Abstract
Visible light communication (VLC) is considered an important complementary technology for extremely high sixth-generation (6G) data transmission and has become part of a hybrid 6G indoor network architecture with an ultradense deployment of VLC access points (APs) that presents severe challenges to user mobility. An adaptive handover mechanism, which includes a seamless handover protocol and a selection algorithm optimized with a deep reinforcement learning (DRL) method, is proposed to overcome these challenges. Experimental simulation results reveal that the average downlink data rate with the proposed algorithm is up to 48% better than those with traditional RL algorithms and that this algorithm also outperforms the deep Q-network (DQN), Sarsa and Q-learning algorithms by 8%, 13% and 13%, respectively.
Highlights
The upcoming sixth-generation 6G networks will no longer use single-frequency bands as channels, and they may have great potential to achieve high throughput, extremely low latency, strong connectivity and high reliability for meeting the requirements of multi-scenario communication [1]
SIMULATION EXPERIMENTS AND RESULTS ANALYSIS An experimental system was established with experimental visible light communication (VLC) data and typical parameters reported for commercially available devices to simulate the indicators for our proposed access points (APs) selection algorithm, and the results were compared with the performance of three classic reinforcement learning (RL) algorithms: Q-learning, Sarsa and deep Q-network (DQN)
We examined the convergence in the above situation with all four RL-based algorithms for various user device (UD) movement speeds and one thousand training episodes
Summary
The upcoming sixth-generation 6G networks will no longer use single-frequency bands as channels, and they may have great potential to achieve high throughput, extremely low latency, strong connectivity and high reliability for meeting the requirements of multi-scenario communication [1]. Liang et al [7] studied a similar problem using the analytic hierarchy process (AHP) with a cooperative game (CG) model for indoor environments These traditional methods without artificial intelligence (AI) can be adopted to optimize vertical handover strategies under heterogeneous networking conditions, thereby improving the system reliability and performance in a conventional way. The performance of the RL algorithms adopted in the above studies is significantly reduced by the restrictions on the Qtable in a large-scale indoor scene with an ultradense deployment of VLC APs. Other AI algorithms mentioned above are difficult to work with because it is challenging to obtain a sufficiently large amount of training data for a particular application scenario. The proposed DRLbased algorithm with a decreasing experience replay space exhibits the best performance in the training process compared to the other algorithms in the control group
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
