论文标题
评论“通过大型MIMO的物理层密码学”的评论
Comments on "Physical-layer cryptography through massive MIMO"
论文作者
论文摘要
我们基于大量多输入多输出（MIMO）对两个不同版本的物理层密码方案进行了两次攻击。这两个密码系统都采用了单数值分解（SVD）预编码技术。对于第一个，我们表明窃听者（知道自己的频道以及合法用户之间的渠道）可以在与合法接收方相同的条件下解密信息数据。我们研究了在使用发射机处的任意预编码器时，在更概括的方案中研究合法用户对窃听器的解码的信噪比的比率。在负面的一面，我们表明，如果窃听者使用许多接收天线比合法用户天线的数量要大得多，那么就没有优势，而与发射器所采用的预编码方案无关。从积极的一面来看，对于对手的限制为具有与合法用户相同数量的天线的情况，我们给出了$ o \ weles（n^2 \右）$上限的优势，并证明可以使用反编码器来接触该界限。对于第二个密码系统，我们表明所需的安全条件阻止合法用户独特地解码明文。
We present two attacks on two different versions of physical layer cryptography schemes based on massive multiple-input multiple-output (MIMO). Both cryptosystems employ a singular value decomposition (SVD) precoding technique. For the first one, we show that the eavesdropper (who knows its own channel and the channel between legitimate users) can decrypt the information data under the same condition as the legitimate receiver. We study the signal-to-noise advantage ratio for decoding by the legitimate user over the eavesdropper in a more generalized scheme when an arbitrary precoder at the transmitter is employed. On the negative side, we show that if the eavesdropper uses a number of receive antennas much larger than the number of legitimate user antennas, then there is no advantage, independent of the precoding scheme employed at the transmitter. On the positive side, for the case where the adversary is limited to have the same number of antennas as legitimate users, we give an $O\left(n^2\right)$ upper bound on the advantage and show that this bound can be approached using an inverse precoder. For the second cryptosystem, we show that the required security conditions prevent the legitimate user from decoding the plaintext uniquely.
