公开(公告)号：US20240232719A1

公开(公告)日：2024-07-11

申请号：US18515317

申请日：2023-11-21

Applicant: Beijing University of Posts and Telecommunications

Inventor： Hui TIAN ,  Wanli NI ,  Ping ZHANG ,  Keyan LIU ,  Shaoshuai FAN ,  Gaofeng NIE

IPC: G06N20/00

CPC classification number: G06N20/00

Abstract: A semi-federated learning (semiFL) method based on a next-generation multiple access (NGMA) technology is provided. Centralized learning (CL) and FL are integrated such that devices with weak computing capabilities can also participate in training of a global model. A simultaneously transmitting and reflecting reconfigurable intelligent surface (STAR-RIS) is deployed to dynamically change a channel environment such that a system can meet different task requirements of heterogeneous users. Communication-centric CL users and computing-centric FL users can transmit data in parallel on a same time-frequency resource. This avoids a waste of data resources, enriches data obtaining of a base station (BS), and improves accuracy of the global model. The semiFL method also integrates a strategy for jointly optimizing user power allocation and a configuration of the STAR-RIS to reduce total uplink transmit power consumption of the system and prolong a life cycle of an intelligent Internet of Things (IoT) network.

公开(公告)号：US20240297700A1

公开(公告)日：2024-09-05

申请号：US18387068

申请日：2023-11-06

Applicant: Beijing University of Posts and Telecommunications

Inventor： Hui TIAN ,  Wanli NI ,  Ping ZHANG ,  Shaoshuai FAN ,  Gaofeng NIE ,  Shilin TAO

IPC: H04B7/06

CPC classification number: H04B7/0626 ,  H04B7/0639

Abstract: A wireless federated learning (FL) framework and a resource optimization method are provided to resolve a problem that FL is not suitable for many hardware-constrained Internet of Things (IoT) devices with a small amount of computing resources. In the framework, users with sufficient computing resources upload locally trained model parameters to a base station, and users with limited computing resources only need to send training data to the base station. The base station performs data training and model aggregation to obtain a global model. In this way, the users with limited computing resources and the users with sufficient computing resources cooperatively train the global model. To improve a data transmission rate and reduce an aggregation error of FL, a non-convex optimization problem is constructed to jointly design user transmit power and a reception strategy of the base station, and solves the problem through a successive convex approximation (SCA) method.

公开(公告)号：US20240297684A1

公开(公告)日：2024-09-05

申请号：US18387063

申请日：2023-11-06

Applicant: Beijing University of Posts and Telecommunications

Inventor： Hui TIAN ,  Wen WANG ,  Ping ZHANG ,  Gaofeng NIE ,  Xue RONG ,  Wanli NI

IPC: H04B7/04 ,  H04B7/0426

CPC classification number: H04B7/04013 ,  H04B7/043

Abstract: A multi-functional reconfigurable intelligence surface (MF-RIS) integrating signal reflection, refraction and amplification and energy harvesting and an application thereof are provided. The MF-RIS can support wireless signal reflection, refraction and amplification and energy harvesting on one surface, to amplify, reflect, or refract a signal through harvested energy, and further enhance effective coverage of wireless signals. When a signal model of the MF-RIS constructed in the present disclosure is applied to a multi-user wireless network, a non-convex optimization problem of jointly designing operation modes and parameters that include BS transmit beamforming, and different components and a deployment position of the MF-RIS is constructed with an objective of maximizing a sum rate (SR) of a plurality of users in an MF-RIS-assisted non-orthogonal multiple access network. Then, an iterative optimization algorithm is designed to effectively solve the non-convex optimization problem, to maximize the SR of the plurality of users.