Stochastic optimization for power-aware distributed scheduling Michael J. Neely University of Southern California t ω(t)

Slides:



Advertisements
Similar presentations
Ch. 12 Routing in Switched Networks
Advertisements

Mobility Increase the Capacity of Ad-hoc Wireless Network Matthias Gossglauser / David Tse Infocom 2001.
Ch. 12 Routing in Switched Networks Routing in Packet Switched Networks Routing Algorithm Requirements –Correctness –Simplicity –Robustness--the.
Winter 2004 UCSC CMPE252B1 CMPE 257: Wireless and Mobile Networking SET 3f: Medium Access Control Protocols.
Network Utility Maximization over Partially Observable Markov Channels 1 1 Channel State 1 = ? Channel State 2 = ? Channel State 3 = ? Restless.
Delay Reduction via Lagrange Multipliers in Stochastic Network Optimization Longbo Huang Michael J. Neely WiOpt *Sponsored in part by NSF.
EE 685 presentation Optimal Control of Wireless Networks with Finite Buffers By Long Bao Le, Eytan Modiano and Ness B. Shroff.
DYNAMIC POWER ALLOCATION AND ROUTING FOR TIME-VARYING WIRELESS NETWORKS Michael J. Neely, Eytan Modiano and Charles E.Rohrs Presented by Ruogu Li Department.
Stochastic Network Optimization with Non-Convex Utilities and Costs Michael J. Neely University of Southern California
Intelligent Packet Dropping for Optimal Energy-Delay Tradeoffs for Wireless Michael J. Neely University of Southern California
Dynamic Product Assembly and Inventory Control for Maximum Profit Michael J. Neely, Longbo Huang (University of Southern California) Proc. IEEE Conf. on.
Matthew Andrews Bell Labs June 27, 2005 Scheduling High Speed Data in (Adversarial) Wireless Networks.
Dynamic Index Coding Broadcast Station N N Michael J. Neely, Arash Saber Tehrani, Zhen Zhang University of Southern California Paper available.
Universal Scheduling for Networks with Arbitrary Traffic, Channels, and Mobility Michael J. Neely, University of Southern California Proc. IEEE Conf. on.
Kuang-Hao Liu et al Presented by Xin Che 11/18/09.
Efficient Algorithms for Renewable Energy Allocation to Delay Tolerant Consumers Michael J. Neely, Arash Saber Tehrani, Alexandros G. Dimakis University.
Utility Optimization for Dynamic Peer-to-Peer Networks with Tit-for-Tat Constraints Michael J. Neely, Leana Golubchik University of Southern California.
Stock Market Trading Via Stochastic Network Optimization Michael J. Neely (University of Southern California) Proc. IEEE Conf. on Decision and Control.
Delay-Based Network Utility Maximization Michael J. Neely University of Southern California IEEE INFOCOM, San Diego, March.
Queuing Theory For Dummies
Dynamic Optimization and Learning for Renewal Systems Michael J. Neely, University of Southern California Asilomar Conference on Signals, Systems, and.
Dynamic Index Coding User set N Packet set P Broadcast Station N N p p p Michael J. Neely, Arash Saber Tehrani, Zhen Zhang University.
Dynamic Optimization and Learning for Renewal Systems -- With applications to Wireless Networks and Peer-to-Peer Networks Michael J. Neely, University.
Max Weight Learning Algorithms with Application to Scheduling in Unknown Environments Michael J. Neely University of Southern California
Dynamic Data Compression for Wireless Transmission over a Fading Channel Michael J. Neely University of Southern California CISS 2008 *Sponsored in part.
*Sponsored in part by the DARPA IT-MANET Program, NSF OCE Opportunistic Scheduling with Reliability Guarantees in Cognitive Radio Networks Rahul.
Scheduling Algorithms for Wireless Ad-Hoc Sensor Networks Department of Electrical Engineering California Institute of Technology. [Cedric Florens, Robert.
Multi-Hop Networking with Hard Delay Constraints Michael J. Neely, University of Southern California DARPA IT-MANET Presentation, January 2011 PDF of paper.
Cross Layer Adaptive Control for Wireless Mesh Networks (and a theory of instantaneous capacity regions) Michael J. Neely, Rahul Urgaonkar University of.
1 40 th Annual CISS 2006 Conference on Information Sciences and Systems Some Optimization Trade-offs in Wireless Network Coding Yalin E. Sagduyu Anthony.
1 TDMA Scheduling in Competitive Wireless Networks Mario CagaljHai Zhan EPFL - I&C - LCA February 9, 2005.
Optimal Energy and Delay Tradeoffs for Multi-User Wireless Downlinks Michael J. Neely University of Southern California
