Technology recognition and traffic characterization for wireless technologies in ITS band

Girmay, Merkebu; Maglogiannis, Vasilis; Naudts, Dries; Aslam, Muhammad; Shahid, Adnan; Moerman, Ingrid

doi:10.1016/j.vehcom.2022.100563

Cited by 12 publications

(10 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, these approaches assume that IQ samples come from single-user and single-flow scenarios, which requires a mechanism to identify the flows directly on the spectrum. Girmay et al [30] tackled this limitation by proposing a traffic characterization process where the output of an IQ-based TR module is used to identify the traffic characteristics of the technologies in terms of channel occupancy time, transmission pattern, and frame count using binary representations, a representation similar to the one used in [25] and [27]. The obtained results showed that the proposed solution can be used to characterize the identified traffic effectively.…”

Section: A Tc At Any Layermentioning

confidence: 99%

“…In the first dimension, we continue using L1 packets to perform the TC at any layer from our previous work [8] since this approach has demonstrated competitive performance compared to byte-based TC. Compared to recent works such as [30], which employ DL techniques to perform traffic characterization at flow level directly on the spectrum, L1 packet-based TC still provides more flexibility in classifying traffic types at any layer and granularity.…”

Section: Research Gaps and Position Of This Work In The Literaturementioning

confidence: 99%

See 1 more Smart Citation

Resource-Efficient Spectrum-Based Traffic Classification on Constrained Devices

Góez,

Aycan Beyazit,

Fletscher

et al. 2024

IEEE Open J. Commun. Soc.

View full text Add to dashboard Cite

Traffic Classification (TC) systems are designed to identify the applications generating network traffic. Recent advancements in TC leverage Deep Learning (DL) techniques, surpassing traditional methods in complex scenarios, including those with encrypted traffic. Notably, state-of-the-art DL-based TC systems have been developed for wireless networks using Physical Layer (L1) packets. This approach overcomes the common limitation in TC research that assumes traffic flows within a wired network under a single network management domain. Despite their benefits, DL-based TC systems often demand significant computational resources, typically available only in cloud environments. Consequently, deploying models at the edge is often infeasible due to their resource-intensive nature, given their original training and optimization for high-resource environments. The inherent challenge lies in adapting these systems for edge computing scenarios, including deployment at access points. In this paper, we propose a novel methodology that exploits expert knowledge in combination with recent advances in Multi-Task Learning (MTL) and Deep Neural Network (DNN) optimization to allow spectrum-based TC systems to run on constrained devices. Performance evaluations on an NVIDIA Jetson TX2 demonstrate that our most optimized MTL model, handling four TC tasks, can reduce memory requirements by a factor of 2.65x and improve execution time by 3.6x compared to sequential execution of four Single-Task Learning (STL) models in a servergrade configuration, with minimal accuracy impact (less than a 0.5% drop) and energy efficiency of 0.97 millijoules per sample at inference. Compared to other edge platforms such as the Raspberry Pi model 3B+ (RPI3B+) with a low-power Artificial Intelligence (AI)-accelerator such as the Coral Tensor Processing Unit (TPU), the NVIDIA Jetson achieves a 12-fold improvement in energy efficiency with no impact on accuracy. These are the first available results to provide a benchmark for different performance metrics (memory, computing, energy) over heterogeneous constrained devices for this type of TC system.

show abstract

Section: A Tc At Any Layermentioning

confidence: 99%

Section: Research Gaps and Position Of This Work In The Literaturementioning

confidence: 99%

Resource-Efficient Spectrum-Based Traffic Classification on Constrained Devices

Góez,

Aycan Beyazit,

Fletscher

et al. 2024

IEEE Open J. Commun. Soc.

