Interest in unmanned aerial system (UAS) powered solutions for 6G communication networks has grown immensely with the widespread availability of machine learning based autonomy modules and embedded graphical processing units (GPUs). While these technologies have revolutionized the possibilities of UAS solutions, designing an operable, robust autonomy framework for UAS remains a multi-faceted and difficult problem. In this work, we present our novel, modular framework for UAS autonomy, entitled MR-iFLY, and discuss how it may be extended to enable 6G swarm solutions. We begin by detailing the challenges associated with machine learning based UAS autonomy on resource constrained devices. Next, we describe in depth, how MR-iFLY’s novel depth estimation and collision avoidance technology meets these challenges. Lastly, we describe the various evaluation criteria we have used to measure performance, show how our optimized machine vision components provide up to 15X speedup over baseline models and present a flight demonstration video of MR-iFLY’s vision-based collision avoidance technology. We argue that these empirical results substantiate MR-iFLY as a candidate for use in reducing communication overhead between nodes in 6G communication swarms by providing standalone collision avoidance and navigation capabilities.

A novel cross-domain attentional multi-task architecture - xDom - for robust real-world wireless radio frequency (RF) fingerprinting is presented in this work. To the best of our knowledge, this is the first time such comprehensive attention mechanism is applied to solve RF fingerprinting problem. In this paper, we resort to real-world IoT WiFi and Bluetooth (BT) emissions (instead of synthetic waveform generation) in a rich multipath and unavoidable interference environment in an indoor experimental testbed. We show the impact of the time-frame of capture by including waveforms collected over a span of months and demonstrate the same time-frame and multiple time-frame fingerprinting evaluations. The effectiveness of resorting to a multi-task architecture is also experimentally proven by conducting single-task and multi-task model analyses. Finally, we demonstrate the significant gain in performance achieved with the proposed xDom architecture by benchmarking against a well-known state-of-the-art model for fingerprinting. Specifically, we report performance improvements by up to 59.3% and 4.91x under single-task WiFi and BT fingerprinting respectively, and up to 50.5% increase in fingerprinting accuracy under the multi-task setting.

In this work we introduce Constrained Offline Policy Optimization (COPO), an offline policy optimization algorithm for learning in MDPs with cost constraints. COPO is built upon a novel offline cost-projection method, which we formally derive and analyze. Our method improves upon the state-of-the-art in offline constrained policy optimization by explicitly accounting for distributional shift and by offering non-asymptotic confidence bounds on the cost of a policy. These formal properties are superior to those of existing techniques, which only guarantee convergence to a point estimate. We formally analyze our method and empirically demonstrate that it achieves state-of-the-art performance on discrete and continuous control problems, while offering the aforementioned improved, stronger, and more robust theoretical guarantees.

Wireless signal recognition is becoming increasingly more significant for spectrum monitoring, spectrum management, and secure communications. Consequently, it will become a key enabler with the emerging fifth-generation (5G) and beyond 5G communications, Internet of Things networks, among others. State-of-the-art studies in wireless signal recognition have only focused on a single task which in many cases is insufficient information for a system to act on. In this work, for the first time in the wireless communication domain, we exploit the potential of deep neural networks in conjunction with multi-task learning (MTL) framework to simultaneously learn modulation and signal classification tasks. The proposed MTL architecture benefits from the mutual relation between the two tasks in improving the classification accuracy as well as the learning efficiency with a lightweight neural network model. Additionally, we consider the problem of heterogeneous wireless signals such as radar and communication signals in the electromagnetic spectrum. Accordingly, we have shown how the proposed MTL model outperforms several state-of-the-art single-task learning classifiers while maintaining a lighter architecture and performing two signal characterization tasks simultaneously. Finally, we also release the only known open heterogeneous wireless signals dataset that comprises of radar and communication signals with multiple labels.

Multiple-input multiple-output (MIMO) systems find immense potential and applicability in the long term evolution (LTE), 5G, Internet of Things (IoT), vehicular ad hoc networks (VANETs), and tactical communication systems. Jamming poses significant communication hindrance as well as security risks to the wireless communication systems. The achievable rate and reliability are the two most compromised aspects of a wireless link under such severe jamming. Owing to the high capacity and reliability of MIMO systems, they are increasingly used in tactical and critical applications. Therefore, it becomes essential to assess and enhance their sustenance under hostile jamming scenarios. To this end, we address the rate and reliability requirements of a MIMO OFDM system and propose a novel rate-reliability beamformer transceiver design for highly reliable and spectrally efficient operation under the most detrimental jamming attacks. We consider the disguised all band and multiband jamming where the jammer continuously attempts to mimic the legit transmissions. Additionally, we evaluate the rate and reliability performance under barrage jamming.The significant contributions of the proposed rate-reliability beamformer scheme are: (i) achieves a minimum of 2 orders of magnitude better reliability in contrast to the state-of-the-art, (ii) outperforms the state-of-the-art scheme by 1.4× with regards to achievable spectral efficiency, (iii) a very low complexity (O (|Q|)) decoder is presented, and (iv) first work to evaluate the performance of state-of-the-art transmit diversity scheme under hostile jamming attacks.

