miss silvia Cassina

Assessing a binary quantum channel exploiting a silicon photomultiplier based hybrid receiver

Quantum communication has been a topic of strong interest in the last few decades and the improvements in the technological field have given us all the tools to implement several communications protocols. These may be divided in two macro categories: continuous variables (CV) protocols and discrete variables (DV) protocols. Focusing on CV protocols, the most exploited detection scheme is the well-known homodyne detection. This work is meant to be a preliminary study to investigate the feasibility of a new CV detection scheme based on photon-number resolving detectors as a quantum communication channel. The mentioned scheme has many similarities with the standard homodyne, such as the mixing at a balanced beam splitter of the state of interest (in our case a coherent state) and a more intense coherent state, called local oscillator. In our case the exploited detectors are silicon photomultipliers (SiPM), that have already been demonstrated to be suitable for the characterization of several states of light. This study involves a first step of classical characterization of the channel in terms of mutual information and a further investigation of the channel its performance in the case of QKD protocol, under the wiretap channel assumption, in terms of key generation rate . To evaluate the performance of the channel we took into account the possible source of information. In fact, according to how the detection is performed we can access the direct outputs of the detectors or their difference. Here are the possibilities we considered in our work: 1) the sign of the output difference that represents a binary discrimination strategy (BDS); 2) the statistical distribution of the difference of the two outputs (namely the so-called Skellam distribution), which is similar to the standard homodyne detection and therefore referred to as homodyne-like (HL); 3) the statistical distribution of the direct output of each PNR detector, considering the scheme as a weak field receiver (WF). We prove that the experimental data agree with the expected numerical models. In particular, we demonstrate that both in a state-discrimination scenario and in a QKD protocol based on a binary alphabet the WF and HL approaches coincide, while the BDS method leads to worse results.