Skip to Main content Skip to Navigation
Conference papers

A stack cross-layer analytical model for CSMA/CA IEEE 802.15.4 networks

Abstract : Because of the specifications in low-cost, low-power IEEE 802.15.4 wireless sensor networks, comprehensive analytical model is important for evaluating performance under varying wireless channel constraints. The systematic properties of single physical layer and medium access control (MAC) layer protocol have been studied through the techniques based on mathematical models or experiment-based approaches. However, It is insufficient to evaluate network performance on the basis of existing single layer model or cross-layer model with stationary parameters, especially for the multi-variable parameters-based wireless network environment. In this paper, we propose an enhanced stack cross-layer analytical model based on the comprehensive combination and interaction between PHY layer propagation model and MAC layer Markov chain model. Dynamic interaction between sublayer models achieve adaptive performance estimation with hyperparameters sets. Cross-layer performance degradation is analyzed under the varying inputs of multi-parameters vectors, several Quality of Service (QoS) metrics and effective energy consumption metric are proposed and evaluated, respectively. From the simulation results compared with benchmark models, stack cross-layer model offers more comprehensive performance analysis with different cross-layer parameters sets which include distance, transmit power, noise power, and information loads, etc.
Complete list of metadata
Contributor : Pons Patrick <>
Submitted on : Thursday, March 28, 2019 - 5:43:24 PM
Last modification on : Thursday, June 10, 2021 - 3:48:43 AM



Zongyi Liu, Daniela Dragomirescu, Georges da Costa, Thierry Monteil. A stack cross-layer analytical model for CSMA/CA IEEE 802.15.4 networks. International Conference on Internet of Things, Data and Cloud Computing (ICC), Mar 2017, Cambridge, United Kingdom. pp.1-7, ⟨10.1145/3018896.3065839⟩. ⟨hal-02083163⟩



Record views