Abstract-Recent advances in wireless networks on the one hand, and increasing application of sensors in the medical field, On the other hand, have led to extension of studies on wireless network and their applications in treatment of patients. Sensor data and their nature, are of great importance in terms of timely response in wireless sensor networks. Employment of consistent Clustering techniques is one of the approaches used to reduce energy consumption and improve accessibility and fault in Body Area Networks (BAN). The present paper, introduces a new clustering technique for BANs, with improved tolerance in the face of network nodes failure. Assessment of The proposed technique using simulation, shows that this technique outperforms the previous ones in terms of cluster recovery speed in the event of node failure, the average power consumption and average network control messages.Keywords-Body Area Networks, Self-healing Clustering, Fault Tolerance
I. INTRODUCTIONWireless Sensor Networks consist of a large number of sensor nodes embedded in the patient's body. The sensor nodes sense and collect the patients' physiological parameters. This data will be sent to physicians, nurses or the hospital through network access points. These nodes organize and cluster themselves, and then begin to operate. After a while, the nodes or connection-oriented links may face hardware or software problems [1].To improve the quality of life in hospitalized patients, they should be provided with maximum mobility chances. This means that wireless sensor networks used in the medical field, should support the movement of sensor nodes in the patient body. This support is associated with many challenges and cause problems in the evaluation of these networks, and that's why research on wireless body sensor networks is so important [2]. In this paper, the failure of nodes refers to exhaustion of sensor node energy and their in operation. Node failure and energy exhaustion occur frequently. Such a situation may lead to breakdown of the entire network. In addition, clustering in such networks should be associated with efficient use of energy, because sensor nodes energy is usually limited. Therefore, the network optimization techniques usually seek to increase the lifetime of the network by reducing the energy consumption of the network. Many techniques can be used to reduce energy consumption and extend the network lifetime. For example, hardware improvements in node manufacturing technology can help reduce energy consumption in these networks. Clustering is one of the most effective techniques for dealing with faults and becoming aware of node failure in the shortest time possible. In the Clustering-based techniques, some sensor nodes as selected as cluster heads and the rest play their role as cluster members. Sensor nodes consume energy for message sending, Processing and motion purposes and will continue to operate until their energy is over [3]. In case a member of the cluster runs out of energy and stops, the failure must be detec...