INFOCOM 2022 Papers

Block capacity plays a critical role in maintaining blockchain security and improving transactions per second (TPS). Increasing block capacity can help attain higher TPS, but it also prolongs block propagation latency and degrades system security. In the present paper, we argue that existing work compressing the block size to shorten block propagation latency introduces an undesired side effect, which is that the size of compressed blocks will increase with transaction volume. Instead, we propose Dino, a new block transmission protocol between two peers. Once a node receives a Dino block, it can recover the original block with that Dino block and transactions in its transaction pool. Since Dino transmits block construction rules instead of compressed block content, it has good scalability to transmit blocks with larger transaction volume. We deploy Dino into Bitcoin and Bitcoin Cash to compare it with the state-of-art protocols: Compact, XThin, and Graphene. For a block with 3,000 transactions, its corresponding Dino block is no more than 1 KB in size, which is only 4% of a XThin block, 5% of a Compact block, and 20% of a Graphene block. The size of Dino blocks stays constant when the transaction volume reaches Bitcoin and Bitcoin Cash’s protocol limit. Simulation experiments show that Dino scales well with higher transaction generation rates and can reduce block propagation latency.

In this work, we study a status update system with a source node sending timely information to the destination through a channel with random delay. We measure the timeliness of the information stored at the receiver via the Age of Information (AoI), the time elapsed since the freshest sample stored at the receiver is generated. The goal is to design a sampling strategy that minimizes the total cost of the expected time average AoI and sampling cost in the absence of transmission delay statistics. We reformulate the total cost minimization problem as the optimization of a renewal-reward process, and propose an online sampling strategy based on the Robbins-Monro algorithm. Denote K to be the number of samples we have taken. We show that, when the transmission delay is bounded, the expected time average total cost obtained by the proposed online algorithm converges to the minimum cost when K goes to infinity, and the optimality gap decays with rate ${\mathcal{O}}$(ln K/K). Simulation results validate the performance of our proposed algorithm.

Intent-Based Networking mandates that high-level human-understandable intents are automatically interpreted and implemented by network management entities. As a key part in this process, it is required that network orchestrators activate the correct automated decision model to meet the intent objective. In anticipatory networking tasks, this requirement maps to identifying and deploying a tailored prediction model that can produce a forecast aligned with the specific –and typically complex– network management goal expressed by the original intent. Current forecasting models for network demands or network management optimize generic, non-flexible, and manually designed objectives, hence do not fulfil the needs of anticipatory Intent-Based Networking. To close this gap, we propose LossLeaP, a novel forecasting model that can autonomously learn the relationship between the prediction and the target management objective, steering the former to minimize the latter. To this end, LossLeaP adopts an original deep learning architecture that advances current efforts in automated machine learning, towards a spontaneous design of loss functions for regression tasks. Extensive experiments in controlled environments and in practical application case studies prove that LossLeaP outperforms a wide range of benchmarks, including state-of-the-art solutions for network capacity forecasting.

Today’s network is notorious for its complexity and uncertainty. Network operators often rely on network models to achieve efficient network planning, operation, and optimization. The network model is responsible for understanding the complex relationships between the network performance metrics (e.g., latency) and the network characteristics (e.g., traffic). However, we still lack a systematic approach to developing accurate and lightweight network models that are aware of the impact of network configurations (i.e., expressiveness) and provide fine-grained flow-level temporal predictions (i.e., granularity).In this paper, we propose xNet, a data-driven network modeling framework based on graph neural networks (GNN). Unlike the previous proposals, xNet is not a dedicated network model designed for specific network scenarios with constraint considerations. On the contrary, xNet provides a general approach to modeling the network characteristics of concern with relation graph representations and configurable GNN blocks. xNet learns the state transition function between time steps and rolls it out to obtain the full fine-grained prediction trajectory. We implement and instantiate xNet with three use cases. The experiment results show that xNet can accurately predict different performance metrics while achieving over two orders of magnitude of speedup compared with the conventional packet-level simulator.

