A Priority-based Fair Queuing (PFQ) Model for Wireless Healthcare System

Healthcare is a very active research area, primarily due to the increase in the elderly population that leads to increasing number of emergency situations that require urgent actions. In recent years some of wireless networked medical devices were equipped with different sensors to measure and report on vital signs of patient remotely. The most important sensors are Heart Beat Rate (ECG), Pressure and Glucose sensors. However, the strict requirements and real-time nature of medical applications dictate the extreme importance and need for appropriate Quality of Service (QoS), fast and accurate delivery of a patient’s measurements in reliable e-Health ecosystem.

As the elderly age and older adult population is increasing (65 years and above) due to the advancement in medicine and medical care in the last two decades; high QoS and reliable e-health ecosystem has become a major challenge in Healthcare especially for patients who require continuous monitoring and attention. Nevertheless, predictions have indicated that elderly population will be approximately 2 billion in developing countries by 2050 where availability of medical staff shall be unable to cope with this growth and emergency cases that need immediate intervention. On the other side, limitations in communication networks capacity, congestions and the humongous increase of devices, applications and IOT using the available communication networks add extra layer of challenges on E-health ecosystem such as time constraints, quality of measurements and signals reaching healthcare centres.

Hence this research has tackled the delay and jitter parameters in E-health M2M wireless communication and succeeded in reducing them in comparison to current available models. The novelty of this research has succeeded in developing a new Priority Queuing model ‘’Priority Based-Fair Queuing’’ (PFQ) where a new priority level and concept of ‘’Patient’s Health Record’’ (PHR) has been developed and
integrated with the Priority Parameters (PP) values of each sensor to add a second level of priority. The results and data analysis performed on the PFQ model under different scenarios simulating real M2M E-health environment have revealed that the PFQ has outperformed the results obtained from simulating the widely used current models such as First in First Out (FIFO) and Weight Fair Queuing (WFQ).

PFQ model has improved transmission of ECG sensor data by decreasing delay and jitter in emergency cases by 83.32% and 75.88% respectively in comparison to FIFO and 46.65% and 60.13% with respect to WFQ model. Similarly, in pressure sensor the improvements were 82.41% and 71.5% and 68.43% and 73.36% in comparison to FIFO and WFQ respectively. Data transmission were also improved in the Glucose sensor by 80.85% and 64.7% and 92.1% and 83.17% in comparison to FIFO and WFQ respectively. However, non-emergency cases data transmission using PFQ model was negatively impacted and scored higher rates than FIFO and WFQ since PFQ tends to give higher priority to emergency cases.

Thus, a derivative from the PFQ model has been developed to create a new version namely “Priority Based-Fair Queuing-Tolerated Delay” (PFQ-TD) to balance the data transmission between emergency and non-emergency cases where tolerated delay in emergency cases has been considered. PFQ-TD has succeeded in balancing fairly this issue and reducing the total average delay and jitter of emergency and non-emergency cases in all sensors and keep them within the acceptable allowable standards. PFQ-TD has improved the overall average delay and jitter in emergency and non-emergency cases among all sensors by 41% and 84% respectively in comparison to PFQ model.