This demonstration presents a software-defined radio (SDR) testbed for experimentally evaluating air-to-ground (A2G) communication under dynamic flight conditions. By synchronizing UAV telemetry with orthogonal frequency-division multiplexing (OFDM) measurements, the system enables precise frame-to-telemetry alignment and frame-level bit error rate (BER) analysis. Results show that yaw angle is the dominant flight parameter affecting link reliability. Building on this finding, we demonstrate a yaw-aware adaptive modulation and coding (AMC) scheme that maps angular conditions to the optimum modulation selection. The proposed framework provides a reproducible and practical platform for flight-parameter–aware link adaptation, advancing the robustness and efficiency of future UAV communication systems.

Automated warehouses involve robots that move across the floor, avoiding obstacles while remaining connected via an access point (AP) to a central controller that instructs the robots. The complex propagation environment and presence of metallic surfaces results in spotty coverage, which changes over time as the location of stored products and machinery changes. Thus, maintaining an assured connectivity to APs while performing navigation is a challenge, although it is needed to relay local sensor data from the robots to the controller and receive directions from the latter.$\rm{DARWIN}$, involves creating a digital twin of the warehouse for training the robots by jointly optimizing the navigation and avoiding wireless dead-spots.$\rm{DARWIN}$has three key capabilities: First, it captures the features of both physical and RF environments in the digital world. Second, it allows real-time updating of the digital twin if significant disparity is detected compared to the physical environment. Finally, it includes a reinforcement learning algorithm that jointly optimizes navigation and network resource management, while accounting for handover and outage. We validate$\rm{DARWIN}$on an emulation environment consisting of Robot Operating System and Gazebo platforms along with real-world RF measurements. Results reveal that$\rm{DARWIN}$reduces the number of steps by 43% compared to choosing the closest AP, while detecting environmental changes with maximum 96% accuracy to maintain a high-fidelity digital twin.

The capability to detect radar signals autonomously, without reliance on radar transmitters, is pivotal for the advancement of contemporary shared-spectrum wireless networks like the Citizens Broadband Radio Service (CBRS). Recent trends underscore the integration of AI-driven methodologies to address this challenge effectively. In this paper, we present a novel supervised deep learning framework for radar detection, denoted as Spec-SCAN. We design Spec-SCAN to efficiently identify low-power radar signals amidst interference within a condensed timeframe and over a narrower frequency spectrum compared to existing benchmarks. Our approach employs a YOLO-based training strategy tailored for the detection of radar signals and prevalent interference patterns within the CBRS band. We perform rigorous experiments encompassing scenarios involving LTE, 5G, and DSSS signals as interfering signals to evaluate Spec-SCAN. Our findings indicate that Spec-SCAN attains a radar detection recall of 99% for type 1 radar signals, even at Signal-to-Interference-Noise Ratios (SINR) as low as 15 dB, while scanning a 100MHz spectrum within a 15ms timeframe-demonstrating superior performance compared to alternative methodologies. Spec-SCAN framework also offers comparable performance while scanning 50MHz spectrum for 15 milliseconds.

In today's rapidly evolving technological landscape, accurate object characterization is crucial for a wide range of applications, from autonomous systems to smart environments and security. Simultaneously, the growing concern for privacy necessitates innovative approaches that can characterize objects without compromising sensitive visual information. In this paper, we introduce a novel approach for object characterization using Radio Frequency (RF) propagation map generation through ray-tracing within a Digital Twin (DT) framework. We outline a systematic pipeline for leveraging NVIDIA's Sionna RayTracing tool to generate DT propagation maps created in Blender for indoor environments. Using these propagation maps, we propose a machine learning-based approach to facilitate object characterization. Our results demonstrate the feasibility of object characterization through strategic scene configuration using a small dataset that leverages RF maps within DTs. This paper provides valuable insights into the potential of our framework as a reliable and more efficient method for object characterization, offering a promising alternative to traditional vision-based techniques in scenarios where privacy concerns or environmental constraints limit the use of conventional imaging methods.

