Impatient Backoff Algorithm: Fairness in a Distributed Ad-Hoc MAC

Many distributed multiple access (MAC) protocols use an exponential backoff mechanism. In that mechanism, a node picks a random backoff time uniformly in an interval that doubles in size after a collision. When used in an ad-hoc network spanning multiple interference domains, this backoff mechanism is unfair towards nodes in the middle of the network. Indeed, such nodes tend to experience more collisions than nodes with fewer neighbors; consequently they choose larger backoff delays than those other nodes; and as a result, get lesser throughput. We propose a different backoff mechanism that achieves a fairer allocation of the available bandwidth, by decreasing the backoff delay upon collision or failure to send a packet. That is, a node becomes more aggressive after each failure. Accordingly, we call this mechanism the impatient backoff algorithm (IBA). The nodes maintain stability of the algorithm by resetting, in a distributed way, the average backoff delays if they become too small. We perform a Markov analysis of the system to prove stability and fairness in simple topologies. We also use simulations to study the performance of IBA in random ad-hoc networks, and compare with an exponential backoff scheme. Results show that IBA achieves comparable mean throughput, while delivering significantly better fairness.