It has been proposed that a secure key distribution scheme using correlated random bit sequences can be implemented using common random-signal induced synchronization of semiconductor laser systems. In this scheme it is necessary to use laser systems consisting of multiple cascaded lasers to be secure against a powerful eavesdropper. In this paper, we report the results of an experimental study that demonstrate that the common random-signal induced synchronization is possible in cascaded semiconductor laser systems. We also show that the correlated random bit sequences generated in the synchronized cascaded laser systems can be used to create an information-theoretically secure key between two legitimate users.

No abstract

We experimentally and numerically observe the synchronization between two semiconductor lasers induced by common optical injection with constant-amplitude and random-phase modulation in configurations with and without optical feedback. Large cross correlation (~0.9) between the intensity oscillations of the two response lasers can be achieved although the correlation between the drive laser and either one of the two response lasers is very small (~0.2). High quality synchronization is achieved in the presence of optical feedback in response lasers with matched feedback phase offset. We investigate the dependence of synchronization on parameter values over wide parameter ranges.

We demonstrate physical implementation of information-theoretic secure oblivious transfer based on bounded observability using optical correlated randomness in semiconductor lasers driven by common random light broadcast over optical fibers. We demonstrate that the scheme can achieve one-out-oftwo oblivious transfer with effective key generation rate of 110 kb/s. The results show that this scheme is a promising approach to achieve information-theoretic secure oblivious transfer over long distances for future applications of secure computation such as privacy-preserving database mining, auctions and electronic-voting.With the rapid evolution of big data and cloud computing systems there is increasing interest in practical schemes for secure operations on information on large scales. One example is secure computation which would allow computation of functions over data without revealing the data [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15] . Practical large scale implementations of secure computation are needed to realize applications such as private information retrieval, privacy preserving database mining, auctions, and electronic voting systems.A key component for secure computation is oblivious transfer. Oblivious transfer is message transfer in which a sender sends encoded messages in such a way that the receiver can only decode some of the messages and the sender does not know which messages were decoded. The original notion of oblivious transfer using an erasure channel was given by Rabin 16 . Later, one-out-of-two oblivious transfer was considered by Even et al. 17 . Naor and Pinkas 18, 19 gave an oblivious transfer protocol based on the Diffie-Hellman assumption, where the protocol relies on computational complexity. It has been known that the Naor-Pinkas protocol is time-consuming, and large amount of computation is required. The oblivious transfer extension technique of Ishai et al. 20 and follow up work has been aimed at achieving faster and more efficient oblivious transfer.Various schemes for oblivious transfer based on information-theoretic security have also been proposed. Information-theoretic oblivious transfer can be secure with respect to adversaries that are computationally unbounded. Moreover, information-theoretic oblivious transfer can be future proof in the sense that secrets will not be revealed by future advances in computational power. Information-theoretic schemes are based on the idea of distilling a secret bit, or string of secret bits, from a statistical advantage in correlation of bits acquired from a probabilistic system. Different models can be distinguished based on specific features of the probabilistic model of the system. Following the original notion of the erasure channel 16 , there have been schemes proposed based on noisy channels [21][22][23][24] , bounded storage 25 , wireless communication systems 26 , quantum mechanical systems [27][28][29][30] , and network behaviors 31,32 . In the noisy channel model, users observe a random sequence from a common source (such...

No abstract