Entanglement at distance: qubits versus continuous variables

We consider the problem of obtaining maximally entangled photon states at distance in the presence of loss. We compare the efficiency of two different schemes in establishing $N$ shared ebits: i) $N$ single ebit states with the qubit encoded on polarization; ii) a single continuous variable entangled state (emode) assisted by optimal local operation and classical communication (LOCC) protocol in order to obtain a $2^N$-dimensional maximally entangled state, with qubits encoded on the photon number.