Research On Trust-Based Video Content Delivery Method In D2D-Enabled Heterogeneous Networks

With the further increase in users’ demand for video content,mobile video traffic has gradually become the primary network traffic,putting cellular networks under greater pressure.How to alleviate the contradiction between demand and capacity is one of the urgent problems to be solved.In a D2D-enabled Heterogeneous Network(Het Net),D2 D communication provides a new method for video delivery between devices through direct communication without a base station to relay.Existing research shows that D2 D technology can effectively relieve network load and improve network efficiency.However,the presence of malicious devices(MDs)in Het Net makes the network face the following challenges: First,MDs may disseminate forged videos and it is difficult to verify the authenticity of these videos;second,the complex and diverse attack behavior of MD makes it more difficult for the user equipment to identify them;finally,how the base station identifies trusted devices and caches content on trusted devices is also a key issue that needs to be solved.Most previous studies employ social similarity to assess device trust,but they cannot guarantee honest recommendations.Furthermore,considering that devices may not interact with other devices frequently in the real world,it is difficult to guarantee accurate trust assessment results.This thesis proposes a Trust Evaluation method joint Active Detection(TEAD).This method can enable the device to quickly and accurately obtain the trust degree of the neighbor device,thereby a trusted D2D-enabled Het Net can be built.The TEAD permits devices to verify the authenticity of the received videos via the base station.Therefore,the trust evaluation is based on accurate interaction results.TEAD also designs an active detection method to obtain the trust value of devices actively.It can effectively increase the interaction between devices and make the evaluation results more accurate.The experimental results show that the TEAD method can make the trust knowledge correctness of normal devices is as high as 99%,and the trust value of normal devices is much higher than MD.In addition,TEAD is significantly better than other methods in several key scenarios.When the proportion of MDs exceeds 50%,TEAD can still resist malicious attacks between MDs.In order to solve the problem of how to cache video content on trusted devices,this thesis proposes a Trust-based Cache Placement Approach(TCPA).Firstly,an adaptive weighted trust evaluation method is adopted to enable the base station to identify the reliability of the device.Secondly,a trust-based greedy cache placement algorithm is proposed to effectively store the content on the trusted device.Comparative experiments demonstrate the effectiveness of TCPA,which can effectively and accurately identify MDs existing in the network.It can effectively protect the quality of experience of users while reducing the system cache cost and the cache hit ratio obtained by TCPA is better than other methods.