Traffic splitting is a required functionality in networks, for example for load balancing over multiple paths or among different servers. The capacities of the servers determine the partition by which traffic should be split. A recent approach implements traffic splitting within the ternary content addressable memory (TCAM), which is often available in switches. It is important to reduce the amount of memory allocated for this task since TCAMs are power consuming and are also required for other tasks such as classification and routing. Previous work showed how to compute the smallest prefix-matching TCAM necessary to implement a given partition exactly. In this paper we solve the more practical case, where at most n prefix-matching TCAM rules are available, restricting the ability to implement exactly the desired partition. We consider the L1distance between partitions, which is of interest when overloaded requests are simply dropped, and we want to minimize the total loss. We prove that the Niagara algorithm [1] can be used to find the closest partition in L1to the desired partition, that can be realized with n TCAM rules. Moreover, we prove it for arbitrary partitions, with (possibly) non-integer parts.

Attacks against industrial control systems (ICSs) often exploit the insufficiency of authentication mechanisms. Verifying whether the received messages are intact and issued by legitimate sources can prevent malicious data/command injection by illegitimate or compromised devices. However, the key challenge is to introduce message authentication for various ICS communication models, including multicast or broadcast, with a messaging rate that can be as high as thousands of messages per second, within very stringent latency constraints. For example, certain commands for protection in smart grids must be delivered within 2 milliseconds, ruling out public-key cryptography. This paper proposes two lightweight message authentication schemes, named CMA and its multicast variant CMMA, that perform precomputation and caching to authenticate future messages. With minimal precomputation and communication overhead, C(M)MA eliminates all cryptographic operations for the source after the message is given, and all expensive cryptographic operations for the destinations after the message is received. C(M)MA considers the urgency profile (or likelihood) of a set of future messages for even faster verification of the most time-critical (or likely) messages. We demonstrate the feasibility of C(M)MA in an ICS setting based on a substation automation system in smart grids.

The next generation of cellular networks will be characterized by softwarized, open, and disaggregated architectures exposing analytics and control knobs to enable network intelligence via innovative data-driven algorithms. How to practically realize this vision, however, is largely an open problem. For a given network optimization/automation objective, it is currently unknown how to select which data-driven models should be deployed and where, which parameters to control, and how to feed them appropriate inputs. In this paper, we take a decisive step forward by presenting and prototyping OrchestRAN, a novel orchestration framework for next generation systems that embraces and builds upon the Open Radio Access Network (RAN) paradigm to provide a practical solution to these challenges. OrchestRAN has been designed to execute in the non-Real-time (RT) RAN Intelligent Controller (RIC) and allows Network Operators (NOs) to specify high-level control/inference objectives (i.e., adapt scheduling, and forecast capacity in near-RT, e.g., for a set of base stations in Downtown New York). OrchestRAN automatically computes the optimal set of data-driven algorithms and their execution location (e.g., in the cloud, or at the edge) to achieve intents specified by the NOs while meeting the desired timing requirements and avoiding conflicts between different data-driven algorithms controlling the same parameters set. We show that the intelligence orchestration problem in Open RAN is NP-hard, and design low-complexity solutions to support real-world applications. We prototype OrchestRAN and test it at scale on Colosseum, the world’s largest wireless network emulator with hardware in the loop. Our experimental results on a network with 7 base stations and 42 users demonstrate that OrchestRAN is able to instantiate data-driven services on demand with minimal control overhead and latency.

Behavior recognition plays an essential role in numerous behavior-driven applications (e.g., virtual reality and smart home) and even in the security-critical applications (e.g., security surveillance and elder healthcare). Recently, WiFi-based behavior recognition (WBR) technique stands out among many behavior recognition techniques due to its advantages of being non-intrusive, device-free, and ubiquitous. However, existing WBR research mainly focuses on improving the recognition precision, while neglecting the security aspects. In this paper, we reveal that WBR systems are vulnerable to manipulating physical signals. For instance, our observation shows that WiFi signals can be changed by jamming signals. By exploiting the vulnerability, we propose two approaches to generate physically online adversarial samples to perform untargeted attack and targeted attack, respectively. The effectiveness of these attacks are extensively evaluated over four real-world WBR systems. The experiment results show that our attack approaches can achieve 80% and 60% success rates for untargeted attack and targeted attack in physical world, respectively. We also propose three methods to mitigate the hazard of such attacks.

