## Joint congestion control, routing and MAC for stability and fairness in wireless networks (2006)

Venue: | IEEE Journal on Selected Areas in Communications |

Citations: | 58 - 8 self |

### BibTeX

@ARTICLE{Eryilmaz06jointcongestion,

author = {A. Eryilmaz and R. Srikant},

title = {Joint congestion control, routing and MAC for stability and fairness in wireless networks},

journal = {IEEE Journal on Selected Areas in Communications},

year = {2006},

volume = {24},

pages = {1514--1524}

}

### Years of Citing Articles

### OpenURL

### Abstract

In this work, we describe and analyze a joint scheduling, routing and congestion control mecha-nism for wireless networks, that asymptotically guarantees stability of the buffers and fair allocation of the network resources. The queue lengths serve as common information to different layers of the network protocol stack. Our main contribution is to prove the asymptotic optimality of a primal-dual congestion controller, which is known to model different versions of TCP well.

### Citations

1371 |
Rate control for communication networks: shadow prices, proportional fairness and stability
- Kelly, Maulloo, et al.
- 1998
(Show Context)
Citation Context ...ormation and Decision Systems at the Massachusetts Institute of Technology {eryilmaz@mit.edu}. R. Srikant is with the University of Illinois at Urbana, Champaign {rsrikant@uiuc.edu}. was developed in =-=[12]-=-, and followed by others in [18], [34], [14], [1], [30]; see [24] for a survey. In these works, the main contribution was the design of a distributed congestion control mechanism to drive the rates of... |

538 | Stability Properties of Constrained Queueing Systems and Scheduling Policies for Maximum throughput in Multihop Radio Networks
- Tassiulas, Ephremides
- 1992
(Show Context)
Citation Context ...ers, and drives the mean flow rates to a system-wide fair allocation point. The question of designing stable scheduling algorithms for wireless networks was first addressed by Tassiulas and Ephremides=-=[28]-=- under the assumption that the incoming flows are inelastic, i.e., the flow rates are fixed as for voice or video traffic. They showed that scheduling transmissions as a function of the buffer occupan... |

517 | Optimization Flow Control, I: Basic Algorithm and Convergence
- Low, Lapsley
- 1999
(Show Context)
Citation Context ... the Massachusetts Institute of Technology {eryilmaz@mit.edu}. R. Srikant is with the University of Illinois at Urbana, Champaign {rsrikant@uiuc.edu}. was developed in [12], and followed by others in =-=[18]-=-, [34], [14], [1], [30]; see [24] for a survey. In these works, the main contribution was the design of a distributed congestion control mechanism to drive the rates of elastic flows towards the syste... |

465 | Fair end-to-end window-based congestion control
- Mo, Walrand
- 2000
(Show Context)
Citation Context ...nctions. Uf (x) = βf x1−αf (1 − αf ) w 2 ∀ αf > 0. (2) This class of utility functions is known to characterize a large class of fairness concepts including weighted-proportional and max-min fairness =-=[19]-=-. Next, we describe the capacity region of the network as in [21], [15]. Definition 1 (Capacity region): The capacity region, Λ, of the network contains the set of flow rates x ≥ 0 for which � there e... |

301 |
The Mathematics of Internet Congestion Control
- Srikant
- 2004
(Show Context)
Citation Context ...echnology {eryilmaz@mit.edu}. R. Srikant is with the University of Illinois at Urbana, Champaign {rsrikant@uiuc.edu}. was developed in [12], and followed by others in [18], [34], [14], [1], [30]; see =-=[24]-=- for a survey. In these works, the main contribution was the design of a distributed congestion control mechanism to drive the rates of elastic flows towards the system-wide optimum. In [35], [6], the... |

219 | The impact of imperfect scheduling on cross-layer congestion control in wireless networks - Lin, Shroff - 2006 |

