INFOCOM 2021 Papers

This paper presents a data-driven localization framework with high precision in time-varying complex multi-path environments, such as dense urban areas and indoors, where GPS and model-based localization techniques come short. We consider the angle-delay profile (ADP), a linear transformation of channel state information (CSI), in massive MIMO systems and show that ADPs preserve users' motion when stacked temporally. We discuss that given a static environment, future frames of ADP time-series are predictable employing a video frame prediction algorithm. We express that a deep convolutional neural network (DCNN) can be employed to learn the background static scattering environment. To detect foreground changes in the environment, corresponding to path blockage or addition, we introduce an algorithm taking advantage of the trained DCNN. Furthermore, we present DyLoc, a data-driven framework to recover distorted ADPs due to foreground changes and to obtain precise location estimations. We evaluate the performance of DyLoc in several dynamic scenarios employing DeepMIMO dataset [1] to generate geo-tagged CSI datasets for indoor and outdoor environments. We show that previous DCNN-based techniques fail to perform with desirable accuracy in dynamic environments, while DyLoc pursues localization precisely. Moreover, simulations show that as the environment gets richer in terms of the number of multipath, DyLoc gets more robust to foreground changes.

Caching content over CDNs or at the network edge has been solidified as a means to improve network cost and offer better streaming experience to users. Furthermore, nudging the users towards low-cost content has recently gained momentum as a strategy to boost network performance. We focus on the problem of optimal policy design for Network Friendly Recommendations (NFR). We depart from recent modeling attempts, and propose a Markov Decision Process (MDP) formulation. MDPs offer a unified framework that can model a user with random session length. As it turns out, many state-of-the-art approaches can be cast as subcases of our MDP formulation. Moreover, the approach offers flexibility to model users who are reactive to the quality of the received recommendations. In terms of performance, for users consuming an arbitrary number of contents in sequence, we show theoretically and using extensive validation over real traces that the MDP approach outperforms myopic algorithms both in session cost as well as in offered recommendation quality. Finally, even compared to optimal state-of-art algorithms targeting specific subcases, our MDP framework is significantly more efficient, speeding the execution time by a factor of 10, and enjoying better scaling with the content catalog and recommendation batch sizes.

Coronavirus disease 2019 (COVID-19) has resulted in an ongoing pandemic. Since COVID-19 spreads mainly via close contact among people, social distancing has become an effective manner to slow down the spread. However, completely forbidding close contact can also lead to unacceptable damage to the society. Thus, a system that can effectively monitor people's social distance and generate corresponding alerts when a high infection probability is detected is in urgent need. In this paper, we propose SmartDistance, a smartphone based software framework that monitors people's interaction in an effective manner, and generates a reminder whenever the infection probability is high. Specifically, SmartDistance dynamically senses both the relative distance and orientation during social interaction with a well-designed relative positioning system. In addition, it recognizes different events (e.g., speaking, coughing) and determines the infection space through a droplet transmission model. With event recognition and relative positioning, SmartDistance effectively detects risky social interaction, generates an alert immediately, and records the relevant data for close contact reporting. We prototype SmartDistance on different Android smartphones, and the evaluation shows it reduces the false positive rate from 33% to 1% and the false negative rate from 5% to 3% in infection risk detection.

Elastic Content Delivery Networks (Elastic CDNs) have been introduced to support explosive Internet traffic growth by providing small Content Providers (CPs) with just-in-time services. Due to the diverse requirements of small CPs, they need customized adaptive caching modules to help them adjust the cached contents to maximize their long-term utility. The traditional adaptive caching module is usually a built-in service in a cloud CDN. They adaptively change cache contents using size-scaling-only methods or strategy-adaptation-only methods. A natural question is: can we jointly optimize size and strategy to achieve tradeoff and better performance for small CPs when renting services from elastic CDNs? The problem is challenging because the two decision variables could involve both discrete and categorical variables, where discrete variables have an intrinsic order while categorical variables do not. In this paper, we propose a distribution-guided reinforcement learning framework JEANA to learn the joint cache size scaling and strategy adaptation policy. We design a distribution-guided regularizer to keep the intrinsic order of discrete variables. More importantly, we prove that our algorithm has a theoretical guarantee of performance improvement. Trace-driven experimental results demonstrate our method can improve the hit ratio while reducing the rental cost.

