A subjective survey of user experience for data applications for future cellular wireless networks

We present results from a subjective survey of user experience for data applications in future cellular wireless networks. Using a network emulator, we tested various computing scenarios with different channel sharing schemes (dedicated and shared bandwidth), computing models (local processing with remote files and thin-client server based computing), applications (Web browsing, MS Word, and MS Power Point), and devices (laptops and palm devices). Subjective quality ratings (1 to 5) from 100+ users were collected to obtain an average quality score for the user experience in each scenario. They are used to generate a set of utility curves based on channel rate and number of users sharing the system. Our survey results show that, with laptop-type devices, the preferred computing model changes with the application, and may also change as the available bandwidth changes. For palm devices, the ability to handle general computing tasks is primarily limited by the display and is not sensitive to the wireless channel conditions