A Lyapunov Optimization Approach to Repeated Stochastic Games Michael J. Neely University of Southern California Proc.
Seyed Mohamad Alavi, Chi Zhou, Yu Cheng Department of Electrical and Computer Engineering Illinois Institute of Technology, Chicago, IL, USA ICC 2009.
Resource Allocation for E-healthcare Applications
DaVinci: Dynamically Adaptive Virtual Networks for a Customized Internet Jennifer Rexford Princeton University With Jiayue He, Rui Zhang-Shen, Ying Li,
1 11 Subcarrier Allocation and Bit Loading Algorithms for OFDMA-Based Wireless Networks Gautam Kulkarni, Sachin Adlakha, Mani Srivastava UCLA IEEE Transactions.
Delay Analysis for Maximal Scheduling in Wireless Networks with Bursty Traffic Michael J. Neely University of Southern California INFOCOM 2008, Phoenix,
By Avinash Sridrahan, Scott Moeller and Bhaskar Krishnamachari.
A Non-Monetary Protocol for P2P Content Distribution in Wireless Broadcast Networks with Network Coding I-Hong Hou, Yao Liu, and Alex Sprintson Dept. of.
Utility-Optimal Scheduling in Time- Varying Wireless Networks with Delay Constraints I-Hong Hou P.R. Kumar University of Illinois, Urbana-Champaign 1/30.
Michael J. Neely, University of Southern California CISS, Princeton University, March 2012 Wireless Peer-to-Peer Scheduling.
Michael J. Neely, University of Southern California CISS, Princeton University, March 2012 Asynchronous Scheduling for.
Utility Maximization for Delay Constrained QoS in Wireless I-Hong Hou P.R. Kumar University of Illinois, Urbana-Champaign 1 /23.
Stochastic Optimal Networking: Energy, Delay, Fairness Michael J. Neely University of Southern California
DaVinci: Dynamically Adaptive Virtual Networks for a Customized Internet Jiayue He, Rui Zhang-Shen, Ying Li, Cheng-Yen Lee, Jennifer Rexford, and Mung.
Energy-Aware Wireless Scheduling with Near Optimal Backlog and Convergence Time Tradeoffs Michael J. Neely University of Southern California INFOCOM 2015,
Super-Fast Delay Tradeoffs for Utility Optimal Scheduling in Wireless Networks Michael J. Neely University of Southern California
ITMANET PI Meeting September 2009 ITMANET Nequ-IT Focus Talk (PI Neely): Reducing Delay in MANETS via Queue Engineering.
Information Theory for Mobile Ad-Hoc Networks (ITMANET): The FLoWS Project Competitive Scheduling in Wireless Networks with Correlated Channel State Ozan.
Fairness and Optimal Stochastic Control for Heterogeneous Networks Time-Varying Channels     U n (c) (t) R n (c) (t) n (c) sensor.
Order Optimal Delay for Opportunistic Scheduling In Multi-User Wireless Uplinks and Downlinks Michael J. Neely University of Southern California
Content caching and scheduling in wireless networks with elastic and inelastic traffic Group-VI 09CS CS CS30020 Performance Modelling in Computer.
Stochastic Optimization for Markov Modulated Networks with Application to Delay Constrained Wireless Scheduling Michael J. Neely University of Southern.
Delay Analysis for Max Weight Opportunistic Scheduling in Wireless Systems Michael J. Neely --- University of Southern California
October 28, 2005 Single User Wireless Scheduling Policies: Opportunism and Optimality Brian Smith and Sriram Vishwanath University of Texas at Austin October.
Energy Optimal Control for Time Varying Wireless Networks Michael J. Neely University of Southern California
Chance Constrained Robust Energy Efficiency in Cognitive Radio Networks with Channel Uncertainty Yongjun Xu and Xiaohui Zhao College of Communication Engineering,
Asynchronous Control for Coupled Markov Decision Systems Michael J. Neely University of Southern California Information Theory Workshop (ITW) Lausanne,
Self-Organized Resource Allocation in LTE Systems with Weighted Proportional Fairness I-Hong Hou and Chung Shue Chen.
Online Fractional Programming for Markov Decision Systems
Delay Efficient Wireless Networking
Howard Huang, Sivarama Venkatesan, and Harish Viswanathan
IEEE Student Paper Contest
energy requests a(t) renewable source s(t) non-renewable source x(t)
Scheduling Algorithms in Broad-Band Wireless Networks
Throughput-Optimal Broadcast in Dynamic Wireless Networks
Utility Optimization with “Super-Fast”
Data and Computer Communications
Javad Ghaderi, Tianxiong Ji and R. Srikant
Presentation transcript:

