Centralized Versus Heuristic-Based Distributed Channel Allocation to Minimize Packet Transmission Delay for Multiband Relay Networks

Mughal, Bushra; Fadlullah, Zubair Md.; Ikki, Salama

doi:10.1109/lnet.2020.3030870

Cited by 5 publications

(6 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…where, T sr = F sr /D sr and T rd = F rd /D rd [6], defined as: the packet time span calculated as frame size divided by data rate as described in Table I. T tot denotes the total time needed for RN to complete its simultaneous reception and re-transmission of F sr and F rd , respectively.…”

Section: System Model and Problem Formulationmentioning

confidence: 99%

“…, where rewards are being assigned against actions (data rates) in an inversely proportional manner, giving higher reward to the lesser data rate, as it utilizes lesser power and lesser reward to higher data rate as it utilizes more power. In other words: in this state, the faster the data rate, the more the buffsize and the more the power is required to re-transmit data packet, without effecting total transmission time [6], hence in this state, we are assigning the least reward to the fastest data rate and the greatest reward to the slowest data rate. If the selected data rate requires power more than the P avail of the RN, the environment assigns a zero or negative reward (also known as bad rewards) to the selected action and leads the agent to next state of S 5 .…”

Section: B Proposed Mdp Modelmentioning

confidence: 99%

“…, where rewards are being assigned to actions in a directly proportional manner, giving higher rewards to actions resulting in shorter total transmission time (higher data rates) and assigning lesser rewards to actions resulting in longer total transmission time (lower data rates). In other words: in this state, the faster the data rate, the lesser the buffsize and the lesser the total transmission time is required to forward the data packet [6], hence in this state, we are assigning the greatest reward to the fastest data rate and the least reward to the slowest data rate. In this state, after an action is taken, the environment follows the same criteria for choosing the next states as discussed in case of state S 2 .…”

Section: B Proposed Mdp Modelmentioning

confidence: 99%

“…In contrast, in a distributed approach, the RNs need to make their own individual decisions in accordance with their local channel environment. The centralized approach requires complete information and coordination from/with RNs to provide a one-shot (static) optimal solution, which are more likely invalid by the time the central oracle decides them [6], [7]. Such an approach is impractical and not suitable for a time-sensitive and dynamically-changing network.…”

Section: Introductionmentioning

confidence: 99%

“…Our previous work considered distributed algorithms [6], [7] to solve this problem under the assumption of stable channel conditions and static radios. In this paper, we consider a dynamic relay network scenario with unstable channel environments and mobile radios.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Optimizing Packet Forwarding Performance in Multiband Relay Networks via Customized Reinforcement Learning

Mughal

Fadlullah

Fouda

et al. 2022

IEEE Open J. Commun. Soc.

Self Cite

View full text Add to dashboard Cite

In next-generation wireless networks, relay-based packet forwarding, emerged as an appealing technique to extend network coverage while maintaining the required service quality. The incorporation of multiple frequency bands, ranging from MHz/GHz to THz frequencies, and their opportunistic and/or simultaneous exploitation by relay nodes can significantly improve system capacity, however at the risk of increased packet latency. Since a relay node can use different bands to send and receive packets, there is a pressing need to design an efficient channel allocation algorithm without a central oracle. While existing greedy heuristics and game-theoretic techniques, which were developed for multi-band channel assignment to relay nodes, achieve minimum packet latency, their performance drops significantly when network dynamism (i.e., user mobility, non-quasi-static channel conditions) is introduced. Since this problem involves multiple relay nodes, we model it as a Markov Decision Process (MDP) involving various stages, which essentially means that achieving an optimal and stable solution is a computationally hard problem. Since solving the MDP, traditionally, consumes a great deal of time and is intractable for relay nodes, we explore how to approximate the optimal solution in a distributed manner by reformulating a reinforcement learning-based, smart channel adaptation problem in the considered multi-band relay network. By customizing a Q-Learning algorithm that adopts an epsilon-greedy policy, we can solve this re-formulated reinforcement learning problem. Extensive computer-based simulation results demonstrate that the proposed reinforcement learning algorithm outperforms the existing methods in terms of transmission time, buffer overflow, and effective throughput. We also provide the convergence analysis of the proposed model by systematically finding and setting the appropriate parameters.

show abstract

Section: System Model and Problem Formulationmentioning

confidence: 99%

Section: B Proposed Mdp Modelmentioning

confidence: 99%

Section: B Proposed Mdp Modelmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations