1,721,289 research outputs found
Private wireless federated learning with anonymous over-the-air computation
No full textIn conventional federated learning (FL), differential privacy (DP) guarantees can be obtained by injecting additional noise to local model updates before transmitting to the parameter server (PS). In the wireless FL scenario, we show that the privacy of the system can be boosted by exploiting over-the-air computation (OAC) and anonymizing the transmitting devices. In OAC, devices transmit their model updates simultaneously and in an uncoded fashion, resulting in a much more efficient use of the available spectrum. We further exploit OAC to provide anonymity for the transmitting devices. The proposed approach improves the performance of private wireless FL by reducing the amount of noise that must be injected
Strong Converse for Testing Against Independence over a Noisy channel
No full textA distributed binary hypothesis testing (HT) problem over a noisy (discrete and memoryless) channel studied previously by the authors is investigated from the perspective of the strong converse property. It was shown by Ahlswede and Csiszar that a strong converse holds in the above setting when the channel is rate-limited and noiseless. Motivated by this observation, we show that the strong converse continues to hold in the noisy channel setting for a special case of HT known as testing against independence (TAI), under the assumption that the channel transition matrix has non-zero elements. The proof utilizes the blowing up lemma and the recent change of measure technique of Tyagi and Watanabe as the key tools
Wireless source transmission with time-varying side information
No full textMany applications in wireless networks require the transmission of an analog source over a fading channel to be reconstructed with the minimum distortion possible, i.e. multimedia signals over cellular networks or the accumulation of local measurements at a fusion center in sensor networks. In many practical scenarios, the destination receives additional correlated side information either form other transmitters in the network or through its own sensing devices. For example, signals from repeaters in digital TV broadcasting or relay signals in future mobile networks. However, similar to the channel state information at the transmitter, it is costly to provide an estimate of the available source side information to the transmitter. © 2011 IEEE
Linear transmission of correlated gaussian sources over MIMO channels
No full textLinear transmission of correlated L-dimensional Gaussian vector sources over multiple-input multiple-output (MIMO) channels is considered. For a static MIMO channel necessary and sufficient conditions are obtained for the optimality of zero-delay linear transmission in terms of average squared error distortion. Linear transmission is shown to be optimal when the transmission power constraint is below a threshold determined by the channel and the source parameters. Then, a fast fading MIMO channel is considered, and linear transmission is shown to achieve the optimal decay rate of the average distortion in the low SNR regime. In the high SNR regime, linear transmission is shown to achieve the optimal exponential decay rate of the expected distortion in certain settings. © VDE Verlag GMBH
Source coding under secrecy constraints
No full textDistributed compression involves compressing multiple data sources by exploiting the underlying correlation structure of the sources at separate non-cooperating encoders, while decoding is done jointly at a single decoder. Recent years have witnessed an increasing amount of research on the theoretical and practical aspects of distributed source codes, which find applications in distributed video compression, peer-to-peer data distribution systems, and sensor networks [1-3]. In many practical scenarios, limited network resources such as power and bandwidth, or physical limitations of the devices as in the case of sensor networks, pose challenges in terms of network performance and security. Oftentimes, the data aggregated in distributed compression systems may have commercial value as in the case of warehouse inventory monitoring systems, may contain sensitive information as in the case of distributed video surveillance systems, or might infringe personal privacy concerns as in the case of human body sensors measuring various health indicators. In all these scenarios, it is essential to develop distributed compression and communication protocols which exploit the limited power and bandwidth resources efficiently as well as satisfying the security requirements. Our goal in this chapter is to review fundamental limitations and tradeoffs for the overall performance optimization taking into account the quality and the security considerations jointly. There are two fundamental approaches to guarantee security in wireless networks. In the approach based on computational complexity [4], on which most practical cryptographic applications are based, the security of the system depends on the intractability assumption for a problem such as prime factorization. On the other hand, in the approach based on information theoretic secrecy introduced by Shannon in [5], the emphasis is on unconditional secrecy, which requires that, an eavesdropper with unbounded time and computational resources, and the knowledge of the encryption algorithm, does not gain any additional information about the underlying secret message upon intercepting the encrypted cryptogram. For a general review of recent progress in information theoretic security, see [6]. Although the complexity based approach has been successful in satisfying the security concerns of many practical networking applications such as the Internet, wireless networks pose additional limitations and threats that cannot be solved solely through encryption. The broadcast nature of the wireless medium makes it particularly vulnerable to eavesdropping and authentication attacks, and the energy and bandwidth limitations of wireless devices restrict their computational power, hence rendering high complexity encryption techniques undesirable. Furthermore, especially in the sensor network scenario, where the sensor nodes are generally deployed in remote locations highly vulnerable to tampering, secure key management becomes impractical. Issues such as mobility and lack of infrastructure (e.g., in mobile ad hoc networks) also pose significant challenges to traditional approaches based on maintaining secret keys. In such applications information theoretic security can support and enhance the computational complexity based approach. In this chapter, we survey information theoretic security in distributed source compression, and in particular how compression and