The concept of digital twin (DT) has recently gained momentum in the context of wireless networks. In this paper, we propose a tiny blockage twin (TBT) for generating blockage map/data for a millimeter wireless network environment using geometric information of the buildings and streets. The generated data will enable the DT to understand coverage map of particular area/location on real-time basis and therefore will serve the service provider with important information such as blocked location, base station (BS) selection etc., to enhance users' quality of service (QoS). Due to severe blockage suffered by propagation in millimeter-wave (mmWave) range, modeling and estimation of mmWave blockage remain at the heart of the challenges, especially in the case of DT, where the DT has to make realtime rather proactive decisions. The proposed framework finds blocked/unblocked location by considering buildings dimensions, locations, as well as their orientation with respect to BS. Also, the modularized framework by its inherent expansion capability is efficient in dealing with changes in infrastructure and obtaining the resultant blockage data. To evaluate the accuracy of the data, we compare with blockage data obtained using NVIDIA Sionna simulator. The results indicate that the proposed model generates blockage data with higher accuracy in less time and with minimum requirement of resources, thus making it a lightweight twin.

This study presents a comparative analysis of various types of communication latency in networked robotic setup. We consider three distinct communication link: Robot-to-Robot (R2R), Network-to-Robot (N2R), and Robot-to-Network (R2N) for enabling communication within a swarm of robots. To conduct our study we design an experimental testbed in a laboratory setup featuring a grid layout with strategically placed obstacles emulating a warehouse scenario. We use two different type of state-of-the-art experimental robots, transmitting multimedia data using the frequency bands 2.4 GHz and 5 GHz. Our experimental results indicate that the R2R communication using 5 GHz frequency achieves the lowest average latency of 0.4 milliseconds, compared to the other types of communication.

Digital twins (DT) offer a low-overhead evaluation platform and the ability to generate rich datasets for training machine learning (ML) models before actual deployment. Specifically, for the scenario of ML-aided millimeter wave (mmWave) links between moving vehicles to roadside units, we show how DT can create an accurate replica of the real world for model training and testing. The contributions of this paper are twofold: First, we propose a framework to create a multimodal Digital Twin (DT), where synthetic images and LiDAR data for the deployment location are generated along with RF propagation measurements obtained via ray-tracing. Second, to ensure effective domain adaptation, we leveragemeta-learning, specificallyModel-Agnostic Meta-Learning(MAML), withtransfer learning(TL) serving as a baseline validation approach. The proposed framework is validated using a comprehensive dataset containing both real and synthetic LiDAR and image data for mmWave V2X beam selection. It also enables the investigation of how each sensor modality impacts domain adaptation, taking into account the unique requirements of mmWave beam selection. Experimental results show that models trained on synthetic data using transfer learning and meta-learning, followed by minimal fine-tuning with real-world data, achieve up to 4.09× and 14.04× improvements in accuracy, respectively. These findings highlight the potential of synthetic data and meta-learning to bridge the domain gap and adapt rapidly to real-world beamforming challenges.

In recent years, there has been a notable surge in interest focused on refining autonomous systems, particularly in advancing robot navigation through the strategic utilization of sensing data. However, this heightened attention has also raised significant privacy concerns. To effectively address these challenges while capitalizing on the benefits of sensing-based robot navigation, this paper introduces a novel concept of Radio Frequency (RF) map creation derived from ray-tracing within a digital twin of an unstructured environment. We present a systematic pipeline for utilizing NVIDIA’s Sionna Ray-Tracing tool to generate propagation models of the digital twin generated in Blender by taking inputs from the real world. Through the integration of the RF map, we propose a reinforcement learning based approach to facilitate robot navigation. This integration process enables the formulation of RF propagation models tailored for mobile robots operating within indoor environments. The validation process shows the feasibility of the proposed algorithm in an indoor lab setup with the robot navigating through the various obstacles avoiding any collision. Our work represents a significant advancement towards the practical implementation of robot navigation by harnessing RF propagation data generated through ray-tracing. Through our proposed framework, we contribute to the development of a robust and privacy-preserving approach for robot navigation in autonomous environments.