Stochastic optimization for power-aware distributed scheduling Michael J. Neely University of Southern California t ω(t)

Outline Lyapunov optimization method Power-aware wireless transmission – Basic problem – Cache-aware peering – Quality-aware video streaming Distributed sensor reporting and correlated scheduling

A single wireless device R(t) = r(P(t), ω(t)) Timeslots t = {0, 1, 2, …} ω(t) = Random channel state on slot t P(t) = Power used on slot t R(t) = Transmission rate on slot t (function of P(t), ω(t)) observedchosen

Example R(t) = log(1 + P(t)ω(t)) observed chosen t ω(t) t R(t)

Example R(t) = log(1 + P(t)ω(t)) observed chosen t ω(t) t R(t)

Example R(t) = log(1 + P(t)ω(t)) observed chosen t ω(t) t R(t)

Optimization problem Maximize: R Subject to: P ≤ c Given: Pr[ω(t)=ω] = π(ω), ω in {ω 1, ω 2, …, ω 1000 } p(t) in P = {p 1, p 2, …, p 5 } c = desired power constraint

Consider randomized decisions Pr[p k | ω i ] = Pr[P(t) = p k | ω(t)=ω i ] ω(t) in {ω 1, ω 2, …, ω 1000 } P(t) in P = {p 1, p 2, …, p 5 } ∑ Pr[p k | ω i ] = 1 ( for all ω i in {ω 1, ω 2, …, ω 1000 } ) k=1 5

Linear programming approach Max: R S.t. : P ≤ c Given parameters: π(ω i ) (1000 probabilities) r(p k, ω i ) (5*1000 coefficients) Optimization variables: Pr[p k |ω i ] (5*1000 variables) ∑ ∑ π(ω i ) Pr[p k |ω i ] r(p k,ω i ) 1000 i=1 k=1 5 ∑ ∑ π(ω i ) Pr[p k |ω i ] p k ≤ c i=1 k=1 Max: S.t.:

Multi-dimensional problem 1 1 Access Point Access Point 2 2 N N Observe (ω 1 (t), …, ω N (t)) Decisions: -- Choose which user to serve -- Choose which power to use R 1 (t) R 2 (t) R N (t)

Goal and LP approach Maximize: R 1 + R 2 + … + R N Subject to: P n ≤ c for all n in {1, …, N} LP has given parameters: π(ω 1, …, ω N ) (1000 N probabilities) r n (p k, ω i ) (N*5N*1000 N coefficients) LP has optimization variables: Pr[p k |ω i ] (5N*1000 N variables)

Advantages of LP approach Solves the problem of interest LPs have been around for a long time Many people are comfortable with LPs

Disadvantages of LP approach

Need to estimate an exponential number of probabilities. LP has exponential number of variables. What if probabilities change? Fairness? Delay? Channel errors?

