To enable massive machine type communication (mMTC), data aggregation is apromising approach to reduce the congestion caused by a massive number ofmachine type devices (MTDs). In this work, we consider a two-phasecellular-based mMTC network where MTDs transmit to aggregators (i.e.,aggregation phase) and the aggregated data is then relayed to base stations(i.e., relaying phase). Due to the limited resources, the aggregators not onlyaggregate data, but also schedule resources among MTDs. We consider twoscheduling schemes: random resource scheduling (RRS) and channel-aware resourcescheduling (CRS). By leveraging the stochastic geometry, we present a tractableanalytical framework to investigate the signal-to-interference ratio (SIR) foreach phase, thereby computing the MTD success probability, the average numberof successful MTDs and probability of successful channel utilization, which arethe key metrics characterizing the overall mMTC performance. Our numericalresults show that, although the CRS outperforms the RRS in terms of SIR at theaggregation phase, the simpler RRS has almost the same performance as the CRSfor most cases with regards to the overall mMTC performance. Furthermore, theprovision of more resources at the aggregation phase is not always beneficialto the mMTC performance.
展开▼