This paper presents COPILOT, a ML-based approach that allows vehicles requiring ubiquitous high bandwidth connectivity to identify the most suitable road side units (RSUs) through proactive handoffs. By cooperatively exchanging the data obtained from local 3D Lidar point clouds within adjacent vehicles and with coarse knowledge of their relative positions, COPILOT identifies transient blockages to all candidate RSUs along the path under study. Such cooperative perception is critical for choosing RSUs with highly directional links required for mmWave bands, which majorly degrade in the absence of LOS. COPILOT proposes three modules that operate in an inter-connected manner: (i) As an alternative to sending raw Lidar point clouds, it extracts and transmits low-dimensional intermediate features to lower the overhead of inter-vehicle messaging; (ii) It utilizes an attention-mechanism to place greater emphasis on data collected from specific vehicles, as opposed to nearest neighbor and distance-based selection schemes, and (iii) it experimentally validates the outcomes using an outdoor testbed composed of an autonomous car and Talon AD7200 60GHz routers emulating the RSUs, accompanied by the public release of the datasets. Results reveal COPILOT yields upto 69.8% and 20.42% improvement in latency and throughput compared to traditional reactive handoffs for mmWave networks, respectively.

Robotic factory floors will revolutionize the future of manufacturing and the service industry by automating tasks. However, to fully supplement human effort, these robots will need low-latency, reliable connectivity throughout the work zone through links established by wireless access points (APs). This will allow the robot to assuredly respond to programming directives that rely on the real-time relaying of robot-generated sensor data to the Mobile Edge Computing (MEC) server. In this paper, we propose L-NORM, a multi-AP and multi-robot coordination framework, as a multi-tiered solution for such autonomous edge networks. First, multi-robot motion planning through reinforcement learning occurs at the MEC, using as input multi-modal robot sensor data. Second, multi-AP resource orchestration is performed using another reinforcement learning-based method that maps a subset of available APs to each robot toward meeting their sensor data delivery requirements. Furthermore, we suggest diversity combination of uplink channels with the 802.11ax scheduled access mode that will (i) support high reliability of multi-robot uplink sensor packets and (ii) enable multi-AP coordination, for optimized resource utilization. Through extensive simulation studies, we show that the probability of robot deviation to remain within 0.5 m from its optimal path, is 19% more in L-NORM compared to classical 802.11ax based edge network solution, considering$\sim$1 MB of sensor data per robot.

Fast sector-steering in the mmWave band for vehicular mobility scenarios is a challenge because standard-defined exhaustive search over predefined antenna sectors cannot be assuredly completed within short contact times. This paper proposes machine learning to speed up sector selection using data from multiple non-RF sensors, such as LiDAR, GPS, and camera images in the mmWave radios with large codebooks. The contributions in this paper are threefold: First, we propose a multimodal deep learning architecture that fuses the inputs from these data sources and locally predicts the sectors for best alignment at a vehicle. Second, we propose FLASH-and-Prune, which combines the knowledge from multiple vehicles by aggregating the local model parameters and exploits model pruning to optimize the model parameter exchange overhead. Third, we present a pruning strategy that takes into account the distributed nature of federated learning to adaptively prune or retrieve model weights. We validate the proposed architecture on a real-world multimodal dataset collected from an autonomous car. We observe that FLASH-and-Prune incurs 29.25% and 35.89% less overhead in the uplink and downlink, respectively, compared to standard federated learning.

Creating a digital world that closely mimics the real world with its many complex interactions and outcomes is possible today through advanced emulation software and ubiquitous computing power. Such a software-based emulation of an entity that exists in the real world is called a ‘digital twin’. In this paper, we consider a twin of a wireless millimeter-wave band radio that is mounted on a vehicle and show how it speeds up directional beam selection in mobile environments. To achieve this, we go beyond instantiating a single twin and propose the ‘$\MV$’ paradigm, with several possible digital twins attempting to capture the real world at different levels of fidelity. Towards this goal, this paper describes (i) a decision strategy at the vehicle that determines which twin must be used given the latency limitation, and (ii) a self-learning scheme that uses the$\MV$-guided beam outcomes to enhance DL-based decision-making in the real world over time. Our work is distinguished from prior works as follows: First, we use a publicly available RF dataset collected from an autonomous car for creating different twins. Second, we present a framework with continuous interaction between the real world and$\MV$of twins at the edge, as opposed to a one-time emulation that is completed prior to actual deployment. Results reveal that$\MV$offers up to$79.43\%$and$85.22\%$top-$10$beam selection accuracy for LOS and NLOS scenarios, respectively. Moreover, we observe$67.70-90.79\%$improvement in beam selection time compared to 802.11ad standard and 5G-NR standards.

