In this paper, we consider wireless caching helper networks (WCHNs) consisting of cache-enabled device-to-device (D2D) transmitters and caching helpers (CHs), which deliver data by exploiting cached contents. We consider two types of modes at a typical user, namely D2D and CH modes. In the D2D and CH modes, after requesting a content, the user receives the content from a D2D transmitter and a CH caching the content, respectively. In practical scenarios, to mitigate interference, the CHs may not be placed close to each other, and thus there exists a form of repulsion among the CHs' locations. In this context, we model the spatial distribution of the CHs as a β-Ginibre point processe, which reflects the repulsive behavior and contains the Poisson point process as a special case. Then, we provide analytical expressions for the coverage probabilities in the WCHNs.

In this paper, we consider fog radio access networks (F-RANs) consisting of cache-enabled device-to-device (D2D) transmitters and fog access points (F-APs), which deliver data by exploiting cached contents or leveraging cloud processing. We consider three types of modes at a typical user, namely, D2D, F-AP, and cooperative modes. In the D2D and the F-AP modes, when the user requests content, the user receives the content from a D2D transmitter and an F-AP caching the content, respectively. In the cooperative mode, F-APs located near the user send data aided by a centralized cloud processing unit. We also examine a mode selection algorithm in which the user adaptively selects one of the three modes. In practical scenarios, to mitigate interference, the transmitters may not be placed close to each other, and thus, there may exist a form of repulsion among the transmitters' locations. In this context, we model the spatial distributions of the D2D transmitters and the F-APs as $\beta $ -Ginibre point processes, which reflect the repulsive behavior and contain the Poisson point process as a special case. Then, we provide analytical expressions for the coverage probabilities in the F-RANs. Our results are corroborated by Monte Carlo simulations.

Due to its mathematical tractability, the homogeneous Poisson point process (PPP) has been employed to model wireless networks and analyze their performance. The PPP has the fundamental property that in a network with n nodes, the n nodes are distributed independently from each other. As such the PPP is not a suitable model for many networks where there exists a repulsion among the nodes. In order to address this limitation, in this paper we model the spatial distribution of transmitters in wireless networks as a Poisson hard-core process (PHCP) in which no two nodes can be closer to each other than a given repulsion radius from one another. We first provide an exact expression of the coverage probability of the networks and then introduce the method to efficiently evaluate the derived expression. Additionally, we derive approximations of the coverage probability which have low computational complexities. The accuracy and efficiency of our analytical results are validated by our simulations.

Wireless radio frequency energy harvesting has been adopted in wireless networks as a method to supply energy to wireless nodes. In this paper, we analyze a wireless energy harvesting network based on a Boolean-Poisson model. This model assumes that energy sources are distributed according to a Poisson point process and have disc-shaped coverage regions with random radii. We introduce a distribution for the coverage radii, which takes aggregated harvested power into account. The union of the coverage regions of the energy sources forms the energy harvesting zone. We derive the transmission success probability of single-hop networks characterized by the probability that two sensor nodes are located in the energy harvesting zone. Then, we analyze the performance of multi-hop networks in the cases, where the locations of the sensor nodes are either fixed or randomly distributed. Moreover, we consider a star-shaped topology, which reflects the scenario wherein some sensor nodes simultaneously transmit data to a data collector. In this setting, we derive an approximation of the average throughput at the data collector. Numerical results validate the accuracy of our analysis in the single-hop and multi-hop networks and confirm the tightness of our approximation in the case of the star-shaped topology.

The Poisson point process (PPP) has been widely employed to model wireless networks and analyze their performance. The PPP has the property that nodes are conditionally independent from each other. As such, it may not be a suitable model for many networks, where there exists repulsion among the nodes. In order to address this limitation, we adopt a Poisson hardcore process (PHCP), in which no two nodes can be closer than a repulsion radius from one another. We consider two-tier heterogeneous networks, where the spatial distributions of transmitters in the first-tier and the second-tier networks follow a PHCP and a PPP, respectively. To alleviate inter-tier interference, we consider a guard zone for the first-tier network and presume that the second-tier transmitters located in the zone are deactivated. Under this setup, the activated second-tier transmitters form a Poisson hard-core hole process. We first derive exact computable expressions of the coverage probability and introduce a method to efficiently evaluate the expressions. Then, we provide approximations of the coverage probability, which have lower computational complexities. In addition, as a special case, we investigate the coverage probability of single-tier networks by modeling the locations of transmitters as a PHCP.