207 | Dynamic power allocation and routing for time varying wireless networks
- Neely, Modiano, et al.
- 2005
(Show Context)
Citation Context ...ity (stability) region of the network. Subsequently, there has been a large body of work that extended the same idea to different scenarios and more general settings[29], [26], [27], [2], [23], [11], =-=[21]-=-, [10]. However, these works do not consider the case of traffic whose rate can be adjusted online. In the context of wireline networks, the idea of a distributed flow control based on a system-wide o... |

183 | Dynamic server allocation to parallel queues with randomly varying connectivity
- Tassiulas, Ephremides
- 1993
(Show Context)
Citation Context ... arrival rates lie within the capacity (stability) region of the network. Subsequently, there has been a large body of work that extended the same idea to different scenarios and more general settings=-=[29]-=-, [26], [27], [2], [23], [11], [21], [10]. However, these works do not consider the case of traffic whose rate can be adjusted online. In the context of wireline networks, the idea of a distributed fl... |

178 | A game theoretic framework for bandwidth allocation and pricing in broadband networks
- Yaïche, Mazumdar, et al.
(Show Context)
Citation Context ...assachusetts Institute of Technology {eryilmaz@mit.edu}. R. Srikant is with the University of Illinois at Urbana, Champaign {rsrikant@uiuc.edu}. was developed in [12], and followed by others in [18], =-=[34]-=-, [14], [1], [30]; see [24] for a survey. In these works, the main contribution was the design of a distributed congestion control mechanism to drive the rates of elastic flows towards the system-wide... |

148 | Fairness and optimal stochastic control for heterogeneous networks
- Neely, Modiano, et al.
(Show Context)
Citation Context ...l characteristics of wireless networks are not fully addressed there. More recently, the problem of serving elastic traffic over wireless networks has been investigated in [22], [25], [15], [7], [9], =-=[20]-=-, [16], [5]. Here, the queues and the wireless characteristics of the network are included in the system model. The main idea in these works has been to combine the results on scheduling inelastic tra... |

145 |
Linear complexity algorithms for maximum throughput in radionetworks and input queued switches
- Tassiulas
- 1998
(Show Context)
Citation Context ...es lie within the capacity (stability) region of the network. Subsequently, there has been a large body of work that extended the same idea to different scenarios and more general settings[29], [26], =-=[27]-=-, [2], [23], [11], [21], [10]. However, these works do not consider the case of traffic whose rate can be adjusted online. In the context of wireline networks, the idea of a distributed flow control b... |

133 | Scheduling for multiple flows sharing a time-varying channel: The exponential rule
- Shakkottai, Stolyar
- 2002
(Show Context)
Citation Context ...in the capacity (stability) region of the network. Subsequently, there has been a large body of work that extended the same idea to different scenarios and more general settings[29], [26], [27], [2], =-=[23]-=-, [11], [21], [10]. However, these works do not consider the case of traffic whose rate can be adjusted online. In the context of wireline networks, the idea of a distributed flow control based on a s... |

132 | Maximizing queueing network utility subject to stability: Greedy primal-dual algorithm - Stolyar - 2005 |

125 | Fair resource allocation in wireless networks using queue-length-based scheduling and congestion control
- Eryilmaz, Srikant
- 2005
(Show Context)
Citation Context ...ential characteristics of wireless networks are not fully addressed there. More recently, the problem of serving elastic traffic over wireless networks has been investigated in [22], [25], [15], [7], =-=[9]-=-, [20], [16], [5]. Here, the queues and the wireless characteristics of the network are included in the system model. The main idea in these works has been to combine the results on scheduling inelast... |

106 |
Nonlinear Systems, 3rd ed
- Khalil
- 2002
(Show Context)
Citation Context ...at will be useful in our subsequent analysis. Consider the differential equation: , where is a locally Lipschitz function. Then, the following theorem can be used to determine its asymptotic behavior =-=[13]-=-. Theorem 1 (LaSalle’s Invariance Principle): Let be a radially unbounded, 1 continuously differentiable, positive definite2 function such that for all . Let be the set of points in where . Let be the... |

