13th New England Workshop on Software-Defined Radio
Unity Hall (Room UH400), Worcester Polytechnic Institute, 27 Boynton St, Worcester, MA, USA Main Event: Friday 2 June 2023, 9:00 AM (US Eastern) – 5:00 PM (US Eastern) Tutorials: Thursday 1 June 2023, 5:00 PM (US Eastern) – 9:00 PM (US Eastern)
The 2023 New England Workshop on Software-Defined Radio (NEWSDR’23) is the thirteenth installment of an annual workshop series organized by the Boston SDR User Group (SDR-Boston). We are very excited about this year’s NEWSDR event being hosted in-person for the first time in three years, and that it will be taking place on the beautiful campus of Worcester Polytechnic Institute (WPI) in Worcester, MA, USA. The primary goal of this workshop is to provide a forum that enables SDR enthusiasts to get together, collaborate, and introduce SDR concepts to those interested in furthering their knowledge of SDR capabilities and available resources. NEWSDR 2023 welcomes both experienced SDR enthusiasts as well as individuals who are interested in getting started with SDR.
Getting Here
The event is being held in Unity Hall on the campus of Worcester Polytechnic Institute (WPI). Here are the directions to campus. Attendee parking is located at the WPI Park Avenue Garage (map) and access to the event itself is from the campus side (upper) entrance to Unity Hall (map) for both days.
Registration
Online registration for this event is closed. On site registration will be available the day of the event but with no guarantee of lunch, parking, and other services.
Final Agenda
NEWSDR 2023 activities are distributed over Thursday June 1st (evening tutorials) and Friday June 2nd (main event) on the fourth and fifth floors of Unity Hall. Unless otherwise indicated, all activities are held in Room UH400.
Invited Talk Dr. Tommaso Melodia (William Lincoln Smith Chair Professor, Northeastern University ECE) “Open 6G: Toward a Reference Architecture for Programmable and AI-Driven NextG Open RAN Systems“
Friday 2 June 2023 — Afternoon Session
01:30pm – 02:15pm EDT
Keynote Talk Mr. Scott Fox (Lead Senior Technical Advisor for 5G/FutureG Initiative, OUSD R&E) “DOD & FutureG – Opportunities Ahead“
02:15pm – 03:00pm EDT
Information Session Dr. James Li (Director of Electronic Systems Microelectronics, BAE Systems, Inc.) “Northeast Microelectronics Coalition“
Scott Fox serves as the Lead Senior Technical Advisor to The Department of Defense Office of the Under Secretary for Research and Development and Engineering, (DOD OUSD (R&E)) “5G / FutureG Initiative” at the Pentagon where he leads the ‘5G & Cyber SME Team’. In this role, Scott provides strategic, technical, and business advice related to the DoD’s efforts to modernize our nation’s warfighting capabilities and enhance National Security, specifically leveraging and applying 5G’s (and beyond) advanced capabilities and solutions.
Having spent over four decades in key executive leadership positions across the wireless, telecommunications, and start-up worlds, Scott knows what it takes to build successful companies and deliver high-quality complex solutions in environments of rapid change. A visionary and forward-thinking strategist, Scott has been the Chief Technology Officer (CTO) for BellSouth & Cingular, Chairman of the GSM Association (the global Industry Association of Wireless Operators), Group President – Wireless Facilities Inc (WFII) (Took public on the NASDAQ), Lead Consultant for FirstNet / National Telecommunications & Information Association (NTIA) and has served as a Board Member for over 30+ public & private companies.
Scott also previously served as the founding Director – Collaboration, Innovation, and Commercialization for SpectrumX, a National Science Foundation (NSF)-funded “National Center for Spectrum Innovation” tasked with helping solve our nation’s most vexing spectrum-related challenges (www.SpectrumX.org).
In his personal time, Scott is an Entrepreneur-in-Residence (EIR) at the University of Colorado Boulder’s Venture Partners. Scott also donated significant time working as a member of the ‘Boulder County Amateur Radio Emergency Services’ (BCARES) providing emergency communications support during major emergencies, disasters, and other large-scale incidents. Scott recently moved from Colorado to Florida to be closer to his family.
