User Scheduling Based on Multi-Agent Deep Q-Learning for Robust Beamforming in Multicell MISO Systems

Braga, Iran M.; Cavalcante, Eduardo de Olivindo; Fodor, Gábor; Silva, Yuri C. B.; Silva, Carlos F. M. e; Freitas, Walter C.

doi:10.1109/lcomm.2020.3015462

Cited by 6 publications

(4 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For that, we modeled the CSI imperfection by assuming that the BS estimates the channel using an minimum mean square error (MMSE) estimator. The estimated channels are given by: 45,46

\begin{align} {\overset{g}{true}}_{i, k}^{DL} & = ξ g_{i, k}^{DL} + \sqrt{1 prefix- ξ^{2}} e_{1}, \end{align}

\begin{align} {\overset{g}{true}}_{j, k}^{UL} & = ξ g_{j, k}^{UL} + \sqrt{1 prefix- ξ^{2}} e_{2}, \end{align}

\begin{align} {\overset{g}{true}}_{i, j, k} & = ξ g_{i, j, k} + \sqrt{1 prefix- ξ^{2}} e_{3}, \end{align}

where

e_{1}

e_{2}

, and

e_{3}

are complex Gaussian i.i.d. entries with zero mean and unit variance, which model the channel estimation error, while

0 \leq ξ \leq 1

denotes the reliability of the channel estimation.…”

Section: Performance Evaluationmentioning

confidence: 99%

See 1 more Smart Citation

Rate maximization in full‐duplex systems: Gains and complexity of employing adaptive power allocation in downlink and/or uplink

Braga

Lima

Maciel

et al. 2021

Trans Emerging Tel Tech

Self Cite

View full text Add to dashboard Cite

The next generations of mobile communications should deal with a set of new use cases, leading to a higher demand for spectral efficiency. Full-duplex (FD) arises as a promising technology since it has the potential to double the system capacity without additional spectrum. However, FD networks are hindered by self-interference (SI) and co-channel interference caused by the reuse of frequency resources. Therefore, radio resource allocation (RRA) in FD networks by means of user pairing, frequency assignment, and power allocation are of utmost importance in order to improve spectral/energy efficiency and ensure quality of service (QoS). Although previous works investigated RRA to improve spectral

show abstract

“…For that, we modeled the CSI imperfection by assuming that the BS estimates the channel using an minimum mean square error (MMSE) estimator. The estimated channels are given by: 45,46

\begin{align} {\overset{g}{true}}_{i, k}^{DL} & = ξ g_{i, k}^{DL} + \sqrt{1 prefix- ξ^{2}} e_{1}, \end{align}

\begin{align} {\overset{g}{true}}_{j, k}^{UL} & = ξ g_{j, k}^{UL} + \sqrt{1 prefix- ξ^{2}} e_{2}, \end{align}

\begin{align} {\overset{g}{true}}_{i, j, k} & = ξ g_{i, j, k} + \sqrt{1 prefix- ξ^{2}} e_{3}, \end{align}

where

e_{1}

e_{2}

, and

e_{3}

are complex Gaussian i.i.d. entries with zero mean and unit variance, which model the channel estimation error, while

0 \leq ξ \leq 1

denotes the reliability of the channel estimation.…”

Section: Performance Evaluationmentioning

confidence: 99%

“…For that, we modeled the CSI imperfection by assuming that the BS estimates the channel using an minimum mean square error (MMSE) estimator. The estimated channels are given by: 45,46…”

Section: Imperfect Channel State Informationmentioning

confidence: 99%

Rate maximization in full‐duplex systems: Gains and complexity of employing adaptive power allocation in downlink and/or uplink