Per-flow spread measurement in high-speed networks can provide indispensable information to many practical applications. However, it is challenging to measure millions of flows at line speed because on-chip memory modules cannot simultaneously provide large capacity and large bandwidth. The prior studies address this mismatch by entirely using on-chip compact data structures or utilizing off-chip space to assist limited on-chip memory. Nevertheless, their on-chip data structures record massive transient elements, each of which only appears in a short time interval in a long-period measurement task, and thus waste significant on-chip space. This paper presents short-term memory sampling, a novel spread estimator that samples new elements while only holding elements for short periods. Our estimator can work with tiny on-chip space and provide accurate estimations for online queries. The key of our design is a short-term memory duplicate filter that reports new elements and filters duplicates effectively while allowing incoming elements to override the stale elements to reduce on-chip memory usage. We implement our approach on a NetFPGA-equipped prototype. Experimental results based on real Internet traces show that, compared to the state-of-the-art, short-term memory sampling reduces up to 99% of on-chip memory usage when providing the same probabilistic assurance on spread-estimation error.

In a commercial cloud, service providers (e.g., video streaming service provider) rent resources from cloud vendors (e.g., Google Cloud Platform) and provide services to cloud users, making a profit from the price gap. Cloud users acquire services by forwarding their requests to corresponding servers. In practice, as a common scenario, traffic dynamics will cause server overload or load-unbalancing. Existing works mainly deal with the problem by two methods: elastic resource provisioning and request updating. Elastic resource provisioning is a fast and agile solution but may cost too much since service providers need to buy extra resources from cloud vendors. Though request updating is a free solution, it will cause a significant delay, resulting in a bad users’ QoS. In this paper, we present a new scheme, called real-time request updating with elastic resource provisioning (TRUST), to help service providers pay less cost with users’ QoS guarantee in clouds. In addition, we propose an efficient algorithm for TRUST with a bounded approximation factor based on randomized rounding. Both small-scale experiment results and large-scale simulation results show the superior performance of our proposed algorithm compared with state-of-the-art benchmarks.

As acoustic communication systems become more common in homes and offices, eavesdropping brings significant security and privacy risks. Current approaches of acoustic eavesdropping either provide low resolution due to the use of sub-6 GHz frequencies, work only for limited words using classification, or cannot work through-wall due to the use of optical sensors. In this paper, we present MILLIEAR, a mmWave acoustic eavesdropping system that leverages the high-resolution of mmWave FMCW ranging and generative machine learning models to not only extract vibrations but to reconstruct the audio. MILLIEAR combines speaker vibration estimation with conditional generative adversarial networks to eavesdrop with unconstrained vocabulary. We implement and evaluate MIL-LIEAR using off-the-shelf mmWave radar deployed in different scenarios and settings. We find that it can accurately reconstruct the audio even at different distances, angles and through the wall with different insulator materials. Our subjective and objective evaluations show that the reconstructed audio has a strong similarity with the original audio.

Parallel systems divide jobs into smaller tasks that can be serviced by many workers at the same time. Some parallel systems have blocking barriers that require all of their tasks to start and/or depart in unison. This is true of many parallelized machine learning workloads, and the popular Apache Spark processing engine has recently added support for Barrier Execution Mode, which allows users to add such barriers to their jobs. The drawback of these barriers is reduced performance and stability compared to equivalent non-blocking systems.We derive analytical expressions for the stability regions for parallel systems with blocking start and/or departure barriers. We extend results from queueing theory to derive waiting and sojourn time bounds for systems with blocking start barriers. Our results show that for a given system utilization and number of servers, there is an optimal degree of parallelism that balances waiting time and job execution time. This observation leads us to propose and implement a class of self-adaptive schedulers, we call "Take-Half", that modulate the allowed degree of parallelism based on the instantaneous system load, improving mean performance and eliminating stability issues.

