INFOCOM 2022 Papers

With the explosive growth of mobile and Internet of Things (IoT) applications, increasing Mobile Edge Computing (MEC) systems have been developed by diverse Edge Service Providers (ESPs), opening a new computing market with stiff competition. However, considering the spatiotemporally varying features of computation tasks, taking over all the received tasks alone may greatly degrade the service performance of the MEC system and lead to poor economical benefit. To this end, this paper proposes a two-layer collaboration model for ESPs. Each ESP can balance the computation workload among the internal edge nodes from the ESP and offload part of computation tasks to the ESP external edge servers from other ESPs. For internal load balancing, we propose a task balancing scheme based on the Alternating Direction Method of Multipliers (ADMM) to manage the computation tasks within the edge nodes of the ESP, such that the computation delay can be minimized. For external task offloading, we formulate a game-based pricing and task allocation scheme to derive the best game strategy, aiming at maximizing the total revenue of each ESP. Extensive simulation results demonstrate that the proposed schemes can achieve improved performance in terms of system revenue and stability, as well as computation delay.

Mobile Edge Computing (MEC) is a new paradigm where mobile users can offload computation tasks to the nearby MEC server to reduce their resource consumption. Some works have pointed out that the true amount of offloaded tasks may reveal the sensitive information (e.g., device usage pattern and location information) of users, and proposed several privacy-preserving offloading mechanisms. However, to the best of our knowledge, none of them can provide strict and provable privacy guarantee. In this paper, we focus on the privacy leakage issue in computation offloading in MEC with a honest-but-curious server, and propose a novel online privacy-preserving computation offloading mechanism, called OffloadingGuard, to generate efficient offloading strategies for users in real time, which provide strict user privacy guarantee while minimizing the total cost of task computation. To this end, we design a deep reinforcement learning-based offloading model which allows each user to adaptively determine the satisfactory perturbed offloading ratio according to the time-varying channel state at each time slot to achieve trade-off between user privacy and computation cost. In particular, to strictly protect the true amount of offloaded tasks and prevent the untrusted MEC server from revealing mobile users’ privacy, a range-constrained Laplace distribution is designed to obfuscate the original offloading ratio of each user and restrict the perturbed offloading ratio in a rational range. OffloadingGuard is proved to satisfy ϵ-differential privacy, and extensive experiments demonstrate its effectiveness.

While deep learning (DL)-based network traffic classification has demonstrated its success in a range of practical applications, such as network management and security control to just name a few, it is vulnerable to adversarial attacks. This paper reports TrojanFlow, a new and practical neural backdoor attack to DL-based network traffic classifiers. In contrast to traditional neural backdoor attacks where a designated and sample-agnostic trigger is used to plant backdoor, TrojanFlow poisons a model using dynamic and sample-specific triggers that are optimized to efficiently hijack the model. It features a unique design to jointly optimize the trigger generator with the target classifier during training. The trigger generator can thus craft optimized triggers based on the input sample to efficiently manipulate the model’s prediction. A well-engineered prototype is developed using Pytorch to demonstrate TrojanFlow attacking multiple practical DL-based network traffic classifiers. Thorough analysis is conducted to gain insights into the effectiveness of TrojanFlow, revealing the fundamentals of why it is effective and what it does to efficiently hijack the model. Extensive experiments are carried out on the well-known ISCXVPN2016 dataset with three widely adopted DL network traffic classifier architectures. TrojanFlow is compared with two other backdoor attacks under five state-of-the-art backdoor defenses. The results show that the TrojanFlow attack is stealthy, efficient, and highly robust against existing neural backdoor mitigation schemes.

PIN inference attack leveraging keystroke-induced side-channel information poses a substantial threat to the security of people’s privacy and properties. Among various PIN inference attacks, video-assisted method provide more intuitive and robust side-channel information to infer PINs. But it usually requires there is no visual occlusion between the attacker and the victims or their hand gestures, making the attackers either easy to expose themselves or inapplicable to the scenarios such as ATM or POS terminals. In this paper, we present a novel and practical video-assisted PIN inference system, ArmSpy, which infers victim’s PIN by observing from behind the victims in a stealthy way. Specifically, ArmSpy explores the subtle keystroke-induced arm posture changes, including elbow bending angle changes and the spatial relationship between different arm joints, to infer the PIN entries. We develop the keystroke inference mechanism to detect the keystroke events and pinpoint the keystroke positions, and then accurately infer the PINs with the proposed inferred PIN coordination mechanism. Extensive experimental results demonstrate that ArmSpy can achieve over 67% average accuracy on inferring the PIN with 3 attempts and even over 80% for some victims, indicating the severity of the threat posed by ArmSpy.

