Including dynamic aspects in the environmental assessment of power systems allows computing the environmental benefits of demand-side management strategies for the smart grid which could not be assessed with static data such as shifting part of the demand from one period to another. Several methodological approaches have been developed in life cycle assessment to account for dynamic aspects, but none has given much attention to the demand side of the equation. However, demand is also prone to fluctuate in time and its misrepresentation may lead to additional errors. In this study, a stochastic approach was applied to model the fluctuating residential power demand of Canadians' homes. An hourly and a yearly average electricity mix were then used to compute the environmental impacts of the hourly or yearly average homes' electricity demand. Finally, an approach combining an average and a marginal hourly electricity mix was then proposed to assess the benefits of a simple demand side management strategy: the shifting of homes' dryers loads up to two hours later than usual. Results show that assuming a constant demand or electricity mix both leads to errors which may be as high as 150% depending on the period of the month assessed. Moreover, using an hourly average electricity mix to set up the demand side strategy increases climate change impact by 0.6% whereas using a marginal mix decreases climate change impact by 10%.
To design effective control schemes for energy efficiency in data centers, it is crucial to have a thermal model of the system. Constructing thermal models of data centers for temperature prediction is extremely challenging, due to inherent complexity. Computational fluid dynamics (CFD) simulations or physical heat transfer equations are conventionally used to construct such thermal models. More recent approaches combine physical heat transfer rules and data-driven methods in an effort to obtain more accurate models. Our proposed adaptive learning-based thermal model (ALTM) is fast, adapts to thermal changes in the data center environment, and does not require prior knowledge of heat transfer rules between data center entities. Unlike other methods, ALTM is a holistic thermal model that predicts temperature of critical zones using data center operational variables as inputs. The operational variables are the controllable parameters and easily obtained measurements from IT and cooling units. A key use case for ALTM is that it can be effectively used for thermal-aware workload schedulers or cooling system controllers. Our results confirm the accuracy and adaptability of the model.
Around the world, vast improvements in public transportation methods in urban environments have been made. However, in densely populated areas, the bicycle remains a very useful means of transportation. Its small size and minimal environmental impact are the critical factors that maintain its relevance. Moreover, the advancement of connected devices and sharing-based services have allowed private vendors to develop bike-sharing programs, giving millions access to bike transportation around the globe. These bike-sharing programs rely on the user to check out and return the bike to a designated bike-holding station. With the growth of Internet of Things (IoT) services and wirelessly connected devices, there is a major benefit in enabling vendors to track their bicycle assets. Satellite navigation has come a long way, however, it requires a large power overhead. This paper proposes an energy-efficient bicycle tracking system that utilizes bicycle powered Bluetooth Low Energy (BLE) beacons and Long Range (LoRa) type base-stations in order to track and maintain a real-time location-based inventory of all assets. The BLE beacons are used to track individual bicycle assets based on Received Signal Strength Indicator (RSSI) proximity and the LoRa base stations exploit longer range communication capabilities to transmit asset location information between each other, for added management capabilities. Preliminary proximity estimations using BLE beacons in an urban outdoor environment show promising results with proximity accuracy consistently under 2 meters.
In the last decade Cloud computing has seen a surge of popularity. Clouds, with their scale and high functionality, are used to outsource various infrastructure, platform, and software services. However, relaying solely on distant Cloud Data Centers (DCs) can be inefficient for many applications concerning mobile devices and Internet of Things (IoT) in general. A more decentralized Fog computing paradigm has been proposed to augment Cloud availability and execution. This work addresses latency and power consumption in Fog computing networks. Models for power consumption and delay are proposed. Performance of Fog computing is estimated using parameters setting based on real-world equipment and traffic. Our results tackle the balance between Fog and Cloud. Applications requiring heavy computations (relative to size of offloaded data) are best served by Cloud DCs, while it is faster (and more power-efficient) to compute "lighter" requests in the Fog Nodes (FNs). However, where is the trade-off between power consumption and delay in the context of Fog and Cloud? We answer this question modeling multiple architectures and using various network scenarios.
Smart cities bring new technological advances that help improve everyday life. One such improvement is the ability to map out a city based on a characteristic. The amount of acoustic noise in an environment has many impacts on human life and has the potential to be collected wirelessly. Unfortunately, systems made today would not have the battery life capabilities to handle such a high demand if continuous transmission was used. In this paper, the design of a testbed for a smart microphone system is presented. In order to promote power savings, an analog-to-digital converter (ADC) which dynamically switches between high and low power modes in response to environmental noise is presented. Specifically, the high power ADC mode is triggered from a spike in the acoustic noise level. Ideally, this would be configurable and allow for detection of different types of sound such as human voice or music. A framework for basic environmental sound collection is presented along with preliminary results of the testbed.
