Video Analytics System has emerged as a promising technology to realize deep neural network based intelligent applications for video streams. Its objective is to maximize the video analytics performance of video streams, such as accuracy, while utilizing the limited computing resource capacity efficiently. Existing video analytics systems attempt to adapt configurations to optimize resource-accuracy tradeoffs under limited resource capacity. Especially, several works recently propose a configuration adaptation algorithm based on online profiling to overcome the inefficiency of one-time offline profiling caused by the dynamics of the configuration’s impact on video analytics accuracy. However, their systems are inefficient or limited to process the video analytics of multiple video streams in a GPU-enabled edge server. Furthermore, their online profiling methods still leads to a high profiling cost. In this paper, we design a video analytics system to adapt configurations for optimizing the resource-accuracy tradeoffs of multiple video streams with respect to frame rate and resolution fully under limited resource capacity of a GPU-enable edge server. In addition, we propose a lightweight online profiling method, utilizing the underlying characteristics of the video objects. Then, based on it, we propose a configuration adaptation algorithm to find the best configuration of each video stream and minimize accuracy degradation of multiple video streams under limited resource capacity of a GPU-enable edge server. To evaluate the proposed algorithm, we use a subset of surveillance videos and annotations from VIRAT 2.0 Ground dataset. The experimental results show that, in a GPU-enabled edge server, our video analytics system achieves the optimal configuration adaptation on resource-accuracy tradeoffs and our algorithm reduce the profiling cost of existing systems significantly while achieving the video analytics performance comparable to them.

The growth of advanced metering infrastructure (AMI) deployment has enabled intelligent power control and management at the customer level; Highly accurate individual short-term load forecasting is crucial for this precise control per customer. Recently, deep learning has been widely adopted to improve forecasting accuracy. However, training an individual deep network (local training) has overfitting issues due to data paucity per customer. Thus, a pooling scheme has been introduced to augment a training dataset by batching several customers’ data. Nevertheless, there is room for existing pooling approaches to further improve accuracy by considering distribution heterogeneity within a customer dataset. In addition, their static pool assignment only with a customer’s training dataset may cause accuracy degradation under concept drift in serving time. To overcome these, we propose a Temporal Data Pooling (TDP) that constructs data pools at the data sample level with a novel distribution inference method and theoretical analysis. It allows the most probable forecasting model to serve predictions while resolving data shortage issues in local training. The TDP outperforms the other six competing methods for point and probabilistic forecasting; it shows robust accuracy under concept drift. Moreover, it demonstrates superior accuracy for unseen customers without additional training, proving its scalability.

As smart metering technology evolves, power suppliers can make low-cost, low-risk estimation of customer-side power consumption by analyzing energy demand data collected in real-time. With advances network infrastructure, smart sensors, and various monitoring technologies, a standardized energy metering infrastructure, called advanced metering infrastructure (AMI), has been introduced and deployed to urban households to allow them to develop efficient power generation plans. Compared to traditional stochastic approaches for time-series data analysis, deep-learning methods have shown superior accuracy on many prediction applications. Because smart meters and infrastructure monitors produce a series of measurements over time, a large amount of data is accumulated, creating a large data stream, which takes much time from data generation to deployment of deep-learning model training. In this article, we propose an accelerated computing system that considers time-variant properties for accurate prediction of energy demand by processing the AMI stream data. The proposed system is a real-time training/inference system that deploys AMI data over a distributed edge cloud. It comprises two core components: an adaptive incremental learning solver and deep-learning acceleration with FPGA-GPU resource scheduling. An adaptive incremental learning scheme adjusts the batch/epoch in training iteration to reduce the time delay of the latest trained model, while trying to prevent biased-training due to the sub-optimal optimizer of incremental learning. In addition, a resource scheduling scheme manages various accelerator resources for accelerated deep-learning processing while minimizing the computational cost. The experimental results demonstrated that our method achieved good performance for adaptive batch size and epoch for incremental learning while guaranteeing a low inference error, a high model score, and queue stability with cost efficient processing.

A joint selected strategy is proposed for wireless sensor networks to address the energy consumption problem in this letter. The proposed strategy saves energy by selecting the sensor and modulation constellation size. Results from simulations show that the proposed strategy obtains significant energy reduction compared with the existing strategy under the same environment.