Opening Talk: Adaptive Radio Science from Earth to Space
Abstract: In this presentation I will discuss the design and architecture of software radio sensors for scientific applications. Such sensors allow us to study the Earth’s atmosphere, near space environment, and the larger astronomical universe. Many radio science observations are focused on physics measurements or the characterization of the physical properties of astronomical or geophysical systems. Scientific instruments often sample wide bandwidths, require precise calibration, need great dynamic range, and have very high absolute sensitivity. Often the signals used for scientific measurements can be extremely weak. Because of this fact, radio science measurements are very susceptible to interference.
Tolerance and mitigation of interference is increasingly a major consideration in sensor design and deployment. Easy and low cost access to wide bandwidths and large frequency ranges is both enabling and problematic for radio science. As an example, the launch of large scale satellite communications constellations is creating a new class of globally visible and highly mobile radio sources. The signals from these systems cannot be easily avoided and their interference impact is not yet well understood. What is clear, however, is that not even distant locations on Earth or in Space will be without anthropogenic radio signals. Mitigation of interference from terrestrial and orbiting systems will require approaches that include technological solutions as well as efforts to modernize policy and the cooperative coordination of radio spectrum usage.
Future scientific systems will need to be much more adaptable both in their capabilities, modes of operation, algorithms, and network interactions with Society. I will discuss the technologies most relevant to near future systems and discuss areas needing significant further development. I will illustrate the discussion with examples drawn from our recent work at MIT Haystack Observatory. In particular I will highlight the interaction of hardware, software, high performance computing, and mathematics in providing solutions to radio science and spectrum related challenges.
Dr. Frank D. Lind was born near Portland Oregon. After attending high school in Seattle, he studied at the University of Washington where he received a Bachelor of Science degree in Physics and a Bachelor of Science degree in Computer Science in 1994. He then joined the UW Geophysics Program and pursued studies leading to the Doctor of Philosophy in Geophysics in 1999. His work there focused on Passive Radar Observations of the Aurora Borealis. Currently he is a Research Engineer at MIT Haystack Observatory where he develops and operates ground and space based radio science instrumentation.
Invited Talk: Open 6G: Toward a Reference Architecture for Programmable and AI-Driven NextG Open RAN Systems
Abstract: This talk will present an overview of our work laying the basic architectural and algorithmic principles for new approaches to design open, programmable, AI-driven, and virtualized next-generation cellular networks. We will cover in detail challenges and opportunities associated with the evolution of cellular system into cloud-native softwarized architectures enabling fine grained control of end-to-end functionalities. We will discuss architectural aspects, automation principles, and algorithmic frameworks enabling fine-grained end-to-end control of wireless system from low-level RAN functionalities to orchestration and management. We will also explore a number of enabling technologies including network slicing, spectrum sharing, security, and energy efficiency, and discuss the way forward.
Tommaso Melodia is the William Lincoln Smith Chair Professor with the Department of Electrical and Computer Engineering at Northeastern University in Boston. He is also the Founding Director of the Institute for the Wireless Internet of Things and the Director of Research for the PAWR Project Office. He received his Laurea (integrated BS and MS) from the University of Rome – La Sapienza and his Ph.D. in Electrical and Computer Engineering from the Georgia Institute of Technology in 2007. He is an IEEE Fellow and recipient of the National Science Foundation CAREER award. He was named a College of Engineering Faculty Fellow in 2017 and received the Søren Buus Outstanding Research Award in 2018 – the highest research award in the College of Engineering at Northeastern University. Prof. Melodia has served as Associate Editor fo IEEE Transactions on Wireless Communications, IEEE Transactions on Mobile Computing, Elsevier Computer Networks, among others. He has served as Technical Program Committee Chair for IEEE Infocom 2018, General Chair for IEEE SECON 2019, ACM Nanocom 2019, and ACM WUWnet 2014. Prof. Melodia is the Director of Research for the Platforms for Advanced Wireless Research (PAWR) Project Office, a $100M public-private partnership to establish 4 city-scale platforms for wireless research to advance the US wireless ecosystem in years to come. The PAWR Project Office is co-lead by Northeastern University and US Ignite and is overseeing the overall deployment and operation of the PAWR Program. Prof. Melodia’s research on modeling, optimization, and experimental evaluation of Internet-of-Things and wireless networked systems has been funded by the National Science Foundation, the Air Force Research Laboratory the Office of Naval Research, DARPA, and the Army Research Laboratory.