Wireless radio frequency (RF) energy harvesting has been adopted in wireless networks as a method to supply energy to wireless nodes, e.g., sensors. In this paper, we present a new analysis of the wireless energy harvesting network based on a Boolean-Poisson model. This model considers that the energy sources have a fixed coverage range. The energy sources are distributed according to a Poisson point process (PPP) while their radii of coverage are random and are assumed to follow a given probability distribution. We derive the performance measures consisting of the energy harvesting probability and the transmission success probability both in the cases of two nodes and multiple nodes. Our analysis is validated by simulation.

Ambient radio frequency (RF) energy harvesting methods have drawn significant interests due to their ability to provide energy to wireless devices from ambient RF sources. This paper considers ambient RF energy harvesting wireless sensor networks where a sensor node transmits data to a data sink using the energy harvested from the signals transmitted by the ambient RF sources. We analyze the performance of the network, i.e., the mean of the harvested energy, the power outage probability, and the transmission outage probability. In many practical networks, the locations of the ambient RF sources are spatially correlated and the ambient sources exhibit repulsive behaviors. Therefore, we model the spatial distribution of the ambient sources as an α-Ginibre point process (α-GPP), which reflects the repulsion among the RF sources and includes the Poisson point process as a special case. We also assume that the fading channel is Nakagami-m distributed, which also includes Rayleigh fading as a particular case. In this paper, by exploiting the Laplace transform of the α-GPP, we introduce semi-closed-form expressions for the considered performance metrics and provide an upper bound of the power outage probability. The derived expressions are expressed in terms of the Fredholm determinant, which can be computed numerically. In order to reduce the complexity in computing the Fredholm determinant, we provide a simple closed-form expression for the Fredholm determinant, which allows us to evaluate the Fredholm determinant much more efficiently. The accuracy of our analytical results is validated through simulation results.

RF-enabled wireless power transfer and energy harvesting has recently emerged as a promising technique to provision perpetual energy replenishment for low-power wireless networks. The network devices are replenished by the RF energy harvested from the transmission of ambient RF transmitters, which offers a practical and promising solution to enable self-sustainable communications. This paper adopts a stochastic geometry framework based on the Ginibre model to analyze the performance of self-sustainable communications over cellular networks with general fading channels. Specifically, we consider the point-to-point downlink transmission between an access point and a battery-free device in the cellular networks, where the ambient RF transmitters are randomly distributed following a repulsive point process, called Ginibre α-determinantal point process (DPP). Two practical RF energy harvesting receiver architectures, namely time-switching and power-splitting, are investigated. We perform an analytical study on the RF-powered device and derive the expectation of the RF energy harvesting rate, the energy outage probability and the transmission outage probability over Nakagami-m fading channels. These are expressed in terms of so-called Fredholm determinants, which we compute efficiently with modern techniques from numerical analysis. Our analytical results are corroborated by the numerical simulations, and the efficiency of our approximations is demonstrated. In practice, the accurate simulation of any of the Fredholm determinant appearing in the manuscript is a matter of seconds. An interesting finding is that a smaller value of α (corresponding to larger repulsion) yields a better transmission outage performance when the density of the ambient RF transmitters is small. However, it yields a lower transmission outage probability when the density of the ambient RF transmitters is large. We also show analytically that the power-splitting architecture outperforms the time-switching architecture in terms of transmission outage performances. Lastly, our analysis provides guidelines for setting the time-switching and power-splitting coefficients at their optimal values.