Braga

Lima

Maciel

et al. 2021

Trans Emerging Tel Tech

Self Cite

View full text Add to dashboard Cite

show abstract

“…For that, we modeled the CSI imperfection by assuming that the BSs estimate the channel using an MMSE estimator. Thus, the estimated channel matrix satisfies [54], [55]: Ĥ𝑏 𝑖 ,𝑢,𝑛 = 𝜚H 𝑏 𝑖 ,𝑢,𝑛 + √︁ 1 − 𝜚 2 𝚪, where 𝚪 ∈ C 𝑁 𝑅 ×𝑁 𝑇 is an error matrix with complex Gaussian i.i.d. entries with zero mean and unit variance, while 0 ≤ 𝜚 ≤ 1 denotes the reliability of the channel estimation.…”

Section: E Imperfect Channel State Informationmentioning

confidence: 99%

Joint Resource Allocation and Transceiver Design for Sum-Rate Maximization under Latency Constraints in Multicell MU-MIMO Systems

Braga¹,

Antonioli²,

Fodor³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Due to the continuous advancements of orthogonal frequency division multiplexing (OFDM) and multiple antenna techniques, multiuser multiple input multiple output (MU-MIMO) OFDM is a key enabler of both fourth and fifth generation networks. In this paper, we consider the problem of weighted sum-rate maximization under latency constraints in finite buffer multicell MU-MIMO OFDM systems.Unlike previous works, the optimization variables include the transceiver beamforming vectors, the scheduled packet size and the resources in the frequency and power domains. This problem is motivated by the observation that multicell MU-MIMO OFDM systems serve multiple quality of service classes and the system performance depends critically on both the transceiver design and the scheduling algorithm.Since this problem is non-convex, we resort to the max-plus queuing method and successive convex approximation. We propose both centralized and decentralized solutions, in which practical design aspects, such as signaling overhead, are considered. Finally, we compare the proposed framework with state-of-the-art algorithms in relevant scenarios, assuming a realistic channel model with space, frequency and time correlations. Numerical results indicate that our design provides significant gains over designs based on the wide-spread saturated buffers assumption, while also outperforming algorithms that consider a finite-buffer model.

show abstract

“…In a recent work [8], another type of machine learning strategy is applied to solve the rate maximization problem in a multi-user multiple-input single-output (MU-MISO) scenario. The authors show that deep Q-learning algorithms can learn a policy that can provide a robust BF structure, improving the performance of models based on perfect channel state information (CSI) in scenarios with ICSI.…”

Section: Introductionmentioning

confidence: 99%

On the Robustness of Deep Learning Based Beamforming for MU-MISO Systems

Santana¹,

Guerreiro²,

Silva³

et al. 2021

Anais Do XXXIX Simpósio Brasileiro De Telecomunicações E Processamento De Sinais

Self Cite

View full text Add to dashboard Cite

In this paper, we address the problem of robust beamforming in multi-user multiple-input single-output (MU-MISO) systems. To this end, we assess the performance of a deep neural network (DNN)-based beamforming strategy under imperfect channel state information (ICSI) conditions in a MU-MISO scenario. A robustness analysis is provided by adding a controlled additive error term in the fast fading component of the channel. Such an additive error is independent of the channel realization and does not change the statistics of the channel. The DNN-based beamforming strategy is compared with the classical weighted minimum mean square error (WMMSE) solution in terms of the impact of ICSI on the average sum rate. Numerical results show that ICSI affects in a similar way both the DNNbased strategy and the WMMSE solution. Thus, besides being an alternative method with less computation complexity due to its offline training phase, the DNN-based beamforming strategy can be said to be as robust as the WMMSE solution under ICSI.

show abstract

User Scheduling Based on Multi-Agent Deep Q-Learning for Robust Beamforming in Multicell MISO Systems

Cited by 6 publications

References 17 publications

Rate maximization in full‐duplex systems: Gains and complexity of employing adaptive power allocation in downlink and/or uplink

Rate maximization in full‐duplex systems: Gains and complexity of employing adaptive power allocation in downlink and/or uplink

Joint Resource Allocation and Transceiver Design for Sum-Rate Maximization under Latency Constraints in Multicell MU-MIMO Systems

On the Robustness of Deep Learning Based Beamforming for MU-MISO Systems

Contact Info

Product

Resources

About