| With the rapid development of wireless communications,multi-modal services,formed by the fusion of traditional audio-visual services and emerging haptic informations,are gradually becoming the mainstream of future applications.In multi-modal services,audio and video services mainly require high-speed transmission,while haptic informations require ultra-high reliability and ultra-low latency transmission.However,most of the traditional multi-modal services transmission strategies design different transmission mechanisms separately for different modal services,so it is difficult to meet the diversified transmission needs for multi-modal services.In addition,with the continuous growth of the amount of data,the problems caused by signal loss and latency in the transmission process have become increasingly obvious,which have a serious impact on the quality of service of multi-modal services,and significantly affect the user experience.Therefore,it is very meaningful to design an effective cross-modal communication transmission strategy to realize the reasonable bearing for multi-modal services.Network slicing technology can divide the same network environment into multiple independent slices,and each slice can be individually configured according to the needs of different modal services.Thus,it can provide a reasonable and efficient solution for the cooperative transmission of multi-modal services.This paper studies the multi-modal service transmission strategy based on network slicing technology in cross-modal communication,explores the efficient and high-quality transmission method of multi-modal services,and comprehensively enhances the immersive experience of users.The main research works of this paper are as follows:(1)To solve the problem that different modal services affect each other during transmission,which leads to the decline of service performance,we design a network slicing isolated transmission strategy in cross-modal communications.This transmission strategy can reserve the slice resources allocated to each modal service before the start of transmission,effectively reducing the interaction between various modal services,and realizing the efficient transmission of multi-modal services.At the same time,we establish a return on investment model in the transmission process to maximize the quality of service of multi-modal services under the limited investment in communication resources,to save resources and reduce waste as much as possible.We then decompose the optimization problem into the power control problem for data flows and the resource allocation problem for slices,and solve them separately.First,by determining the range of power,we transform the data flow power control problem into a convex optimization problem to solve it.Furthermore,we solve the resource allocation problem of slices in the optimal power control case,and design a heuristic algorithm to solve it to obtain the optimal return on investment of the system.(2)To address the characteristics of short packet transmission,sporadic occurrence,and ultra-low latency requirements of haptic informations in multi-modal services,we design a network slicing puncture scheduling strategy in cross-modal communication to effectively solve the problem of slice resource waste caused by isolated transmission.Meanwhile,we introduce the intelligent reflecting surface technology to optimize the transmission environment to reduce the amount of resources required for puncturing as much as possible,and to reduce the loss of multi-modal services caused by puncture scheduling.We then construct a resource allocation model based on different time scales to minimize the data rate loss of audio and video flows while maximizing the packet acceptance rate of haptic flows.First,we actively design the phase shift matrix configuration at the beginning of each time slot to enhance the gain of all slices in the network used to transmit multi-modal services.Then,we jointly optimize the power and frequency of multi-modal services based on mini time slot,and propose a heuristic algorithm to obtain the optimal resource allocation,in order to minimize the data rate loss of audio and video flows while maximizing the packet acceptance rate of haptic flows. |
