Considering the active/sleep dynamics and the contention backoscheme of SMAC protocol in multi-hop wireless sensor networks (WSNs), we model each node as a finite single server queue with server shutdown, and model each node’s states as a two-dimensional continuous-time Markov chain. Based on the model, we derive the network performance in terms of average packet loss ratio, network throughput, average packet delay and average power consumption. Experiment comparisons show that the analytical results match well with the simulation results, which validates the accuracy of the analytical model. Furthermore, using the analytical model can enable us to investigate the performance tradeobetween energy efficiency and QoS requirement, and give us theoretical insight into the optimal parameters such as duty cycle, mean active period and buffer size in multi-hop wireless sensor networks. Considering the active / sleep dynamics and the contention backscheme of SMAC protocol in multi-hop wireless sensor networks (WSNs), we model each node as a finite single server queue with server shutdown, and model each node states states as two-dimensional continuous- based on the model, we derive the network performance in terms of average packet loss ratio, network throughput, average packet delay and average power consumption. Experiment comparisons show that the analytical results match well with the simulation results, which validates the accuracy of the analytical model. Furthermore, using the analytical model can enable us to investigate the performance trade obetween energy efficiency and QoS requirement, and give us theoretical insight into the optimal parameters such as duty cycle, mean active period and buffer size in multi-hop wireless sensor networks.