Lightweight control planes are techniques that create a control plane in WiFi networks without any additional spectrum requirements. Flash signals are an example of such control signals that exploit the link margin that typically exists in WiFi communication. In this paper we consider the problem of allowing transmitters and receivers of a transmission to exploit such control channels while the communication is ongoing. We present a mechanism called switch that facilitates switches in communication modes (Tx to Rx and vice-versa). We then use switch as the core building block to solve problems in WiFi networks such as starvation due to hidden terminals, early collision termination, and frequency backoffs. We rely on WARP radios to experimentally verify that switch is indeed possible, and use ns-3 simulations to study the impact of using switch to solve the aforementioned WiFi problems.

We ask the following question in this paper: Can the goals of centralized WiFi scheduling be achieved using purely distributed operations? We present a solution called Look Who's Talking (LWT) that allows for arbitrary schedules to be distributed to nodes in a WiFi network. The nodes in the network then use purely local and distributed operations to achieve the prescribed schedule. The scope of LWT in this paper is restricted to a single collision domain (single or multiple cells), but we discuss how LWT can be extended to multiple collision domains. We use both experimental evaluations (using a WARP-based testbed) and simulation-based analysis (using ns3) to evaluate LWT.

The ubiquitous adoption of WiFi implicitly introduces large diversity in types of application requirements and topological characteristics. Consequently, considerable attention is being devoted to making WiFi networks controllable without compromising their scalability. Within this broad paradigm, we propose Rhythm, a MAC protocol that achieves scheduled WiFi efficiently and that is subject to the following constraints: (i) It does not need fine-grained time synchronization, (ii) it adds no "active listening" time, (iii) it does not need to gather the queue status from clients, and (iv) it requires no additional hardware. It also has the following properties: (i) low overhead, (ii) work conservation, (iii) robustness to partial connectivity, and (iv) backward compatibility.

As the popularity of wireless communication increases rapidly, the unlicensed Industrial Scientific and Medical (ISM) band becomes crowded. Interference avoidance thus becomes an important issue when designing wireless communication in the unlicensed ISM band. As shown in related works, channel hopping with blacklisting is a promising mechanism addressing interference in 802.15.4e time slotted channel hopping (TSCH) networks. However, the blacklisting mechanism changes the number of channels used, and thus requiring the frequency hopping sequence (FHS) be regenerated. A time-random channel selection mechanism avoids regeneration overhead; however, it breaks close-to-optimality property of the default FHSs, which in turns leads to high probability of interference. In this paper, we propose a new algorithm that generates FHSs in presence of interference without regeneration overhead, and at the same time maintains the optimality property. With the new proposed mechanism, we can generate FHSs with one designed original sequence. Analysis and simulation results show that optimal/near optimal properties are achieved by these new FHSs.

In this paper, we introduce the concept of "route robustness" for path selection in multi-hop cognitive radio networks. We demonstrate that the aggregate throughput and the robustness of routes determined by the proposed route selection strategy are superior to existing rate-based selection strategies. The rationale behind our approach is to guarantee a basic level of robustness for a set of routes (referred to as skeletons in this paper). Then, we select some routes from this robust route set and determine the spectrum to be allocated on each link along these routes such that the system throughput is maximized. We also design a polynomial time algorithm for this problem, and evaluate our proposed mechanism via simulations. The results show that our proposed algorithm indeed achieves a near optimal solution of this problem for multi-hop overlay CR networks.

In this paper, we introduce the concept of "route robustness" for path selection in multi-hop cognitive radio networks. We demonstrate that the aggregate throughput and robustness for routes determined by the introduced route robustness selection strategy are much better than those determined by other selection strategies as proposed in existing work. The rationale behind our approach is to guarantee a basic level of robustness for each route. We then jointly determine the routes and the spectrum allocation on each link along the routes such that the system throughput is maximized. The performance of the proposed mechanism is evaluated via simulations. The results show that our proposed mechanism can achieve better performance than other routing strategies.