Cellular systems have to resist smart jammers that can optimize their selection of jamming channels and powers based on the estimated ongoing network states. In this article, we present an unmanned aerial vehicle (UAV) aided cellular framework against jamming, in which an UAV uses reinforcement learning to choose the relay policy for a mobile user whose serving base station is attacked by a jammer. More specifically, the UAV applies deep reinforcement learning and transfer learning to help cellular systems resist smart jamming without knowing the cellular topology, the message generation model, the server computation model and the jamming model, based on the previous anti-jamming relay experiences and the observed current communication status. The performance bound in terms of the bit error rate and the UAV energy consumption is derived from the Nash equilibrium of the studied dynamic relay game and verified via simulations. Simulation results show that this scheme can reduce the bit error rate and save the UAV energy consumption in comparison with the benchmark.

UAV-Aided Cellular Communications with Deep Reinforcement Learning Against Jamming

By using smart radio devices, a jammer can dynamically change its jamming policy based on opposing security mechanisms; it can even induce the mobile device to enter a specific communication mode and then launch the jamming policy accordingly. On the other hand, mobile devices can exploit spread spectrum and user mobility to address both jamming and interference. In this paper, a two-dimensional (2-D) antijamming mobile communication scheme is proposed in which a mobile device leaves a heavily jammed/interfered-with frequency or area. It is shown that, by applying reinforcement learning techniques, a mobile device can achieve an optimal communication policy without the need to know the jamming and interference model and the radio channel model in a dynamic game framework. More specifically, a hotbooting deep Q-network based 2-D mobile communication scheme is proposed that exploits experiences in similar scenarios to reduce the exploration time at the beginning of the game, and applies deep convolutional neural network and macro-action techniques to accelerate learning in dynamic situations. Several real-world scenarios are simulated to evaluate the proposed method. These simulation results show that our proposed scheme can improve both the signal-to-interference-plus-noise ratio of the signals and the utility of the mobile devices against cooperative jamming compared with benchmark schemes.

Two-Dimensional Antijamming Mobile Communication Based on Reinforcement Learning

This paper considers the shortest link scheduling problem in wireless networks under the signal-to-interference-plus-noise ratio (SINR) model. We propose an $O({}\log({}l_{\max}/l_{\min}{}){})$ -approximation algorithm called shortest link scheduling with power control (SLSPC) with oblivious power assignment and an $O(\log_{1+\varphi}(l_{\max}/l_{\min})$ -approximation algorithm called shortest link scheduling with uniform or mean power assignment (SLSUM) with uniform or mean power control, where $\varphi > 0$ is a constant serving as a regulatory factor for slight transmit power adjustment, and where $l_{\max}$ and $l_{\min}$ denote the lengths of the longest and shortest links, respectively. We conduct a rigorous theoretical performance analysis to analyze the feasibility and approximation factors of the proposed algorithms. We also carry out an extensive comparison-based simulation study, whose results indicate that the performances of SLSPC and SLSUM are superior over the state of the art as the set of the so-called “black and gray” links, which are difficult to schedule and should be sequentially scheduled, is completely removed by adjusting the transmit power appropriately via $\varphi$ . Our numerical analysis demonstrates that the approximation ratios of our algorithms are tighter than the best known ratios.

Shortest Link Scheduling Algorithms in Wireless Networks Under the SINR Model

In this paper, we consider a jammed wireless scenario where a network operator aims to schedule users to maximize network performance while guaranteeing a minimum performance level to each user. We consider the case where no information about the position and the triggering threshold of the jammer is available. We show that the network performance maximization problem can be modeled as a finite-horizon joint power control and user scheduling problem, which is NP-hard. To find the optimal solution of the problem, we exploit dynamic programming techniques. We show that the obtained problem can be decomposed, i.e., the power control problem and the user scheduling problem can be sequentially solved at each slot. We investigate the impact of uncertainty on the achievable performance of the system and we show that such uncertainty leads to the well-known exploration-exploitation tradeoff. Due to the high complexity of the optimal solution, we introduce an approximation algorithm by exploiting state aggregation techniques. We also propose a performance-aware online greedy algorithm to provide a low-complexity sub-optimal solution to the joint power control and user scheduling problem under minimum quality-of-service requirements. The efficiency of both solutions is evaluated through extensive simulations, and our results show that the proposed solutions outperform other traditional scheduling policies.

Optimal Power Allocation and Scheduling Under Jamming Attacks

In this paper, we examine the maximization of energy efficiency (EE) in next-generation multiuser MIMO–OFDM networks that vary dynamically over time—e.g., due to user mobility, fluctuations in the wireless medium, modulations in the users’ load, etc. Contrary to the static/stationary regime, the system may evolve in an arbitrary manner, so users must adjust “on the fly,” without being able to predict the state of the system in advance. To tackle these issues, we propose a simple and distributed online optimization policy that leads to no regret , i.e., it allows users to match (and typically outperform) even the best fixed transmit policy in hindsight, irrespective of how the system varies with time. Moreover, to account for the scarcity of perfect channel state information (CSI) in massive MIMO systems, we also study the algorithm’s robustness in the presence of measurement errors and observation noise. Importantly, the proposed policy retains its no-regret properties under very mild assumptions on the error statistics: on average, it enjoys the same performance guarantees as in the noiseless deterministic case. Our analysis is supplemented by extensive numerical simulations, which show that, in realistic network environments, users track their individually optimum transmit profile even under rapidly changing channel conditions, achieving gains of up to 600% in energy efficiency over uniform power allocation policies.

