A Partial Amplify and Forward Cooperative Communication Strategy Based on Blind Signal Detection

Chang Li,Qibo Cai,Zhili Zhou,Guihua Cui,Zhenzhou Tang,Xiukai Ruan

doi:10.1109/access.2020.3008760

Chang Li, Qibo Cai + Show 4 more

Open Access

https://doi.org/10.1109/access.2020.3008760

Copy DOI

Abstract

Due to the great potential in extending propagation distance and enhancing the link reliability, amplify and forward (AF) cooperative communications have attracted much attention recently. However, the existing studies mainly focus on the energy consumption analysis and optimization, and also the channels used in communications are based on Rayleigh/Rician fading model, and/or it was usually supposed that the channel state information (CSI) is known or can be acquired by a training sequence. In this paper, blind signal detection for AF cooperative communication systems based on time division multiple access (TDMA) is studied, in which the channels are modeled as frequency selective fading and the CSI are unknown. A blind signal detection algorithm of partial amplify and forward (PAF) cooperative communication is proposed based on semidefinite relaxation (SDR). This method can recover the data without the training sequence and need not to implement blind equalization for the unknown channels. With the correlation between data of different links, the proposed method can obtain the optimal solution efficiently and achieve better performance of signal detection than the traditional algorithm. By simulations, the energy consumption efficiency as well as the overall system performance of the proposed strategy are verified.

Highlights

With the rapid construction of 5G mobile communication networks, mobile Internet and ubiquitous IoT networks have been deployed urgently [1]
To address the problems above, in this paper, we study blind signal detection for a class of AF cooperative communication systems in which the channels are modeled as frequency selective fading and the channel state information (CSI) are unknown
A novel partial amplify and forward (PAF) cooperative communication strategy based on blind signal detection is proposed, which can recover the data without the training sequence and need not to implement blind equalization for the unknown channels

Summary

INTRODUCTION

With the rapid construction of 5G mobile communication networks, mobile Internet and ubiquitous IoT networks have been deployed urgently [1]. In [16]–[19], methods were proposed based on tensor for joint semi-blind symbol and channel estimation in AF cooperative communication systems. The method of blind signal detection recovers unknown transmitted signals from linearly distortive channel outputs without training sequences It can improve the bandwidth efficiency of the system, and make the communication more intelligent [21], [22]. To address the problems above, in this paper, we study blind signal detection for a class of AF cooperative communication systems in which the channels are modeled as frequency selective fading and the CSI are unknown. A novel PAF cooperative communication strategy based on blind signal detection is proposed, which can recover the data without the training sequence and need not to implement blind equalization for the unknown channels. Matrices are denoted by boldface uppercase letters, e.g., X, Y, and X−1, XT , rank(X), Tr(X) diag(X) stand for inverse, transpose, rank, trace and diagonal elements of the matrix X, respectively

SYSTEM MODEL

DEMONSTRATION AND RESULTS

CONCLUSION