96 | Hop-by-hop congestion control over a wireless multi-hop network
- Yi, Shakkottai
- 2004
(Show Context)
Citation Context ...[30]; see [24] for a survey. In these works, the main contribution was the design of a distributed congestion control mechanism to drive the rates of elastic flows towards the system-wide optimum. In =-=[35]-=-, [6], the authors use this idea to develop congestion control algorithms for wireless environments by reducing the available capacity region and converting the network into essentially a wireline net... |

96 | Crosslayer congestion control, routing and scheduling design in ad hoc wireless networks
- Chen, Low, et al.
- 2006
(Show Context)
Citation Context ...reline network. The essential characteristics of wireless networks are not fully addressed there. More recently, the problem of serving elastic traffic over wireless networks has been investigated in =-=[5,7,8,14,15,19,23,28]-=-. Here, the queues and the wireless characteristics of the network are included in the system model. The main idea in these works has been to combine the results on scheduling inelastic traffic in wir... |

86 | Joint rate control and scheduling in multihop wireless networks
- Lin, Shroff
- 2004
(Show Context)
Citation Context ...rk. The essential characteristics of wireless networks are not fully addressed there. More recently, the problem of serving elastic traffic over wireless networks has been investigated in [22], [25], =-=[15]-=-, [7], [9], [20], [16], [5]. Here, the queues and the wireless characteristics of the network are included in the system model. The main idea in these works has been to combine the results on scheduli... |

83 | Maxmin fair scheduling in wireless networks
- Tassiulas, Sarkar
(Show Context)
Citation Context ...reline network. The essential characteristics of wireless networks are not fully addressed there. More recently, the problem of serving elastic traffic over wireless networks has been investigated in =-=[22]-=-, [25], [15], [7], [9], [20], [16], [5]. Here, the queues and the wireless characteristics of the network are included in the system model. The main idea in these works has been to combine the results... |

79 | Joint congestion control and media access control design for ad hoc wireless networks
- Chen, Low, et al.
- 2005
(Show Context)
Citation Context ...see [24] for a survey. In these works, the main contribution was the design of a distributed congestion control mechanism to drive the rates of elastic flows towards the system-wide optimum. In [35], =-=[6]-=-, the authors use this idea to develop congestion control algorithms for wireless environments by reducing the available capacity region and converting the network into essentially a wireline network.... |

75 | Throughput guarantees through maximal scheduling in wireless networks
- Chaporkar, Kar, et al.
- 2005
(Show Context)
Citation Context ...s implemented at the nodes. Here, it must be stressed that even though the congestion control is distributed, the scheduling is still assumed to be centralized in this work. In [16], [31], [3], [32], =-=[4]-=-, the impact of decentralized implementations of the scheduler is studied. We note that the results of this work can be extended to distributed and asynchronous implementations for a special class of ... |

72 |
Scheduling in a queueing system with asynchronously varying service rates,” Prob
- Andrews, Kumaran, et al.
- 2004
(Show Context)
Citation Context ... within the capacity (stability) region of the network. Subsequently, there has been a large body of work that extended the same idea to different scenarios and more general settings[29], [26], [27], =-=[2]-=-, [23], [11], [21], [10]. However, these works do not consider the case of traffic whose rate can be adjusted online. In the context of wireline networks, the idea of a distributed flow control based ... |

70 |
A Time-Scale Decomposition Approach to Adaptive Explicit Congestion Notification (ECN
- Kunniyur, Srikant
- 2002
(Show Context)
Citation Context ...usetts Institute of Technology {eryilmaz@mit.edu}. R. Srikant is with the University of Illinois at Urbana, Champaign {rsrikant@uiuc.edu}. was developed in [12], and followed by others in [18], [34], =-=[14]-=-, [1], [30]; see [24] for a survey. In these works, the main contribution was the design of a distributed congestion control mechanism to drive the rates of elastic flows towards the system-wide optim... |

