There has been burgeoning interest in wireless technologies that can use wider frequency spectrum. Technology advances, such as 802.11n and ultra-wideband (UWB), are pushing toward wider frequency bands. The analog-to-digital TV transition has made 100-250 MHz of digital whitespace bandwidth available for unlicensed access. Also, recent work on WiFi networks has advocated discarding the notion of channelization and allowing all nodes to access the wide 802.11 spectrum in order to improve load balancing. This shift towards wider bands presents an opportunity to exploit frequency diversity. Specifically, frequencies that are far from each other in the spectrum have significantly different SNRs, and good frequencies differ across sender-receiver pairs. This paper presents FARA, a combined frequency-aware rate adaptation and MAC protocol. FARA makes three departures from conventional wireless network design: First, it presents a scheme to robustly compute per-frequency SNRs using normal data transmissions. Second, instead of using one bit rate per link, it enables a sender to adapt the bitrate independently across frequencies based on these per-frequency SNRs. Third, in contrast to traditional frequency-oblivious MAC protocols, it introduces a MAC protocol that allocates to a sender-receiver pair the frequencies that work best for that pair. We have implemented FARA in FPGA on a wideband 802.11-compatible radio platform. Our experiments reveal that FARA provides a 3.1 × throughput improvement in comparison to frequency-oblivious systems that occupy the same spectrum.

Abstract – With the advent of new FCC policies on spectrum allocation for next generation wireless devices, we have a rare opportunity to redesign spectrum access protocols to support demanding, latency-sensitive applications such as high-def media streaming in home networks. Given their low tolerance for traffic delays and disruptions, these applications are ill-suited for traditional, contention-based CSMA protocols. In this paper, we explore an alternative approach to spectrum access that relies on frequency-agile radios to perform interference-free transmission across orthogonal frequencies. We describe Jello, a MAC overlay where devices sense and occupy unused spectrum without central coordination or dedicated radio for control. We show that over time, spectrum fragmentation can significantly reduce usable spectrum in the system. Jello addresses this using two complementary techniques: online spectrum defragmentation, where active devices periodically migrate spectrum usage, and non-contiguous access, which allows a single flow to utilize multiple spectrum fragments. Our prototype on an 8-node GNU radio testbed shows that Jello significantly reduces spectrum fragmentation and provides high utilization while adapting to client flows ’ changing traffic demands. 1

Today’s local-area, mesh and cellular networks assign a single narrow-band channel to a node, and this assignment remains fixed over long time scales. Using network traces, we show that the load within a network can vary significantly even over short time scales on the order of tens of seconds. Therefore, we make the case for allocating spectrum ondemand to nodes and regions of the network that need it. We present an architecture that shares the entire spectrum on-demand using spread-spectrum codes. If implemented, the system will achieve fine-grained spectrum allocation for bursty traffic without requiring inter-cell coordination. Preliminary experiments suggest a throughput improvement of 75 % over commodity 802.11b networks. By eschewing the notion of channelization, and matching demand bursts with spectrum dynamically, better wireless networks that sustain higher throughputs may be designed. 1

Abstract — Demand-driven spectrum allocation can drastically improve performance for WiFi access points struggling under increasing user demands. While their frequency agility makes cognitive radios ideal for this challenge, performing adaptive spectrum allocation is a complex and difficult process. In this work, we propose FLEX, an efficient spectrum allocation architecture that efficiently adapts to dynamic traffic demands. FLEX tunes network-wide spectrum allocation by access points coordinating with peers, minimizing network resets through local adaptations. Through detailed analysis and experimental evaluation, we show that FLEX converges quickly, provides users with proportional-fair spectrum usage and significantly outperforms existing spectrum allocation proposals.

Over the past few years, researchers have developed many crosslayer wireless protocols to improve the performance of wireless networks. Experimental evaluations of these protocols have been carried out mostly using software-defined radios, which are typically two to three orders of magnitude slower than commodity hardware. FPGA-based platforms provide much better speeds but are quite difficult to modify because of the way high-speed designs are typically implemented. Experimenting with cross-layer protocols requires a flexible way to convey information beyond the data itself from lower to higher layers, and a way for higher layers to configure lower layers dynamically and within some latency bounds. One also needs to be able to modify a layer’s processing pipeline without triggering a cascade of changes. We have developed Airblue, an FPGA-based software radio platform, that has all these properties and runs at speeds comparable to commodity hardware. We discuss the design philosophy underlying Airblue that makes it relatively easy to modify it, and present early experimental results.

