Jinsung Lee

Implementing utility-optimal CSMA

Hundreds of papers over the last two decades have studied the theory of distributed scheduling in wireless networks, including a number of them on stability or utility maximizing random access. Several publications in 2008 studied an adaptive CSMA that in theory can approach utility optimality without any message passing under a number of assumptions. This paper reports the results from the first deployment of such random access algorithms through an implementation over conventional 802.11 hardware, an on-going effort that started in summer 2009. It shows both a confirmation that utility optimal CSMA may work well in practice even with an implementation over legacy equipment, and a wide array of gaps between theory and practice in the field of wireless scheduling. This paper therefore also brainstorms the discovery of and bridging over these gaps, and the implementation-inspired questions on modeling and analysis of scheduling algorithms.

Making 802.11 DCF near-optimal: design, implementation, and evaluation

This paper proposes a new protocol called Optimal DCF (O-DCF). O-DCF modifies the rule of adapting CSMA parameters, such as backoff time and transmission length, based on a function of the demand--supply differential of link capacity captured by the local queue length. O-DCF is fully compatible with 802.11 hardware, so that it can be easily implemented only with a simple device driver update. O-DCF is inspired by the recent analytical studies proven to be optimal under assumptions, which often generates a big gap between theory and practice. O-DCF effectively bridges such a gap, which is implemented in off-the-shelf 802.11 chipset. Through extensive simulations and real experiments with a 16-node wireless network testbed, we evaluate the performance of O-DCF and show that it achieves near-optimality in terms of throughput and fairness and outperforms other competitive ones, such as 802.11 DCF, optimal CSMA, and DiffQ for various scenarios. Also, we consider the coexistence of O-DCF and 802.11 DCF and show that O-DCF fairly shares the medium with 802.11 via its parameter control.

Making 802.11 DCF near-optimal: Design, implementation, and evaluation

This paper proposes a new protocol called Optimal DCF (O-DCF). O-DCF modifies the rule of adapting CSMA parameters, such as backoff time and transmission length, based on a function of the demand-supply differential of link capacity captured by the local queue length. O-DCF is fully compatible with 802.11 hardware, so that it can be easily implemented only with a simple device driver update. O-DCF is inspired by the recent analytical studies proven to be optimal under assumptions, which often generates a big gap between theory and practice. O-DCF effectively bridges such a gap, which is implemented in off-the-shelf 802.11 chipset. Through extensive simulations and real experiments with a 16-node wireless network testbed, we evaluate the performance of O-DCF and show that it achieves near-optimality in terms of throughput and fairness and outperforms other competitive ones, such as 802.11 DCF, optimal CSMA, and DiffQ for various scenarios. Also, we consider the coexistence of O-DCF and 802.11 DCF and show that O-DCF fairly shares the medium with 802.11 via its parameter control.

Improving TCP Performance over Optimal CSMA in Wireless Multi-Hop Networks

Optimal CSMA (oCSMA) has been receiving extensive attentions for its provable optimality in throughput and fairness without message passing over wireless multi-hop networks. However, recent studies suggest that TCP over oCSMA performs poorly, hindering the deployment of oCSMA in real networks. In this letter, we show that just a simple, additional virtual queue at the MAC layer can significantly improve TCP performance when oCSMA is used as the underlying MAC. Through testbed-based experiments, we demonstrate that with our virtual queueing scheme, TCP flows achieve near-optimal throughput performance in various scenarios.

3D MIMO Outdoor to Indoor Macro/Micro-Cellular Channel Measurements and Modeling

3-dimensional Multiple-Input Multiple-Output (3D MIMO) systems have received great interest recently because of the spatial diversity advantage and capability for full-dimensional beamforming, making them promising candidates for practical realization of massive MIMO. For 3D MIMO system design, it is important to have full characterization of the 3D MIMO propagation channel, i.e., characterize both the elevation and azimuth characteristics of the wireless propagation channel, especially at the base station end. In this paper, we present first results from a measurement campaign for obtaining these characteristics. For those measurements we use a hybrid switched/virtual cylindrical array with 480 antenna elements at the base station (BS), and a switched array with 24 antenna elements at the user equipment (UE). We perform outdoor-to-indoor (O2I) channel measurements in an urban macro-cellular (UMa) and a micro-cellular (UMi) environments. We provide the elevation and azimuth angular spreads at the transmitter and receiver, and study their dependence on the UE height. With the increase in the UE height, the BS elevation spreads decreased from 1.24 deg to 1 deg and 1.15 deg to 0.78 deg respectively for UMa and UMi; the azimuth spreads remained approximately the same (7.5 deg). Based on the measurements done with the UE placed on different floors, we study the feasibility of separating users in the elevation domain. Users were separable in 44% and 54.17% scenarios respectively for the UMa and UMi environments.

A Group of People Acts like a Black Body in a Wireless Mesh Network

A wireless mesh network (WMN) is being considered for commercial use in spite of several unaddressed issues. In this paper we focus on one of the most critical issues: the impact of ambient motion of entities like people on the channel characteristics and on the WMN performance. A human body in an electro-magnetic (EM) field acts as an scatterer that absorbs 60% of incident EM energy, thereby shadowing the receiver. This human body model along with the human mobility behavior gives rise to a black body (a group movement) effect that traps the incident EM wave with repetitive internal reflections. The black body theory is verified by simulating the WiSEMesh testbed in picoKAIST, a tool based on deterministic ray tube method. Experimental results show each link exhibiting a unique channel variation pattern in presence of the black body. Based on the pattern we provide several insights in WMN deployment and protocol design.

Commoncode: a code-reuse platform for wireless network experimentation

Experimentation of a wireless network protocol over the air is of significant interest. However, it is more rarely performed than simulation because of the difficulties in coding and debugging as well as lack of scalability and repeatability. In this article, the concept of a code-reuse platform making use of a simulation code directly for real experiments is revisited as an efficient and easy means of experimentation. Furthermore, an architecture and key components of an idealistic code-reuse platform are suggested, and then CommonCode, the most advanced codereuse platform, is proposed. Through extensive simulations and experiments using Common-Code for the same codes, we demonstrate that CommonCode is valid and accurate in terms of protocol performance, and simultaneously fast and easy in terms of protocol development.

CommonCode: a code reuse platform for co-simulation and experimentation

Experimentation of a wireless network protocol over the air is of significant interest. However it is rarely performed compared to the simulation because of the difficulties in coding and debugging as well as scalability and repeatability. CommonCode is a unique protocol validation platform reusing the simulation codes for the experiments. It helps fast, convenient and accurate validation of protocols.

Towards optimal MAC without message passing in wireless networks

One of the key messages in significant research on cross-layer design of wireless networks over the past 15 years is that extensive message passing may be necessary to achieve optimal performance. Heavy message passing is obviously an undesirable feature for practical implementation due to its reduced actual throughput and security issues. In this paper, we first summarize our preliminary work that does not need message passing, yet achieves near-optimality. Then, we present our recent efforts to realize such an algorithmic solution in a real network, e.g., IEEE 802.11-based systems, and discuss possible future research directions.

A TCP starvation problem in combining TCP and max-weight scheduling of cross-layer algorithms in WMNs

In this paper, we investigate the problem coming from the adaptation of legacy protocol TCP to the cross-layer protocols that is likely happen in future Internet. We point out named a TCP starvation problem in combining TCP and max-weight scheduling of cross-layer algorithms in wireless mesh networks. TCP starvation occurs when neighbor nodes are already scheduled with large queue length. To overcome TCP starvation as denoted in this paper, we propose a simple remedy named fake-ACK algorithm.