Cooperative reinforcement learning for adaptive power allocation in device-to-device communication

Khan, Muhidul Islam; Alam, Muhammad Mahtab; Moullec, Yannick Le; Yaacoub, Elias

doi:10.1109/wf-iot.2018.8355169

Cited by 18 publications

(14 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, the last few years have seen the emergence of machine learning based approaches for resource allocation and interference mitigation in D2D enabled networks (e.g., Reference [ 14 ] and the references therein). However, in the literature [ 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 ], we observe that light-weight, time critical on-line mechanisms for adapting resource allocation and enhancing ESE are not available. Furthermore, such works use typically centralized approaches that do not exploit the OSA resources available in this type of scenario; when the approach is partly distributed, as in Reference [ 23 ], Q-learning is exploited just for the system throughput.…”

Section: Related Workmentioning

confidence: 95%

Q-Learning Based Joint Energy-Spectral Efficiency Optimization in Multi-Hop Device-to-Device Communication

Khan

Reggiani

Alam

et al. 2020

Sensors

Self Cite

View full text Add to dashboard Cite

In scenarios, like critical public safety communication networks, On-Scene Available (OSA) user equipment (UE) may be only partially connected with the network infrastructure, e.g., due to physical damages or on-purpose deactivation by the authorities. In this work, we consider multi-hop Device-to-Device (D2D) communication in a hybrid infrastructure where OSA UEs connect to each other in a seamless manner in order to disseminate critical information to a deployed command center. The challenge that we address is to simultaneously keep the OSA UEs alive as long as possible and send the critical information to a final destination (e.g., a command center) as rapidly as possible, while considering the heterogeneous characteristics of the OSA UEs. We propose a dynamic adaptation approach based on machine learning to improve a joint energy-spectral efficiency (ESE). We apply a Q-learning scheme in a hybrid fashion (partially distributed and centralized) in learner agents (distributed OSA UEs) and scheduler agents (remote radio heads or RRHs), for which the next hop selection and RRH selection algorithms are proposed. Our simulation results show that the proposed dynamic adaptation approach outperforms the baseline system by approximately 67% in terms of joint energy-spectral efficiency, wherein the energy efficiency of the OSA UEs benefit from a gain of approximately 30%. Finally, the results show also that our proposed framework with C-RAN reduces latency by approximately 50% w.r.t. the baseline.

show abstract

Section: Related Workmentioning

confidence: 95%

Q-Learning Based Joint Energy-Spectral Efficiency Optimization in Multi-Hop Device-to-Device Communication

Khan

Reggiani

Alam

et al. 2020

Sensors

Self Cite

View full text Add to dashboard Cite

show abstract

“…Some studies on inband D2D communication networks in sub-6 GHz band have applied machine learning techniques. In [66], a cooperative reinforcement learning algorithm was developed to perform adaptive power allocation in D2D communication to maximize D2D and cellular users' throughput by maintaining a proper interference level. In [67], joint power adaptation and mode selection strategy were developed based on multiagent Q-learning algorithm was considered.…”

Section: Mobile Information Systemsmentioning

confidence: 99%

Resource Allocation in Millimeter-Wave Device-to-Device Networks

Ombongi

Absaloms

Kibet

2019

Mobile Information Systems

View full text Add to dashboard Cite

Recently, the mobile wireless communication has seen explosive growth in data traffic which might not be supported by the current Fourth Generation (4G) networks. The Fifth Generation (5G) networks will overcome this challenge by exploiting a higher spectrum available in millimeter-wave (mmwave) band to improve network throughput. The integration of the millimeter-wave communication with device-to-device communication can be an enabling 5G scheme in providing bandwidth-intensive proximity-based services such as video sharing, live streaming of data, and socially aware networking. Furthermore, the current cellular network traffic can also be offloaded by the D2D user devices thereby reducing loading at Base Stations (BSs), which would then increase the system capacity. However, the mmwave D2D communication is associated with numerous challenges, which include signal blockages, user mobility, high-computational complexity resource allocation algorithms, and increase in interuser interference for dense D2D user scenario. The paper presents review of existing channel and power allocation approaches and mathematical resource optimization solution techniques. In addition, the paper discusses the challenges hindering the realization of an effective allocation scheme in mmwave D2D communication and gives open research issues for further study.

show abstract

“…In [48], a special neural network was proposed and reinforcement learning was used to minimize the total secondary user interference for OFDMA systems without NOMA. A cooperative adaptive power allocation based on deep learning in D2D communications was proposed in [49], and a deep learning approach that aimed to maximize the spectrum efficiency (SE) in underlay cognitive radio networks was reported in [50]. To learn the suboptimal scheme (weighted minimum MSE, WMMSE), the authors of [51] treated the input and output of a WMMSE algorithm as an unknown nonlinear mapping using a DNN and supervised learning, and achieved performance that was close to that of a suboptimal WMMSE algorithm.…”

Section: Introductionmentioning

confidence: 99%

Outage-Capacity-Based Cross Layer Resource Management for Downlink NOMA-OFDMA Video Communications: Non-Deep Learning and Deep Learning Approaches

Tseng

Tsai

Yu³

2020

IEEE Access

View full text Add to dashboard Cite

Prior works either considered outage capacity of wireless video transmission systems but did not consider NOMA which is a key technology for 5G ultra-reliable low-latency (URLLC), or concerned the ergodic capacity of NOMA-OFDMA systems but did not consider the outage capacity emphasized in 5G URLLC scenario. In this paper, outage capacity (as well as ergodic capacity) maximization in a 5G URLLC scenario, are considered, using two proposed resource management schemes (i.e. B, C) and finally, proposed deep-learning-based versions of Schemes B and C (i.e. Schemes B' and C') to reduce the complexity and the latency for 5G URLLC. The proposed schemes re-allocate subcarrier according to outage capacityinstead of ergodic capacity only-maximization objective and choose the candidate user to gain subcarrier in a new way to improve the outage capacity. The numerical results show the proposed Scheme B and C increase the outage capacity (the percentage of satisfied users) from 79.2% for a prior work (Scheme A) to 85.2% and 92.6%, respectively. Scheme C also increases the ergodic capacity (average PSNR) from 34.7dB for Scheme A to 35.5dB. The deep learning based Schemes B' and C' perform slightly poorer than the corresponding non deep learning based Schemes B and C but the execution time/latency of Scheme B'/C' is less than Schemes B/C.

show abstract

Cooperative reinforcement learning for adaptive power allocation in device-to-device communication

Cited by 18 publications

References 12 publications

Q-Learning Based Joint Energy-Spectral Efficiency Optimization in Multi-Hop Device-to-Device Communication

Q-Learning Based Joint Energy-Spectral Efficiency Optimization in Multi-Hop Device-to-Device Communication

Resource Allocation in Millimeter-Wave Device-to-Device Networks

Outage-Capacity-Based Cross Layer Resource Management for Downlink NOMA-OFDMA Video Communications: Non-Deep Learning and Deep Learning Approaches

Contact Info

Product

Resources

About