Breakout Sessions: “Building and Securing Future Spectrum”
Session 1: Design Strategies & Methodologies for Developing SDR-Based Wireless Systems and Testbeds
Given the continually increasing complexity of the wireless ecosystem, significant attention must be given to the procedures applied when designing and developing advanced system architectures and experimental testbeds. In this session, panelists will discuss tools, methodologies, and experiences related to bringing complex SDR-based test systems from concept to practice. The wide-ranging conversation will include design practices, deployment and testing methodologies, and open-access strategies. The panel will also discuss opportunities to continue advancing the experimental platforms and resources available to the wireless research community. Interactive Session: The communications industry has rapidly advanced over the past two decades, introducing technologies such as RF sampling ADCs and DACs, low-power Zero-IF transceivers, FPGAs, millimeter-wave devices, GaN devices, AI, and machine learning. However, many systems companies are struggling to integrate these devices into their systems while keeping pace with market demands, resulting in slower development times. Exacerbating the issue, many communications engineers have been slow to adopt the development methods that computer scientists have developed over the past 30-plus years that could make things easier, faster, and bug free. Model-Based Design, test-driven development, and continuous integration are proven methodologies that accelerate development and deliver high-quality products with fewer bugs. With modern tools, these methodologies can now be utilized for software-defined communication systems, such as software-defined radio and radar, and are expected to accelerate wireless field growth and evolution. These developments have the potential to revolutionize the way we communicate and drive innovation.
Moderator: Siddhartan Govindasamy, Boston College Panelist: Mike McLernon, MathWorks Panelist: Pedram Johari, Northeastern University Panelist: Leonardo Bonati, Northeastern University
Session 2: The Science Side of the Spectrum
Scientists have used the RF spectrum since the beginning of radio to better understand our world. Even today radio and especially software defined radio is playing a critical role in exploring the world around us. This panel will bring together scientists and engineers to talk about how SDRs are used in their work and what type of research and scientific questions we can begin to answer with this technology. Moderator: John Swoboda, MIT Haystack Observatory
Session 3: ChatGPT: Threat or Opportunity in SDR Development?
Over the past year, there has been a growing amount of attention focused on the challenges associated with ChatGPT and other similar AI-based generative tools, such as essays being plagiarized, and human jobs being replaced by AI. At the same time, this disruptive technology also opens new opportunities to enhance educational experiences and work productivity, especially when it comes to activities involving software development. The focus of this breakout session is to explore the pros and cons of ChatGPT and other AI tools from both technical and ethical perspectives, as well as to see how this technology can potentially be wielded to enhance software defined radio development. Moderator: Alexander Wyglinski, Worcester Polytechnic Institute Panelist: Ermal Toto, Worcester Polytechnic Institute Panelist: Yunus Telliel, Worcester Polytechnic Institute
Authors: Stefan Gvozdenovic, Johannes K Becker, John Mikulskis, and David Starobinski Affiliation: Boston University Abstract: Network reconnaissance is a core security functionality, which can be used to detect hidden unauthorized devices or to identify missing devices. Currently, there is a lack of network reconnaissance tools capable of discovering Internet of Things (IoT) devices across multiple protocols. To bridge this gap, we introduce IoT-Scan, an extensible IoT network reconnaissance tool.
Poster 2: Implementation of a radiofrequency signal generator for ionosonde radar transmitter using red pitaya
Authors: Brayan Lui Estalla Quinteros, Marco Antonio Milla Bravo, Joaquin Blas Verastegui Walqui, and Juan Carlos Espinoza Guerra Affiliation: Jicamarca Radio Observatory Abstract: The radiofrequency signal generator for ionosonde radar transmitter is based on the Red Pitaya Signal Lab 250-12 SDR which comprises a Zynq 7020 SoC FPGA and can transmit modulated signals with a frequency sweep from 1 MHz to 30 MHz. Xilinx-AMD’s Vivado development environment was used for the design.The synthesis of the hardware was based on the VHDL language and the behavioral description style for the modules SPI controller, register map, numerically controlled oscillator (NCO), BPSK and OOK modulator, multiplexer and a synchronization module with GPS clock and trigger inputs, to start sending the signals.