RAN energy consumption is a major OPEX source for mobile telecom operators, and 5G is expected to increase these costs by several folds. Moreover, paradigm-shifting aspects of the 5G RAN architecture like RAN disaggregation, virtualization and cloudification introduce new traffic-dependent resource management decisions that make the problem of energy-efficient 5G RAN orchestration harder. To address such a challenge, we present a first comprehensive virtualized RAN (vRAN) system model aligned with 5G RAN specifications, which embeds realistic and dynamic models for computational load and energy consumption costs. We then formulate the vRAN energy consumption optimization as an integer quadratic programming problem, whose NP-hard nature leads us to develop GreenRAN, a novel, computationally efficient and distributed solution that leverages Lagrangian decomposition and simulated annealing. Evaluations with real-world mobile traffic data for a large metropolitan area are another novel aspect of this work, and show that our approach yields energy efficiency gains up to 25% and 42%, over state-of-the-art and baseline traditional RAN approaches, respectively.

This work proposes AMIS, an edge computing-based adaptive video streaming system. AMIS explores the power of edge computing in three aspects. First, with video contents pre-cached in the local buffer, AMIS is content-aware which adapts the video playout strategy based on the scene features of video contents and quality of experience (QoE) of users. Second, AMIS is channel-aware which measures the channel conditions in real-time and estimates the wireless bandwidth. Third, by integrating the content features and channel estimation, AMIS applies the deep reinforcement learning model to optimize the playout strategy towards the best QoE. Therefore, AMIS is an intelligent content- and channel-aware scheme which fully explores the intelligence of edge computing and adapts to general environments and QoE requirements. Using trace-driven simulations, we show that AMIS can succeed in improving the average QoE by 14%-46% as compared to the state-of-the-art adaptive bitrate algorithms.

A distributed machine learning platform needs to recruit many heterogeneous worker nodes to finish computation simultaneously. As a result, the overall performance may be degraded due to straggling workers. By introducing redundancy into computation, coded machine learning can effectively improve the runtime performance by recovering the final computation result through the first k (out of the total n) workers who finish computation. While existing studies focus on designing efficient coding schemes, the issue of designing proper incentives to encourage worker participation is still under-explored. This paper studies the platform's optimal incentive mechanism for motivating proper workers' participation in coded machine learning, despite the incomplete information about heterogeneous workers' computation performances and costs. A key contribution of this work is to summarize workers' multi-dimensional heterogeneity as a one-dimensional metric, which guides the platform's efficient selection of workers under incomplete information with a linear computation complexity. Moreover, we prove that the optimal recovery threshold k is linearly proportional to the participator number n if we use the widely adopted MDS codes for data encoding. We also show that the platform's increased cost due to incomplete information disappears when worker number is sufficiently large, but it does not monotonically decrease in worker number.

Distributed training performs data-parallel training of DNN models which is a necessity for increasingly complex models and large datasets. Recent works are identifying major communication bottlenecks in distributed training. These works seek possible opportunities to speed-up the training in systems supporting distributed ML workloads. As communication reduction, compression techniques are proposed to speed up this communication phase. However, compression comes at the cost of reduced model accuracy, especially when compression is applied arbitrarily. Instead, we advocate a more controlled use of compression and propose DC2, a delay-aware compression control mechanism. DC2 couples compression control and network delays in applying compression adaptively. DC2 not only compensates for network variations but can also strike a better trade-off between training speed and accuracy. DC2 is implemented as a drop-in module to the communication library used by the ML toolkit and can operate in a variety of network settings. We empirically evaluate DC2 in network environments exhibiting low and high delay variations. Our evaluation of different popular CNN models and datasets shows that DC2 improves training speed-ups of up to 41× and 5.3 × over baselines with no-compression and uniform compression, respectively.

There has been significant interest in leveraging limited look-ahead to achieve low competitive ratios for online convex optimization (OCO). However, existing online algorithms (such as Averaging Fixed Horizon Control (AFHC)) that can leverage look-ahead to reduce the competitive ratios still produce competitive ratios that grow unbounded as the coefficient ratio (i.e., the maximum ratio of the switching-cost coefficient and the service-cost coefficient) increases. On the other hand, the regularization method can attain a competitive ratio that remains bounded when the coefficient ratio is large, but it does not benefit from look-ahead. In this paper, we propose a new algorithm, called Regularization with Look-Ahead (RLA), that can get the best of both AFHC and the regularization method, i.e., its competitive ratio decreases with the look-ahead window size when the coefficient ratio is small, and remains bounded when the coefficient ratio is large. We also provide a matching lower bound for the competitive ratios of all online algorithms with look-ahead, which differs from the achievable competitive ratio of RLA by a factor that only depends on the problem size. The competitive analysis of RLA involves a non-trivial generalization of online primal-dual analysis to the case with look-ahead.