This paper aims to integrate two synergetic technologies, federated learning (FL) and width-adjustable slimmable neural network (SNN) architectures. FL preserves data privacy by exchanging the locally trained models of mobile devices. By adopting SNNs as local models, FL can flexibly cope with the time-varying energy capacities of mobile devices. Combining FL and SNNs is however non-trivial, particularly under wireless connections with time-varying channel conditions. Furthermore, existing multi-width SNN training algorithms are sensitive to the data distributions across devices, so are ill-suited to FL. Motivated by this, we propose a communication and energy efficient SNN-based FL (named SlimFL) that jointly utilizes superposition coding (SC) for global model aggregation and superposition training (ST) for updating local models. By applying SC, SlimFL exchanges the superposition of multiple width configurations that are decoded as many as possible for a given communication throughput. Leveraging ST, SlimFL aligns the forward propagation of different width configurations, while avoiding the inter-width interference during back propagation. We formally prove the convergence of SlimFL. The result reveals that SlimFL is not only communication-efficient but also can counteract non-IID data distributions and poor channel conditions, which is also corroborated by simulations.

Caching popular contents at the network edge is promising to reduce the retrieval latency, the network congestion, and the number of requests to the remote content provider during peak hours. In general, edge caching resource is costly and highly limited. Nevertheless, it is possible to provide cost-effective caching services using unreliable resources, which are resources reserved for other applications but have not been fully used or resources on vulnerable servers. In this paper, we consider the problem of caching popular contents over unreliable resources as a less expensive solution to limited edge caching capacity. In particular, to address the unreliability of edge resources, erasure coding is leveraged to increase the availability of cached contents. We formulate the problem as a discrete optimization problem and prove it is NP-hard. We start with two special cases of the problem and provide optimal algorithms for them. We then design an algorithm for the general version of the proposed problem and provide a provable performance guarantee. Real-world data-driven simulations demonstrate that the proposed algorithms significantly outperform popular baselines, and the rewards for the general version of the problem are near-optimal.

The traditional migration methods are confronted with formidable challenges when data deduplication technologies are incorporated. Firstly, the deduplication creates data-sharing dependencies in the stored files; breaking such dependencies in migration would attach extra space overhead. Secondly, the redundancy elimination heightens the risk of data unavailability during server crashes. The existing methods fail to tackle them at one shot. To this end, we propose Jingwei, an efficient and adaptable data migration strategy for deduplicated storage systems. To be specific, Jingwei tries to minimize the extra space cost in migration for space efficiency. Meanwhile, Jingwei realizes the service adaptability by encouraging replicas of hot data to spread out their data access requirements. We first model such a problem as an integer linear programming (ILP) and solve it with a commercial solver when only one empty migration target server is allowed. We then extend this problem to a scenario wherein multiple non-empty target servers are available for migration. We solve it by effective heuristic algorithms based on the Bloom Filter-based data sketches. Trace-driven experiments show that Jingwei fortifies the file replicas by 25%, while only 5.7% of the extra storage space is occupied compared with the latest "Goseed" method.

Quantum networks are considered as a promising future platform for quantum information exchange and quantum applications, which have capabilities far beyond the traditional communication networks. Remote quantum entanglement is an essential component of a quantum network. How to efficiently design a multi-routing entanglement protocol is a fundamental yet challenging problem. In this paper, we study a quantum entanglement routing problem to simultaneously maximize the number of quantum-user pairs and their expected throughput. Our approach is to formulate the problem as two sequential integer programming steps. We propose efficient entanglement routing algorithms for the two integer programming steps and analyze their time complexity and performance bounds. Results of evaluation highlight that our approach outperforms existing solutions in both served quantum-user pairs numbers and the network expected throughput.

