The Cognitive Internet of Vehicles (CIoV) introduces a cognitive engine in the traditional Internet of Vehicles, which can realize more intelligent functions such as vehicle deployment and resource allocation. Especially in terms of content caching, the cognitive engine can perceive the content requirements of users and match the content providers and content requesters to improve the caching hit rate. However, during this process, users may worry that their privacy data may be leaked. Content requesters need to submit data about points of interest in contents, which is part of their sensitive information. In addition, in terms of rapid speed of vehicles, the connection time is limited, such as vehicle-to-vehicle and vehicle-to-roadside unit (RSU), which leads to limited transaction time during obtaining contents. In order to solve these problems, in this article, we propose a blockchain-based privacy-aware content caching architecture in CIoV. In general, when vehicles need contents, in order to protect the privacy of vehicles, it is no longer necessary to submit a request to an RSU, but through broadcasting contents from an RSU or surrounding vehicles, where vehicles can selectively obtain contents. To improve the cache hit rate, the cognitive engine will perceive content requirements and recommend relevant content requirements to an RSU or content-providing vehicles based on machine learning or deep learning methods. However, this method of content acquisition will connect different vehicles, which brings the untrusted problem. Both content requesters and providers may worry about untrusted users connected with them. To this end, we adopt the blockchain technology to record the completed content transactions, which are written into the block after the consensus mechanism is completed, thus solving the problem of distrust between vehicles. Experiments demonstrate the privacy-aware content caching architecture based on blockchains effectiveness.

中文翻译：

---

车辆认知互联网中基于区块链的隐私感知内容缓存

车辆认知互联网（CIoV）在传统的车辆互联网中引入了认知引擎，可以实现更智能的功能，例如车辆部署和资源分配。特别是在内容缓存方面，认知引擎可以感知用户的内容需求，并匹配内容提供者和内容请求者，以提高缓存命中率。但是，在此过程中，用户可能担心其隐私数据可能会泄漏。内容请求者需要提交有关内容兴趣点的数据，这是其敏感信息的一部分。另外，就车辆的快速速度而言，诸如车辆到车辆和车辆到路边单元（RSU）的连接时间受到限制，这导致在获取内容期间的交易时间有限。为了解决这些问题，在本文中，我们提出了CIoV中基于区块链的隐私感知内容缓存体系结构。通常，当车辆需要内容时，为了保护车辆的隐私，不再需要向RSU提交请求，而是通过广播RSU或周围车辆的内容，在该处车辆可以选择性地获取内容。为了提高缓存命中率，认知引擎将基于机器学习或深度学习方法感知内容需求，并向RSU或提供内容的车辆推荐相关的内容需求。但是，这种内容获取方法将连接不同的车辆，这带来了不可信的问题。内容请求者和提供者都可能担心与他们连接的不受信任的用户。为此，我们采用区块链技术记录完成的内容交易，在共识机制完成后将其写入区块，从而解决了车辆之间的不信任问题。实验证明了基于区块链有效性的隐私感知内容缓存架构。