The Digital Power Network (DPN) is an energy-on-demand approach. In terms of Internet of Things (IoT), it treats the energy itself as a `thing' to be manipulated (in contrast to energy as the `thing's enabler'). The approach is mostly appropriate for energy starving micro-grids with limited capacity, such as a generator for the home while the power grid is down. The process starts with a request of a user (such as, appliance) for energy. Each appliance, energy source or energy storage has an address which is able to communicate its status. A network server, collects all requests and optimizes the energy dissemination based on priority and availability. Energy is then routed in discrete units to each particular address (say air-condition, or, A/C unit). Contrary to packets of data over a computer network whose data bits are characterized by well-behaved voltage and current values at high frequencies, here we deal with energy demands at highvoltage, low-frequency and fluctuating current. For example, turning a motor ON requires 8 times more power than the level needed to maintain a steady states operation. Our approach is seamlessly integrating all energy resources (including alternative sources), energy storage units and the loads since they are but addresses in the network. Optimization of energy requests and the analysis of satisfying these requests is the topic of this paper. Under energy constraints and unlike the current power grid, for example, some energy requests are queued and granted later. While the ultimate goal is to fuse information and energy together through energy digitization, in its simplest form, this micro-grid can be realized by overlaying an auxiliary (communication) network of controllers on top of an energy delivery network and coupling the two through an array of addressable digital power switches.
more »
« less
Improving Communication through Overlay Detours: Pipe Dream or Actionable Insight?
It has been long observed that communication between a client and a content server using overlay detours may result in substantially better performance than a native path offered by IP routing. Yet the use of detours has been limited to distributed platforms such as Akamai. This paper poses a question - how can clients practically take advantage of overlay detours without modification to content servers (which are obviously outside clients' control)? We have posited elsewhere that the emergence of gigabit-to-the-home access networks would precipitate a new home network appliance, which would maintain permanent presence on the Internet for the users and have general computing and storage capabilities. Given such an appliance, our vision is that Internet users may form cooperatives in which members agree to serve as waypoints points to each other to improve each other's Internet experience. To make detours transparent to the server, we leverage MPTCP, which normally allows a device to communicate with the server on several network interfaces in parallel but we use it to communicate through external waypoint hosts. The waypoints then mimic MPTCP's subflows to the server, making the server oblivious to the overlay detours as long as it supports MPTCP.
more »
« less
- Award ID(s):
- 1647145
- PAR ID:
- 10088045
- Date Published:
- Journal Name:
- IEEE 38th International Conference on Distributed Computing Systems (ICDCS)
- Page Range / eLocation ID:
- 1422 to 1431
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
With the acceleration of ICT technologies and the Internet of Things (IoT) paradigm, smart residential environments , also known as smart homes are becoming increasingly common. These environments have significant potential for the development of intelligent energy management systems, and have therefore attracted significant attention from both academia and industry. An enabling building block for these systems is the ability of obtaining energy consumption at the appliance-level. This information is usually inferred from electric signals data (e.g., current) collected by a smart meter or a smart outlet, a problem known as appliance recognition . Several previous approaches for appliance recognition have proposed load disaggregation techniques for smart meter data. However, these approaches are often very inaccurate for low consumption and multi-state appliances. Recently, Machine Learning (ML) techniques have been proposed for appliance recognition. These approaches are mainly based on passive MLs, thus requiring pre-labeled data to be trained. This makes such approaches unable to rapidly adapt to the constantly changing availability and heterogeneity of appliances on the market. In a home setting scenario, it is natural to consider the involvement of users in the labeling process, as appliances’ electric signatures are collected. This type of learning falls into the category of Stream-based Active Learning (SAL). SAL has been mainly investigated assuming the presence of an expert , always available and willing to label the collected samples. Nevertheless, a home user may lack such availability, and in general present a more erratic and user-dependent behavior. In this paper, we develop a SAL algorithm, called K -Active-Neighbors (KAN), for the problem of household appliance recognition. Differently from previous approaches, KAN jointly learns the user behavior and the appliance signatures. KAN dynamically adjusts the querying strategy to increase accuracy by considering the user availability as well as the quality of the collected signatures. Such quality is defined as a combination of informativeness , representativeness , and confidence score of the signature compared to the current knowledge. To test KAN versus state-of-the-art approaches, we use real appliance data collected by a low-cost Arduino-based smart outlet as well as the ECO smart home dataset. Furthermore, we use a real dataset to model user behavior. Results show that KAN is able to achieve high accuracy with minimal data, i.e., signatures of short length and collected at low frequency.more » « less
-
The advent of ultrabroadband Internet connectivity brings a 2-3 orders of magnitude jump in the capacity of access networks (a.k.a. the “last mile”). Beyond mere capacity increase, this leap represents a qualitative shift in the overall Internet environment. Therefore, we argue that only by seizing the opportunity to re-think the way we structure network applications and services can we realize the full potential ultrabroadband provides. Specifically, with ultrabroadband residential networks, we have the opportunity to re-center our digital lives around our residence, similar to how our physical lives generally center around our homes. To this end, we introduce a new appliance in home networks–a “home point of presence”–that provides a variety of services to the users in the house regardless of where they are physically located and connected to the network. We illustrate the utility of this appliance by discussing a range of new services that both bring new functionality to the users and improve performance of existing applications.more » « less
-
Monitoring performance and availability are critical to operating successful content distribution networks. Internet measurements provide the data needed for traffic engineering, alerting, and network diagnostics. While there are significant benefits to performing end-user active measurements, these capabilities are limited to a small number of content providers with application control. In this work, we present a solution to the long-standing problem of issuing active measurements from clients without requiring application control, e.g., injecting JavaScript to the content served. Our approach uses server-side programmable features of the Network Error Logging specification that allow a CDN to induce a browser connection to an HTTPS server of the CDN's choosing without application control.more » « less
-
We study stochastic gradient descent (SGD) with local iterations in the presence of malicious/Byzantine clients, motivated by the federated learning. The clients, instead of communicating with the central server in every iteration, maintain their local models, which they update by taking several SGD iterations based on their own datasets and then communicate the net update with the server, thereby achieving communication-efficiency. Furthermore, only a subset of clients communicate with the server, and this subset may be different at different synchronization times. The Byzantine clients may collaborate and send arbitrary vectors to the server to disrupt the learning process. To combat the adversary, we employ an efficient high-dimensional robust mean estimation algorithm from Steinhardt et al.~i̧te[ITCS 2018]Resilience_SCV18 at the server to filter-out corrupt vectors; and to analyze the outlier-filtering procedure, we develop a novel matrix concentration result that may be of independent interest. We provide convergence analyses for strongly-convex and non-convex smooth objectives in the heterogeneous data setting, where different clients may have different local datasets, and we do not make any probabilistic assumptions on data generation. We believe that ours is the first Byzantine-resilient algorithm and analysis with local iterations. We derive our convergence results under minimal assumptions of bounded variance for SGD and bounded gradient dissimilarity (which captures heterogeneity among local datasets). We also extend our results to the case when clients compute full-batch gradients.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1109/ICDCS.2018.00143
