Accuracy Requirements Of Applications Research Articles

Mez: An Adaptive Messaging System for Latency-Sensitive Multi-Camera Machine Vision at the IoT Edge

Mez is a novel publish-subscribe messaging system for latency sensitive multi-camera machine vision applications at the IoT Edge. The unlicensed wireless communication in IoT Edge systems are characterized by large latency variations due to intermittent channel interference. To achieve user specified latency in the presence of wireless channel interference, Mez takes advantage of the ability of machine vision applications to temporarily tolerate lower quality video frames if overall application accuracy is not too adversely affected. Control knobs that involve lossy image transformation techniques that modify the frame size, and thereby the video frame transfer latency, are identified. Mez implements a network latency feedback controller that adapts to channel conditions by dynamically adjusting the video frame quality using the image transformation control knobs, so as to simultaneously satisfy latency and application accuracy requirements. Additionally, Mez uses an application domain specific design of the storage layer to provide low latency operations. Experimental evaluation on an IoT Edge testbed with a pedestrian detection machine vision application indicates that Mez is able to tolerate latency variations of up to 10x with a worst-case reduction of 4.2% of the application accuracy F1 score metric. The performance of Mez is also experimentally evaluated against state-of-the-art low latency NATS messaging system.

High Performance GPU Tensor Completion With Tubal-Sampling Pattern

Data completion is a problem of filling missing or unobserved elements of partially observed datasets. Data completion algorithms have received wide attention and achievements in diverse domains including data mining, signal processing, and computer vision. We observe a ubiquitous tubal-sampling pattern in big data and Internet of Things (IoT) applications, which is introduced by many reasons such as high data acquisition cost, downsampling for data compression, sensor node failures, and packet losses in low-power wireless transmissions. To meet the time and accuracy requirements of applications, data completion methods are expected to be accurate as well as fast. However, the existing methods for data completion with the tubal-sampling pattern are either accurate or fast, but not both. In this article, we propose high-performance graphics processing unit (GPU) tensor completion for data completion with the tubal-sampling pattern. First, by exploiting the convolution theorem, we split a tensor least-squares minimization problem into multiple least-squares sub-problems in the frequency domain. In this way, massive parallelisms are exposed for many-core GPU architectures while still preserving high recovery accuracy. Second, we propose computing slice-level and tube-level tasks in batches to improve GPU utilization. Third, we reduce the data transfer cost by eliminating the accesses to the CPU memory inside algorithm loop structures. The experimental results show that the proposed tensor completion is both fast and accurate. Using synthetic data of varying sizes, the proposed GPU tensor completion achieves maximum 248.18×, 7, 403.27×, and 33.27× speedups over the CPU MATLAB implementation, GPU element-sampling tensor completion in the cuTensor-tubal library, and GPU high-performance matrix completion, respectively. With a 50 percent sampling rate, the proposed GPU tensor completion achieves a recovery error of 1.40e-5, which is comparable with that of the GPU element-sampling tensor completion and three orders of magnitude better than that of the GPU high-performance matrix completion. To utilize multiple GPUs in servers, we design a multi-GPU scheme for tubal-sampling tensor completion. The multi-GPU tensor completion achieves maximum 1.89× speedup on two GPUs versus on a single GPU for medium or big tensors. We further evaluate the performance of the proposed GPU tensor completion in three real applications, namely, video transmission in wireless camera networks, RF fingerprint-based indoor localization, and seismic data completion, and it achieves maximum speedups of 448.68×, 24.63×, and 311.54×, respectively. We integrate this high-performance GPU tensor completion implementation into the cuTensor-tubal library to support various applications.

SNR Degradation in Undersampled Phase Measurement Systems.

A wide range of measuring applications rely on phase estimation on sinusoidal signals. These systems, where the estimation is mainly implemented in the digital domain, can generally benefit from the use of undersampling to reduce the digitizer and subsequent digital processing requirements. This may be crucial when the application characteristics necessarily imply a simple and inexpensive sensor. However, practical limitations related to the phase stability of the band-pass filter prior digitization establish restrictions to the reduction of noise bandwidth. Due to this, the undersampling intensity is practically defined by noise aliasing, taking into account the amount of signal-to-noise ratio (SNR) reduction caused by it considering the application accuracy requirements. This work analyzes the relationship between undersampling frequency and SNR reduction, conditioned by the stability requirements of the filter that defines the noise bandwidth before digitization. The effect of undersampling is quantified in a practical situation where phase differences are measured by in-phase and quadrature (I/Q) demodulation for an infrared ranging application.

Achieving Guaranteed Anonymity in GPS Traces via Uncertainty-Aware Path Cloaking

The integration of Global Positioning System (GPS) receivers and sensors into mobile devices has enabled collaborative sensing applications, which monitor the dynamics of environments through opportunistic collection of data from many users' devices. One example that motivates this paper is a probe-vehicle-based automotive traffic monitoring system, which estimates traffic congestion from GPS velocity measurements reported from many drivers. This paper considers the problem of achieving guaranteed anonymity in a locational data set that includes location traces from many users, while maintaining high data accuracy. We consider two methods to reidentify anonymous location traces, target tracking, and home identification, and observe that known privacy algorithms cannot achieve high application accuracy requirements or fail to provide privacy guarantees for drivers in low-density areas. To overcome these challenges, we derive a novel time-to-confusion criterion to characterize privacy in a locational data set and propose a disclosure control algorithm (called uncertainty-aware path cloaking algorithm) that selectively reveals GPS samples to limit the maximum time-to-confusion for all vehicles. Through trace-driven simulations using real GPS traces from 312 vehicles, we demonstrate that this algorithm effectively limits tracking risks, in particular, by eliminating tracking outliers. It also achieves significant data accuracy improvements compared to known algorithms. We then present two enhancements to the algorithm. First, it also addresses the home identification risk by reducing location information revealed at the start and end of trips. Second, it also considers heading information reported by users in the tracking model. This version can thus protect users who are moving in dense areas but in a different direction from the majority.