Learning to Be Green: Robust Energy Efficiency Maximization in Dynamic MIMO–OFDM Systems

We study algorithms for wireless spectrum access of $n$ communication requests when interference conditions are given by the Rayleigh-fading model. This model extends the recently popular deterministic interference model based on the signal-to-interference-plus-noise ratio (SINR) using stochastic propagation to address fading effects observed in reality. We consider worst-case approximation guarantees for the two standard problems of capacity maximization (maximize the expected number of successful transmissions in a single slot) and latency minimization (minimize the expected number of slots until all transmissions were successful). Our main result is a generic reduction of Rayleigh fading to the deterministic SINR model. It allows to apply existing algorithms for the non-fading model in the Rayleigh-fading scenario while losing only a factor of O(logast n) in the approximation guarantee. This way, we obtain the first approximation guarantees for Rayleigh fading and, more fundamentally, show that non-trivial stochastic fading effects can be successfully handled using existing and future techniques for the non-fading model. Using a more detailed argument, a similar result applies even for distributed and game-theoretic capacity maximization approaches. For example, it allows to show that regret learning yields an O(log* n)-approximation with uniform power assignments. Our analytical treatment is supported by simulations illustrating the performance of regret learning and, more generally, the relationship between both models.

Scheduling in wireless networks with rayleigh-fading interference

We study approximation algorithms for optimization of wireless spectrum access with $n$ communication requests when interference conditions are given by the Rayleigh-fading model. This model extends the deterministic interference model based on the signal-to-interference-plus-noise ratio (SINR) using stochastic propagation to address fading effects observed in reality. We consider worst-case approximation guarantees for the two standard problems of capacity maximization and latency minimization. Our main result is a generic reduction of Rayleigh fading to the deterministic non-fading model. It allows to apply existing algorithms for the non-fading model in the Rayleigh-fading scenario while losing only a factor of $O(\log ^\ast\, n)$ in the approximation guarantee. This way, we obtain the first approximation guarantees for Rayleigh fading and, more fundamentally, show that non-trivial stochastic fading effects can be successfully handled using existing and future techniques for the non-fading model. We generalize these results in two ways. First, the same results apply for capacity maximization with variable data rates, when links obtain (non-binary) utility depending on the achieved SINR. Second, for binary utilities, we use a more detailed argument to obtain similar results even for distributed and game-theoretic approaches. Our analytical treatment is supported by simulations illustrating the performance of regret learning and, more generally, the relationship between both models.

Scheduling in Wireless Networks with Rayleigh-Fading Interference

We consider capacity maximization in wireless networks under adversarial interference conditions. There are $n$ links, i.e., sender-receiver pairs, which repeatedly try to perform a successful transmission. In each time step, the success of attempted transmissions depends on interference conditions, which are captured by an interference model (e.g., the SINR model). Additionally, an adversarial jammer can render a $(1-\delta)$ -fraction of time steps in a time window unsuccessful. For this scenario, we analyze a framework for distributed no-regret learning algorithms to get provable approximation guarantees. We obtain an $O(1/\delta)$ -approximation for the problem of maximizing the number of successful transmissions. Our approach provides even a constant-factor approximation when the jammer exactly blocks a $(1-\delta)$ -fraction of time steps. In addition, we consider the parameters of the jammer being partially unknown to the algorithm, and we also consider a stochastic jammer, for which we obtain a constant-factor approximation after a polynomial number of time steps. We extend our results to more general settings, in which links arrive and depart dynamically, and where each sender tries to reach multiple receivers. Our algorithms perform favorably in simulations.

Jamming-Resistant Learning in Wireless Networks

We consider capacity maximization algorithms for wireless networks with changing availabilities of spectrum. There are n sender-receiver pairs (called links) and k channels. We consider an iterative round-based scenario, where in each round the set of channels available to each link changes. Each link independently decides about access to one available channel in order to implement a successful transmission. Transmissions are subject to interference and noise, and we use a general approach based on affectance to define which attempts are successful. This includes recently popular interference models based on SINR.

Sleeping Experts in Wireless Networks

We consider capacity maximization in wireless networks under adversarial interference conditions. There are n links, each consisting of a sender and a receiver, which repeatedly try to perform a successful transmission. In each time step, the success of attempted transmissions depends on interference conditions, which are captured by an interference model (e.g. the SINR model). Additionally, an adversarial jammer can render a (1 − δ)-fraction of time steps unsuccessful. For this scenario, we analyze a framework for distributed no-regret learning algorithms. We obtain an $O\left(1/\delta\right)$-approximation for the problem of maximizing the number of successful transmissions. Our approach provides even a constant-factor approximation when the jammer exactly blocks a (1 − δ)-fraction of time steps. In addition, we consider the parameters of the jammer being unknown to the algorithm, and we also consider a stochastic jammer, for which we obtain a constant-factor approximation after a polynomial number of time steps. We extend our results to more general settings, in which links arrive and depart dynamically.

Johannes Dams

Papers

Scheduling in wireless networks with rayleigh-fading interference

Scheduling in Wireless Networks with Rayleigh-Fading Interference

Jamming-Resistant Learning in Wireless Networks

Sleeping Experts in Wireless Networks

Jamming-Resistant Learning in Wireless Networks