Base Station Deployment and Resource Allocation in Sustainable Wireless Networks

1Zhongming Zheng, 1Shibo He, 2Lin X. Cai, and 1Xuemin (Sherman) Shen

1Department of Electrical and Computer Engineering

University of Waterloo2School of Engineering and Applied Science

Princeton University

HANDBOOK ON GREEN INFORMATION AND COMMUNICATION SYSTEMS

Chapter 16

2

• Introduction

• Literature Review

• System Model

• Problem Formulation

• TCGBP Algorithm

• Numerical Results

• Conclusion & Future Work

Outline

Introduction

• Energy Sources

– Renewable Energy

• Repeatedly replenished

• Examples: wind, solar

– Non-renewable Energy:

• Once depleted, no more available

• Examples: coal, natural gas

3

Introduction

• Green Wireless Communication Networks– WLAN mesh network structure

4

5

• Introduction

• Literature Review

• System Model

• Problem Formulation

• TCGBP Algorithm

• Numerical Results

• Conclusion & Future Work

Outline

Literature Review

• Device Design– PV systems

• [1] Probabilistic methods

• [2] Simulation model

– Energy charging and discharging models• [3] Battery/energy buffer

• [4] Power consumption model of BSs

6

[1] H. A. M. Maghraby, M. H. Shwehdi, and G. K. Al-Bassam, “Probabilistic assessment of photovoltaic (pv) generation systems,” Power Systems, IEEE Transactions on, vol. 17, no. 1, pp. 205–208, Feb. 2002.[2] E. Lorenzo and L. Navarte, “On the usefulness of stand-alone pv sizing methods,” Progress in Photovoltaics: Research and Applications, vol. 8, no. 4, pp. 391–409, Aug. 2000.[3] L. X. Cai, Y. Liu, H. T. Luan, X. Shen, J. W. Mark, and H. V. Poor, “Adaptive resource management in sustainable energy powered wireless mesh networks,” in IEEE Globecom, Houston, TX, USA, Dec. 5-9 2011, pp. 1–5.[4] O. Arnold, F. Richter, G. Fettweis, and O. Blume, “Power consumption modeling of different base station types in heterogeneous cellular networks,” in Future Network & Mobile Summit, Florence, IT, Jun. 16-18 2010, pp. 1–8.

Literature Review

• Minimal Device Deployment– Continuous Case

• Direct search

• [5] Quasi-Newton methods

– Discrete Case• [6] Sustainability

• [7] Outage free

7

[5] G. L. Z. Wei and L. Qi, “New quasi-newton methods for unconstrained optimization problems,” Applied Mathematics and Computation, vol. 175, no. 2, pp. 1156–1188, Apr. 2006.[6] Z. Zheng, L. X. Cai, M. Dong, X. Shen, and H. V. Poor, “Constrained energyaware ap placement with rate adaptation in wlan mesh networks,” in IEEE GLOBECOM, Houston, TX, USA, Dec. 5-9 2011, pp. 1–5. [7] S. A. Shariatmadari, A. A. Sayegh, and T. D. Todd, “Energy aware basestation placement in solar powered sensor networks,” in IEEE WCNC, Sydney, AUS, Apr. 18-21 2010, pp. 1–6.

Literature Review

• Resource Allocation– Scheme Design

• [8] Traffic scheduling

• [9] Admission control and routing

• [10] Power control

8

[8] A. A. Hammad, G. H. Badawy, T. D. Todd, A. A. Sayegh, and D. Zhao, “Traffic scheduling for energy sustainable vehicular infrastructure,” in IEEE GLOBECOM, Miami, FL, USA, Dec. 6-10 2010, pp. 1–6.[9] L. Lin, N. B. Shroff, and R. Srikant, “Asymptotically optimal energy-aware routing for multihop wireless networks with renewable energy sources,” Networking, IEEE/ACM Transactions on, vol. 15, no. 5, pp. 1021–1034, Oct. 2007.[10] A. Farbod and T. D. Todd, “Resource allocation and outage control for solarpowered wlan mesh networks,” Mobile Computing, IEEE Transactions on, vol. 6, no. 8, pp. 960–970, Aug. 2007.

9

• Introduction

• Literature Review

• System Model

• Problem Formulation

• TCGBP Algorithm

• Numerical Results

• Conclusion & Future Work

Outline

System Model

• Given a set of BSs, users and candidate locations

• All users are associated with a BS

• BSs are powered by renewable energy

• BSs and users may have different power levels of charging and transmission

• In a WLAN, BS and its associated users use the same transmission power

10

System Model

• No inter-WLAN interference with orthogonal channels assigned to BSs for inter-WLAN communication

• BSs can only be placed at a given set of candidate locations

• BSs at different candidate locations have different charging capabilities

11

12

• Introduction

• Literature Review

• System Model

• Problem Formulation

• TCGBP Algorithm

• Numerical Results

• Conclusion & Future Work

Outline

Problem Formulation

13

The number of deployed BSs

Full coverage & Each user is associated with only one BS

Achieved throughput ≥ Traffic demandHarvested energy ≥ Consumed energy

Problem Formulation

• Initialization:

• Output:

14

Problem Formulation

• Problem Analysis

– Minimal BS placement problem with power allocation

– NP-hard problem• Sub-problems are NP-hard

– Optimal placement of BSs with a fixed power

– Power allocation of BSs

15

Problem Formulation

• Algorithm Design Strategy

– NP-hard → No solution in polynomial time– Design an effective heuristic algorithm

• Achieve good performance

• Reduce the time complexity

16

17

• Introduction

• Literature Review

• System Model

• Problem Formulation

• TCGBP Algorithm

• Numerical Results

• Conclusion & Future Work

Outline

TCGBP Algorithm

• First Phase

– Partition the whole network region into several VPs (Voronoi Polygons)

– Place one BS in each candidate location– Connect users to the BS in the same VP region

18

TCGBP Algorithm

• First Phase

19

TCGBP Algorithm

• Second Phase

– Connect BSs and users in neighboring VP regions until constraints can not be held

– Return the result when all users are connected

20

TCGBP Algorithm

• Second Phase

21

TCGBP Algorithm

22

Phase II

Phase I

TCGBP Algorithm

23

24

• Introduction

• Literature Review

• System Model

• Problem Formulation

• TCGBP Algorithm

• Numerical Results

• Conclusion & Future Work

Outline

Numerical Results

• Simulation Configurations

25

Parameter Value

WLAN mesh networks 100 m × 100 m

Transmission power levels 10 dBm, 15 dBm, 20 dBm

Charging capability [20, 30] mW per slot

Time duration 1000 slots

Channel bandwidth 40 MHz

Path loss exponent 4

Background noise -20 dBm

Numerical Results

Different numbers of users and traffic demands

26

Numerical Results

Different numbers of candidate locations and charging capabilities

27

28

• Introduction

• Literature Review

• System Model

• Problem Formulation

• TCGBP Algorithm

• Numerical Results

• Conclusion & Future Work

Outline

Conclusion

• Green energy sources

• Formulate an optimal green BS placement problem

• Propose TCGBP algorithm

– Approach the optimal solution with significantly reduced time complexity

29

Future Work

• Study the impacts of dynamics in the energy charging and discharging process

• Analyze the network capacity bounds under different deployment strategies

30