arXiv subjects

Thorsten Herfet

Publications and source records attributed to Thorsten Herfet.

10 recordsLinked to original sources

arXiv2025

vNV-Heap: An Ownership-Based Virtually Non-Volatile Heap for Embedded Systems

The Internet of Batteryless Things might revolutionize our understanding of connected devices by harvesting required operational energy from the environment. These systems come with the system-software challenge that the intermittently powered IoT devices have to checkpoint their state in non-volatile memory to later resume with this state when sufficient energy is available. The scarce energy resources demand that only modified data is persisted before a power failure, which requires precise modification tracking. We present vNV-Heap, the first ownership-based virtually Non-Volatile Heap for intermittently powered systems with guaranteed power-failure resilience. The heap exploits ownership systems, a zero-cost (i.e., compile-time) abstraction for example implemented by Rust, to track modifications and virtualize object persistence. To achieve power-failure resilience, our heap is designed and implemented to guarantee bounded operations by static program code analysis: For example, the heap allows for determining a worst-case energy consumption for the operation of persisting modified and currently volatile objects. The evaluation of our open-source implementation on an embedded hardware platform (i.e., ESP32-C3) shows that using our heap abstraction is more energy efficient than existing approaches while also providing runtime guarantees by static worst-case bounds.

cs.OS↗

arXiv2025

Reverse Engineering the ESP32-C3 Wi-Fi Drivers for Static Worst-Case Analysis of Intermittently-Powered Systems

The Internet of Batteryless Things revolutionizes sustainable communication as it operates on harvested energy. This harvested energy is dependent on unpredictable environmental conditions; therefore, device operations, including those of its networking stack, must be resilient to power failures. Reactive intermittent computing provides an approach for solving this by notifications of impending power failures, which is implemented by monitoring the harvested energy buffered in a capacitor. However, to use this power-failure notification and guarantee forward progress, systems must break down tasks into atomic transactions that can be predictably finished before the energy runs out. Thus, static program-code analysis must determine the worst-case energy consumption (WCEC) of all transactions. In Wi-Fi-capable devices, drivers are often closed-source, which avoids the determination of WCEC bounds for transactions since static analysis requires all code along with its semantics. In this work, we integrate an energy-aware networking stack with reverse-engineered Wi-Fi drivers to enable full-stack WCEC analysis for physical transmission and reception of packets. Further, we extended a static worst-case analysis tool with a resource-consumption model of our Wi-Fi driver. Our evaluation with the RISC-V-based ESP32-C3 platform gives worst-case bounds with our static analysis approach for the transactions of the full communication stack, therefore showing that Wi-Fi-based reactive intermittent computing is feasible.

cs.NI↗

arXiv2025

A 1-D CNN inference engine for constrained platforms

1D-CNNs are used for time series classification in various domains with a high degree of accuracy. Most implementations collect the incoming data samples in a buffer before performing inference on it. On edge devices, which are typically constrained and single-threaded, such an implementation may interfere with time-critical tasks. One such task is that of sample acquisition. In this work, we propose an inference scheme that interleaves the convolution operations between sample intervals, which allows us to reduce the inference latency. Furthermore, our scheme is well-suited for storing data in ring buffers, yielding a small memory footprint. We demonstrate these improvements by comparing our approach to TFLite's inference method, giving a 10% reduction in the inference delay while almost halving the memory usage. Our approach is feasible on common consumer devices, which we show using an AVR-based Arduino board and an ARM-based Arduino board.

cs.LG↗

arXiv2022

Reducing FEC-Complexity in Cross-Layer Predictable Data Communication

The PRRT protocol enables applications with strict performance requirements such as Cyber-Physical Systems, as it provides predictably low, end-to-end delay via cross-layer pacing and timely error correction via Hybrid ARQ (HARQ). However, the implemented HARQ uses computationally complex Maximum Distance Separable (MDS) codes to generate redundancy. In this paper we propose code partitioning for the complexity reduction of MDS codes, thereby enabling their deployment on constrained embedded devices.

cs.IT↗

arXiv2022

Polar Coding for Efficient Transport Layer Multicast

In this paper, we shed light on how an adaptive, efficient error coding in the transport layer helps ensure the application requirements. We recap the use of MDS codes and show that binary coding can significantly reduce the complexity and hence increase the applicability also for embedded devices. We exploit the persymmetric structure of the generator matrix in polar codes to establish a duality of dispersion over channels (the polarization effect) and over packets (the generality required for multicast transmission), thereby constructing systematic polar codes for incremental redundancy whose performance, despite a much lower complexity, is near to MDS codes for medium-range residual loss rates.

