A significant challenge for computation offloading in wireless multi-hop networks is the complex interactions among traffic flows in the presence of interference. Existing approaches often ignore these key effects and/or rely on outdated queueing and channel state information. To fill these gaps, we reformulate joint offloading and routing as a routing problem on an extended graph with physical and virtual links. We adopt the state-of-the-art shortest path-biased Backpressure routing algorithm, which allows the destination and the route of a job to be dynamically adjusted at every time step based on network-wide long-term information and real-time states of local neighborhoods. In large networks, our approach achieves smaller makespan than existing approaches, such as separated Backpressure offloading, and joint offloading and routing based on linear programming.

Federated learning (FL) is a useful tool that enables the training of machine learning models over distributed data without having to collect data centrally. When deploying FL in constrained wireless environments, however, intermittent connectivity of devices, heterogeneous connection quality, and non-i.i.d. data can severely slow convergence. In this paper, we consider FL with arbitrary device participation probabilities for each round and show that by weighing each device’s update by the reciprocal of their per-round participation probability, we can guarantee convergence to a stationary point. Our bound applies to non-convex loss functions and non-i.i.d. datasets and recovers state-of-the-art convergence rates for both full and uniform partial participation, including linear speedup, with only a single-sided learning rate. Then, using the derived convergence bound, we develop a new online client selection and power allocation algorithm that utilizes the Lyapunov drift-plus-penalty framework to opportunistically minimize a function of the convergence bound and the average communication time under a transmit power constraint. We use optimization over manifold techniques to obtain a solution to the minimization problem. Thanks to the Lyapunov framework, one key feature of the algorithm is that knowledge of the channel distribution is not required and only the instantaneous channel state information needs to be known. Using the CIFAR-10 dataset with varying levels of data heterogeneity, we show through simulations that the communication time can be significantly decreased using our algorithm compared to uniformly random participation, especially for heterogeneous channel conditions.

Computational offloading has become an enabling component for edge intelligence in mobile and smart devices. Existing offloading schemes mainly focus on mobile devices and servers, while ignoring the potential network congestion caused by tasks from multiple mobile devices, especially in wireless multi-hop networks. To fill this gap, we propose a low-overhead, congestion-aware distributed task offloading scheme by augmenting a distributed greedy framework with graph-based machine learning. In simulated wireless multi-hop networks with 20-110 nodes and a resource allocation scheme based on shortest path routing and contention-based link scheduling, our approach is demonstrated to be effective in reducing congestion or unstable queues under the context-agnostic baseline, while improving the execution latency over local computing.

When making decisions in a network, it is important to have up-to-date knowledge of the current state of the system. Obtaining this information, however, comes at a cost. In this paper, we determine the optimal finite-time update policy for monitoring the binary states of remote sources with a reporting rate constraint. We first prove an upper and lower bound of the minimal probability of error before solving the problem analytically. The error probability is defined as the probability that the system performs differently than it would with full system knowledge. More specifically, an error occurs when the destination node incorrectly determines which top- K priority sources are in the “free” state. We find that the optimal policy follows a specific ordered 3-stage update pattern. We then provide the optimal transition points for each stage for each source.

We propose a novel data-driven approach to allocate transmit power for federated learning (FL) over interference-limited wireless networks. The proposed method is useful in challenging scenarios where the wireless channel is changing during the FL training process and when the training data are not independent and identically distributed (non-i.i.d.) on the local devices. Intuitively, the power policy is designed to optimize the information received at the server end during the FL process under communication constraints. Ultimately, our goal is to improve the accuracy and efficiency of the global FL model being trained. The proposed power allocation policy is parameterized using graph convolutional networks (GCNs), and the associated constrained optimization problem is solved through a primal-dual (PD) algorithm. Theoretically, we show that the formulated problem has a zero duality gap and, once the power policy is parameterized, optimality depends on how expressive this parameterization is. Numerically, we demonstrate that the proposed method outperforms existing baselines under different wireless channel settings and varying degrees of data heterogeneity.

Federated learning (FL) enables distributed model training from local data collected by users. In distributed systems with constrained resources and potentially high dynamics, e.g., mobile edge networks, the efficiency of FL is an important problem. Existing works have separately considered different configurations to make FL more efficient, such as infrequent transmission of model updates, client subsampling, and compression of update vectors. However, an important open problem is how to jointly apply and tune these control knobs in a single FL algorithm, to achieve the best performance by allowing a high degree of freedom in control decisions. In this paper, we address this problem and propose FlexFL – an FL algorithm with multiple options that can be adjusted flexibly. Our FlexFL algorithm allows both arbitrary rates of local computation at clients and arbitrary amounts of communication between clients and the server, making both the computation and communication resource consumption adjustable. We prove a convergence upper bound of this algorithm. Based on this result, we further propose a stochastic optimization formulation and algorithm to determine the control decisions that (approximately) minimize the convergence bound, while conforming to constraints related to resource consumption. The advantage of our approach is also verified using experiments.

Federated learning (FL) is a useful tool in distributed machine learning that utilizes users’ local datasets in a privacy-preserving manner. When deploying FL in a constrained wireless environment; however, training models in a time-efficient manner can be a challenging task due to intermittent connectivity of devices, heterogeneous connection quality, and non-i.i.d. data. In this paper, we provide a novel convergence analysis of non-convex loss functions using FL on both i.i.d. and non-i.i.d. datasets with arbitrary device selection probabilities for each round. Then, using the derived convergence bound, we use stochastic optimization to develop a new client selection and power allocation algorithm that minimizes a function of the convergence bound and the average communication time under a transmit power constraint. We find an analytical solution to the minimization problem. One key feature of the algorithm is that knowledge of the channel statistics is not required and only the instantaneous channel state information needs to be known. Using the FEMNIST and CIFAR-10 datasets, we show through simulations that the communication time can be significantly decreased using our algorithm, compared to uniformly random participation.