In this paper, we present the preliminary work towards providing a complete end-to-end solution that can connect survivors of a disaster with each other and public safety authorities using a completely self-sufficient ad hoc network. Accordingly, we develop a Heterogeneous Efficient Low Power Radio (HELPER) that acts as a WiFi (Wireless Fidelity) access point for end-users to connect using custom application. These HELPERs then coordinate with each other to form a LoRa based ad hoc network. The proposed solution will use a distributed optimized cross-layer routing algorithm that aims to maximize the network lifetime. This aspect is critical especially in energy-limited scenarios after a disaster. Some of the envisioned services include text and voice messages, live map updates, ability to send distress messages (like 911 calls) to authorities. HELPER network can also be used by authorities to remotely monitor the connectivity of the affected area, alert users of imminent dangers and share resource information. We intend to provide resources (code and instructions) that will enable the researchers of the community to set up a HELPER in a cost-effective (<; $ 150) manner using commercial off-the-self components and advance it further. Overall, we hope this technology will become instrumental in improving the efficiency and effectiveness of public safety activities.

A low cost, portable, robust jammer detection, and localization device is proposed and developed in this work. Intentional or unintentional use of jammers is detrimental to the seamless operation of emergency rescue and public safety missions as it disrupts the critical communication devices. The proposed device employs robust parallel detection algorithms based on Kurtosis and FRactional Fourier Transform (FRFT) to detect the most common types of Radio Frequency Interference (RFI) that affects critical communication signals. As part of preliminary performance analysis, the proposed detection technique is compared to the conventional energy detectors (employed in many commercial interference detectors) and shown to achieve significant improvement (~ 40dB) in probability of detection. The developed device is envisioned to revolutionize the low cost, portable spectrum interference monitoring sector.

The upcoming 5G (5th Generation) networks demand high-speed and high spectral-efficiency communications to keep up with the proliferating traffic demands. To this end, Massive multiple-input multiple-output (MIMO) techniques have gained significant traction owing to its ability to achieve these without increasing bandwidth or density of base stations. The preexisting space-time block code (STBC) designs cannot achieve a rate of more than 1 for more than two transmit antennas while preserving the orthogonality and full diversity conditions. In this paper, we present Jagannath codes - a novel complex modulation STBC, that achieves a very high rate of 2 for three and four transmit antennas. The presented designs achieve full diversity and overcome the previously achieved rates with the three and four antenna MIMO systems. We present a detailed account of the code construction of the proposed designs, orthogonality and full diversity analysis, transceiver model and conditional maximum likelihood (ML) decoding. In an effort to showcase the improvement achieved with the presented designs, we compare the rates and delays of some of the known STBCs with the proposed designs. The effective spectral efficiency and coding gain of the presented designs are compared to the Asymmetric Coordinate Interleaved design (ACIOD) and Jafarkhani code. We presented an effective spectral efficiency improvement by a factor of 2 with the proposed Jagannath codes. Owing to the full diversity of the presented designs, we demonstrate significant coding gains (6 dB and 12 dB) with the proposed designs.

Visible Light Ad Hoc Networks (LANETs)is being perceived as an emerging technology to complement Radio Frequency (RF)based ad hoc networks to reduce congestion in the overloaded RF spectrum. LANET is intended to support scenarios requiring dense deployment and high data rates. In Visible Light Communication (VLC), most of the attention has been centered around physical layer with emphasis on point-to-point communication. In this work, we focus on designing a routing protocol specifically to overcome the unique challenges like blockage and deafness that render routes in LANETs highly unstable. Therefore, we propose a cross-layer optimized routing protocol (VL-ROUTE)that interacts closely with the Medium Access Control (MAC)layer to maximize the throughput of the network by taking into account the reliability of routes. To accomplish this in a distributed manner, we carefully formulate a Route Reliability Score (RRS)that can be computed by each node in the network using just the information gathered from its immediate neighbors. Each node computes an RRS for every known sink in the network. RRS of a given node can be considered as an estimate of the probability of reaching a given sink via that node. The RSS value is then integrated to the utility based three-way handshake process used by the MAC protocol (VL-MAC)to mitigate the effects of deafness, blockage, hidden node, and maximize the probability of establishing full-duplex links. All these factors contribute towards maximizing the network throughput. Extensive simulation of VL-ROUTE shows 124% improvement in network throughput over a network that uses Carrier Sense Multiple Access/Collision Avoidance (CSMA/CA)along with shortest path routing. Additionally, VL-ROUTE also showed up to 21% improvement in throughput over the network that uses VL-MAC along with a geographic routing.