Website fingerprinting (WF) attacks allow an attacker to eavesdrop on the encrypted network traffic between a victim and an anonymous communication system so as to infer the real destination websites visited by a victim. Recently, the deep learning (DL) based WF attacks are proposed to extract high level features by DL algorithms to achieve better performance than that of the traditional WF attacks and defeat the existing defense techniques. To mitigate this issue, we propose a-genetic-programming-based variant cover traffic search technique to generate defense strategies for effectively injecting dummy Tor cells into the raw Tor traffic. We randomly perform mutation operations on labeled original traffic traces by injecting dummy Tor cells into the traces to derive variant cover traffic. A high level feature distance based fitness function is designed to improve the mutation rate to discover successful variant traffic traces that can fool the DL-based WF classifiers. Then the dummy Tor cell injection patterns in the successful variant traces are extracted as defense strategies that can be applied to the Tor traffic. Extensive experiments demonstrate that we can introduce 8.1% of bandwidth overhead to significantly decrease the accuracy rate below 0.4% in the realistic open-world setting.

Today’s radio access networks (RANs) are monolithic entities which often operate statically on a given set of parameters for the entirety of their operations. To implement realistic and effective spectrum sharing policies, RANs will need to seamlessly and intelligently change their operational parameters. In stark contrast with existing paradigms, the new O-RAN architectures for 5G-and-beyond networks (NextG) separate the logic that controls the RAN from its hardware substrate, allowing unprecedented real-time fine-grained control of RAN components. In this context, we propose the Channel-Aware Reactive Mechanism (ChARM), a data-driven O-RAN-compliant framework that allows (i) sensing the spectrum to infer the presence of interference and (ii) reacting in real time by switching the distributed unit (DU) and radio unit (RU) operational parameters according to a specified spectrum access policy. ChARM is based on neural networks operating directly on unprocessed I/Q waveforms to determine the current spectrum context. ChARM does not require any modification to the existing 3GPP standards. It is designed to operate within the O-RAN specifications, and can be used in conjunction with other spectrum sharing mechanisms (e.g., LTE-U, LTE-LAA or MulteFire). We demonstrate the performance of ChARM in the context of spectrum sharing among LTE and Wi-Fi in unlicensed bands, where a controller operating over a RAN Intelligent Controller (RIC) senses the spectrum and switches cell frequency to avoid Wi-Fi. We develop a prototype of ChARM using srsRAN, and leverage the Colosseum channel emulator to collect a large-scale waveform dataset to train our neural networks with. To collect standard-compliant Wi-Fi data, we extended the Colosseum testbed using system-on-chip (SoC) boards running a modified version of the OpenWiFi architecture. Experimental results show that ChARM achieves accuracy of up to 96% on Colosseum and 85% on an over-the-air testbed, demonstrating the capacity of ChARMto exploit the considered spectrum channels.

We revisit the problem of missing tag identification in RFID networks by making three contributions. Firstly, we quantitatively compare and gauge the existing propositions spanning over a decade on missing tag identification. We show that the expected execution time of the best solution in the literature is $\Theta \left( {N + \frac{{{{(1 - \alpha )}^2}{{(1 - \delta )}^2}}}{{{\varepsilon ^2}}}} \right)$, where δ and ϵ are parameters quantifying the required identification accuracy, N denotes the number of tags in the system, among which αN tags are missing. Secondly, we analytically establish the expected execution time lower-bound for any missing tag identification algorithm as $\Theta \left( {\frac{N}{{\log N}} + \frac{{{{(1 - \delta )}^2}{{(1 - \alpha )}^2}}}{{{\varepsilon ^2}\log \frac{{(1 - \delta )(1 - \alpha )}}{\varepsilon }}}} \right)$, thus giving the theoretical performance limit. Thirdly, we develop a novel missing tag identification algorithm by leveraging a tree structure with the expected execution time of $\Theta \left( {\frac{{\log \log N}}{{\log N}}N + \frac{{{{(1 - \alpha )}^2}{{(1 - \delta )}^2}}}{{{\varepsilon ^2}}}} \right)$, reducing the time overhead by a factor of up to log N over the best algorithm in the literature. The key technicality in our design is a novel data structure termed as collision-partition tree (CPT), built on a subset of bits in tag pseudo-IDs, leading to more balanced tree structure and reducing the time complexity in parsing the entire tree.