Poster 3: Segmented Spline Curve Neural Network for Digital Predistortion Optimized for Edge Hardware
Authors: Yiyue Jiang, Andrius Vaicaitis, John Dooley, and Miriam Leeser Affiliation: Northeastern University Abstract: Neural networks have been applied to Digital Predistortion in the last decades due to their promising capability for nonlinear modeling. The implementation of neural networks on edge devices enables local processing of wireless data but faces challenges such as high computational complexity and memory requirements when deep neural networks are used. In this poster, we present a customized shallow neural network named Segmented Spline Curve Neural Network (SSCNN) implementation based on AMD/Xilinx RFSoC ZCU111 board. The SSCNN implementation includes a segmented spline curve layer with a normalization function and an index calculation block to reduce computational complexity with very low hardware costs. Our proposed neural network structure meets the quality of previous neural network-based DPD solutions while requiring a small fraction of the hardware resources.
Poster 4: Colosseum as a Digital Twin: Bridging Real-World Experimentation and Wireless Network Emulation
Authors: Davide Villa, Miead Tehrani-Moayyed, Clifton Paul Robinson, Leonardo Bonati, Pedram Johari, Michele Polese, Stefano Basagni, and Tommaso Melodia Affiliation: Northeastern University Abstract: Wireless network emulators are being increasingly used for developing and evaluating new solutions for Next Generation (NextG) wireless networks. However, the reliability of the solutions tested on emulation platforms heavily depends on the precision of the emulation process, model design, and parameter settings. To address, obviate or minimize the impact of errors of emulation models, in this work we apply the concept of Digital Twin (DT) to large-scale wireless systems. Specifically, we demonstrate the use of Colosseum, the world’s largest wireless network emulator with hardware-in-the-loop, as a DT for NextG experimental wireless research at scale. As proof of concept, we leverage the Channel emulation scenario generator and Sounder Toolchain (CaST) to create the DT of a publicly-available over-the-air indoor testbed for sub-6 GHz research, namely, Arena. Then, we validate the Colosseum DT through experimental campaigns on emulated wireless environments, including scenarios concerning cellular networks and jamming of Wi-Fi nodes, on both the real and digital systems. Our experiments show that the DT is able to provide a faithful representation of the real-world setup, obtaining an average accuracy of up to 92.5% in throughput and 80% in Signal to Interference plus Noise Ratio (SINR).
Poster 5: Detection of Co-existing RF Signals in CBRS using ML: Dataset and API-based Collection Testbed
Authors: Chinenye Tassie *, Abdo Gaber†, Vini Chaudhary *, Nasim Soltani , Mauro Belgiovine, Michael Loehning†, Vincent Kotzsch†, Charles Schroeder† and Kaushik R. Chowdhury * Affiliation: *Northeastern University and †National Instruments Corporation Abstract: Opening up the Citizen Radio Broadband Service (CBRS) band, offers unprecedented opportunities for allowing commercial operators to operate in frequencies otherwise reserved for federal use only. However, the challenge of detecting the incumbent radar reliably forces restrictions on the transmit power for operators deploying LTE networks. While Machine Learning (ML)-based solutions have demonstrated the potential for detecting weak radar signals in fully overlapping secondary signals, there exists a fundamental gap in porting these methods for practical, real-world conditions due to a key reason: There are no accessible datasets or even controlled methods to generate such datasets today over-the-air (OTA), where radar and LTE ‘overlap’ in a number of challenging SINR conditions. Our contributions are as follows: (i) We provide the first publicly available CBRS overlapping and non-overlapping LTE and radar OTA dataset in the 3.5 GHz band using †National Instruments Corporation, Austin, Texasa testbed composed of SDRs, (ii) We develop a generic API-based solution that can be easily adapted for other dataset generation needs while offering unified control over SDRs configurations; (iii) We demonstrate the efficacy of this real-world dataset in detecting and localizing RF signals in time-frequency space in the CBRS band using the popular ’You Only Look Once’ (YOLO) object detection model.