As the scale and speed of modern networks continue to increase, traffic sampling has become an indispensable tool in network management. While there exist a plethora of sampling solutions, they either provide limited flow visibility or have poor scalability in large networks. This paper presents the design and evaluation of FlowShark, a high-visibility per-flow sampling system for Software-Defined Networks (SDNs). The key idea in FlowShark is to separate sampling decisions on short and long flows, whereby sampling short flows is managed locally on edge switches, while a central controller optimizes sampling decisions on long flows. To this end, we formulate flow sampling as an optimization problem and design an online algorithm with a bounded competitive ratio to solve the problem efficiently. To show the feasibility of our design, we have implemented FlowShark in a small OpenFlow network using Mininet. We present experimental results of our Mininet implementation as well as performance benchmarks obtained from packet-level simulations in larger networks. Our experiments with a machine learning based Traffic Classifier application show up to 27% and 19% higher classification recall and precision, respectively, with FlowShark compared to existing sampling approaches.

Function-as-a-Service (FaaS) offers a fine-grained resource provision model, enabling developers to build highly elastic cloud applications. User requests are handled by a series of serverless functions step by step, which forms a function-based workflow. The developers are required to set proper resource configuration for functions, so as to meet service level objectives (SLOs) and save cost. However, developing the resource configuration strategy is challenging. It is mainly because execution of cloud functions often suffers from cold start and performance fluctuation, which requires a dynamic configuration strategy to guarantee the SLOs. In this paper, we present StepConf, a framework that automates the resource configuration for functions as the workflow runs. StepConf optimizes memory size for each function step in the workflow and takes inter and intra-function parallelism into consideration. We evaluate StepConf on AWS Lambda. Compared with baselines, the experimental results show that StepConf can save cost up to 40.9% while ensuring the SLOs.

With the development of Mobile Edge Computing (MEC) and Internet of Things (IoT) technology, various real-time monitoring and control applications are deployed to benefit people’s daily life. The performance of these applications relies heavily on the timeliness of collected environmental information, which can be effectively quantified by the recently introduced metric named age of information (AoI). Although extensive researches have been conducted to optimize AoI under various circumstances, these works commonly require a priori information about the system dynamics that is usually unknown in realistic situations. To design a more practical scheduling algorithm, in this paper, we formulate the AoI minimization problem as a Constrained Markov Decision Process (CMDP) which can be solved by Reinforcement Learning (RL) algorithms without prior knowledge. To improve the running efficiency, we (1) introduce post-decision states (PDSs) to exploit the partial knowledge of the system’s dynamics, (2) perform a batch update in every learning step, (3) decompose the system-level value function into multiple device-level value functions, and (4) propose a heuristic algorithm to find the greedy action. Numerical results demonstrate that our algorithm is highly efficient and outperforms the benchmarks under various scenarios.

As video streaming services become more popular, users desire high perceived video quality, which has placed more stringent requirements on the quality of connection. Existing issues of cellular networks encourage users to seek alternative connections such as public Wi-Fi networks; however, expectations of both users and owners of Wi-Fi networks are not sufficiently satisfied and various concerns are yet to be addressed by a better Wi-Fi access system. Based on a go-as-you-pay scheme, we design and implement DiFi, a per-user-based system with dynamic resource allocation and pricing. DiFi offers data burst that accommodates user requirements on the burstiness of traffic, in addition to bandwidth. It better caters to the various individual requirements of users, and better utilizes the limited network resources for the owners. We leverage the blockchain-based smart contract to address realistic concerns on decentralized control, privacy and trustiness and our implementation is compatible with existing Wi-Fi infrastructures.

An Autonomous Driving System (ADS) uses a plethora of sensors and many deep learning based tasks to aid its perception, prediction, motion planning, and vehicle control. To ensure road safety, those tasks should be synchronized and use the latest sensing data, which is challenging since 1) different sensors have different sensing periods, 2) the tasks are interdependent, 3) computing resource is limited. This work is the first that uses Age of Information (AoI) as the performance metric for task scheduling in an ADS. We show that minimizing AoI is equivalent to jointly minimizing the response time and maximizing the throughput. We formally formulate the AoI-centric task scheduling problem. To derive practical scheduling solutions, we extend the formulation and formulate the optimal AoI-centric periodic scheduling problem with a given cycle. A reinforcement learning-based solution is designed accordingly. With experiments simulated according to the Apollo driving system, we compare the scheduling performance of the AoI-centric task scheduling with Apollo’s schedulers from the perspective of AoI, throughput, and worst case response time. The experiment results show that the maximum AoI in the proposed scheduling solution with 4 cores is lower than that in Apollo’s schedulers with 8 cores.

