Adaptive channel bonding in wireless LANs under demand uncertainty

Channel bonding is one promising approach to cope with rising WLAN data demand, given scarce spectrum resources. An access point (AP) can aggregate multiple contiguous channels to satisfy demand. Optimizing the network performance under deterministic demands has been well-studied. It is still an open question of how to optimally utilize available frequency bands under uncertainty in AP demands. We propose two approaches to tackle this problem using a stochastic optimization framework. First, we develop a static joint stochastic center frequency and bandwidth allocation scheme. The goal of this scheme is to minimize the total occupied bandwidth while satisfying the demand of each AP with probability at least ß. Second, we develop a two-stage joint stochastic center frequency and adaptive bandwidth allocation scheme. In contrast to the static scheme, the two-stage allocation scheme allows adaptability of the bandwidth allocated to each AP in response to its demand variation, while keeping the total occupied bandwidth fixed. Our numerical results demonstrate the advantages of (i) stochastic compared to naive allocation and (ii) adaptive compared to static allocation. They also explain the bandwidth-user satisfaction trade-off provided by the adaptive allocation approach.