Tan, Walls, and Collett reply.
Tan, Walls, and Collett reply.
复制标题
谭、沃尔斯和科利特回答。
DOI:
10.1103/physrevlett.68.895
复制
发表时间:
1992
影响因子:
8.6
通讯作者:
Collett
中科院分区:
文献类型:
--
作者:
Tan;Walls;Collett
The first condition relates the probability of a joint detec-tion to the locally determined probabilities of single detections and is equivalent to Eq.(2) in the Comment. The second condition is a statement of the way in which the phase of the local oscillator affects the photodetection probabilities at that homodyne detector. A change in the phase Hp for a given set of parameters is only permitted to redistribute the probabilities of detection between detectors ck and dk but is not allowed to change the total detection probability. This is related to the" no-enhancement hypothesis" of Clauser et al.[31. It is easy to see that for the form of PJ 0; a, ek) given in theComment, this condition is not fulfilled since the sum P,(A „ak, Hk)+ Pgo., ak, ek) depends on the value of Hk. Santos's homodyne detectors must function in a non-probability-conserving way, and in fact forsome values of A. and H& will completely fail to detect an incident photon. We had included (following Clauser et al. and Grangier, Potasek, and Yurke) the requirement of local conserva-tion of probability within each homodyne detector as part of the classical concept of locality; but as Santos's exam-ple shows itcan be regarded as an additional assumption to the factorization of the joint probability density. As discussed by Grangier, Potasek, and Yurke [2] in the ap-pendix to their article, the two conditions given above are sufficient to ensure that the form of Bell's inequality using intensity correlation coefficients is satisfied.