Adaptive Routing in Wireless Mesh Networks Using Hybrid Reinforcement Learning Algorithm

Abstract

Wireless mesh networks are popular due to their adaptability, easy-setup, flexibility, cost, and transmission time-reductions. The routing algorithm plays a vital role in transferring the data between the nodes. The network's performance is significantly impacted by the route opted by the algorithm. The router takes the decision to send the packet to the next router as per the policy of that algorithm. So even though that decision does not favor the right path selection, the router tends to follow its policy. This can be avoided by having intelligent routers that can make routing decisions on the fly.This paper presents the QL-Feed Forward routing algorithm(QFFR), a new generation of routing algorithms that combines reinforcement learning based on the Q-learning algorithm with a Feed Forward neural network. This algorithm(QFFR) can learn from the network environment and make routing decisions based on the algorithm's learnings. The AI agent's ability to select the fastest path, which enhances the efficiency of the routing operation, is demonstrated by the working of the suggested QFFR algorithm. This paper also evaluates the performance of traditional algorithms, namely, Ad-hoc On-Demand Distance-Vector, Optimized-Link-State-routing, Destination-Sequenced Distance-Vector and Distance Source routing. The evaluation parameters include throughput, packet delivery ratio, and delay. The parameters are the outcomes of the time the information takes to reach from source to destination. This analysis highlights the improvement in the routing decision ability of a router. As per analysis, Ad hoc On-Demand Distance Vector Algorithm outperforms with throughput 723.13 Kbps, delay 343.73 ns. Q-learning agent identifies the route and reaches the destination in average of 3.7s in non-grid architecture. The Q-learning agent takes 0.49sec with a grid size ten by ten and 0.53sec in three by four grid size. The suggested QFFR takes 7.62s score-over time with stable, consistent performance. INDEX TERMSDeep Learning , Reinforcement Learning, Q-learning , Markov Decision Process , routing algorithms, wireless mesh networks I. INTRODUCTION D URING this era of digital development, networks play a critical role. Figure 1 shows Wireless Mesh Networks with a mesh architecture. Wireless mesh networks' key advantages are their versatility and customizing capabilities. Any future modifications would be simple to accommodate, resulting in lower expenses and upkeep of the network. Wireless mesh networking is a recent development that emerged from a decade of Ad-hoc networking development [1]. A wireless mesh network (WMN) is a decentralized network system built on existing wireless technology, namely the 802.11 standards, that works on an Ad-hoc communication mode [2] [3]. The wireless mesh network is ideal for nextgeneration communication with its flexibility and extensive coverage. Multi-radio mesh routers and single-radio mesh clients form Wireless Mesh Networks(WMN) [4]. As a result, such networks are developed by connecting wireless access...

LCSMIBR: Design of a Low complexity Sleep-scheduling Model with Iterative Bioinspired learning for improved Routing efficiency under heterogeneous network condition

“…when compared with EF MSS [3], 8.3% when compared with RLB EEP [4], and 6.4% when compared with QFFR [12] even under faults because of the use of delay metric during estimation of optimal sleep schedules via BFO. The use of distance between nodes during the GWPSO process, which helps in the selection of highspeed node con gurations even under large-scale network scenarios, is another factor contributing to the reduction in delay.…”

“…19.5% when compared with RLB EEP [4], and 8.3% when compared with QFFR [12] due to the use of residual energy metric during estimation of optimal sleep schedules via BFO. The use of temporal energy consumption in previous communications between nodes during the GWPSO process, which aids in the selection of high-energy e cient node con gurations even under large-scale network scenarios, is another reason for this improvement in lifetime.…”

“…16 & 17 for an examination of the performance of the model with 5,000 node-to-node communications, which were used to calculate the performance of various routing and sleep-scheduling models in comparable scenarios. For this purpose, models proposed in EF MSS [3], RLB EEP [4], and QFFR [12] were evaluated, and their QoS parameters for 1k to 5k communications under 4k nodes were averaged. Out of these nodes, 10%…”

“…Where, represents the timestamps for completion & starting the communications, are the residual energies of nodes while starting and completing the communications, are the received and transmitted packets during these communications. Due to the use of delay metric during estimation of optimal sleep schedules via BFO, the model is able to improve communication speed by 10.5% when compared with EF MSS [3], 8.3% when compared with RLB EEP [4], and 6.4% when compared with QFFR [12] even under faults. Another reason for this reduction in delay as observed in Fig.…”

Gateway selection and placement for Multi-radio multi-channel Wireless Mesh Networks with Garter Snake Optimization algorithm

“…For instance, a hyperlink may have a less than ideal signal-to-noise ratio (SNR) or increased bit error rates due to physical obstacles, such as barriers, even in the absence of any external interference. To address this issue, the ETX metric employs a strategy as described in references [35,37]. Wireless mesh networks (WMNs) have had a substantial surge in popularity in recent years due to their ability to provide reliable and e cient communication in many environments.…”