Contextual bandit learning selects actions (i.e., arms) based on context information to maximize rewards while balancing exploitation and exploration. In many applications (e.g., cloud resource management with dynamic workloads), before arm selection, the agent/learner can either predict context information online based on context history or selectively query the context from an outside expert. Motivated by this practical consideration, we study a novel contextual bandit setting where context information is either predicted online or queried from an expert. First, considering predicted context only, we quantify the impact of context prediction on the cumulative regret (compared to an oracle with perfect context information) by deriving an upper bound on regret, which takes the form of a weighted combination of regret incurred by standard bandit learning and the context prediction error. Then, inspired by the regret's structural decomposition, we propose context query algorithms to selectively obtain outside expert's input (subject to a total query budget) for more accurate context, decreasing the overall regret. Finally, we apply our algorithms to virtual machine scheduling on cloud platforms. The simulation results validate our regret analysis and shows the effectiveness of our selective context query algorithms.

Malicious data manipulation reduces the effectiveness of machine learning techniques, which rely on accurate knowledge of the input data. Motivated by real-world applications in network flow classification, we address the problem of robust online learning with delayed feedback in the presence of malicious data generators that attempt to gain favorable classification outcome by manipulating the data features. We propose online algorithms termed ROLC-NC and ROLC-C when the malicious data generators are non-clairvoyant and clairvoyant, respectively. We derive regret bounds for both algorithms and show that they are sub-linear under mild conditions. We further evaluate the proposed algorithms in network flow classification via extensive experiments using real-world data traces. Our experimental results demonstrate that both algorithms can approach the performance of an optimal static offline classifier that is not under attack, while outperforming the same offline classifier when tested with a mixture of normal and manipulated data.

Container, as a light-weight virtualization technology with the advantages of continuous integration and easy deployment, has been widely adopted to support diverse microservices. At runtime, non-local container images need to be frequently pulled from remote registries to local servers, resulting in large pulling traffic and hence long startup time. A distinctive feature in container-based microservice, which has not been exploited, is that container images are in layered structure and some common base layers can be shared between co-located microservices. In this paper, we propose a layer sharing microservice deployment and image pulling strategy which explores the advantage of layer sharing to speedup microservice startup and lower image storage consumption. The problem is formulated into an Integer Linear Programming (ILP) form. An Accelerated Distributed Augmented Lagrangian (ADAL) based distributed algorithm executed cooperatively by registries and servers is proposed. Through extensive trace driven experiments, we validate the high efficiency of our ADAL based algorithm as it accelerates the microservice startup by 2.30 times in average and reduces the storage consumption by 55.33%.

The stable path problem (SPP) is a unified model for analyzing the convergence of distributed routing protocols (e.g., BGP), and a foundation for many network verification tools. Although substantial progress has been made on finding solutions (i.e., stable path assignments) for particular subclasses of SPP instances and analyzing the relation between properties of SPP instances and the convergence of corresponding routing policies, the non-trivial challenge of finding stable path assignments to generic SPP instances still remains. Tackling this challenge is important because it can enable multiple important, novel routing use cases. To fill this gap, in this paper we introduce a novel data structure called solvability digraph, which encodes key properties about stable path assignments in a compact graph representation. Thus SPP is equivalently transformed to the problem of finding in the solvability digraph a maximum independent set (MIS) of size equal to the number of autonomous systems (ASes) in the given SPP instance. We leverage this key finding to develop a heuristic polynomial algorithm GREEDYMIS that solves strictly more SPP instances than state-of-the-art heuristics. We apply GREEDYMIS to designing two important, novel use cases: (1) a centralized interdomain routing system that uses GREEDYMIS to compute paths for ASes and (2) a secure multi-party computation (SMPC) protocol that allows ASes to use GREEDYMIS collaboratively to compute paths without exposing their routing preferences. We demonstrate the benefits and efficiency of these use cases via evaluation using real-world datasets.

Task replication has been advocated as a practical solution to reduce response times in parallel systems. The analysis of replication-based systems typically rests on some strong assumptions: Poisson arrivals, exponential service times, or independent service times of the replicas. This study is motivated not only by several studies which indicate that these assumptions are unrealistic, but also by some elementary observations highlighting some contriving behaviour. For instance, when service times are not exponential, adding a replication factor can stabilize an unstable system, i.e., having infinite delays, but a tempting higher replication factor can push the system back in a perilous unstable state. This behaviour disappears however if the replicas are sufficiently correlated, in which case any replication factor would even be detrimental.Motivated by the need to dispense with such common yet unrealistic and misleading assumptions, we provide a robust theoretical framework to compute stochastic bounds on response time distributions in general replication systems subject to Markovian arrivals, quite general service times, and correlated replicas. Numerical results show that our bounds are accurate and improve state-of-the-art bounds in the case of Markovian arrivals by as much as three orders of magnitude. We apply our results to a practical application and highlight that correctly setting the replication factor crucially depends on both the service time distributions of the replicas and the degree of correlation amongst.