The use of alternate modalities, like images, for fast beamforming in the millimeter wave (mmWave)-band is being proposed to ensure high bandwidth connectivity in vehicular scenarios typically seen in the context of autonomous cars. Considering the dynamic deployment conditions, a car may encounter new environments which were not explicitly included in an apriori training dataset. In this paper, we propose to use the Model-Agnostic Meta-Learning (MAML) framework on the image data of the mmWave vehicle-to-infrastructure beam selection FLASH dataset, to overcome the generalization issues of a pre-trained model in unseen non-line-of-sight (NLOS) connectivity environments. MAML has additional advantages over traditional deep-learning techniques: (i) it uses a fraction of the data which, in turn, simplifies data collection and storage, and (ii) it results in equal or higher accuracy in optimal beam selection compared to the case when the new environment dataset is fully available during initial training. We show that our MAML implementation improves test accuracy of beam selection by up to 86% with fine-tuning when encountering an unseen NLOS environment compared to conventional supervised learning.

The use of non-RF data can potentially speed up millimeter wave-band sector-steering in vehicular mobility scenarios by gaining contextual knowledge of the environment. While several works have demonstrated the benefits of this approach, especially applying machine learning models on inputs from LiDAR and image sensors, adapting such models in ‘unseen’ environments remains an open problem. State-of-the-art techniques generally use a single, pretrained model for all different scenarios, which assumes that the network has ‘seen’ representative examples of all future scenarios. In this paper, we propose the TUNE framework, which solves this problem by: ($a$) transfer learning (TL) for better performance with similar convergence times in comparison to non-TL-generated model testing, (b) utilizing statistical properties to select the best-suited starting ‘seen’ scenario (and by extension the model trained for it), and (c) a refinement of the transfer learning framework by dynamically selecting the most pertinent layers for retaining, thus reducing the overhead compared to fully retraining a model. We validate TUNE on publicly available synthetic and real-world datasets for mmWave beam selection for V2X communication, revealing that TUNE generally outperforms non-TL methods in a variety of tasks where a different number of beams is available between the training and testing environments.

The RF environment in a secure space can be compromised by intentional transmissions of hard-to-detect underlay signals that overlap with a high-power baseline transmission. Specifically, we consider the case where a direct sequence spread spectrum (DSSS) signal is the underlay signal hiding within a baseline 4G Long-Term Evolution (LTE) signal. As compared to overt actions like jamming, the DSSS signal allows the LTE signal to be decodable, which makes it hard to detect. ICARUS presents a machine learning based framework that offers choices at the physical layer for inference with inputs of (i) in-phase and quadrature (IQ) samples only, (ii) cycle-frequency features obtained via cyclostationary signal processing (CSP), and (iii) fusion of both, to detect the underlay DSSS signal and its modulation type within LTE frames. ICARUS chooses the best inference method considering both the expected accuracy and the computational overhead. ICARUS is rigorously validated on multiple real-world datasets that include signals captured in cellular bands in the wild and the NSF POWDER testbed for advanced wireless research (PAWR). Results reveal that ICARUS can detect DSSS anomalies and its modulation scheme with 98-100% and 67 − 99% accuracy, respectively, while completing inference within 3 − 40 milliseconds on an NVIDIA A100 GPU platform.

This paper proposes and experimentally validates a new paradigm for computing with wireless signals over-the-air (OTA). It demonstrates the first fully connected (FC) neural network (NN) constructed entirely using channel propagation and signal interference principles. Our design is based on architecting the desired linear operation of an FC layer through the superposition of signals emitted from multiple transmitters and received at a single receiver, similar to multiple input single output (MISO) systems. Our design takes into account several practical considerations, such as the impact of multiple subcarriers, the number of transmit antennas, and the changing wireless channel. The key outcome of our work is developing a principled methodology that transforms a given trained digital FC NN into its OTA equivalent. This novel computational paradigm, which we call AirFC, allows us to run NN tasks without compute-specific hardware during tests. We validate our design using 9 time-synchronized software-defined radios (SDRs) available on the ORBIT testbed, emulating a 16 antenna array. We use the MNIST dataset as input to our wireless FC NN and demonstrate classification with 92.61% accuracy, which proves that our NN with OTA FC layers performs similar to the conventional, all-digital version with an accuracy decrease of only 0.73%.

