Random Media Access Control For Probability Matching Avoidance In Underwater Acoustic Networks

The development of maritime communication network is of great significance to the strategic goal of building China into a maritime power.In particular,the extensive underwater application scenario based on underwater acoustic network,which has important military use of national defense and huge potential economic value.The MAC layer determines how the channel is shared and is dedicated to coordinating nodes to obtain limited channel resources.Because of the uniqueness of the underwater acoustic shared channel,the MAC protocol design has a very important impact on the overall performance of the underwater acoustic network.However,the complexity of the underwater environment makes the design of MAC protocols for underwater acoustic networks face many difficulties and challenges.Random media access control protocol is widely used in underwater acoustic network because of its advantages of fast access and simple design.Due to the lack of interaction of handshake signaling to schedule orderly transmission,nodes contention for access to the channel may cause collisions.Collisions would lead to data re-transmission,which not only results in the degradation of throughput performance,but also increases the delay and energy consumption.Therefore,an avoidance mechanism should be introduced to reduce the probability of collision.Most of the existing random media access control protocol back-off algorithms are designed for the terrestrial wireless network.Because of the significant difference between the underwater environment and the land environment,they cannot be directly applied to the underwater acoustic network.Moreover,there is another common problem in the design of back-off algorithm for such protocols.The updating and selection of back-off strategy is often based on experience.Adopting this heuristic transmission strategy will make the nodes fall into the probability matching trap.To solve this problem,this paper proposes a random medium access control protocol with probability matching trap avoidance.By establishing the non-cooperative game model,nodes can calculate the expected return of transmission strategy and back-off strategy respectively,and then adopt the behavior strategy with high expected return to achieve the purpose of maximizing long-term returns.Since the calculation of the expected return of pure strategy needs to be based on the effective estimation of collision probability,this paper proposes an update method for the estimation of collision probability.This method can adjust the estimated value smoothly and quickly according to the feedback information sent by the nodes in the past,so that the protocol has a better ability to deal with changes in network load.For testing the network performance of the protocol,several important network performance indicators are selected to simulate and experiment on the NS-3 network simulator and the self-developed underwater acoustic network protocol stack platform.Network throughput and end-to-end delay have always been two important indicators of network performance.The former measures the ability of a network to transmit data,while the latter reflects the timeliness of data transmission.At the same time,due to the difficulty of charging underwater nodes,the MAC protocol design of underwater acoustic network will focus more on reducing the energy consumption of nodes.Packet re-transmission on account of node collisions consumes a lot of energy and leads to an increase in delay and a decrease in throughput.Therefore,improving the packet delivery rate also has a very important impact on the global network performance.The simulation results of NS-3 show that the conflict probability estimation update method proposed in this paper,which is used to dynamically adjust the transmission strategy of a node,has a better adaptability than a protocol that supports a static transmission strategy,and therefore also shows a better global network performance.Based on the experimental test results of the underwater acoustic communication network protocol stack platform,the protocol with the probability matching avoidance strategy has improved the overall network performance compared to the protocol with the probability matching strategy.