In latency-sensitive file caching systems such as Content Delivery Networks (CDNs) and Mobile Edge Computing (MEC), the latency of fetching a missing file to the local cache can be significant. Recent studies have revealed that successive requests of the same missing file before the fetching completes could still suffer latency (so-called delayed hits).Motivated by the practical scenarios, we study the online general file caching problem with delayed hits and bypassing, i.e., a request may be bypassed and processed directly at the remote data center. The objective is to minimize the total request latency. We show a general reduction that turns a traditional file caching algorithm to one that can handle delayed hits. We give an O(Z3/2logK)-competitive algorithm called CaLa with this reduction, where Z is the maximum fetching latency of any file and K is the cache size, and we show a nearly-tight lower bound Ω(Z logK) for our ratio. Extensive simulations based on the production data trace from Google and the Yahoo benchmark illustrate that CaLa can reduce the latency by up to 9.42% compared with the state-of-the-art scheme dealing with delayed hits without bypassing, and this improvement increases to 32.01% if bypassing is allowed.

Age of Information (AoI), emerged as a new metric to quantify the data freshness, has attracted increasing interests recently. Most existing works try to optimize the system AoI from the point of data transmission. Unfortunately, at wireless-powered network edge, the charging schedule of the source nodes also needs to be decided besides data transmission. Thus, in this paper, we investigate the joint scheduling problem of data transmission and energy replenishment to optimize the peak AoI at network edge with directional chargers. To the best of our knowledge, this is the first work that considers such two problems simultaneously. Firstly, the theoretical bounds of the peak AoI with respect to the charging latency are derived. Secondly, for the minimum peak AoI scheduling problem with a single charger, an optimal scheduling algorithm is proposed to minimize the charging latency, and then a data transmission scheduling strategy is also given to optimize the peak AoI. The proposed algorithm is proved to have a constant approximation ratio of up to 1.5. When there exist multiple chargers, an approximate algorithm is also proposed to minimize the charging latency and peak AoI. Finally, the simulation results verify the high performance of proposed algorithms in terms of AoI.

Bi-objective online stochastic bipartite matching can capture a wide range of real-world problems such as online ride-hailing, crowdsourcing markets, and internet adverting, where the vertices in the left side are known in advance and that in the right side arrive from a known identical independent distribution (KIID) in an online manner. Mutual interest and limited attention-span are two common conditions and can be modeled as the edge existence probability and two-sided limited patience. Existing works fail to take them into bi-objective online optimization. This paper establishes a unified model for bi-objective online stochastic bipartite matching that can provide a general tradeoff among the matched edges (OBJ-1) and vertices (OBJ-2). We formulate two linear programs (LP) and accordingly design four LP-based parameterized online algorithms to tradeoff OBJ-1 and OBJ-2, with the best competitive ratio of (0.3528α, 0.3528β), where α, β are two positive input parameters and α + β = 1. Our hardness analysis proves that any non-adaptive algorithm cannot achieve (δ1, δ2)-competitive such that ${\delta _1} + {\delta _2} > 1 - \frac{1}{e}$. Trace-driven experiments show that our algorithms can always achieve better performance and provide a flexible tradeoff.

WiFi 5/6 relies on a key feature, Multi-User Multiple-In-Multiple-Out (MU-MIMO), to offer high-volume network throughput and spectrum efficiency. MU-MIMO uses a user selection algorithm, based on each user's channel state information (CSI), to schedule transmission opportunities for a group of users to maximize the service quality and efficiency. In this paper, we discover that such algorithm creates a subtle attack surface for attackers to subvert user selection in MU-MIMO, causing severe disruptions in today's wireless networks. We develop a system, named MU-MIMO user selection strategy inference and subversion (MUSTER), to systematically study the attack strategies and further to seek efficient mitigation. MUSTER is designed to include two major modules: (i) strategy inference, which leverages a new neural group-learning strategy named MC-grouping via combining Recurrent Neural Network (RNN) and Monte Carlo Tree Search (MCTS) to reverseengineer a user selection algorithm, and (ii) user selection subversion, which proactively fabricates CSI to manipulate user selection results for disruption. Experimental evaluation shows that MUSTER achieves a high accuracy rate around 98.6% in user selection prediction and effectively launches the attacks to disrupt the network performance. Finally, we create a Reciprocal Consistency Checking technique to defend against the proposed attacks to secure MU-MIMO user selection.

We propose and implement Directory-Based Access Control (DBAC), a flexible and systematic access control approach for geographically distributed multi-administration IoT systems. DBAC designs and relies on a particular module, IoT directory, to store device metadata, manage federated identities, and assist with cross-domain authorization. The directory service decouples IoT access into two phases: discover device information from directories and operate devices through discovered interfaces. DBAC extends attribute-based authorization and retrieves diverse attributes of users, devices, and environments from multi-faceted sources via standard methods, while user privacy is protected. To support resource-constrained devices, DBAC assigns a capability token to each authorized user, and devices only validate tokens to process a request.