Poster 6: Analyzing and Countering LoS Blockage in 5G+ Technologies
Authors: James Onyejizu, Zirou Jin, and Koushik Kar Affiliation: Rensselaer Polytechnic Institute (RPI) Abstract: Performance of 5G+ technologies, particularly at mmWave and higher frequencies, is known to be vulnerable to Line-of-Sight (LoS) blockage as well as the weather conditions. In this spot-light talk/poster, we analyze the impact of LoS blockage and propagation conditions in the 1-100 GHz frequency range through simulations and limited experiments. First, we compare the SNR results from MATLAB simulations with experimental measurements in the sub-6GHz frequencies at different distances and antenna orientations in order to calibrate the simulation model. We observe that the simulation results generally predict the trend of SNR variation with distance well, up to a 10dB difference due to antenna orientation and other uncontrollable factors. Through simulations using the ray-tracing model, we then quantify the impact of distance and LoS blockage on SNR at different frequencies by simulating a mobile receiver moving around the Manhattan buildings, while the transmitter remains static and mounted on a street-light. We also observe the impact of weather conditions (rain and fog) on the SNR. Finally, we outline some ideas on how higher layer protocols may need to be designed to mitigate the impact of sudden loss of signal strength due to LoS blockage or beam misalignment of 5G+ technologies.
Poster 7: SDR Testbed Design and Implementation of a Remotely Controlled DCO-OFDMA System for Multi-Cell Multi-User OWC
Authors: Maryam Aminu Mukhtar, Ashley Lemus, Lenny Martinez, Theodore Semash, Victoria Planchart, Chukwunodebem Onwuchekwa (Christopher), Kiersten Kerby-Patel, Honggang Zhang, and Michael Rahaim Affiliation: University of Massachusetts Boston Abstract: The wireless communications field is dominated by radio frequency (RF) based technologies, but indoor optical wireless communication (OWC) technologies are of interest for future generations of the wireless communications ecosystem. Software Defined Radio (SDR) has made experimental RF research more accessible to the general research community, but SDR tools and technologies are not directly applicable to OWC. We bring the accessibility of SDR to OWC systems research by implementing a test system that provides dynamic subcarrier assignment in an OWC network with DCO-OFDM and DCO-OFDMA. By developing this implementation within GNURadio, we offer an open-source tool for other researchers to use for further research with novel resource allocation techniques and/or iterative improvements to the baseline optical OFDM/OFDMA techniques. We are developing a proof-of-concept system implementation consisting of two OWC transmitters and three receivers; therefore, resource allocation is required across the two overlapping OWC cells and across receivers within the same cell. We are exploring resource allocation techniques using Optical OFDMA within a cell using a central controller to adjust parameters at the transmitters and receivers. Our main objective is to develop the capability to implement a more automated subcarrier allocation process for a set of devices using DCO-OFDM or DCO-OFDMA.
Poster 8: A Scalable Multi-Node SDR Testbed with Centralized Control and Programmable Mobility
Authors: Lenny Martinez, Victoria Planchart, Kevin Holland, Honggang Zhang, and Michael Rahaim Affiliation: University of Massachusetts Boston Abstract: The increasing trend of wireless communication has led to the development of new technologies, including the upcoming 6G. As the technology advances, there is a need for an efficient testbed that examines both the network and physical layers of wireless communication systems. In this work, we propose a Software Defined Radio testbed that enables researchers to keep analyzing and improving current signal processing skills as well as investigate new signal processing techniques. Our open-source testbed provides a flexible platform that allows for the examination of network capacity, while also enabling the exploration of physical layer parameters by having different nodes such as stationary, terrestrial, and aerial devices operated by a Testbed Controller (TC), that will be able to generate, transmit, receive, measure, and analyze waveforms and signals, while also being able to modify transmission parameters. The proposed testbed can serve as a tool for researchers to validate theoretical models and algorithms, leading to the development of more efficient wireless communication systems.
Poster 9: Adapting Machine Learning Classifier Models to Unknown Environments through Open World Learning
Authors: Todd Morehouse, Charles Montes, and Ruolin Zhou Affiliation: University of Massachusetts Dartmouth Abstract: The ability of machines to adapt to changes in the open world is an exciting but difficult task. Autonomous adaptation after initial modelling allows a device to handle unknown scenarios, incorporating new knowledge, reducing human maintenance, and overcoming emerging challenges. This is especially important for cognitive radio and radio frequency communications, as new technologies are constantly introduced, channel conditions create diverse and unique environments, and devices are often expected to operate without upgrades for many years. However, there still exist many challenges in achieving open world learning in this space. Typical systems have rigidly constructed code that is not capable of adapting, or can only adapt through soft means such as statistical measurements. The advent and introduction of machine learning opens new and exciting avenues to achieve open world learning to greater extents. We demonstrate open world learning using a combination of novelty detection and incremental learning. This is applied to automatic modulation classification as a demonstration, and tested over-the-air using software defined radio.