66 | Stable scheduling policies for fading wireless channels
- Eryilmaz, Srikant, et al.
- 2005
(Show Context)
Citation Context ...tability) region of the network. Subsequently, there has been a large body of work that extended the same idea to different scenarios and more general settings[29], [26], [27], [2], [23], [11], [21], =-=[10]-=-. However, these works do not consider the case of traffic whose rate can be adjusted online. In the context of wireline networks, the idea of a distributed flow control based on a system-wide optimiz... |

56 |
Scheduling and performance limits of networks with constantly changing topology
- Tassiulas
- 1997
(Show Context)
Citation Context ...es are fixed as for voice or video traffic. They showed that scheduling transmissions as a function of the buffer occupancies (queue-lengths) naturally leads to the stability of the buffers. Tassiulas=-=[26]-=- extended this technique to derive a joint routing and scheduling algorithm that ensures the stability of the queues. These results showed that the queue-length-based resource allocation guarantees st... |

52 | A utility-based congestion control scheme for Internet-style networks with delay
- Alpcan, Basar
- 2005
(Show Context)
Citation Context ... Institute of Technology {eryilmaz@mit.edu}. R. Srikant is with the University of Illinois at Urbana, Champaign {rsrikant@uiuc.edu}. was developed in [12], and followed by others in [18], [34], [14], =-=[1]-=-, [30]; see [24] for a survey. In these works, the main contribution was the design of a distributed congestion control mechanism to drive the rates of elastic flows towards the system-wide optimum. I... |

52 | A unifying passivity framework for network flow control
- Wen, Arcak
- 2003
(Show Context)
Citation Context ...itute of Technology {eryilmaz@mit.edu}. R. Srikant is with the University of Illinois at Urbana, Champaign {rsrikant@uiuc.edu}. was developed in [12], and followed by others in [18], [34], [14], [1], =-=[30]-=-; see [24] for a survey. In these works, the main contribution was the design of a distributed congestion control mechanism to drive the rates of elastic flows towards the system-wide optimum. In [35]... |

51 |
Regulated maximal matching: A distributed scheduling algorithm for multi-hop wireless networks with nodeexclusive spectrum sharing
- Wu, Srikant
- 2005
(Show Context)
Citation Context ...on problem that is implemented at the nodes. Here, it must be stressed that even though the congestion control is distributed, the scheduling is still assumed to be centralized in this work. In [16], =-=[31]-=-, [3], [32], [4], the impact of decentralized implementations of the scheduler is studied. We note that the results of this work can be extended to distributed and asynchronous implementations for a s... |

46 | To layer or not to layer: Balancing transport and physical layers in wireless multihop networks
- Chiang
- 2004
(Show Context)
Citation Context ...e essential characteristics of wireless networks are not fully addressed there. More recently, the problem of serving elastic traffic over wireless networks has been investigated in [22], [25], [15], =-=[7]-=-, [9], [20], [16], [5]. Here, the queues and the wireless characteristics of the network are included in the system model. The main idea in these works has been to combine the results on scheduling in... |

38 | Joint Asynchronous Congestion Control and Distributed Scheduling for Multi-Hop Wireless Networks
- Bui, Eryilmaz, et al.
- 2006
(Show Context)
Citation Context ...blem that is implemented at the nodes. Here, it must be stressed that even though the congestion control is distributed, the scheduling is still assumed to be centralized in this work. In [16], [31], =-=[3]-=-, [32], [4], the impact of decentralized implementations of the scheduler is studied. We note that the results of this work can be extended to distributed and asynchronous implementations for a specia... |

34 |
Bounds on the capacity region of multi-hop wireless networks under distributed greedy scheduling
- Wu, Srikant
- 2006
(Show Context)
Citation Context ...that is implemented at the nodes. Here, it must be stressed that even though the congestion control is distributed, the scheduling is still assumed to be centralized in this work. In [16], [31], [3], =-=[32]-=-, [4], the impact of decentralized implementations of the scheduler is studied. We note that the results of this work can be extended to distributed and asynchronous implementations for a special clas... |

