Link Quality Analysis Based Channel Selection in High-Frequency Asynchronous Automatic Link Establishment: A Matrix Completion Approach

Asynchronous automatic link establishment (ALE) has been utilized for decades in high-frequency communications due to its operational simplicity. During the process of ALE, the nodes select the channel with the best channel quality based on a link quality analysis (LQA) matrix, which is derived mainly from channel sounding. However, some elements in the matrix can be missing because of the failure of the channel sounding caused by interference or noise in the sounding phase. The incomplete LQA matrix may lead to bad channel selection. In this paper, by exploiting characteristics of low rank and periodicity in LQA matrix, a matrix completion approach is introduced to improve the channel selection in the asynchronous ALE with the goal of achieving maximum throughput in the data transmission. Simulation results show that, compared with the algorithm in current high-frequency ALE standard, the proposed approach of channel selection not only effectively selects the channel with the better channel quality, but also reduces the link establishment time, increasing the throughput of data transmission. In addition, the performance of the proposed algorithm approaches the channel selection algorithm with perfect LQA matrix.