communication can be achieved in an information theoretically secure way. Consider, for example, a sensor network in which correlated sensor observations are to be reconstructed at an access point either in a lossless fashion or within a prescribed distortion requirement. While some sensors might have secure (possibly wired) connections to the access point, others might be transmitting over the wireless medium, which can be accessed by an adversary trying to obtain information about the underlying phenomenon. Furthermore, this adversary might have her own observation of the main source. Our goal is to explore the fundamental information theoretic limitations for secure distributed compression and communication in this kind of situation. In practical applications, encryption is considered to be a separate block in the protocol stack applied in concatenation with source compression and channel transmission. The information theoretic unconditional secrecy obtained through secure source and/or channel coding or joint source-channel coding hence can be used in parallel with the existing computational encryption schemes enhancing the overall level of security. In order to fully exploit this concept of information theoretic security practical secure source and channel codes need to be developed. While there are many recent developments in this direction for channel coding [7-9] little is known for secure compression. However, design of such secure source codes is beyond the scope of this chapter, and constitutes a potential research direction. The chapter is organized as follows. After reviewing Shannon's model and the preliminaries of information theoretic secrecy in Sect. 8.2, in Sect. 8.3 we analyze distributed lossless compression under security constraints and present related fundamental results. In Sect. 8.4, we focus on lossy reconstruction at the legitimate receiver, and analyze the achievable distortion for given secrecy and communication rate constraints. Section 8.5 focuses on secure joint source-channel coding followed by the Conclusions and the Appendix. © 2010 Springer Science+Business Media, LLC
Speeding up Private Distributed Matrix Multiplication via Bivariate Polynomial Codes
No full textWe consider the problem of private distributed matrix multiplication under limited resources. Coded computation has been shown to be an effective solution in distributed matrix multiplication, both providing privacy against the workers and boosting the computation speed by efficiently mitigating stragglers. In this work, we propose the use of recently-introduced bivariate polynomial codes to further speed up private distributed matrix multiplication by exploiting the partial work done by the stragglers rather than completely ignoring them. We show that the proposed approach reduces the average computation time of private distributed matrix multiplication compared to its competitors in the literature while improving the upload communication cost and the workers' storage efficiency
Multi-Access Communications With Energy Harvesting: A Multi-Armed Bandit Model and the Optimality of the Myopic Policy
Full text linkA multi-access wireless network with N transmitting nodes, each equipped with an energy harvesting (EH) device and a rechargeable battery of finite capacity, is studied. At each time slot (TS) a node is operative with a certain probability, which may depend on the availability of data, or the state of its channel. The energy arrival process at each node is modelled as an independent two-state Markov process, such that, at each TS, a node either harvests one unit of energy, or none. At each TS a subset of the nodes is scheduled by the access point (AP). The scheduling policy that maximises the total throughput is studied assuming that the AP does not know the states of either the EH processes or the batteries. The problem is identified as a restless multi-armed bandit (RMAB) problem, and an upper bound on the optimal scheduling policy is found. Under certain assumptions regarding the EH processes and the battery sizes, the optimality of the myopic policy (MP) is proven. For the general case, the performance of MP is compared numerically to the upper bound
PRIVACY-AWARE COMMUNICATION OVER A WIRETAP CHANNEL WITH GENERATIVE NETWORKS
No full textWe study privacy-aware communication over a wiretap channel using end-to-end learning. Alice wants to transmit a source signal to Bob over a binary symmetric channel, while passive eavesdropper Eve tries to infer some sensitive attribute of Alice's source based on its overheard signal. Since we usually do not have access to true distributions, we propose a data-driven approach using variational autoencoder (VAE)-based joint source channel coding (JSCC). We show through simulations with the colored MNIST dataset that our approach provides high reconstruction quality at the receiver while confusing the eavesdropper about the latent sensitive attribute, which consists of the color and thickness of the digits. Finally, we consider a parallel-channel scenario, and show that our approach arranges the information transmission such that the channels with higher noise levels at the eavesdropper carry the sensitive information, while the non-sensitive information is transmitted over more vulnerable channels
Fast Federated Edge Learning with Overlapped Communication and Computation and Channel-Aware Fair Client Scheduling
No full textWe consider federated edge learning (FEEL) over wireless fading channels taking into account the downlink and uplink channel latencies, and the random computation delays at the clients. We speed up the training process by overlapping the communication with computation. With fountain coded transmission of the global model update, clients receive the global model asynchronously, and start performing local computations right away. Then, we propose a dynamic client scheduling policy, called MRTP, for uploading local model updates to the parameter server (PS), which, at any time, schedules the client with the minimum remaining upload time. However, MRTP can lead to biased participation of clients in the update process, resulting in performance degradation in non-iid data scenarios. To overcome this, we propose two alternative schemes with fairness considerations, termed as age-aware MRTP (A-MRTP), and opportunistically fair MRTP (OF-MRTP). In A-MRTP, the remaining clients are scheduled according to the ratio between their remaining transmission time and the update age, while in OF-MRTP, the selection mechanism utilizes the long term average channel rate of the clients to further reduce the latency while ensuring fair participation of the clients. It is shown through numerical simulations that OFMRTP provides significant reduction in latency without sacrificing test accuracy
- …