On top of the Resource Public Key Infrastructure (RPKI), the Route Origin Authorization (ROA) creates a cryptographically verifiable binding of an autonomous system to a set of IP prefixes it is authorized to originate. By their design, ROAs can protect the inter-domain routing system against prefix and sub-prefix hijacks. However, inappropriate configurations bring in vulnerabilities to other types of routing security attacks. As such, the state-of-the-art approach implements the minimal-ROA principle, eliminating the risk of using ROAs at the cost of system scalability. This paper proposes the hanging ROA, a novel bitmap-based encoding scheme for ROAs, that not only ensures strong security, but also significantly improves system scalability. According to the performance evaluation with real-world data sets, the hanging ROA outperforms the state-of-the-art approach 2.4 times in terms of the compression ratio, and it can reduce the cost of a router to synchronize all validated ROA payloads by 44.5% ~ 64.7%.

More and more people turn to online voice communication with loudspeaker-equipped devices due to its convenience. To prevent speech leakage, soundproof rooms are often adopted. This paper presents mmPhone, a novel acoustic eavesdropping system that recovers loudspeaker speech protected by soundproof environments. The key idea is that properties of piezoelectric films in mmWave band can change with sound pressure due to the piezoelectric effect. If the property changes are acquired by an adversary (i.e., characterizing the piezoelectric effect with mmWaves), speech leakage can happen. More importantly, the piezoelectric film can work without a power supply. Base on this, we proposed a methodology using mmWaves to sense the film and decoding the speech from mmWaves, which turns the film into a passive "microphone". To recover intelligible speech, we further develop an enhancement scheme based on a denoising neural network, multi-channel augmentation, and speech synthesis, to compensate for the propagation and penetration loss of mmWaves. We perform extensive experiments to evaluate mmPhone and conduct digit recognition with over 93% accuracy. The results indicate mmPhone can recover high-quality and intelligible speech from a distance over 5m and is resilient to incident angles of sound waves (within 55 degrees) and different types of loudspeakers.

Despite the evolution of distributed machine learning (ML) systems in recent years, the communication overhead induced by their data transfers remains a major issue that hampers the efficiency of such systems, especially in edge networks with poor wireless link conditions. In this paper, we propose to explore a new paradigm of error-tolerant distribute ML to mitigate the communication overhead. Unlike generic network traffic, ML data exhibits an intrinsic error-tolerant capability which helps the model yield fair performance even with errors in the data transfers. We first characterize the error tolerance capability of state-of-art distributed ML frameworks. Based on the observations, we propose NeuroMessenger, a lightweight mechanism that can be built into the cellular network stack, which can enhance and utilize the error tolerance in ML data to reduce communication overhead. NeuroMessenger does not require per-model profiling and is transparent to application layer, which simplifies the development and deployment. Our experiments on a 5G simulation framework demonstrate that NeuroMessenger reduces the end-to-end latency by up to 99% while maintaining low accuracy loss under various link conditions.

Reconfigurable optical topologies promise to improve the performance in datacenters by dynamically optimizing the physical network in a demand-aware manner. State-of-the-art optical technologies allow to establish and update direct connectivity (in the form of edge-disjoint matchings) between top-of-rack switches within microseconds or less. However, to fully exploit temporal structure in the demand, such fine-grained reconfigurations also require fast algorithms for optimizing the interconnecting matchings.Motivated by the desire to offload a maximum amount of demand to the reconfigurable network, this paper initiates the study of fast algorithms to find k disjoint heavy matchings in graphs. We present and analyze six algorithms, based on iterative matchings, b-matching, edge coloring, and node-rankings. We show that the problem is generally ${\mathcal{N}}{\mathcal{P}}{\text{ - hard}}$ and study the achievable approximation ratios.An extensive empirical evaluation of our algorithms on both real-world and synthetic traces (88 in total), including traces collected in Facebook datacenters and in HPC clusters reveals that all our algorithms provide high-quality matchings, and also very fast ones come within 95 % or more of the best solution. However, the running times differ significantly and what is the best algorithm depends on k and the acceptable runtime-quality tradeoff.

Federated Learning (FL) incurs high communication overhead, which can be greatly alleviated by compression for model updates. Yet the tradeoff between compression and model accuracy in the networked environment remains unclear and, for simplicity, most implementations adopt a fixed compression rate only. In this paper, we for the first time systematically examine this tradeoff, identifying the influence of the compression error on the final model accuracy with respect to the learning rate. Specifically, we factor the compression error of each global iteration into the convergence rate analysis under both strongly convex and non-convex loss functions. We then present an adaptation framework to maximize the final model accuracy by strategically adjusting the compression rate in each iteration. We have discussed the key implementation issues of our framework in practical networks with representative compression algorithms. Experiments over the popular MNIST and CIFAR-10 datasets confirm that our solution effectively reduces network traffic yet maintains high model accuracy in FL.