In cellular networks, a recent trend is to make spectrum access dynamic in the spatial and temporal dimensions, for the sake of efficient utilization of spectrum. In such a model, the spectrum is divided into channels, and periodically allocated to competing base stations using an auction-based market mechanism, under interference constraints. An efficient auction mechanism is essential to the success of such a dynamic spectrum access model. The desired objectives of an efficient auction mechanism are, viz., (i) maximizing revenue (sum of the payments by the bidders), (ii) truthfulness (which encourages bidders to truthfully declare their true valuations), (iii) maximizing social-welfare (i.e., the total valuation, so that the spectrum is allocated to the bidders who value it the most). We note that maximization of revenue is not feasible if truthfulness is to be guaranteed. In the talk, I will present two auction mechanisms, viz., one that delivers near-optimal revenue (and social-welfare, but is not truthful), and the other that is truthful and delivers near-optimal social-welfare. We consider general (pairwise and physical) interference and bidding models. We demonstrate the performance of our designed technique through simulations over random and real cellular networks. To the best of our knowledge, ours is the first work to design a truthful spectrum auction mechanism that guarantees near-optimal social-welfare.