With the recent surge in autonomous driving vehicles, the need for accurate vehicle detection and tracking is critical now more than ever. Detecting vehicles from visual sensors fails in non-line-of-sight (NLOS) settings. This can be compensated by the inclusion of other modalities in a multi-domain sensing environment. We propose several deep learning based frameworks for fusing different modalities (image, radar, acoustic, seismic) through the exploitation of complementary latent embeddings, incorporating multiple state-of-the-art fusion strategies. Our proposed fusion frameworks considerably outperform unimodal detection. Moreover, fusion between image and non-image modalities improves vehicle tracking and detection under NLOS conditions. We validate our models on the real-world multimodal ESCAPE dataset, showing 33.16% improvement in vehicle detection by fusion (over visual inference alone) over test scenarios with 30-42% NLOS conditions. To demonstrate how well our framework generalizes, we also validate our models on the multimodal NuScene dataset, showing$\sim$22% improvement over competing methods.

In IEEE 802.11 WiFi-based waveforms, the receiver performs coarse time and frequency synchronization using the first field of the preamble known as the legacy short training field (L-STF). The L-STF occupies upto 40% of the preamble length and takes upto$32 \mu \text{s}$of airtime. With the goal of reducing communication overhead, we propose a modified waveform, where the preamble length is reduced by eliminating the L-STF. To decode this modified waveform, we propose a neural network (NN)-based scheme called PRONTO that performs coarse time and frequency estimations using other preamble fields, specifically the legacy long training field (L-LTF). Our contributions are threefold: (i) We present PRONTO featuring customized convolutional neural networks (CNNs) for packet detection and coarse carrier frequency offset (CFO) estimation, along with data augmentation steps for robust training. (ii) We propose a generalized decision flow that makes PRONTO compatible with legacy waveforms that include the standard L-STF. (iii) We validate the outcomes on an over-the-air WiFi dataset from a testbed of software defined radios (SDRs). Our evaluations show that PRONTO can perform packet detection with 100% accuracy, and coarse CFO estimation with errors as small as 3%. We demonstrate that PRONTO provides upto 40% preamble length reduction with no bit error rate (BER) degradation. We further show that PRONTO is able to achieve the same performance in new environments without the need to re-train the CNNs. Finally, we experimentally show the speedup achieved by PRONTO through GPU parallelization over the corresponding CPU-only implementations.

We are being increasingly surrounded by electronic devices that need reliable authentication before they can be accessed. While typed passwords, face and fingerprint-based authentication are popular today, we explore the possibility of using additional bio-origin signals, especially electrocardiogram (ECG) signals that can be collected without an active engagement and attention of the human user. Our proposed approach, abbreviated as FERST, uses a two-stage deep learning framework for signal processing and classification. The acquired and pre-equalized ECG signal is transferred from a wearable device through arm-wrist-palm galvanic coupled channel. The first part of the reception framework consists of a denoising autoencoder (DAE) that filters this noisy ECG signal. Then, the second stage includes a convolutional neural network (CNN) to classify and authenticate the denoised ECG signal as present/absent in a prior database used for training. The overall design goal of our approach is to achieve acceptable authentication performance with simple hardware, which further decreases network complexity and enables miniaturization of the end-to-end system. Results reveal that the FERST provides (i) 82% improvement in inference time relative to state-of-the-art adaptive filtration methods, (ii) outperforms the state-of-the-art with 99.7% classification accuracy on a standardized ECG library dataset and (iii) a relatively high classification accuracy of 99.2% for different arm-wrist-palm channel dimensions.