Poster 10: Unsupervised SNR Estimation Using Prototype-Based Multi-Stage Deep Neural Network
Authors: Charles Montes, Todd Morehouse, and Ruolin Zhou Affiliation: University of Massachusetts Dartmouth Abstract: This presentation is for a novel unsupervised learning approach for signal to noise ratio (SNR) estimation combined with supervised modulation classification. Unsupervised learning consists of training a network while the input data labels are not provided. Previous work has shown that knowing the SNR or being within some range of SNR improves performance when performing modulation classification by using multiple networks trained separately. Existing methods are either supervised or have very specific requirements of a dataset that might not be possible to obtain in the implementation environment. Current modulation classification methods perform poorly at low or negative SNR values which previous works have shown is due to the difference in frames’ SNR. Our proposed method is a frame-level SNR estimator which uses a custom prototype-based objective function that is minimized using a regression deep neural network. The estimator network partitions a dataset by estimating SNR ranges and each range is trained on a separate network for modulation classification. We explore multiple splits of a hierarchical clustering method to evaluate the separability of the SNR and determine feasibility of multi-network approaches. The performance of our method is evaluated on DeepSig RadioML2016, which consists of multiple modulation types and SNR values. Results show the ability to effectively estimate and separate multiple SNR ranges in a dataset.
Poster 11: Machine-learning-enabled Angle of Arrival Estimation with FPGA Acceleration
Authors: Jin Feng Lin, Todd Morehouse, Charles Montes,Erika Caushi,Artem Dudko, Noah Oikarinen, Samuel Rouillard, and Ruolin Zhou Affiliation: University of Massachusetts Dartmouth Abstract: This paper is to develop an RF angle-of-arrival (AoA) estimator using signal processing techniques and machine learning (ML). Various methods, including MUSIC and ESPRIT algorithms, were evaluated, alongside different ML architectures, to compare their performance. Machine learning proves to be a powerful tool for this task, offering the capability to learn complex mappings, extract features, and handle challenging scenarios, such as high noise levels. A linear antenna array was chosen for its simplicity and ease of construction. The experimental setup involved the use of two USRP X310 software-defined radios (SDRs) to receive signals through the antenna array. To ensure synchronized sampling and phase alignment, an Octoclock external synchronization system was utilized. Additionally, attempts were made to accelerate ML algorithms using a field programmable gate array (FPGA). The inference cycle can be reduced by 28.2% on Xilinx ZCU102 FPGA.
Poster 12: Development and Characterization of a Heterogenous Wireless Broadband Testbed
Authors: Joseph Murphy, Alexander M. Wyglinski Affiliation: Worcester Polytechnic Institute Abstract: As Internet-enabled services and devices become more ubiquitous, the need for inexpensive and easy deployment of broadband infrastructure becomes essential. This project utilizes multiple wireless technologies to facilitate the creation of a broadband infrastructure testbed that can be used to determine the feasibility and resilience of such a system. A mmWave backhaul, Ubiquiti LTU distribution network, and Minim home Wi-Fi equipment are all combined to characterize the behavior of this heterogeneous wireless system. Intelligent and dynamic routing systems are also developed and examined to improve the shortcomings present in the proposed system and mitigate the risks and challenges often present in fully wireless broadband deployment infrastructure.
Poster 13: The Kraken: A USRP Testbed for RF Fingerprinting Research
Authors: Adriyel Nieves, Mitchell Jacobs, Joseph Murphy, and Alexander M. Wyglinski Affiliation: Worcester Polytechnic Institute Abstract: At WPI’s Wireless Innovation Laboratory, we have built an 8 radio USRP testbed hosted on a Virtual Machine server for RF Fingerprinting over-the-air experiments. The radios used are the USRP N210 and SBX daughter card frontend where each radio acts as a transmitter of digital waveforms at ISM frequencies with controllable power levels to generate RF Fingerprinting datasets. The testbed has server rack mounting options and can easily be removed for portable experiments or modifications to the radio hardware. The poster presents the design of the testbed and preliminary RF Fingerprinting experiments.
