Abstract: ‍ ‍To solve the establishment problem of multi-channel wireless ad hoc networks under practical conditions， a Multi-channel Adaptive Network Establishment （MANE） algorithm was proposed. In MANE， we considered the practical conditions before the network was established， where each node didn’t know the whole network topology， the available spectrums of other nodes， and the synchronization information. Firstly， a network establishment strategy based on neighborhood information was designed， so that nodes could make autonomous decisions by using their own and limited neighborhood information， such as available channel conditions and neighbor cluster head conditions. Secondly， the channel quality evaluation algorithm and the neighbor cluster head evaluation algorithm were proposed， where they were based on the maximum entropy principle and the Physarum-inspired model， respectively. During network establishment， the two evaluation algorithms were utilized to sort the available channels and neighbor cluster heads of each node and facilitate the nodes to select proper channels and cluster heads. Extensive simulation results demonstrate that MANE can enable nodes to complete network establishment with less communication overhead than the state-of-the-art works while achieving a good balance in the number of common channels and cluster size.