Fronthaul Compression and Precoding Design for Full-Duplex Cloud Radio Access Network
| Autor: | Ali Cagatay Cirik, Omid Taghizadeh, Lutz Lampe, Rudolf Mathar |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
Radio access network
Mathematical optimization 021103 operations research Computer Networks and Communications Iterative method Computer science 0211 other engineering and technologies 02 engineering and technology Maximization Precoding Computer Science Applications Control and Systems Engineering Channel state information Convex optimization Convergence (routing) Telecommunications link Electrical and Electronic Engineering Computer Science::Information Theory Information Systems |
| Zdroj: | IEEE Systems Journal. 13:1113-1124 |
| ISSN: | 2373-7816 1932-8184 |
| DOI: | 10.1109/jsyst.2019.2900996 |
| Popis: | In this paper, joint design of fronthaul compression and precoding is studied for full-duplex (FD) cloud radio access networks. Multiple uplink and downlink users equipped with multiple antennas communicate with a control unit (CU) in the “cloud” through a set of multiantenna FD radio units that are connected to the CU through limited capacity fronthaul links. In the first part of this paper, we address the weighted sum-rate maximization problem, to compute the optimal precoding and the quantization noise covariance matrices. By exploiting the relationship between weighted-sum-rate maximization and weighted minimum-mean-square-error minimization problems, and leveraging the successive convex approximation (SCA) method, we propose an iterative algorithm that guarantees convergence to a stationary point. In the second part of this paper, we address the stochastic sum-rate maximization problem under fast-fading channels, where only the statistics of the channel state information is available. Casting this nonconvex problem as a difference of convex problem, an iterative algorithm based on the combination of stochastic successive upper bound minimization and SCA approaches that guarantees convergence to a stationary point is proposed. Numerical results demonstrate the advantage of the proposed algorithms. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|