Collaborative object localization and sharing at smart intersections promises to improve situational awareness of traffic participants in key areas where hazards exist due to visual obstructions. By sharing a moving object's location between different camera-equipped devices, it effectively extends the vision of traffic participants beyond their field of view. However, accurately sharing objects between moving clients is extremely challenging due to the high accuracy requirements for localizing both the client position and positions of its detected objects. Therefore, we introduce EdgeSharing, a localization and object sharing system leveraging the resources of edge cloud platforms. EdgeSharing holds a real-time 3D feature map of its coverage region to provide accurate localization and object sharing service to the client devices passing through this region. We further propose several optimization techniques to increase the localization accuracy, reduce the bandwidth consumption and decrease the offloading latency of the system. The result shows that the system is able to achieve a mean vehicle localization error of 0.28-1.27 meters, an object sharing accuracy of 82.3%-91.4%, and a 54.7% object awareness increment in urban streets and intersections. In addition, the proposed optimization techniques reduce bandwidth consumption by 70.12% and end-to-end latency by 40.09%.

As a leading industrial wireless standard, WirelessHART has been widely implemented to build wireless sensor-actuator networks (WSANs) in industrial facilities, such as oil refineries, chemical plants, and factories. For instance, 54,835 WSANs that implement the WirelessHART standard have been deployed globally by Emerson process management, a WirelessHART network supplier, to support process automation. While the existing research to improve industrial WSANs focuses mainly on enhancing network performance, the security aspects have not been given enough attention. We have identified a new threat to WirelessHART networks, namely smart selective jamming attacks, where the attacker first cracks the channel usage, routes, and parameter configuration of the victim network and then jams the transmissions of interest on their specific communication channels in their specific time slots, which makes the attacks energy efficient and hardly detectable. In this paper, we present this severe, stealthy threat by demonstrating the step-by-step attack process on a 50-node network that runs a publicly accessible WirelessHART implementation. Experimental results show that the smart selective jamming attacks significantly reduce the network reliability without triggering network updates.

Time series has numerous application scenarios. However, since many time series data are personal data, releasing them directly could cause privacy infringement. All existing techniques to publish privacy-preserving time series perturb the values while retaining the original temporal order. However, in many value-critical scenarios such as health and financial time series, the values must not be perturbed whereas the temporal order can be perturbed to protect privacy. As such, we propose "local differential privacy in the temporal setting" (TLDP) as the privacy notion for time series data. After quantifying the utility of a temporal perturbation mechanism in terms of the costs of a missing, repeated, empty, or delayed value, we propose three mechanisms for TLDP. Through both analytical and empirical studies, we show the last one, Threshold mechanism, is the most effective under most privacy budget settings, whereas the other two baseline mechanisms fill a niche by supporting very small or large privacy budgets.

Data collection by mobile crowdsensing (MCS) is emerging as data sources for smart city applications, however how to ensure data freshness has sparse research exposure but quite important in practice. In this paper, we consider to use a group of mobile agents (MAs) like UAVs and driverless cars which are equipped with multiple antennas to move around in the task area to collect data from deployed sensor nodes (SNs). Our goal is to minimize the age of information (AoI) of all SNs and energy consumption of MAs during movement and data upload. To this end, we propose a centralized deep reinforcement learning (DRL)-based solution called "DRL-freshMCS" for controlling MA trajectory planning and SN scheduling. We further utilize implicit quantile networks to maintain the accurate value estimation and steady policies for MAs. Then, we design an exploration and exploitation mechanism by dynamic distributed prioritized experience replay. We also derive the theoretical lower bound for episodic AoI. Extensive simulation results show that DRL-freshMCS significantly reduces the episodic AoI per remaining energy, compared to five baselines when varying different number of antennas and data upload thresholds, and number of SNs. We also visualize their trajectories and AoI update process for clear illustrations.

We introduce a framework and provably-efficient schemes for `fresh' caching at the (front-end) local cache of content that is subject to `dynamic' updates at the (back-end) database. We start by formulating the hard-cache-constrained problem for this setting, which quickly becomes intractable due to the limited cache. To bypass this challenge, we first propose a flexible time-based-eviction model to derive the average system cost function that measures the system's cost due to the service of aging content in addition to the regular cache miss cost. Next, we solve the cache-unconstrained case, which reveals how the refresh dynamics and popularity of content affect the optimal caching. Then, we extend our approach to a soft-cache-constrained version, where we can guarantee that the cache use is limited with arbitrarily high probability. The corresponding solution reveals the interesting insight that `whether to cache an item or not in the local cache?' depends primarily on its popularity level, whereas `how long the cached item should be held in the cache before eviction?' depends primarily on its refresh rate. Moreover, we investigate the cost-cache saving tradeoffs and prove that substantial cache gains can be obtained while also asymptotically achieving the minimum cost as the database size grows.