapped onto two different physically separated processing elements (usually low-power microcontrollers), allowing each to be optimized according to their individual requirements. Such hardware partitioning is a standard approach, frequently found in more complex sensor system implementations [2]. While the communication processor handles wireless packet transmission and reception, the application processor is dedicated to the sensor data acquisition, processing and actuation. But this strict separation of resources and function also requires a processor interconnect: BOLT [7], a stateful processor interconnect specifically designed based on this paradigm allows a strict decoupling of the power, clock and time domains of the two processing elements by allowing only asynchronous message passing between the two. The strict limitation to an asynchronous interface allows for predictable run-time behavior that, in addition to typical behavior observed from experiments, has been formally verified [7]. The most notable advantages of this approach are:

This abstract describes the first public demonstration of feedback control and coordination of multiple physical systems over a dynamic multi-hop low-power wireless network with update intervals of tens of milliseconds. Our running system can dynamically change between different sets of application tasks (e.g., sensing, actuation, control) executing on the spatially distributed embedded devices, while closed-loop stability is provably guaranteed even across those so-called mode changes. Moreover, any subset of the devices can move freely, which does not affect closed-loop stability and control performance as long as the wireless network remains connected.

Wired fieldbuses have long been proven effective in supporting Cyber-Physical Systems (CPS). However, various domains are now striving for wireless solutions due to ease of deployment or novel functionality requiring the ability to support mobile devices. Low-power wireless protocols have been proposed in response to this need, but requirements of a large class of CPS applications can still not be satisfied. We thus propose Time-Triggered Wireless (TTW), a distributed low-power wireless system design that minimizes communication energy consumption and offers end-to-end timing predictability, runtime adaptability, reliability, and low latency. Evaluation shows a 2× reduction in communication latency and 33-40% lower radio-on time compared with DRP, the closest related work, validating the suitability of TTW for new exciting wireless CPS applications.

If visions and forecasts of industry come true then we will be soon surrounded by billions of interconnected embedded devices. We will interact with them in a cyber-human symbiosis, they will not only observe us but also our environment, and they will be part of many visible and ubiquitous objects around us. The information that is collectively gathered and analyzed is supposed to help us in our daily live, in making faithful decisions, but it will also directly be used for actuation and it will cause changes by means of local and global control loops.

In cyber-physical systems (CPS), the communication among the sensing, actuating, and computing elements is often subject to hard real-time constraints. Real-time communication among wireless network interfaces and real-time scheduling for complex, dynamic applications have been intensively studied. Despite these major efforts, there is still a significant gap to fill. In particular, the integration of several real-time components to provide end-to-end real-time guarantees between interfaces of distributed applications in wireless CPS is an unsolved problem. We thus present a distributed protocol that considers the complete transmission chain including peripheral busses, memory accesses, networking interfaces, and the wireless real-time protocol. Our protocol provably guarantees that message buffers along this chain do not overflow and that all messages received at the destination application interface meet their end-to-end deadlines. To achieve this while being adaptive to unpredictable changes in the system and the real-time traffic requirements, our protocol establishes at run-time a set of contracts among all major elements of the transmission chain based on a worst-case delay and buffer analysis of the overall system. Using simulations, we validate that our analytic bounds are both safe and tight.