Efficient HMAC-based secure communication for VANETs
Efficient HMAC-based secure communication for VANETs
复制标题
DOI：
10.1016/j.comnet.2012.04.002
复制
发表时间：
2012-06
期刊：
影响因子：
--
通讯作者：
Changhui Hu;T. W. Chim;S. Yiu;L. Hui;V. Li
中科院分区：
文献类型：
--
作者：
Changhui Hu;T. W. Chim;S. Yiu;L. Hui;V. Li
车载自组网(VANET)是一种新兴的网络，它方便道路上的车辆进行安全驾驶通信。它需要一种机制来帮助验证消息，识别有效的车辆，并删除不遵守规则的恶意车辆。大多数现有的解决方案要么没有有效的消息验证方案，要么使用公钥基础设施(PKI)。在这个网络中，车辆能够向其他车辆广播消息，一组已知的车辆之间也可以安全地通信。因此，群组通信对于网络来说是必要的。然而，现有的大多数解决方案要么没有考虑到这一点，要么使用配对操作来实现这一点。它们要么不安全，要么无效。本文针对车载自组网中基于散列的消息认证码(HMAC)的不足，提出了一套更全面的安全方案。当然，我们仍然需要在某些地方使用配对操作。我们的方案由三个方案组成：(1)车辆与路旁单元(RSU)之间的通信；(2)群内的一对一通信；(3)无群的一对一通信。基于我们的仿真研究，我们证明了我们的方案是有效的，并且所引起的时延要小得多。我们的第一个方案造成的平均时延比以前的方案低几千倍。第二种方案的平均时延为0.312ms，而前一种方案的平均时延为12.3ms。第三种方案的平均时延为0.312ms，前一种方案的平均时延约为9s。
Vehicular Ad Hoc Network (VANET) is an emerging type of network which facilitates vehicles on roads to communicate for driving safety. It requires a mechanism to help authenticate messages, identify valid vehicles, and remove malevolent vehicles which do not obey the rules. Most existing solutions either do not have an effective message verification scheme, or use the public key infrastructure (PKI). In this network, vehicles are able to broadcast messages to other vehicles and a group of known vehicles can also communicate securely among themselves. So group communication is necessary for the network. However, most existing solutions either do not consider this or use pairing operation to realize this. They are either not secure or not effective. In this paper, we provide a more comprehensive set of secure schemes with Hash-based Message Authentication Code (HMAC) in VANETs to overcome their shortcomings. Of course, we still need to use Pairing operation in some place. Our scheme is composed of three schemes: (1) Communications between Vehicles and Road-Side Units (RSUs), (2) One to One Communications within a Group, (3) One to One Communications without a Group. Based on our simulation study, we show that our schemes are effective and the delay caused is much lower. The average delay caused by our first scheme is nearly thousands of times lower than prior schemes. The average delay caused by our second scheme is 0.312ms, while the delay caused by prior scheme is 12.3ms. Meanwhile the average delay caused by our third scheme is 0.312ms, and the delay caused by prior scheme is about 9s.