View full text Add to dashboard Cite

show abstract

“…1 shows the process of the adaptive MBSFN resource allocation algorithm that chooses the MBSFN frame pattern based on the LTE traffic queue and COT of Wi-Fi transmissions. Initially, the LTE eNB determines the COT of the channel using the TRTC model in [8]. Initially, the system starts with 5 muted MBSFN subframes, giving an almost equal spectrum share to the two technologies.…”

Section: Use Initial Configurationmentioning

confidence: 99%

“…The LTE eNB uses the LTE traffic queue and estimated Wi-Fi Channel Occupancy Time (COT) to configure the number and periodicity of muted LTE MBSFN subframes. The LTE eNB uses the Technology Recognition and Traffic Characterization (TRTC) model proposed in our previous work in [8] to estimate the Wi-Fi COT and schedule its resources accordingly. On the other hand, Wi-Fi uses its Carrier-Sense Multiple Access with Collision Avoidance (CSMA-CA) protocol to sense and transmit in the muted LTE MBSFN subframes.…”

Section: Introductionmentioning

confidence: 99%

Intelligent Spectrum Sharing Between LTE and Wi-Fi Networks using Muted MBSFN Subframes

Girmay

Avila-Campos

Maglogiannis

et al. 2023

2023 IEEE Wireless and Microwave Technology Conference (WAMICON)

Self Cite

View full text Add to dashboard Cite

Due to the fast growth of diverse wireless network deployments, the radio spectrum is becoming scarce. Hence, it is beneficial that different radio access technologies share the spectrum in a harmonious way. In this paper, we propose a coexistence scheme between Long Term Evolution (LTE) and Wi-Fi networks that utilizes a Multimedia Broadcast Multicast Service (MBMS) over a Single Frequency Network (MBSFN) feature of an LTE network. MBSFN is an LTE feature that provides support for multicast/broadcast traffic. We propose an adaptive scheme that configures muted subframes, initially intended for MBSFN operation, to allow Wi-Fi transmissions. For the adaptive configuration of muted MBSFN subframes, the LTE eNB uses its traffic queue and the Wi-Fi spectrum occupancy information, which is determined by a convolutional neural network-based technology recognition and traffic characterization system. The standard LTE System Information Blocks are used to convey the updated configuration to the LTE UE. Hence, the proposed coexistence scheme doesn't require any modifications to a standard MBSFN-compliant LTE UE. Performance analysis is done for various traffic situations, and the results show that muted MBSFN subframe-based coexistence gives a 15% improvement in average aggregated throughput as compared to using Almost Blank Subframe-based coexistence.

show abstract

“…DNN-based wireless technology recognition models follow a series of steps to identify wireless signals [4]. First, a dataset is collected by capturing wireless signals from the devices connected to the considered RATs.…”

Section: Introductionmentioning

confidence: 99%

Over-the-air Aggregation-based Federated Learning for Technology Recognition in Multi-RAT Networks

Girmay,

Seif,

Maglogiannis

et al. 2023

Preprint

View full text Add to dashboard Cite

<p>With the continuous evolution of wireless communication and the explosive growth in data traffic, decentralized spectrum sensing has become essential for the optimal utilization of wireless resources. In this direction, we propose an over-the-air aggregation-based Federated Learning (FL) for a technology recognition model that can identify signals from multiple Radio Access Technologies (RATs), including Wi-Fi, Long Term Evolution (LTE), 5G New Radio (NR), Cellular Vehicle-to-Everything PC5 (C-V2X PC5), and Intelligent Transport Systems G5 (ITS-G5). In the proposed FL-based technology recognition framework, we consider edge network elements as clients to train local models and a central server to create the global model. In each client, a Convolutional Neural Network (CNN)-based model is trained from Inphase and Quadrature (IQ) samples collected from a certain combination of RATs. The possible combination of RATs considered in the clients is selected based on the capabilities of the real-world network elements that can be used as a client.</p> <p>The FL framework involves a process where multiple clients periodically send updates derived from local data to a central server, which then integrates these contributions to enhance a shared global model. This method ensures that the system stays current with the evolving real-world environment while also minimizing bandwidth required for training data transfer and allowing for the maintenance of personalized local models on each client’s end.</p>

show abstract

Technology recognition and traffic characterization for wireless technologies in ITS band

Cited by 12 publications

References 36 publications

Resource-Efficient Spectrum-Based Traffic Classification on Constrained Devices

Resource-Efficient Spectrum-Based Traffic Classification on Constrained Devices

Intelligent Spectrum Sharing Between LTE and Wi-Fi Networks using Muted MBSFN Subframes

Over-the-air Aggregation-based Federated Learning for Technology Recognition in Multi-RAT Networks

Contact Info

Product

Resources

About