On Analysis of Network Coding-Based One-Shot Random Access for Low-Latency MTC

For low-latency communication in massive machine-type communication (MTC), since re-transmissions may not be allowed, one-shot random access (OSRA) such as
$K$
-repetition or network coding (NC)-based OSRA can be considered. In our previous study, it was shown that NC-based OSRA can always perform better than
$K$
-repetition for some limited cases. In this article, more realistic conditions are considered to derive the packet loss rate (PLR) of NC-based OSRA for a fair performance comparison in terms of PLR. Interestingly, it is shown that
$K$
-repetition can outperform NC-based OSRA when the traffic intensity is sufficiently low, which might be necessary to provide a high reliability in terms of PLR. For a low-medium traffic intensity, the PLR of NC-based OSRA is lower than that of
$K$
-repetition. Simulations are also carried out to confirm our derivations.