Mobile QoS tomography using compressed sensing

The use of mobile devices such as smartphones is becoming very widespread, and the variety of usage patterns is increasing. Consequently, the factors that affect the end-to-end quality of service (QoS) of mobile communications are becoming much more diverse. These factors are not only due to the wireless communication environment but also to other attributes of mobile communications such as the type of access network (e.g., wireless local area network or cellular network), application, and mobile device used. We propose a mobile QoS tomography method that enables us to identify the cause of QoS degradation when such degradation is detected. This method involves using a signal processing technique called compressed sensing to reconstruct a signal from a series of sampled measurements, assuming that the signal vector is sparse, which means that it consists of a few non-zero elements and the remaining elements' value is zero. Assuming that the number of dominant attribute(s) (factor(s)) causing the QoS degradations is generally small, we expect to be able to identify such attributes based on the compressed sensing. Our method uses the pairs of measured QoS and the multiple attribute values of mobile communications as input to the compressed sensing and estimates a multiple attribute vector in which each element indicates the degree of QoS degradation caused by the corresponding attribute. We evaluated our method to verify its effectiveness using two sets of mobile QoS measurement data.