Abstract — In recent years we have witnessed a great interest in large distributed computing platforms, also known as clouds. While these systems offer enormous computing power, they are however major energy consumers. In the existing data centers CPUs are responsible for approximately half of the energy consumed by the servers. A promising technique for saving CPU energy consumption is dynamic speed scaling, in which the speed at which the processor is ran is adjusted based on demand and performance constraints. In this paper we look at the problem of allocating the demand in the network of processors (each being capable to perform dynamic speed scaling) to minimize the global energy consumption/cost. The nonlinear dependence between the energy consumption and the performance as well as the high variability in the energy prices result in a nontrivial resource allocation. The problem can be abstracted as a fully distributed convex optimization with a linear constraint. On the theoretical side, we propose two low-overhead fully decentralized algorithms for solving the problem of interest and provide closedform conditions that ensure stability of the algorithms. Then we evaluate the efficacy of the optimal solution using simulations driven by the real-world energy prices. Our findings indicate a possible cost reduction of 10 − 40 % compared to power-oblivious 1/N load balancing, for a wide range of load factors. I.

The throughput of a wireless network is often limited by interference caused by multiple concurrently active nodes. The conventional approach of using a “one-transmission-ata-time” MAC protocol to combat interference leads to a significant loss of achievable throughput compared to schemes such as interference cancellation that keep all transmitters active simultaneously. Unfortunately, interference cancellation incurs significant computational complexity, and cannot be implemented with commodity hardware. In this paper, we propose a practical approach for improving the throughput of interfering nodes using variable-width frequency allocation. We show that variable-width channels provide significant theoretical capacity improvements, comparable to interference cancellation for infrastructure networks. We design an algorithm that reduces interference by assigning orthogonal variable-width channels to transmitters. We evaluate a prototype implementation of this algorithm on an outdoor wireless network with ten long-distance links configured into point-to-point and point-to-multipoint topologies. We observe a throughput improvement of between 30 % and 110 % compared to the existing fixed-width channel allocation. 1

Spectrum frequency allocation problems are fundamental problems in wireless spectrum auctions and wireless LAN management. Due to their complexity, most existing proposals simplify the allocation problems by reducing physical interference to a graph-based interference model. In this paper, we propose LIGHTHOUSE, a new line of efficient approximation algorithms that operate directly on physical interference models and perform within a constant bound from the optimum under geometric signal propagation. Our design is motivated by the fact that conventional greedy methods become brittle under physical interference models although they perform well under graph interference models. We overcome such brittleness by building a new globalized optimization path, first reducing the optimization constraints into linear constraints to produce a starting solution and then applying iterative improvements to approach the global optimum. To our best knowledge, our solution is the first to achieve a constant approximation bound. It also has low complexity and supports a wide-range of optimization goals. Experiments show that our solution outperforms existing solutions by 50 + % in utilization, and is within 10 % gap from the optimal solution. 1

Abstract—IEEE 802.11n is a newly emerging wireless standard which, relative to former legacy 802.11 protocols, such as 802.11a/b/g, has a number of new mechanisms that enable multifold increases in transmission speeds. These new mechanisms include multiple-input-multiple-output (MIMO) transmission schemes as well as the option for channel bonding two 20MHz channels into a single 40MHz channel. Furthermore, 802.11n has the option of operating in the 5GHz range, which offers a larger number of subchannels to operate on in comparison to operating in the 2.4GHz range, as is the case for 802.11b/g. Given the vast bandwidth improvements 802.11n brings, vendors as well as users in wireless deployments are quickly migrating towards this new technology. The features available in 802.11n provide greater opportunities to exploit the existing bandwidth. Management of available bandwidth can be done through efficient channel management strategies. Significant work has been done on channel management schemes for legacy clients in wireless deployments; however, these strategies fall short to 802.11n features. Therefore, in this work, we propose a baseline 802.11n channel management scheme that utilizes methods from previous channel assignment solutions designed for legacy clients. We evaluate the behavior of our system under changing environment conditions, and, in so doing, we study the effects of environment conditions on 802.11n channel performance. As a result, we define the characteristics of an 802.11n channel management strategy and propose an algorithm REVIVE that meets these characteristics and that is both robust and adaptive. The contribution of our work is twofold: we provide important insight into the behavior of 802.11n devices and its implications on how the channel management scheme could address such behavior, and we propose REVIVE which is the first adaptive, robust channel management strategy for 802.11n wireless deployments. I.

How to distribute radio spectrum across network nodes is a critical problem in spectrum auctions and management. In this paper, we consider the problem of distributing spectrum using SINR-driven physical interference models. We propose Optimus, a new line of approximation algorithms that perform within a constant distance of min {2α + 1, 10} from the optimum in terms of spectrum usage efficiency, where α ≥ 2 is the pathloss exponent. Different from conventional greedy solutions, Optimus applies a global optimization mechanism that transforms the spatial interference constraints into a set of linear constraints, reducing the original optimization into a linear/convex/separableprogramming problem. While linearization techniques have been applied in prior works, Optimus makes a new and important contribution by deriving a highly efficient constraint transformation applicable to general network configurations. Experiments using real network measurements and sophisticated propagation models show that Optimus outperforms existing solutions by 20–50 % in spectrum utilization and is within 20 % gap from the optimum. Optimus supports a wide variety of objective functions, and is applicable to many spectrum-driven applications such as spectrum auctions and spectrum admission control. 1.