Opening the Citizen Broadband Radio Service (CBRS) band in the US to secondary users offers unprecedented opportunities to LTE and 5G networks, as long as incumbent radar signals are protected from interference. Towards this aim, the US Federal Communications Commission (FCC) requires Environmental Sensing Capabilities (ESCs) to be installed along the coastal regions. Furthermore, FCC mandates that the secondary users transmit with low power levels, such that the aggregated interference and noise power in the vicinity of ESC sensors remains below −109 dBm/MHz. At this interference level, the ESC must detect 99 % of radar pulses with peak power of at least −89 dBm/MHz. In this paper, we design an enhanced ESC sensor, called ESC+, that leverages the deep learning framework called 'you only look once’ (YOLO) for signal detection using spectrograms. We propose a two-stage spectrogram-based coarse and fine signal analysis method for: (i) detecting, and characterizing radar pulses in environments where the aggregated noise and interference level goes beyond FCC restrictions, and (ii) detecting and characterizing other signal types (e.g., 5G and LTE) in the CBRS band, with a goal of determining unauthorized users. We generate a realistic spectrogram dataset in MATLAB consisting of three signal types of radar, 5G, and LTE where the aggregated interference and noise power occurring concurrently with the radar pulse is varied upto −104 dBm/MHz. We show 100% radar pulse detection in interference and noise levels of up to 3 dB higher than what is required today.

Fast sector-steering in the mmWave band for vehicular mobility scenarios remains an open challenge. This is because standard-defined exhaustive search over predefined antenna sectors cannot be assuredly completed within short contact times. This paper proposes machine learning to speed up sector selection using data from multiple non-RF sensors, such as LiDAR, GPS, and camera images. The contributions in this paper are threefold: First, a multimodal deep learning architecture is proposed that fuses the inputs from these data sources and locally predicts the sectors for best alignment at a vehicle. Second, it studies the impact of missing data (e.g., missing LiDAR/images) during inference, which is possible due to unreliable control channels or hardware malfunction. Third, it describes the first-of-its-kind multimodal federated learning framework that combines model weights from multiple vehicles and then disseminates the final fusion architecture back to them, thus incorporating private sharing of information and reducing their individual training times. We validate the proposed architectures on a live dataset collected from an autonomous car equipped with multiple sensors (GPS, LiDAR, and camera) and roof-mounted Talon AD7200 60GHz mmWave radios. We observe 52.75% decrease in sector selection time than 802.11ad standard while maintaining 89.32% throughput with the globally optimal solution.

We propose a deep learning framework for solving the problem of position and orientation estimation (DeePOE) of a radio frequency (RF) transmitter using the in-phase$(I)$and quadrature-phase$(Q)$components of the RF signal data. Our goal is to demonstrate a proof of concept system with an end-to-end implementation in order to overcome the shortcomings of state-of-the-art joint position and orientation estimation systems. The proposed DeePOE framework consists of a convolutional neural network (CNN) which is designed to exploit latent features present within the received raw I/Q signal data. This enables receivers equipped with the DeePOE framework to predict the position and orientation of a transmitter, relative to itself in a predefined coordinate system, solely from physical layer information. DeePOE jointly optimizes the position and orientation estimation objectives using transfer learning, iteratively over the training epochs. In order to validate and refine the DeePOE framework, we perform real-world (indoor and outdoor) experiments using 16 GB of raw I/Q data collected with directional emitters placed in various orientations and at different distances (positions) from both directional and omnidirectional receivers. The framework achieves on average a F1score of 0.922 for the task of predicting 12 orientations from the data collected using an omnidirectional antenna. It also yields F1score of 0.847 for data collected with a directional antenna which involves predicting 48 orientations. DeePOE achieves on average F1score of 0.963 for predicting the transmitter position with respect to the receiver placed at a known location, for all the cases.

Accurate beam alignment in the millimeter-wave (mmWave) band introduces considerable overheads involving brute-force exploration of multiple beam-pair combinations and beam retraining due to mobility. This cost becomes often intractable under high mobility scenarios, where fast beamforming algorithms that can quickly adapt the beam configurations are still under development for 5G and beyond. Besides, blockage prediction is a key capability in order to establish mmWave reliable links. In this paper, we propose a data fusion approach that takes inputs from visual edge devices and localization sensors to (i) reduce the beam selection overhead by narrowing down the search to a small set containing the best possible beam-pairs and (ii) detect blockage conditions between transmitters and receivers. We evaluate our approach through joint simulation of multi-modal data from vision and localization sensors and RF data. Additionally, we show how deep learning based fusion of images and Global Positioning System (GPS) data can play a key role in configuring vehicle-to-infrastructure (V2I) mmWave links. We show a 90% top-10 beam selection accuracy and a 92.86% blockage prediction accuracy. Furthermore, the proposed approach achieves a 99.7% reduction on the beam selection time while keeping a 94.86% of the maximum achievable throughput.