Lyapunov optimization approach Maximize: R 1 + R 2 + … + R N Subject to: P n ≤ c for all n in {1, …, N}

Lyapunov optimization approach Maximize: R 1 + R 2 + … + R N Subject to: P n ≤ c for all n in {1, …, N} Virtual queue for each constraint: Stabilizing virtual queue  constraint satisfied! Q n (t+1) = max[Q n (t) + P n (t) – c, 0] Q n (t) P n (t) c

Lyapunov drift L(t) = ½ ∑ Q n (t) 2 Δ(t) = L(t+1) – L(t) n Q1Q1 Q2Q2

Drift-plus-penalty algorithm Every slot t: Observe (Q 1 (t), …., Q N (t)), (ω 1 (t), …, ω N (t)) Choose (P 1 (t), …, P N (t)) to greedily minimize: Update queues. Δ(t) - (1/ε)(R 1 (t) + … + R N (t)) drift penalty Low complexity No knowledge of π(ω) probabilities is required

Specific DPP implementation Each user n observes ω n (t), Q n (t). Each user n chooses P n (t) in P to minimize: -(1/ε)r n (P n (t), ω n (t)) + Q n (t)P n (t) Choose user n* with smallest such value. User n* transmits with power level P n* (t). Low complexity No knowledge of π(ω) probabilities is required

Performance Theorem Assume it is possible to satisfy the constraints. Then under DPP with any ε>0: All power constraints are satisfied. Average thruput satisfies: Average queue size satisfies: ∑ Q n ≤ O(1/ε) R 1 + … + R N ≥ throughput opt – O(ε)

General SNO problem Minimize: y 0 (α(t), ω(t)) Subject to: y n (α(t), ω(t)) ≤ 0 for all n in {1, …, N} α(t) in A ω(t) for all t in {0, 1, 2, …} Such problems are solved by the DPP algorithm. Performance theorem: O(ε), O(1/ε) tradeoff. ω(t) = Observed random event on slot t π(ω) = Pr[ω(t)=ω] (possibly unknown) α(t) = Control action on slot t A ω(t) = Abstract set of action options

What we have done so far Lyapunov optimization method Power-aware wireless transmission – Basic problem – Cache-aware peering – Quality-aware video streaming Distributed sensor reporting and correlated scheduling

What we have done so far Lyapunov optimization method Power-aware wireless transmission – Basic problem – Cache-aware peering – Quality-aware video streaming Distributed sensor reporting and correlated scheduling

What we have done so far Lyapunov optimization method Power-aware wireless transmission – Basic problem – Cache-aware peering – Quality-aware video streaming Distributed sensor reporting and correlated scheduling

What we have done so far Lyapunov optimization method Power-aware wireless transmission – Basic problem – Cache-aware peering – Quality-aware video streaming Distributed sensor reporting and correlated scheduling

Mobile P2P video downloads

Access Point Access Point

Mobile P2P video downloads Access Point Access Point

Mobile P2P video downloads Access Point Access Point

Mobile P2P video downloads Access Point Access Point Access Point Access Point

Mobile P2P video downloads Access Point Access Point Access Point Access Point Access Point Access Point

Mobile P2P video downloads Access Point Access Point Access Point Access Point Access Point Access Point

Mobile P2P video downloads Access Point Access Point Access Point Access Point Access Point Access Point

Mobile P2P video downloads Access Point Access Point Access Point Access Point Access Point Access Point

Mobile P2P video downloads Access Point Access Point Access Point Access Point Access Point Access Point

Mobile P2P video downloads Access Point Access Point Access Point Access Point Access Point Access Point

Cache-aware scheduling Access points (including “femto” nodes) Typically stationary Typically have many files cached Users Typically mobile Typically have fewer files cached Assume each user wants one “long” file Can opportunistically grab packets from any nearby user or access point that has the file.