The design and implementation of a novel distributed deadline-based routing and spectrum allocation algorithm for tactical ad-hoc networks is reported in this article. Different traffic classes including text, voice, surveillance video, and threat alert among others need to be handled by these networks. Each of these traffic classes have different quality of service (QoS) based deadline requirements. Additionally, these networks are characterized by dynamic channel and traffic conditions that vary with time and location. Even under these conditions, it is critical to receive packets before the deadline expires to make rapid decisions in the battlefield. Therefore, a tactical ad-hoc network should be able to adapt to these requirements and maximize the number of packets delivered to the destination within the specified deadline. A distributed deadline-based routing and spectrum allocation algorithm is designed to maximize the utilization of the available resources and ensure delivery of packets within the deadline constraints. To this end, a weighted virtual queue (VQ) that is used to construct the network utility function is defined. Accordingly, the optimal session, next hop, transmit power, and frequency is determined by the distributed algorithm to ensure efficient utilization of the available resources. Hence, maximizing the delivery of packets to the intended destination within the specified deadline. The 49 node simulation shows up to 35 percent improvement in effective throughput and 26 percent improvement in reliability as compared to joint ROuting and Spectrum Allocation algorithm (ROSA), which does not adapt according to the deadline requirements of the data flowing through the network. As a secondary objective, this work advances the state of the art of the experimental cross-layer framework to address the challenges involved in having such cross-layer algorithms implemented on a testbed. The required flexibility to change the transmission parameters on-the-fly is provided by the proposed framework. The network is designed to enable the data exchange between neighbors using custom designed control packets (which might be different for different algorithms) since this information is critical for nodes to perform optimization. Cross-layer optimization is achieved by means of data management and control entities that enable information exchange between layers. The practicality of the proposed solution was proven by having the novel algorithm implemented on a five-node software defined radio testbed which leverages the proposed cross-layer framework. In contrast to ROSA, the proposed algorithm demonstrated up to 17 percent improvement in terms of throughput and reliability. The performance improvement achieved is expected to increase on a larger network deployment.

In this paper, we design and evaluate a practical AMC system that can be readily deployed to provide robust performance in various real-time commercial scenarios. Thus, our main goal is to develop a robust AMC algorithm with low computational complexity for easy implementation and practical deployment. To this end, we utilize recently revitalized machine learning based approaches used for various classification purposes. In our proposed AMC architecture, we first propose various statistics that serve as features of the AMC signals; next, we design an artificial neural network (ANN) based classifier that performs AMC over a wide range of SNRs. We employ Nesterov accelerated adaptive moment (NADAM) estimation technique to improve the classification performance of our ANN. Further, to establish the practical feasibility of our proposed architecture, we implement it on a SDR testbed. The proposed ANN-based classifier is shown to outperforms the hybrid hierarchical AMC (HH-AMC) system and is flexible enough to easily expand the dictionary of modulation formats for other applications.

In this paper, we propose a novel distributed deadline-based routing and spectrum allocation algorithm for tactical ad-hoc networks. The proposed algorithm will enable nodes to adapt to various deadline requirements unique to each traffic classes. A tactical ad-hoc network needs to handle a variety of data flowing through the network including voice, surveillance video, threat alert among others. Each of these traffic classes may have different quality of service (QoS) based deadline requirements. It is critical to receive these packets before the deadline expires to make crucial decisions in the battlefield. Therefore, the network should be able to adapt to these requirements and maximize the effective throughput. Accordingly, a distributed deadline-based routing and spectrum allocation algorithm is designed to maximize the utilization of the available resources to ensure delivery of packets within the deadline constraints. The simulations show up to 35 % improvement in effective throughput and 26 % improvement in reliability as compared to the routing and spectrum allocation algorithm (ROSA).

The objective of this work is to design and implement a novel framework for automatic signal classification techniques (FACT) for software defined radios (SDR) capable of classifying multiple signals simultaneously. The focus of this work is to create a modular classification framework to facilitate the testing and implementation of new classification methods. The framework is divided into three parts: (i) Sensor resource manager (SRM), which performs the initial signal detection, preprocessing and the delegation of secondary receivers to the corresponding signals of interest (SOIs); (ii) Modulation classifier block (MCB), which takes the received signal from SRM and performs the required modulation classification and (iv) Data library and statistical block contains all the templates required to perform classification, thresholds for signal detection and also known parameters of expected signals. To prove the feasibility of the framework, FACT is implemented and tested on a Universal Software Radio Peripheral (USRP) test bed using GNU radio signal processing toolkit. We evaluate the performance of signal detection based on the probability of detection (Pd) in varying signal-to-noise ratios (SNR). Additionally, the USRP based experiments demonstrate FACT operating as a single unit, preforming both blind detection and classification of multiple SOIs using different classification methods.