Primary user emulation (PUE) attacks can pose a significant threat to the deployment of a robust cognitive radio network implementing dynamic spectrum access, for an intelligent allocation and usage of already crowded spectrum bands. In this paper, we present a solution towards the PUE attacks. We present two generative adversarial net (GAN) based models to successfully emulate the primary users (PUs) in two ways. We propose a (i) dumb generator model without any "prior" knowledge of PU's feature space, (ii) a smart generator model with some "prior" knowledge about PU's transmission. We also propose two deep neural network based discriminator models to discriminate between the PU and the emulated primary users (EPU) from the corresponding generators. Both the generator and discriminator of each GAN model gets smarter with iterative and sequential GAN training. Through a testbed evaluation, we show that discriminators are able to catch ~50% of PUE attackers without the GAN training during the deployment phase. We also observe 100% accuracy for both the GAN models during training phase. Ultimately, after the GAN training, the discriminators achieved 98% and 99.5% accuracies, for dumb and smart generator models respectively, to distinguish "yet to be seen" PUE attacker.

The recent advances of wireless technologies in RF environments coupled with large scale usage of such technologies has warranted more autonomous deployments of wireless systems. Machine learning techniques, that include recurrent structures, have shown promise in creating such autonomous deployments using the idea of Radio Frequency Machine Learning (RFML). In large scale autonomous deployments of wireless communication networks, the signals received from one component play a crucial role in the decision making process of other components. In order to efficiently implement such systems each component of the network should be uniquely identifiable. In this paper we propose a transmitter fingerprinting technique for radio device identification using recurrent structures, by exploiting the temporal property of the received radio signal. We design and implement three recurrent neural networks (RNNs) using different types of cell models: (i) long short term memory (LSTM); (ii) gated recurrent unit (GRU) and (iii) convolutional long short term memory (ConvLSTM), for this task. We program 8 universal software radio peripheral (USRP) software defined radios (SDRs) as transmitters and collect over-the-air raw in-phase (I) and quadrature (Q) (I/Q) time series data from them using a DVB-T RTL-SDR receiver, in a laboratory setting. We exploit both the temporal variations as well as the inherent spatial dependencies in the collected I/Q time series data, to learn unique feature representations and use these as "fingerprints'" for identifying the transmitters. Experimental results reveal that the RNNs with LSTM, GRU, and ConvLSTM cells are able to correctly distinguish between the 8 transmitters with 92%, 95.3%, 97.2% accuracy respectively.

Understanding and analyzing the radio frequency (RF) environment have become indispensable for various autonomous wireless deployments. To this end, machine learning techniques have become popular as they can learn, analyze and even predict the RF signals and associated parameters that characterize a RF environment. However, classical machine learning methods have their limitations and there are situations where such methods become ineffective. One such setting is where active adversaries are present and try to disrupt the RF environment through malicious activities like jamming or spoofing. In this paper we propose an adversarial learning technique for identifying rogue RF transmitters and classifying trusted ones by designing and implementing generative adversarial nets (GAN). The GAN exploits the in-phase (I) and quadrature imbalance (i.e., the IQ imbalance) present in all transmitters to learn the unique high dimensional features that can be used as “fingerprints” for identifying and classifying the transmitters. We implement a generative model that learns the sample space of the IQ values of the known transmitters and use the learned representation to generate fake signals that imitate the transmissions of the known transmitters. We program 8 universal software radio peripheral (USRP) software defined radios as trusted transmitters and collect over-the-air raw IQ data from them using a RTL-SDR in a laboratory setting. We also implement a discriminator model and show that the discriminator is able to discriminate between the trusted transmitters from fake ones with 99.9% accuracy. Finally, the trusted transmitters are classified using convolutional neural network (CNN) and fully connected deep neural networks (DNN). Results reveal that the CNN and DNN are able to correctly discriminate between the 8 trusted transmitters with 81.6% and 96.6% accuracies respectively.

