NeTS: Small: Social Tie Aware Spectrum Sharing: Physical-Social Game and Cloud-Based Cooperative Sensing

为了满足快速增长的无线应用需求，迫切需要开发创新的频谱共享技术，使认知无线电设备能够动态感知通信环境并适应其传输方案。一个关键的观察是，无线设备是由人类携带的，人们在社会交往中通常表现得很理性。的确，社会信任是建立在人际关系之上的，在许多人类活动中经常可以观察到利他行为。有了这个见解，在这个项目中，认知无线电网络被视为一个覆盖/底层系统，其中“虚拟社会网络”（即用户之间的社会联系结构）覆盖在物理通信网络上。然后，利用社会纽带结构促进合作感知和频谱共享，这种合作有可能实现频谱效率的大幅提高。该项目为探索工程、社会科学和经济学之间的相互作用以提高频谱效率的创新研究提供了一个很好的例子。利用社会联系结构进行频谱共享的研究结果有助于推进认知无线电网络设计的最新技术，并有很大的潜力为加强频谱共享开辟新的途径，从而使整个社会受益。通过创新议程，该项目侧重于开发社会联系感知频谱共享机制，同时考虑认知无线电用户之间的物理耦合和社会耦合。特别地，在这一共同主题下，本项目被组织成两个协调良好的重点：1)重点是在主要用户活动变化相对缓慢时数据库辅助频谱接入；二级用户之间的社会意识频道分配是每个用户进行频道选择以最大化其社会群体效用的方式进行的，其定义为自己的效用和与之有社会关系的其他用户的效用的加权总和。然后，研究了物理-社会博弈中的社会群体效用最大化问题，并设计了分布式算法来实现社会关系感知的纳什均衡。2)推力II围绕设计一个主要用户活动快速变化时的社会意识频谱感知框架，利用云平台，利用用户之间的社会信任来激励次要用户参与感知任务。直观地说，协同感知通过利用群体智慧，使二次用户能够克服个体用户信息不完整和能力有限的挑战，从而更准确地发现频谱机会。除了广泛的仿真研究外，所设计的技术将在实际的无线网络测试台上进行评估。总体而言，该项目旨在开发一个社会群体效用最大化框架，以捕捉移动用户之间复杂的社会结构，包括各种积极的社会关系（例如，朋友和盟友之间）和消极的社会关系（由于恶意行为）。

To meet the rapidly growing demand of wireless applications, there is an urgent need to develop innovative spectrum sharing techniques that enable cognitive radio devices to dynamically sense the communication environment and adapt their transmission schemes. One key observation is that wireless devices are carried by human beings and people typically behave with rationality in social interactions. Indeed, social trust is built upon human relationship, and altruistic behaviors are often observed in many human activities. With this insight, in this project a cognitive radio network is viewed as an overlay/underlay system where a "virtual social network" (i.e., the social tie structure among users) overlays the physical communication network. Then, the social tie structure is leveraged to facilitate cooperative sensing and spectrum sharing, and such cooperation has potential to achieve substantial gains in spectral efficiency. This project serves as an excellent example for exploring innovative research on the interplay among engineering, social sciences, and economics for improving spectrum efficiency. The findings on exploiting social tie structure for spectrum sharing contribute to advancing the state-of-the-art of cognitive radio network design, and have great potential to open a new avenue for enhancing spectrum sharing and hence benefit the society at large.With an innovative agenda, this project focuses on developing social tie aware spectrum sharing mechanisms, while taking into account both physical coupling and social coupling among cognitive radio users. Specially, under this common theme, this project is organized into two well coordinated thrusts: 1) Thrust I focuses on database assisted spectrum access when primary user activities change relatively slowly; and social-aware channel allocation among secondary users is cast in a manner in which each user carries out channel selection to maximize its social group utility, defined as the weighted sum of its own utility and the utilities of other users having social ties with it. Then, social group utility maximization (SGUM) for the physical-social game is investigated and distributed algorithms are devised to achieve social tie aware Nash equilibrium. 2) Thrust II is centered around devising a social-aware spectrum sensing framework when primary user activities change fast, in which a cloud-based platform is employed to incentivize secondary users to participate in sensing tasks by leveraging social trust among them. Intuitively, by leveraging the wisdom of crowds, cooperative sensing enables secondary users to overcome the challenges due to incomplete information and limited capability of individual users, leading to more accurate detection of spectrum opportunities. Besides extensive simulation studies, the devised techniques will be evaluated in a realistic wireless network testbed. Overall, this project aims to develop a social group utility maximization framework to capture complex social structure among mobile users, consisting of diverse positive social ties (e.g., between friends and allies) and negative social ties (due to malicious behavior).