A novel fuzzy logic prediction method to determine scheduling algorithm parameters for resource intensive network dependent applications on digital handheld devices

Most research efforts have concentrated on advancement in scheduling algorithms of soft real time systems to improve user response time for real-time applications like Video Telephony (VT) on handheld devices. However, scheduling algorithms like Rate Monotonie (RM) and Deadline Monotonie (DM) are based on parameters like period and deadline which are network dependent for resource intensive applications like VT and hence are uncertain. This intrinsic uncertainty has become a challenge for system designers to decide the trade-off between computational complexity involved in the task and offering optimally improved user response time. Motivated by these observations, we introduce a prediction method to determine periodicity and deadline for resource intensive network dependent applications. Unlike typical methods, our method considers the trend of receiving packets over network for estimating the periodicity using the moving average technique (MAT). The main contribution is proposing a fuzzy logic approach to calculate the value for deadline of the Video Encoding/Decoding task of the VT application. A simulation is also performed and the results are judged against each other. It is concluded that the proposed approach is very promising and it has the potential to be considered for future research.