23 |
Joint optimal congestion control, routing, and scheduling in wireless ad hoc networks
- Chen, SH, et al.
- 2006
(Show Context)
Citation Context ...stics of wireless networks are not fully addressed there. More recently, the problem of serving elastic traffic over wireless networks has been investigated in [22], [25], [15], [7], [9], [20], [16], =-=[5]-=-. Here, the queues and the wireless characteristics of the network are included in the system model. The main idea in these works has been to combine the results on scheduling inelastic traffic in wir... |

7 | Optimal distributed power control, routing, and congestion control in wireless networks.” Submitted to
- Xi, Yeh
- 2005
(Show Context)
Citation Context ...ed to distributed and asynchronous implementations for a special class of interference models using the approach in [3]. Finally, we note that a related, but different, problem has been considered in =-=[33]-=- where a distributed algorithm has been designed to route inelastic flows to minimize delay costs in a wireless network. The rest of the paper is organized as follows: Section II describes the system ... |

6 |
Throughput optimal scheduling for broadcast channels
- Eryilmaz, Srikant, et al.
- 2001
(Show Context)
Citation Context ... capacity (stability) region of the network. Subsequently, there has been a large body of work that extended the same idea to different scenarios and more general settings[29], [26], [27], [2], [23], =-=[11]-=-, [21], [10]. However, these works do not consider the case of traffic whose rate can be adjusted online. In the context of wireline networks, the idea of a distributed flow control based on a system-... |

5 |
Efficient and fair scheduling for wireless networks
- Eryilmaz
- 2005
(Show Context)
Citation Context ...ome γ ∈ (0, 1) that is determined as a function of σ ∈ (0, 1). Again, this condition is satisfied by the class of utility functions described by (2). Then, we can state the following two results (see =-=[8]-=-): Result 1: There exists positive constants ¯c < ∞ and σ ∈ (0, 1), that depend on Λ, the utility function set {Uf (·)}, and the moments of the arrival processes, such that for each λ ⋆ ∈ KΨ ⋆ , E [�q... |

3 | 2003) Exponential RED: A stabilizing AQM scheme for low and high speed TCP, University of Illinois Tech Report. An earlier version appeared
- Liu, Bas¸ar, et al.
- 2005
(Show Context)
Citation Context ... so called Primal-Dual Congestion Controller in this work. Primal-dual algorithms are well known in the optimization literature and have been studied extensively in different contexts[1], [30], [24], =-=[17]-=-. Since the response of the primal-dual controller is more gradual compared to the dual controller, it is not immediately clear as to whether the buffer stability and rate convergence properties will ... |

2 |
on the capacity region of multi-hop wireless networks under distributed greedy scheduling
- “Bounds
- 2006
(Show Context)
Citation Context ...mplemented at the nodes. Here, it must be stressed that even though the congestion control is distributed, the scheduling is still assumed to be centralized in this work. In [3], [4], [16], [31], and =-=[32]-=-, the impact of decentralized implementations of the scheduler is studied. We note that the results of this work can be extended to distributed and asynchronous implementations for a special class of ... |

1 |
Nonlinear Systems. 3nd edition
- Khalil
- 2002
(Show Context)
Citation Context ...l in our subsequent analysis. Consider the differential equation: ˙y(t) = f(y(t)), where f is a locally Lipschitz function. Then, the following theorem can be used to determine its asymptotic behavior=-=[13]-=-. Theorem 1 (LaSalle’s Invariance Principle): Let Y : D → R be a radially unbounded 1 , continuously differentiable, positive definite 2 function such that Y ˙ (z) ≤ 0 for all z ∈ D. Let E be the set ... |

1 |
optimal scheduling for broadcast channels
- “Throughput
(Show Context)
Citation Context ...es lie within the capacity (stability) region of the network. Subsequently, there has been a large body of work that extended the same idea to different scenarios and more general settings [2], [10], =-=[11]-=-, [21], [23], Manuscript received August 30, 2005; revised April 16, 2006. This work was supported in part by the National Science Foundation under Grant ECS-0401125 and Grant CNS-05-19691. A. Eryilma... |