Poster 14: Federated Learning for Routing in Swarm Based Distributed Multi-Hop Networks
Authors: Martha Cash, Joseph Murphy, and Alexander M. Wyglinski Affiliation: Worcester Polytechnic Institute Abstract: Unmanned Aerial Vehicles (UAVs) are a rapidly emerging technology offering fast and cost-effective solutions for many areas, including public safety, surveillance, and wireless networks. However, due to the highly dynamic network topology of UAVs, traditional mesh networking protocols, such as the Better Approach to Mobile Ad-hoc Networking (B.A.T.M.A.N.), are unsuitable. To this end, we investigate modifying the B.A.T.M.A.N. routing protocol with a machine learning (ML) model and propose implementing this solution using federated learning (FL). This work aims to aid the routing protocol to learn to predict future network topologies and preemptively make routing decisions to minimize network congestion. We also present an FL testbed built on a network emulator for future testing of the proposed ML aided B.A.T.M.A.N. routing protocol.
Poster 15: Implementation and Evaluation of a Smart Uplink Jamming Attack in a Public 5G Network
Authors: Maya Flores, Devon Poisson, Colin Stevens, Adriyel Nieves, and Alexander M. Wyglinski Affiliation: Worcester Polytechnic Institute Abstract: We present a hardware implementation and an evaluation of the effectiveness and feasibility of a smart jamming attack that targets specific uplink physical channels to destabilize 5G communication systems. Using software defined radio, we examine the susceptibility of the 5G Physical Uplink Shared Channel (PUSCH) to a smart jamming attack as well as the impact of such an attack on user equipment (UE) throughput. The smart jamming attack is designed to exploit the radio access procedure by: (1) identifying the user’s cell radio network temporary identifier (C-RNTI); (2) decoding the physical downlink control channel (PDCCH) information containing the specific UE resources; and (3) generating a quadrature phase shift keying (QPSK) modulated Orthogonal Frequency-Division Multiplexing (OFDM) waveform to effectively deny uplink access. The evaluation results show that the smart jamming attack successfully denied uplink access by reducing the throughput of a specific UE by 99.5%.
Poster 16: Latest Insights and Approaches on Hands-On Project-Based Wireless Education Using SDR — Notes from the Field
Authors: Galahad Wernsing and Alexander M. Wyglinski Affiliation: Worcester Polytechnic Institute Abstract: Over the past 15 years, software defined radio (SDR) technology has become an integrative part of wireless communications and networking educational experience of undergraduate students with respect to hands-on project-based learning of the latest techniques and approaches for over-the-air connectivity. With SDR costs having dramatically decreased over this time period in addition to a substantial increase in functionality, new approaches can be employed in the classroom to provide students with new insights on wireless connectivity, spectrum, and propagation. In this poster, we provide the latest approaches and insights on how SDR can be used to introduce students to a commercial standard (Bluetooth Low Energy) in an over-the-air controlled environment to understand the operations of the Physical and Medium Access Control layers across several real-world scenarios.
Poster 17: Adversarial Attacks on Graph Neural Networks in P2P Wireless Communications
Authors: Ahmad Ghasemi, Ehsan Zeraatkar, Majid Moradikia, and Seyed (Reza) Zekavat Affiliation: Worcester Polytechnic Institute Abstract: In recent years, Machine Learning (ML) techniques, particularly deep learning (DL), have shown remarkable success in addressing complex communications challenges in radio resource management, beam prediction, channel estimation, and localization. However, ML algorithms suffer from poor generalization and scalability. To overcome these limitations, graph neural networks (GNNs) have emerged by combining graph theory and DL, demonstrating promising outcomes in various domains. Accordingly, researchers have started applying GNN to wireless communications. Despite their potential, GNN-based wireless communication is susceptible to adversarial attacks, which manipulate or perturb DL systems during training or testing. Adversarial attacks pose significant threats to wireless networks, including point-to-point (P2P) communications, machine-to-machine (M2M), and vehicle-to-vehicle (V2V) communication systems. Such attacks can lead to performance degradation, security risks, compromised communication quality, and unreliable power distribution. Here, we propose four novel adversarial attacks targeting vertices and edges of the trained GNN model while considering constraints related to channel limits, power limitations, and minimization of detectability. Additionally, we introduce a Min Max optimization problem based on the One-Sample Kolmogorov-Smirnov Test to detect attacks that evade the system’s defense mechanisms. Our proposed method analyzes the cumulative distribution function (CDF) of eigenvalues of the channel tensor, revealing changes caused by adversarial attacks. Furthermore, we present an optimization problem that maximizes the destructive impact on the total quality of communication (QoC) within the limited resource budget available to the adversary. Experimental results demonstrate the effectiveness of the proposed adversarial attacks, resulting in a significant 95% decrease in the total QoC. This study sheds light on the vulnerabilities of GNN-based wireless communications and provides insights into detecting and mitigating adversarial attacks in real-world applications. The findings emphasize the need for secure and efficient operation of wireless networks and highlight the potential for applying the proposed approach to various network controllers.