Quality-aware video delivery Video chunks as time progresses Quality Layer 1 Quality Layer 2 Bits: 8176 D: Bits: 7370 D: Quality Layer L Bits: D: 0 Bits: D: Bits: D: Bits: D: Bits: D: Bits: D: 0 Bits: D: 0 Bits: D: Bits: D: Bits: D: 0 D = Distortion. Results hold for any matrices Bits( layer, chunk ), D( layer, chunk ). Bits are queued for wireless transmission.

Fair video quality delivery Minimize: f( D 1 ) + f( D 2 ) + … + f( D N ) Subject to: P n ≤ c for all n in {1, …, N} Video playback rate constraints

Fair video quality delivery Minimize: f( D 1 ) + f( D 2 ) + … + f( D N ) Subject to: P n ≤ c for all n in {1, …, N} Video playback rate constraints Recall the general form: Min: y 0 S.t. : y n ≤ 0 for all n α(t) in A ω(t) for all t

Fair video quality delivery Min: y 0 S.t. : y n ≤ 0 for all n α(t) in A ω(t) for all t Minimize: f( D 1 ) + f( D 2 ) + … + f( D N ) Subject to: P n ≤ c for all n in {1, …, N} Video playback rate constraints Recall the general form: Define Y n (t) = P n (t) - c

Fair video quality delivery Minimize: f( D 1 ) + f( D 2 ) + … + f( D N ) Subject to: P n ≤ c for all n in {1, …, N} Video playback rate constraints Recall the general form: Define auxiliary variable γ(t) in [0, D max ] Min: y 0 S.t. : y n ≤ 0 for all n α(t) in A ω(t) for all t

Equivalence via Jensen’s inequality Minimize: f( D 1 ) + f( D 2 ) + … + f( D N ) Subject to: P n ≤ c for all n in {1, …, N} Video playback rate constraints Minimize: f( γ 1 (t)) + f( γ 2 (t)) + … + f( γ N (t)) Subject to: P n ≤ c for all n in {1, …, N} γ n = D n for all n in {1, …, N} Video playback rate constraints

Example simulation BS Region divided into 20 x 20 subcells (only a portion shown here) mobile devices, 1 base station mobiles/subcell

Phases 1, 2, 3: File availability prob = 5%, 10%, 7% Basestation Average Traffic: 2.0 packets/slot Peer-to-Peer Average Traffic: packets/slot Factor of 77.8 gain compared to BS alone!

What we have done so far Lyapunov optimization method Power-aware wireless transmission – Basic problem – Cache-aware peering – Quality-aware video streaming Distributed sensor reporting and correlated scheduling

Distributed sensor reports ω i (t) = 0/1 if sensor i observes the event on slot t P i (t) = 0/1 if sensor i reports on slot t Utility: U(t) = min[P 1 (t)ω 1 (t) + (1/2)P 2 (t)ω 2 (t),1] Fusion Center Fusion Center Maximize: U Subject to: P 1 ≤ c P 2 ≤ c ω 1 (t) ω 2 (t)

What is optimal? Agree on plan t 4

What is optimal? Agree on plan t 4 Example plan: User 1: t=even  Do not report. t=odd  Report if ω 1 (t)=1. User 2: t=even  Report if ω 2 (t)=1 t=odd:  Report with prob ½ if ω 2 (t)=1

Common source of randomness Example: 1 slot = 1 day Each user looks at Boston Globe every day: If first letter is a “T”  Plan 1 If first letter is an “S”  Plan 2 Etc. Day 1 Day 2

Specific example Assume: Pr[ω 1 (t)=1] = ¾, Pr[ω 2 (t)=1] = ½ ω 1 (t), ω 2 (t) independent Power constraint c = 1/3 Approach 1: Independent reporting If ω 1 (t)=1, user 1 reports with probability θ 1 If ω 2 (t)=1, user 2 reports with probability θ 2 Optimizing θ 1, θ 2 gives u = 4/9 ≈

Approach 2: Correlated reporting Pure strategy 1: User 1 reports if and only if ω 1 (t)=1. User 2 does not report. Pure strategy 2: User 1 does not report. User 2 reports if and only if ω 2 (t)=1. Pure strategy 3: User 1 reports if and only if ω 1 (t)=1. User 2 reports if and only if ω 2 (t)=1.

