In the widely used IEEE 802.11 standard, the so-called performance anomaly is a well-known issue. Several works have tried to solve this problem by introducing mechanisms such as packet fragmentation, backoff adaptation, or packet aggregation during a fixed time interval. In this article, we present and thoroughly analyze PAS, a dynamic and distributed approach solving the performance anomaly problem. PAS is based on packets' aggregation using a dynamic time interval, which depends on the wireless channel occupation time perceived by each node. Since each station senses the medium independently, this makes PAS a totally distributed solution. Even more, PAS may coexist with standard IEEE 802.11 nodes without any particular adaptation, yet being able to improve performance. Our solution differs from other propositions in the literature because of its dynamic and distributed nature, which makes it suitable in the context of multi-hop networks. Furthermore, it allows increasing fairness, reactivity, and in some cases efficiency. In this article, we thoroughly analyze and emphasize the performance evaluation of our proposal.