Thursday Evening Tutorials
Mathworks: AI Workshop for Wireless Applications
Artificial intelligence (AI) is rapidly becoming a critical component of many engineering systems and disciplines today. In the field of wireless communications, AI is being used to design and develop smarter ways to model physical layers, optimize performance of wireless systems and networks, and address new 6G design challenges.
In this hands-on workshop you will write and run code entirely in the browser using MATLAB® Online™. You will learn how to apply principles of AI (machine learning, deep learning, domain-specific processing) to wireless communication workflows.
This interactive hands-on session will include the following: (1) Familiarize yourself with MATLAB Online and AI tools. (2) Create and evaluate necessary components to succeed in AI modeling, by implementing an example of modulation classification. (3) Deep dive into an advanced, domain-specific application that showcases a complete workflow for accomplishing 5G channel estimation.
MathWorks instructors and teaching assistants (TAs) will be available throughout the session to guide you. Please bring your laptop and install the Google Chrome browser beforehand.
NI/Ettus Research: FPGA Programming on the USRP with the RFNoC Framework
This workshop provides a tutorial on the RFNoC framework, including a discussion on its design and capabilities, demonstrations of several practical examples, and a walk-through of implementing a user-defined RFNoC Block and integrating it into both UHD and GNU Radio. The RFNoC (RF Network-on-Chip) framework is the FPGA architecture used in USRP devices, specifically the E310, E312, E320, X300, X310, N300, N310, N320, N321, X410. The RFNoC framework enables users to program the USRP FPGA, and facilitates the integration of custom FPGA-based algorithms into the signal processing chain of the USRP radio. Users can create modular, FPGA-accelerated SDR applications by chaining multiple RFNoC Blocks together and integrating them into both C++ and Python programs using the UHD API, and into GNU Radio flowgraphs. Attendees should gain a practical understanding of how to use the RFNoC framework to implement custom FPGA processing on the USRP radio platform.
Red Wire Technologies: Using Embedded SDRs – Benefits and Limitations
This workshop will cover aspects of using Embedded SDRs, including benefits and limitations. The workshop will begin with an overview of differences between Embedded SDR’s vs. Radio front-ends that are designed for use with a larger computer. The workshop will then focus on applications that demonstrate the use of embedded SDRs. Topics include GNU Radio control from a host and the local device, Yocto/Openembedded, and low-level Linux control of hardware. There will be a few radios available to use during the workshop. A laptop with Linux as the main operating system will be required to use the radio.
TMY Technology: mmW-OAIBOX: An FR2-enabled OAI Testbed
Do you happen to have one or two spare SDRs on hand? Have you ever thought about establishing a complete 5G FR2 end-to-end network with your SDRs?
In the first section of the workshop, TMYTEK co-founder Ethan Lin will introduce the mmW-OAIBOX, an FR2-enabled OAI testbed that TMYTEK worked on with Allbesmart. It is worth noting that this solution has already been delivered to Japan. Incorporating the best of millimeter-wave and OpenAirInterface (OAI), we provide a comprehensive test environment from UE to the core network. The mmW-OAIBOX offers 5G beamformers to mimic gNB and UE array antennas, a frequency converter, a powerful PC installed with the latest OAI stack, including OAI gNB, CN5G, a dashboard, and more.
In the second part of the workshop, we will show you how to use the APIs to control TMYTEK FR2 devices, including a 24-44 GHz up/down converter (UD Box) and a 28 GHz mmWave beamformer (BBox), with your SDR development environment. This will include an API introduction, control calls, DLL imports, and more.
There are many topics that need to be addressed in wireless research. We have built the most advanced tools to unleash your creativity, so you are able to develop innovative solutions for the next generation of wireless technology.
Sponsors / Exhibitors
If your company is interested in participating in NEWSDR 2023, please contact us at gr-newsdr-info@wpi.edu for additional information.
