Enhancing the quality-of-service of wireless body area networks

Abstract

Wireless Body Area Networks (WBANs) open many challenges by placing biosensors on/inside human bodies for collecting various health-related information. Communication in WBANs suffers from high channel attenuation due to human body fat and the low transmission power used by the biosensors present on the body. The low power signals transmitted by the biosensors towards the coordinator node (placed outside the human body), experience high attenuation during the non-line-of-sight (NLOS) communications. The NLOS communications are possible in WBANs when the patient’s body impedes the signals transmitted by a biosensor towards the coordinator node. The reception of poor signals at the coordinator node could further deteriorate the quality-of-service (QoS) provided by a WBAN. Therefore, relay-based communications with data forwarding techniques are used to handle link failures and poor network connectivity. Relay-based communication in a WBAN helps in enhancing the lifetime of the network and improves the quality of the signal delivered at the coordinator. However, the existing works related to relay-based communications did not consider the different human postures. This thesis considers three human postures— sitting, sleeping, and walking/ running. Every posture has different mobility patterns and pace of motion due to which, the duration of NLOS communication could vary for different human postures. In the first work, a remote patient monitoring (RPM) application of WBANs is considered where a patient is sitting on the chair in his residential room. This work analyzes the signal received from RPM sensors when a patient rotates by different angles while sitting on a chair as well as heed the use of a relay node placed on his/her body. Literature suggests many relay-based communication protocols to deliver physio-signals efficiently in an RPM application. However, limited studies have focused on the position of a relay node on the human body. This work empirically analyzes the off-body communication path of sensor nodes by collecting data from different body orientations in a residential room. The collected data is used for estimating the path loss parameters for underweight, normal and overweight body mass index (BMI) categories. The estimated parameters are then used to simulate the physical layer of a home-based indoor RPM application. Further, this work inspects different relay node positions on the human body and allude an optimal position of the relay node that cover the transmission range of all sensors and provides an improved channel quality. Finally, an adaptive cross-layer communication protocol is designed for WBANs using the proposed relay node position and improve the Quality of Service (QoS) during non-line-of-sight (NLOS) situation. Sometimes, the special relay node (node installed in a WBAN for supporting relaying only) kept for a sitting position may fail to support the communication during the human sleeping positions. This is possible when the person sleeps on the same side where the relay node is placed on the body. The NLOS situation lasts for a longer duration during human sleeping postures. In such circumstances, an intermediate biosensor forwards the signal of the occluded biosensor node. The forwarding of messages results in quick depletion of energy resources at the intermediate biosensor which affects the overall WBAN services. To resolve this, first, an adaptive Relay-Node Centric (RNC) relay-based communication protocol for WBANs is proposed, which reduces energy used in relaying and improves the stability period of the network. Second, a novel simulation model is designed using an existing real-life experimental dataset to simulate a WBAN placed on the sleeping patient’s body. This work derive a Discrete Time Markov Chain (DTMC) model from real-life data and use human biomechanisms to simulate biosensors’ connectivity status in four human sleeping positions namely, supine, prone, lying on the left side, and lying on the right side. Lastly, performance of RNC is evaluated against the existing cost-function-based and Analytical Hierarchical Process (AHP) based relay selection protocols. Results show that RNC outperforms the existing methods, primarily when nodes are distributed across all body parts. The mobility of nodes in a WBAN is periodic and highly dynamic during the human walking/running scenarios. Existing related work on WBAN used the characteristics, such as periodical movement of WBAN nodes and improves the Quality of Service (QoS) at the coordinator. For example, a biosensor present on the human wrist moves in a back and forth motion while walking. The sensor communicating with the coordinator, present on the chest of the person experiences best communication channel condition when the arm is in the front side of the body. The WBANs exploit opportunistic communication and perform delivery of the signal at best channel conditions. However, the nodes present on the human torso (i.e., front or back side of the person) are static with respect to the coordinator node and hence, cannot use the advantages of the opportunistic communication. Therefore, in this work, an existing real-life health datasets on human heart rate, diabetes, and body temperature are analyzed to find the reduced sampling frequencies for the static nodes present in a mobile WBAN. Next, a direct communication algorithm is proposed that combines the use of the reduced sampling frequencies and the opportunistic communication for a human walking scenario in WBANs. The results show that the algorithm improves the lifetime and the quality of signal delivered in a WBAN. Additionally, this thesis proposes an adaptive sampling for enhancing the lifetime of the smart wearable devices used for monitoring health in day-to-day life. Smartwatches are used widely by people for recording the running performance. All watches sense heart rate continuously during running for maintaining the record of varying heart rate with the change in running speed. The continuous monitoring of heart rate consumes energy of the device. Therefore, existing literature related to this area proposed different methods for reducing the sampling rate of the heart rate sensor during less active events like sitting or sleeping. However, no efforts have been made for adjusting the sampling rate of the heart rate sensor for high acceleration events like running. In such cases, the energy consumption rate of the wearable device increases for long running events. Therefore, in this work, existing real-life human heart rate datasets are used for finding the trends in varying heart rate with the change in the running speed of the person. The observations made are then used for designing an adaptive sampling algorithm for the heart rate sensor present in heart rate monitoring devices. The results show nearly 50% reduction in the sensed data and hence, improves the lifetime of the device. On the other hand, the adaptive sampling of the heart rate may induce information gaps in the sensed data. Therefore, this work also designs a data regeneration system for regenerating the missing information. The results show that the proposed data regeneration system provides atleast 90% accuracy in the regenerated information.