We address a denial of service attack on low-latency communication, wherein a cognitive adversary, armed with a full-duplex radio, can simultaneously measure the energy levels on the victim’s frequency band besides jamming it. First, we show that traditional countermeasures such as frequency hopping cannot mitigate such attacks. To circumvent this problem, we propose two cooperative relaying schemes, namely: semi-coherent fast-forward half-duplex and semi-coherent fast-forward full-duplex relaying schemes, wherein the victim node seeks the assistance of a helper node in its vicinity, that can either work in half-duplex or full-duplex mode, to instantaneously multiplex the victim’s messages in its message to the destination. In the former scheme, the victim is required to continue transmission on the jammed band, whereas the helper node is asked to overhear the victim’s frequency band to multiplex their information in its frequency band. In the latter scheme, the victim and the helper node cooperatively pour fraction of their power on both frequency bands so as to reliably communicate their messages on the helper’s frequency band. Extensive analysis on the choice of the power splitting factor between the frequency bands is presented besides proving that both schemes successfully engage the adversary on the jammed frequency band.

A number of applications in next-generation multi-hop networks impose low-latency requirements on data transmission thereby necessitating the underlying relays to introduce negligible delay when forwarding the packets. While traditional relaying techniques such as amplify-and-forward protocols may help the packets to satisfy latency-constraints, such strategies do not facilitate the destination in learning the provenance of the packets, which in turn could be used for either learning the topology of the network or detecting security threats on the network. With potential applications to vehicular networks, we address the design of provenance embedding algorithms that reduce the delays on the packets and yet assist the destination in determining their provenance. We propose double-edge (DE) embedding techniques, wherein a subset of the relay nodes in the path strategically skip the provenance embedding process to reduce the delays on the packets. Using bloom filters as tools to implement double-edge embedding, first, we derive upper bounds on the error-rates of the DE embedding techniques, and subsequently present experimental results on a test bed to demonstrate the efficacy of the proposed techniques. We show that the DE embedding techniques offer latency benefits up to 17 % along with remarkable reduction in error-rates in comparison with the baselines

Multi-hop networks comprising line-of-sight dominated wireless channels have found attractive applications in wireless networks involving unmanned aerial vehicles. It is well known that these applications require both high-reliability and low-latency features. While automatic-repeat-request (ARQs) based decode-and-forward (DF) relaying strategies provide the high-reliability feature, they are known to incur delay owing to multiple re-transmissions of packets. Towards addressing the above trade-off between reliability and low-latency of ARQ based DF strategies, we impose a sum constraint on the total number of ARQs across the nodes as it captures an upper bound on the time taken for packet re-transmissions. Subsequently, we propose various cooperative strategies among the nodes to minimize the packet-drop-probability (PDP). At the heart of our cooperative strategies lies the idea that unused ARQs of a node can be used by its neighbors to reduce the PDP without violating the sum constraint on the total number of ARQs. For each of the cooperative strategies, we formulate an optimization problem for distributing a given number of ARQs among the nodes. Then, we provide necessary and sufficient conditions on the optimal ARQ distribution in each of these strategies, and use these conditions to synthesize low-complexity algorithms to solve the problem. In addition to using PDP, we also present a thorough analysis on the delay profiles of the packets to incorporate the effect of cooperation on the latency performance of the cooperative strategies.

In this paper, we propose a new adversarial attack on frequency-hopping-based wireless communication between two users, namely Alice and Bob. In this attack, the adversary referred to as Eve, instantaneously modifies the transmitted signal by Alice before forwarding it to Bob within the symbol period. We show that this attack forces Bob to incorporate Eve's signal in the decoding process; otherwise, treating it as noise would further degrade the performance akin to jamming. Through this attack, we show that Eve can convert a slow-fading channel between Alice and Bob to a rapid-fading one by modifying every transmitted symbol independently. As a result, neither pilot-assisted coherent detection techniques nor blind-detection methods are directly applicable as countermeasures. As potential mitigation strategies, we explore the applicability of frequency hopping along with on-off keying (OOK) and binary frequency-shift keying (BFSK) as modulation schemes. In the case of OOK, the attacker attempts to introduce deep fades on the tone carrying the information bit, whereas in the case of BFSK, the attacker pours comparable energy levels on the tones carrying bit-0 and bit-1, thereby degrading the performance. Based on extensive analyses and experimental results, we show that when using OOK, Bob must be equipped with a large number of receive antennas to reliably detect Alice's signal, and when using BFSK, Alice and Bob must agree upon a secret key to randomize the location of the tones carrying the bits, in addition to randomizing the carrier frequency of communication.

It is well known that physical-layer Group Secret-Key (GSK) generation techniques allow multiple nodes of a wireless network to synthesize a common secret-key, which can be subsequently used to keep their group messages confidential. As one of its salient features, the wireless nodes involved in physical-layer GSK generation extract randomness from a subset of their wireless channels, referred as the common source of randomness (CSR). Unlike two-user key generation, in GSK generation, some nodes must act as facilitators by broadcasting quantized versions of the linear combinations of the channel realizations, so as to assist all the nodes to observe a CSR. However, we note that broadcasting linear combination of channel realizations incurs non-zero leakage of the CSR to an eavesdropper, and moreover, quantizing the linear combination also reduces the overall key-rate. Identifying these issues, we propose a practical GSK generation protocol, referred to as Algebraic Symmetrically Quantized GSK (A-SQGSK) protocol, in a network of three nodes, wherein due to quantization of symbols at the facilitator, the other two nodes also quantize their channel realizations, and use them appropriately over algebraic rings to generate the keys. First, we prove that the A-SQGSK protocol incurs zero leakage to an eavesdropper. Subsequently, on the CSR provided by the A-SQGSK protocol, we propose a consensus algorithm among the three nodes, called the Entropy-Maximization Error-Minimization (EM-EM) algorithm, which maximizes the entropy of the secret-key subject to an upper-bound on the mismatch-rate. We use extensive analysis and simulation results to lay out guidelines to jointly choose the parameters of the A-SQGSK protocol and the EM-EM algorithm.

With potential applications to short-packet communication, we address communication of low-latency messages in fast-fading channels under the presence of a reactive jammer. Unlike a traditional jammer, we assume a full-duplex (FD) jammer capable of detecting pre-existing countermeasures and subsequently changing the target frequency band. To facilitate reliable communication amidst a strong adversary, we propose non-coherent fast-forward full-duplex relaying scheme wherein the victim uses a helper in its vicinity to fast-forward its messages to the base station, in addition to ensuring that the countermeasures are undetected by the FD adversary. Towards designing the constellations for the proposed scheme, we identify that existing non-coherent constellation for fast-fading channels are not applicable owing to the cooperative nature of the fast-forward scheme. As a result, we formulate an optimization problem of designing the non-coherent constellations at the victim and the helper such that the symbol-error-probability at the base station is minimized. We theoretically analyze the optimization problem and propose several strategies to compute near-optimal constellations based on the helper’s data-rate and fast-forwarding abilities. We show that the proposed constellations provide near-optimal error performance and help the victim evade jamming. Finally, we also prove the scheme’s efficacy in deceiving the countermeasure detectors at the jammer.