| 1. |
- Hu, Chung-Hsuan, 1988-
(author)
-
Communication-Efficient Resource Allocation for Wireless Federated Learning Systems
- 2023
-
Licentiate thesis (other academic/artistic)abstract
- The training of machine learning (ML) models usually requires a massive amount of data. Nowadays, the ever-increasing number of connected user devices has benefited the development of ML algorithms by providing large sets of data that can be utilized for model training. As privacy concerns become vital in our society, using private data from user devices for training ML models becomes tricky. Therefore, federated learning (FL) with on-device information processing has been proposed for its advantages in preserving data privacy. FL is a collaborative ML framework where multiple devices participate in training a common global model based on locally available data. Unlike centralized ML architecture wherein the entire set of training data need to be centrally stored, in an FL system, only model parameters are shared between user devices and a parameter server. Federated Averaging (FedAvg) is one of the most representative and baseline FL algorithms, with an iterative process of model broadcasting, local training, and model aggregation. In every iteration, the model aggregation process can start only when all the devices have finished local training. Thus, the duration of one iteration is limited by the slowest device, which is known as the straggler issue. To resolve this commonly observed issue in synchronous FL methods, altering the synchronous procedure to an asynchronous one has been explored in the literature; that is, the server does not need to wait for all the devices to finish local training before conducting updates aggregation. However, to avoid high communication costs and implementation complexity that the existing asynchronous FL methods have brought in, we alternatively propose a new asynchronous FL framework with periodic aggregation. Since the FL process involves information exchanges over a wireless medium, allowing partial participation of devices in transmitting model updates is a common approach to avoid the communication bottleneck. We thus further develop channel-aware data-importance-based scheduling policies, which are theoretically motivated by the convergence analysis of the proposed FL system. In addition, an age-aware aggregation weighting design is proposed to deal with the model update asynchrony among scheduled devices in the considered asynchronous FL system. The effectiveness of the proposed scheme is empirically proved of alleviating the straggler effect and achieving better learning outcomes compared to some state-of-the-art methods. From the perspective of jointly optimizing system efficiency and learning performance, in the rest of the thesis, we consider a scenario of Federated Edge Learning (FEEL) where in addition to the heterogeneity of data and wireless channels, heterogeneous computation capability and energy availability are also taken into account in the scheduling design. Besides, instead of assuming all the local data are available at the beginning of the training process, a more practical scenario where the training data might be generated randomly over time is considered. Hence, considering time-varying local training data, wireless link condition, and computing capability, we formulate a stochastic network optimization problem and propose a dynamic scheduling algorithm for optimizing the learning performance subject to per-round latency requirement and long-term energy constraints. The effectiveness of the proposed design is validated by numerical simulations, showing gains in learning performance and system efficiency compared to alternative methods.
|
|
| 2. |
- Gülgün, Ziya, 1992-
(author)
-
GNSS and Massive MIMO : Spoofing, Jamming and Robust Receiver Design for Impulsive Noise
- 2023
-
Doctoral thesis (other academic/artistic)abstract
- In this thesis, we focus on vulnerabilities and robustness of two wireless communication technologies: global navigation satellite system (GNSS), a technology that provides position-velocity-time information, and massive multiple-input-multiple-output (MIMO), a core cellular 5G technology. In particular, we investigate spoofing and jamming attacks to GNSS and massive MIMO, respectively, and the robust massive MIMO receiver against impulsive noises. In this context, spoofing refers to the situation in which a receiver identifies falsified signals, that are transmitted by the spoofers, as legitimate or trustable signals.Jamming, on the other hand, refers to the transmission of radio signals that disrupt communications by decreasing the signal to interference plus noise ratio (SINR) on the receiver side.The reason why we investigate impulsive noises is that the standard wireless receivers assume that the noise has Gaussian distribution. However, the impulsive noises may appear in any communication link. The difference between impulsive noises and standard Gaussian noises is that it is more likely to observe outliers in impulsive noises. Therefore, we question whether the standard Gaussian receivers are robust against impulsive noises and design robust receivers against impulsive noises.More specifically, in paper A we analyze the effects of distributed jammers on massive MIMO and answer the following questions: Is massive MIMO more robust to distributed jammers compared with previous generation's cellular networks? Which jamming attack strategies are the best from the jammer's perspective, and can the jamming power be spread over space to achieve more harmful attacks?In paper B, we propose a detector for GNSS receivers that is able to detect multiple spoofers without having any prior information about the attack strategy or the number of spoofers in the environment.In paper C and D, we design robust receivers for massive MIMO against impulsive noise. In paper C, we model the noise having a Cauchy distribution and present a channel estimation technique, achievable rates and soft-decision metrics for coded signals. The main observation in paper C is that the proposed receiver works well in the presence of Cauchy and Gaussian noises, although the standard Gaussian receiver performs very bad when the noise has Cauchy distribution. In paper D, we compare two types of receivers, the Gaussian-mixture and the Cauchy-based, when the noise has symmetric alpha-stable (SαS) distributions. Based on the numerical results, the Gaussian-mixture receiver outperforms the Cauchy-based receiver.
|
|
| 3. |
- Wang, Cheng-Xiang, et al.
(author)
-
On the Road to 6G: Visions, Requirements, Key Technologies, and Testbeds
- 2023
-
In: IEEE Communications Surveys and Tutorials. - : IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC. - 1553-877X. ; 25:2, s. 905-974
-
Journal article (peer-reviewed)abstract
- Fifth generation (5G) mobile communication systems have entered the stage of commercial deployment, providing users with new services, improved user experiences as well as a host of novel opportunities to various industries. However, 5G still faces many challenges. To address these challenges, international industrial, academic, and standards organizations have commenced research on sixth generation (6G) wireless communication systems. A series of white papers and survey papers have been published, which aim to define 6G in terms of requirements, application scenarios, key technologies, etc. Although ITU-R has been working on the 6G vision and it is expected to reach a consensus on what 6G will be by mid-2023, the related global discussions are still wide open and the existing literature has identified numerous open issues. This paper first provides a comprehensive portrayal of the 6G vision, technical requirements, and application scenarios, covering the current common understanding of 6G. Then, a critical appraisal of the 6G network architecture and key technologies is presented. Furthermore, existing testbeds and advanced 6G verification platforms are detailed for the first time. In addition, future research directions and open challenges are identified to stimulate the on-going global debate. Finally, lessons learned to date concerning 6G networks are discussed.
|
|