Approach 2: Correlated reporting X(t) = iid random variable (commonly known): Pr[X(t)=1] = θ 1 Pr[X(t)=2] = θ 2 Pr[X(t)=3] = θ 3 On slot t: Users observe X(t) If X(t)=k, users use pure strategy k. Optimizing θ 1, θ 2, θ 3 gives u = 23/48 ≈

Summary of approaches Independent reporting Correlated reporting Centralized reporting Strategy u

Summary of approaches Independent reporting Correlated reporting Centralized reporting Strategy u It can be shown that this is optimal over all distributed strategies!

General distributed optimization Maximize: U Subject to: P k ≤ 0 for k in {1, …, K} ω(t) = (ω 1 (t), …, ω Ν (t)) π(ω) = Pr[ω(t) = (ω 1, …, ω Ν )] α(t) = (α 1 (t), …, α Ν (t)) U(t) = u(α(t), ω(t)) P k (t) = p k (α(t), ω(t))

Pure strategies A pure strategy is a deterministic vector- valued function: g(ω) = (g 1 (ω 1 ), g 2 (ω 2 ), …, g Ν (ω Ν )) Let M = # pure strategies: M = | A 1 | |Ω1| x | A 2 | |Ω2| x... x | A N | |ΩN|

Optimality Theorem There exist: K+1 pure strategies g (m) (ω) Probabilities θ 1, θ 2, …, θ K+1 such that the following distributed algorithm is optimal: X(t) = iid, Pr[X(t)=m] = θ m Each user observes X(t) If X(t)=m  use strategy g (m) (ω).

LP and complexity reduction The probabilities can be found by an LP Unfortunately, the LP has M variables If (ω 1 (t), …, ω Ν (t)) are mutually independent and the utility function satisfies a preferred action property, complexity can be reduced Example N=2 users, | A 1 |=| A 2 |=2 --Old complexity = 2 |Ω1|+|Ω2| --New complexity = (|Ω1|+1)(|Ω2|+1)

Lyapunov optimization approach Define K virtual queues Q 1 (t), …, Q K (t). Every slot t, observe queues and choose strategy m in {1, …, M} to maximize a weighted sum of queues. Update queues with delayed feedback: Q k (t+1) = max[Q k (t) + P k (t-D), 0]

Separable problems If the utility and penalty functions are a separable sum of functions of individual variables (α n (t), ω n (t)), then: There is no optimality gap between centralized and distributed algorithms Problem complexity reduces from exponential to linear.

Simulation (non-separable problem) 3-user problem α n (t) in {0, 1} for n ={1, 2, 3}. ω n (t) in {0, 1, 2, 3, 4, 5, 6, 7, 8, 9} V=1/ε Get O(ε) guarantee to optimality Convergence time depends on 1/ε

Utility versus V parameter (V=1/ε) Utility V (recall V = 1/ε)

Average power versus time Average power up to time t Time t power constraint 1/3 V=10 V=50 V=100

Adaptation to non-ergodic changes

Conclusions Drift-plus-penalty is a strong technique for general stochastic network optimization Power-aware scheduling Cache-aware scheduling Quality-aware video streaming Correlated scheduling for distributed stochastic optimization

Conclusions Drift-plus-penalty is a strong technique for general stochastic network optimization Power-aware scheduling Cache-aware scheduling Quality-aware video streaming Correlated scheduling for distributed stochastic optimization