This paper investigates the secret key-authenticated-capacity region, where information-theoretic authentication is defined by the ability of the decoder to accept and decode messages originating from a valid encoder while rejecting messages from other invalid sources. The model considered here consists of a valid encoder-decoder pairing that can communicate through a channel controlled by an adversary who is also able to eavesdrop on the encoder’s transmissions. Prior to the encoder’s transmission, the adversary decides whether or not to replace the decoder’s observation with an arbitrary one of the adversary’s choosing, with the adversary’s objective being to have the decoder accept and decode their observation to a valid message (different from that of the encoder). To combat the adversary, the encoder and decoder share a secret key. The secret key-authenticated-capacity region is defined as the region of jointly achievable message rate, authentication rate (a to be defined per symbol measure that will generally represent the likelihood that an adversary can fool the decoder), and the key-consumption rate (how many bits of secret key are needed per symbol sent). This is the second of a two-part study, with the parts differing in their measure of the authentication rate. For this second study, the probability of false authentication is considered as a function of the system state, where the system state is defined by the message being transmitted, the value of the secret key, the adversary’s channel observations, and the adversary’s (possibly stochastic) choice for the decoder’s observation. Termed the typical-authentication rate, the authentication measure considered here corresponds to an upper bound on the probability of false authentication for the majority of system states. For this measure, we derive matching inner and outer bounds for the secret key-enabled authenticated capacity region in terms of traditional information-theoretic measures. In doing so, it is shown that the typical-authentication rate and the message rate exhibit a one-to-one trade-off in the capacity region.

This paper investigates the secret-key-authenticated-capacity region, where information-theoretic authentication is defined by the ability of the decoder to accept and decode messages originating from a valid encoder while rejecting messages from other invalid sources. The model considered here consists of a valid encoder-decoder pairing that can communicate through a channel controlled by an adversary who is also able to eavesdrop on the encoder’s transmissions. Over multiple rounds of communication, the adversary first decides whether or not to replace the decoder’s observation with an arbitrary one of the adversary’s choosing, with the goal of the adversary being to have the decoder accept and decode their observation as a valid message (different from that of the encoder). To combat the adversary, the encoder and decoder share a secret key. The secret-key-authenticated-capacity region here is then defined as the region of jointly achievable message rate, authentication rate (a to be defined per symbol measure that will generally represent the likelihood that an adversary can fool the decoder), and the key-consumption rate (how many bits of secret key are needed per symbol sent). This is the first of a two-part study, with the parts differing in their measure of the authentication rate. In this first study, the authentication rate is the exponent of blocklength-normalized exponent of the expected probability of false authentication. For this metric, we provide an inner bound which improves on those existing in the literature. This is achieved by adopting and merging different classical techniques in novel ways. Within these classical secret-key-based authentication techniques, one technique derives authentication capability from secure channel coding to send the secret key with the message, and the other technique derives its authentication capability directly from obscuring the source.

Fingerprint embedding at the physical layer is a highly tunable authentication framework for wireless communication that achieves information-theoretic security by hiding a traditional HMAC tag in noise. In a multiantenna scenario, artificial noise (AN) can be transmitted to obscure the tag even further. The AN strategy, however, relies on perfect knowledge of the channel state information (CSI) between the legitimate users. When the CSI is not perfectly known, the added noise leaks into the receiver's observations. In this article, we explore whether AN still improves security in the fingerprint embedding authentication framework with only imperfect CSI available at the transmitter and receiver. Specifically, we discuss and design detectors that account for AN leakage and analyze the adversary's ability to recover the key from observed transmissions. We compare the detection and security performance of the optimal perfect CSI detector with the imperfect CSI robust matched filter test and a generalized likelihood ratio test (GLRT). We find that utilizing AN can greatly improve security, but suffers from diminishing returns when the quality of CSI knowledge is poor. In fact, we find that in some cases allocating additional power to AN can begin to decrease key security.

The rate regions of many variations of the standard and wire-tap channels have been thoroughly explored. Secrecy capacity characterizes the loss of rate required to ensure that the adversary gains no information about the transmissions. Authentication does not have a standard metric, despite being an important counterpart to secrecy. While some results have taken an information-theoretic approach to the problem of authentication coding, the full rate region and accompanying trade-offs have yet to be characterized. In this paper, we provide an inner bound of achievable rates with an average authentication and reliability constraint. The bound is established by combining and analyzing two existing authentication schemes for both noisy and noiseless channels. We find that our coding scheme improves upon existing schemes.

Authentication via fingerprint embedding at the physical layer utilizes noise in the wireless channel to attain a certain degree of information theoretic security that traditional HMAC methods cannot provide. Fingerprint embedding refers to a key-aided process of superimposing a low-power tag to the primary message waveform for the purpose of authenticating the transmission. The tag is uniquely created from the message and key and successful authentication is achieved when the correct tag is detected by the receiver. This paper generalizes a framework for embedding physical layer fingerprints to create an authenticated side-channel for minimal cost. Side-channel information is conveyed to the receiver through the transmitter's choice of tag from a secret codebook generated by the primary message and a shared secret key. In addition, a new linear coding scheme is introduced which enhances the ability to trade off the performance goals of authentication, side-channel rate, secrecy, and privacy.