In this paper we study and implement real-time adaptation schemes for video encoding and channel selection that work in tandem to facilitate HD video streaming for secondary users in a dynamic spectrum access network. Out-of-band feedbacks on instantaneous pathloss of the signal between the transmitter and the receiver, the received signal strength indicator (RSSI) at the receiver and the quality of the reconstructed video are used to continuously determine the most apt encoding parameters. At the same time, the radio transmitter continuously adjusts the channel parameters (i.e., center frequency and channel bandwidth) based on the transmission activities of the primary users who have prioritized rights on these channels. We consider the physical limitations of the encoder along with the channel statistics to determine when to change the encoder parameters and when to switch to a new channel. We propose a multi-level threshold based mechanism to find the optimal number of encoding bit rates. We also propose a threshold based algorithm to find the best available channel between the transmitter-receiver pair. We validate our theoretical propositions on an indoor testbed using software defined radios (SDRs) and the GNU Radio suite. Live video was captured, encoded using open source H.264 software libraries, streamed using GStreamer and transmitted over the 915 MHz ISM bands with omnidirectional antennas. For the SDRs, we chose the universal software radio peripheral (USRP) B210s from Ettus Research and use them as the transmitter and the receiver. A third B210 was used to sense the energy levels on all the channels to detect the presence of primary transmissions. GNU Radio was used to build the signal processing pipeline, both for the transmitter and the receiver. We use PSNR and SSIM to measure the video quality and report experimental results that show that: (i) the video encoder and the USRP transmitter-receiver pair are able to adapt to the changing RF conditions, (ii) the adaptation schemes yield better video quality than non-adaptive schemes, and (iii) the USRPs can switch the channels fast enough allowing uninterrupted HD video streaming even when primary users preempt the secondary user's transmission.

The high density of interconnects, closer proximity of modules, and routing phase are pivotal during the layout of a performance-centricthree-dimensional integrated circuit(3D IC). Heuristic-based approaches are typically used to handle such NP-complete problems of global routing in 3D ICs. To overcome the inherent limitations of deterministic approaches, a novel methodology for multi-objective global routing based on fuzzy logic has been proposed in this article. The guiding information generated after the placement phase is used during routing with the help of a fuzzy expert system to achieve thermally efficient and congestion-free routing. A complete global routing solution is designed based on the proposed algorithms and the results are compared with selected fully established global routers, namely Labyrinth, FastRoute3.0, NTHU-R, BoxRouter 2.0, FGR, NTHU-Route2.0, FastRoute4.0, NCTU-GR, MGR, and NCTU-GR2.0. Experiments are performed over ISPD 1998 and 2008 benchmarks. The proposed router, calledFuzzRoute, achieves balanced superiority in terms of routability, runtime, and wirelength over others. The improvements on routing time for Labyrinth, BoxRouter 2.0, and FGR are 91.81%, 86.87%, and 32.16%, respectively, for ISPD 1998 benchmarks. It may be noted that, though FastRoute3.0 achieves fastest runtime, it fails to generate congestion-free solutions for all benchmarks, which is overcome by the proposed FuzzRoute of the current article. It also shows wirelength improvements of 17.35%, 2.88%, 2.44%, 2.83%, and 2.10%, respectively, over others for ISPD 1998 benchmarks. For ISPD 2008 benchmark circuits it also provides 2.5%, 2.6%, 1 %, 1.1%, and 0.3% lesser wirelength and averagely runs 1.68×, 6.42×, 2.21×, 0.76×, and 1.54× faster than NTHU-Route2.0, FastRoute4.0, NCTU-GR, MGR, and NCTU-GR2.0, respectively.

In DSM (deep sub-micron) regime, together with the integration density interconnects play a dominant role during layout design of integrated circuits. It eventually increases the importance of global routing problem making it more challenging day by day. To cope up with this ever increasing design complexity, the challenging time faced by researchers provides the important opportunity to explore new ideas to solve it within some reasonable time. Heuristic based approaches are generally used for global routing. Large problem space leads global routing problem to a NP Complete one which is less compatible with modern trends. The proposed multi-objective global routing technique is formulated using fuzzy logic to get rid of the limitations of deterministic approaches. After placement and prior to routing phase a set of guiding information is generated from our approach, which will help routing in subsequent steps. During global routing the decision is taken from a fuzzy logic expert system. A GUI is implemented based on the proposed algorithm which is tested for its feasibility study and experimental validation. Success of our proposed approach will open up an avenue for research in global routing phase.