A power system is considered to be fully observed when all bus voltages and currents are known. Phasor Measurement Unit (PMUs) opt to achieve system's observability when placed at selected buses throughout the system. Optimal PMU placement is concerned with determining the least count and position of PMUs needed to fully observe the system. This paper considers such optimization problem in the presence of zero injection buses (ZIBs). A novel solution is proposed where the system observability is modeled as a graph. A special transformation is presented to reduce the complexity of the optimization while factoring in the contribution of a ZIB to the system's observability. The optimization problem is then formulated as an integer linear program. Validation is conducted using different IEEE bus systems and using randomly generated power grid topologies. The results confirm the performance advantage of the proposed approach over competing scheme in the literature.
In an enterprise Wi-Fi network, Mobile users may be covered by multiple enterprise access points (APs). To optimize resource allocation, a soft handover is require in which the user's device is seamlessly transferred from one AP to another, and this decision made centrally by a Wi-Fi network controller. Unfortunately, state-of-the-art soft handover mechanisms are often designed to optimize resources from the network provider's point of view and do not take into account user's real-time behaviours, which may affect user's Quality of Experience (QoE). In this paper, a new machine learning (ML)-based method presented to find an optimal handover mechanism. This method allows to predict whether the handover that is going to happen will maintain QoE when users are moving inside a building. Our proposed method improves 34% of user throughput compared to state-of-the-art algorithms.
Many interconnected smart devices manage and control different areas in cities using information and communication technologies. Such networked devices, exchanging information through real-time applications, are characterized by their high mobility, which require developing optimized routing metrics. In the literature, several solutions are proposed to solve such an information routing problem. Most of them use many network parameters in routing metrics calculation, such as the node position, the node speed, the link quality and the network density. However, the existing routing solutions may require combining simultaneously the end to end delay, the packet loss and the distance. This adds more efficiency in data transmission since the realtime applications require no packet loss and less delay. In this paper, we consider these parameters to propose a mathematical modeling of new multicriteria routing metric. Subsequently, we solve it with three different methods: exact method, A star method and A star with obstacles method. The simulation results show the efficiency of the A star method in terms of response time and iteration number.
A large fraction of the ICT life cycle impacts are related to the ICT use phase because of electricity consumption. Environmental evaluations of the ICT sector suggests that this sector contributes now to 3% of the global anthropic greenhouse gas (GHG) emissions. It is anticipated that ICT could be used to mitigate the GHG emissions of the other sectors by 12% by 2030. These numbers are however uncertain due to simplifications made in the emission assessment. Especially, the use of annual average emission factors to model the GHG emissions related to the ICT electricity consumption that varies in time is expected to be a large source of uncertainty on the ICT emission assessment. Therefore, in this paper, we present the proposed IEEE 1922.2 standard to include the temporal variation of the power generation and electricity consumption in the computing of ICT emissions. The methodological framework is developed around a series of scopes and considerations to cover most of the situations that could be encountered when assessing ICT emissions. Consequently, the standard is flexible and methodological choices must be made by the user according to its context. Such choices must then be reported and documented to ensure the robustness and transparency of the ICT emission calculation.
In this work, we investigate the energy efficiency of placing virtual machines (VMs) in geo-distributed data centers taking into account inter-VM traffic in addition to users traffic. The problem of VMs placement is formularized as a mixed integer linear programming (MILP) model with an objective to minimize the network and cloud power consumption taking into consideration cooperation traffic between different VMs and synchronization traffic between replicas of the same VM in addition to the download traffic from VMs to users. The model results show that the number of VMs replicas across geo-distributed clouds is limited by the existence of inter-VM traffic in the core network. The total power consumption can potentially increase by a factor of 39 if inter-VM traffic is not taken into consideration when optimizing the placement of VMs.
Railway transportation is one of the most popular options for Urban Massive Transportation Systems (UMTS) because of many attractive features. A robust electric power supply is essential to enable normal operation. However, the power peaks appearing at the start time of the vehicles put heavy pressure on the power grid. Reduction of the power peak is a key issue in improving urban railway transit's power efficiency. Researchers have tried to address this problem by making a delicate timetable, but this method often failed to serve the purpose because of the punctuality problem. In this work, taking advantage of real-time estimation of the single train's power consumption, an online Power peak Reduction by real-time Scheduling (PReS) solution for trains' departure is proposed. Particularly, a Binary Integer Programming (BIP) model is introduced that is able to avoid power consumption peak caused by multiple trains departure simultaneously. The simulation result verified that the proposed real-time scheduling approach can effectively reduce the occurrences of power peak without bringing in additional train travel delay.
Link rate adaptation is an effective means to save energy consumption of network elements by adjusting the link rate according to the carried traffic through a network-level optimization of the flow allocation process. Unfortunately, current adaptation approaches are mainly reactive, in which link speed is changed only when new traffic demand is requested. Once bandwidth has been allocated for a demand, link rate remains constant during the entire session. This approach may result in sub-optimal energy efficiency schemes and requires multiple re-optimizations as traffic flows are fluctuating during the session, hence reducing the overall network performance. In this paper, we propose a multiple-step-ahead method to predictively optimize link rates based on forecasting traffic demand. We formulate the link adaptive energy efficiency as a MIP model and propose a heuristic simulated annealing algorithm to solve it. Our experimental results show our approach provides energy saving while it significantly decreases the number of re-optimizations in the energy-aware routing.
