In this paper, we study the problem of sparse channel estimation via a collaborative and fully distributed approach. The estimation problem is formulated in the angular domain by exploiting the spatially common sparsity structure of the involved channels in a multi-user scenario. The sparse channel estimation problem is solved via an efficient distributed approach in which the participating users collaboratively estimate their channel sparsity support sets, before locally estimate the channel values, under the assumption that global and common support subsets are present. The performance of the proposed algorithm, named WDiOMP, is compared to DiOMP, local OMP and a centralized solution based on SOMP, in terms of the support set recovery error under various experimental scenarios. The efficacy of WDiOMP is demonstrated even in the case in which the underlining sparsity structure is unknown.