Interface Status Monitoring for Wireless Link Aggregation in Cognitive Networks

In wireless communication networks, there is growing interest in using mobile terminals with multiple wireless interfaces in order to achieve increased throughput and reliability. However, the performance of communications using multiple wireless interfaces can be severely impaired by fluctuations in quality of the wireless links. We designed and implemented a novel interface status monitoring scheme which supports the detection of changes in wireless link characteristics for multiple wireless interfaces and proactive decision making about the status of interfaces which can be used for communication. The monitoring scheme acquires information about link quality parameters, such as signal strength, retries and rate, directly from the wireless device driver, which allows it to acquire a large amount of information about wireless transmissions available in the driver, and to reduce latency in deciding the status of interfaces. In this paper we describe the design of the scheme and demonstrate its application to sustaining throughput on aggregated Wi-Fi links.