This paper introduces a generic and scalable framework for automated anomaly detection on large scale time-series data. Early detection of anomalies plays a key role in maintaining consistency of person's data and protects corporations against malicious attackers. Current state of the art anomaly detection approaches suffer from scalability, use-case restrictions, difficulty of use and a large number of false positives. Our system at Yahoo, EGADS, uses a collection of anomaly detection and forecasting models with an anomaly filtering layer for accurate and scalable anomaly detection on time-series. We compare our approach against other anomaly detection systems on real and synthetic data with varying time-series characteristics. We found that our framework allows for 50-60% improvement in precision and recall for a variety of use-cases. Both the data and the framework are being open-sourced. The open-sourcing of the data, in particular, represents the first of its kind effort to establish the standard benchmark for anomaly detection.

The advance in RF energy transfer and harvesting technique over the past decade has enabled wireless energy replenishment for electronic devices, which is deemed as a promising alternative to address the energy bottleneck of conventional battery-powered devices. In this paper, by using a stochastic geometry approach, we aim to analyze the performance of an RF-powered wireless sensor in a downlink simultaneous wireless information and power transfer (SWIPT) system with ambient RF transmitters. Specifically, we consider the point-to-point downlink SWIPT transmission from an access point to a wireless sensor in a network, where ambient RF transmitters are distributed as a Ginibre α-determinantal point process (DPP), which becomes the Poisson point process when a approaches zero. In the considered network, we focus on analyzing the performance of a sensor equipped with the power-splitting architecture. Under this architecture, we characterize the expected RF energy harvesting rate of the sensor. Moreover, we derive the upper bound of both power and transmission outage probabilities. Numerical results show that our upper bounds are accurate for different value of a.

Ambient radio frequency (RF) energy harvesting technique has recently been proposed as a potential solution for providing proactive energy replenishment for wireless devices. This paper aims to analyze the performance of a battery-free wireless sensor powered by ambient RF energy harvesting using a stochastic geometry approach. Specifically, we consider the point-to-point uplink transmission of a wireless sensor in a stochastic geometry network, where ambient RF sources, such as mobile transmit devices, access points and base stations, are distributed as a Ginibre α-determinantal point process (DPP). The DPP is able to capture repulsion among points, and hence, it is more general than the Poisson point process (PPP). We analyze two common receiver architectures: separated receiver and time-switching architectures. For each architecture, we consider the scenarios with and without co-channel interference for information transmission. We derive the expectation of the RF energy harvesting rate in closed form and also compute its variance. Moreover, we perform a worst-case study which derives the upper bound of both power and transmission outage probabilities. Additionally, we provide guidelines on the setting of optimal time-switching coefficient in the case of the time-switching architecture. Numerical results verify the correctness of the analysis and show various tradeoffs between parameter setting. Lastly, we prove that the RF-powered sensor performs better when the distribution of the ambient sources exhibits stronger repulsion.

Ambient RF (Radio Frequency) energy harvesting techniques have recently been proposed as a potential solution to provide proactive energy replenishment for wireless devices. This paper aims to analyze the performance of a battery-free wireless sensor powered by ambient RF energy harvesting using a stochastic-geometry approach. Specifically, we consider a random network model in which ambient RF sources are distributed as a Ginibre α-determinantal point process which recovers the Poisson point process when α approaches zero. We characterize the expected RF energy harvesting rate. We also perform a worst-case study which derives the upper bounds of both power outage and transmission outage probabilities. Numerical results show that our upper bounds are accurate and that better performance is achieved when the distribution of ambient sources exhibits stronger repulsion.

Considering a damaged wireless network, presenting coverage holes or disconnected components, we propose a disaster recovery algorithm repairing the network. It provides the list of locations where to put new nodes to patch the coverage holes and mend the disconnected components. In order to do this we first consider the simplicial complex representation of the network, then the algorithm adds supplementary nodes in excessive number, and afterwards runs a reduction algorithm in order to reach a unimprovable result. One of the novelty of this work resides in the proposed method for the addition of nodes. We use a determinantal point process: the Ginibre point process which has inherent repulsion between vertices, which simulation is new in wireless networks application. We compare both the determinantal point process addition method with other vertices addition methods, and the whole disaster recovery algorithm to the greedy algorithm for the set cover problem.