Federated learning and next generation wireless communications: A survey on bidirectional relationship

Abstract

In order to meet the extremely heterogeneous requirements of the next generation wireless communication networks, research community is increasingly dependent on using machine‐learning solutions for real‐time decision‐making and radio resource management. Traditional machine learning employs fully centralized architecture in which the entire training data is collected at one node for example, cloud server, that significantly increases the communication overheads and also raises severe privacy concerns. Toward this end, a distributed machine‐learning paradigm termed as federated learning (FL) has been proposed recently. In FL, each participating edge device trains its local model by using its own training data. Then, via the wireless channels the weights or parameters of the locally trained models are sent to the central parameter server (PS), that aggregates them and updates the global model. On one hand, FL plays an important role for optimizing the resources of wireless communication networks, on the other hand, wireless communications is crucial for FL. Thus, a “bidirectional” relationship exists between FL and wireless communications. Although FL is an emerging concept, many publications have already been published in the domain of FL and its applications for next generation wireless networks. Nevertheless, we noticed that none of the works have highlighted the bidirectional relationship between FL and wireless communications. Therefore, the purpose of this survey article is to bridge this gap in literature by providing a timely and comprehensive discussion on the interdependency between FL and wireless communications.

Cooperative backscatter NOMA with imperfect SIC: Towards energy efficient sum rate maximization in sustainable 6G networks

“…This shows that the maximum value of the objective function lies on either of the two extremes of Λ. 3 Since U 1 is closer to the BS compared to U 2 , thus, the channel gains can be sorted as g 1 > g 2 . This shows that C 3 in problem ( 8) is always satisfied for any value of Λ, and the feasibility of the constraint depends only on the value of P r .…”

“…The upcoming sixth-generation (6G) systems are expected to connect billions of communication devices all over the world [1], [2]. Most promising 6G technologies are artificial intelligence/machine learning [3], [4], reconfigurable intelligent surfaces [5], backscatter communication [6], nonorthogonal multiple access (NOMA) [7], blockchain [8], [9], Tera-hertz communication [10], and simultaneous wireless information and power transfer [11]. These technologies will integrate to the current communication networks such as unmanned aerial vehicles [12], intelligent transportation systems [13], cognitive radio networks [14], Internet of Things [15], device to device communication [16], and physical layer security [17].…”

“…The diagonal entries of the upper triangular matrix are the eigenvalues of the function. However, for the considered case of single variable Λ, the double derivative of the function gives us the eigenvalues and provides the information about the convexity of the function 3. This means that the maximum value of the sum rate is achieved either when the Λ is assigned the maximum value while satisfying all the constrains or when the value of Λ is set to the minimum, such that, all the constraints are satisfied.…”

A Novel Hybrid Deep Learning Model for Metastatic Cancer Detection

“…Recently, there also exist some other artificial intelligence technologies that can be used in next-generation healthcare systems, that is, federated learning algorithms [52], deep learning algorithms [53], reinforcement, and deep reinforcement learning algorithms [54][55][56][57][58]. In addition, next-generation industries, that is, Industry 4.0 and Industry 5.0, would play a crucial role in developing advanced equipment used in health treatment [59].…”

On Computational Offloading in Massive MIMO-Enabled Next-Generation Mobile Edge Computing

“…Future wireless communication networks are expected to connect massive devices to the Internet [1]. These devices would be intelligent, cost-effective, and energy efficient [2]. Moreover, such devices will provide diverse quality of services [3].…”