A decentralized MAC layer protocol with periodic channel access evaluated with presence of interference

Despite tremendous development in wireless sensor networks, the abilities are far from fully leveraged in the case of the transmission of medical sensor data. We investigate how a heterogeneous medical wireless sensor network, generating periodic data from different medical sensors, can be optimized in terms of throughput and delivery ratio. By utilizing the property of periodicity, we have implemented a packet scheduler that can, in a decentralized manner, predict the channel availability with a given probabilistic model. We evaluate the Periodic MAC protocol for multirate wireless biomedical sensor networks where we show through simulation how the assumption of periodicity will be affected by errors caused by noisy environments and hidden terminals. The results show that, when influenced by random errors, the assumption of periodicity does not cause significantly higher degradation of performance than regular CSMA-CA, and the protocol has a significant advantage over regular CSMA-CA when hidden terminals are present.