cs.IT↗

arXiv2021

Fristograms: Revealing and Exploiting Light Field Internals

In recent years, light field (LF) capture and processing has become an integral part of media production. The richness of information available in LFs has enabled novel applications like post-capture depth-of-field editing, 3D reconstruction, segmentation and matting, saliency detection, object detection and recognition, and mixed reality. The efficacy of such applications depends on certain underlying requirements, which are often ignored. For example, some operations such as noise-reduction, or hyperfan-filtering are only possible if a scene point Lambertian radiator. Some other operations such as the removal of obstacles or looking behind objects are only possible if there is at least one ray capturing the required scene point. Consequently, the ray distribution representing a certain scene point is an important characteristic for evaluating processing possibilities. The primary idea in this paper is to establish a relation between the capturing setup and the rays of the LF. To this end, we discretize the view frustum. Traditionally, a uniform discretization of the view frustum results in voxels that represents a single sample on a regularly spaced, 3-D grid. Instead, we use frustum-shaped voxels (froxels), by using depth and capturing-setup dependent discretization of the view frustum. Based on such discretization, we count the number of rays mapping to the same pixel on the capturing device(s). By means of this count, we propose histograms of ray-counts over the froxels (fristograms). Fristograms can be used as a tool to analyze and reveal interesting aspects of the underlying LF, like the number of rays originating from a scene point and the color distribution of these rays. As an example, we show its ability by significantly reducing the number of rays which enables noise reduction while maintaining the realistic rendering of non-Lambertian or partially occluded regions.

eess.IV↗

arXiv2020

Latency-aware and -predictable Communication with Open Protocol Stacks for Remote Drone Control

In order to create cooperating swarms of Unmanned Autonomous Vehicles (UAVs) that also interact with various other systems and devices, open and free communication systems are mandatory. This paper presents an implementation of such a communication system to incorporate the Crazyflie nano-drone as a UAV platform. The protocol stack leverages the open Predictably Reliable Real-time Transport (PRRT) protocol that adds latency-awareness and -predictability to stacks composed of standard Internet protocols. To enable the drone to receive and reply to control commands via Wi-Fi, it has been extended with a Raspberry Pi that runs two variants of the Crazybridge---a software to connect the control board to the network. To evaluate how practical this solution is for the use in control applications, the communication has been analysed with a focus on the latency properties. Our investigations show that despite using the open protocol stack---and hence opting out of specialised implementations---the resulting latencies are in the same order of magnitude (4 to 9 ms) as the latency of the proprietary link.

cs.NI↗

arXiv2019

$Δ$elta: Differential Energy-Efficiency, Latency, and Timing Analysis for Real-Time Networks

The continuously increasing degree of automation in many areas (e.g. manufacturing engineering, public infrastructure) lead to the construction of cyber-physical systems and cyber-physical networks. To both, time and energy are the most critical operating resources. Considering for instance the Tactile Internet specification, end-to-end latencies in these systems must be below 1ms, which means that both communication and system latencies are in the same order of magnitude and must be predictably low. As control loops are commonly handled over different variants of network infrastructure (e.g. mobile and fibre links) particular attention must be payed to the design of reliable, yet fast and energy-efficient data-transmission channels that are robust towards unexpected transmission failures. As design goals are often conflicting (e.g. high performance vs. low energy), it is necessary to analyze and investigate trade-offs with regards to design decisions during the construction of cyber-physical networks. In this paper, we present $Δ$elta, an approach towards a tool-supported construction process for cyber-physical networks. $Δ$elta extends the previously presented X-Lap tool by new analysis features, but keeps the original measurements facilities unchanged. $Δ$elta jointly analyzes and correlates the runtime behavior (i.e. performance, latency) and energy demand of individual system components. It provides an automated analysis with precise thread-local time interpolation, control-flow extraction, and examination of latency criticality. We further demonstrate the applicability of $Δ$elta with an evaluation of a prototypical implementation.

cs.NI↗

arXiv2018

Cross-Layer Effects on Training Neural Algorithms for Video Streaming

Nowadays Dynamic Adaptive Streaming over HTTP (DASH) is the most prevalent solution on the Internet for multimedia streaming and responsible for the majority of global traffic. DASH uses adaptive bit rate (ABR) algorithms, which select the video quality considering performance metrics such as throughput and playout buffer level. Pensieve is a system that allows to train ABR algorithms using reinforcement learning within a simulated network environment and is outperforming existing approaches in terms of achieved performance. In this paper, we demonstrate that the performance of the trained ABR algorithms depends on the implementation of the simulated environment used to train the neural network. We also show that the used congestion control algorithm impacts the algorithms' performance due to cross-layer effects.

cs.NI↗

arXiv2018

X-Lap: A Systems Approach for Cross-Layer Profiling and Latency Analysis for Cyber-Physical Networks

Networked control applications for cyber-physical networks demand predictable and reliable real-time communication. Applications of this domain have to cooperate with network protocols, the operating system, and the hardware to improve safety properties and increase resource efficiency. In consequence, a cross-layer approach is necessary for the design and holistic optimisation of cyber-physical systems and networks. This paper presents X-Lap, a cross-layer, inter-host timing analysis tool tailored to the needs of real-time communication. We use X-Lap to evaluate the timing behaviour of a reliable real-time communication protocol. Our analysis identifies parts of the protocol which are responsible for unwanted jitter. To system designers, X-Lap provides useful support for the design and evaluation of networked real-time systems.

cs.NI↗