SCSMA: A Smart CSMA/CA Using Blind Learning for Wireless Sensor Networks

A smart carrier sense multiple access (SCSMA) using control messages is proposed for low power and low bandwidth wireless sensor networks (WSNs). In traditional WSNs, the use of media access control (MAC) messages is not recommended since data are of small size. However, if data size is increased with data aggregation or multimedia data, the control messages can be used for reliable data transmission. Thus, an efficient mechanism is needed to avoid the collision of control messages and data. SCSMA employs a blind learning algorithm with low complexity to prevent collision during channel contention and data transmission, even in the presence of hidden nodes. It is shown by simulation and analysis that SCSMA improves throughput by 29.5% to 71.6%, and reduces energy consumption by 20.5% to 36.2%, compared to the recent best approach using MAC messages. Simulation data are also verified by experimental data.