for Efficient Wireless Assistant Professor Department of Electrical Engineering Indian Institute of Technology Madras Joint work with: M. Chandrashekar V. Sandeep Parimal Parag for March 17, 2006
Broadband Wireless for Internet Wireless Cellular Wireless Local Area Network (WLAN) Demand for mobile wireless internet access Need support for multimedia data transfer
Broadband Wireless: Challenges for t t High data rates and limited/expensive spectrum need high spectral efficiency Shared resources and multiple access Multipath fading channel Bursty traffic characteristics
Broadband Wireless: Some Proposals Code-Division Multiple Access (CDMA)-based 1xEV-DO (HDR) HSDPA 1xEV-DV Orthogonal Frequency Division Multiplexing (OFDM)-based: FlashOFDM IEEE 802.16e IEEE 802.20 for
Key Techniques for User 1 User 2 User 3 Periodic reallocation of resources Adaptation to channel and traffic conditions Dynamic resource allocation Reallocation period of the order of a few milliseconds
Downlink Resource User 1 for Traffic Basestation User 2 User K Channel information Physical resources: power and bandwidth Maximize system throughput Total transmit power constraint Fairness or Quality of Service (QoS) constraints
Maximizing Capacity Channel 1 User 1 for Basestation Select the user with best channel Channel 2 User 2 Channel K User K All power and bandwidth resources to one user User with best achievable rate chosen: i = arg max R k, k where R k is the rate that can be supported by user k. No fairness or QoS constraint
Maximizing Capacity: Parallel Channels Parallel Channels to each user User 1 for Basestation For each parallel channel Select the user with best channel User 2 User K Bandwidth resources split to achieve parallel channels For each channel n, user with best channel conditions chosen: i n = arg max R k,n. k Water-filling power allocation No fairness or QoS constraint
Fairness and Quality of Service (QoS) Various notions of fairness or QoS Round-Robin Proportional Fairness [Tse02] i = arg max k R k R k,av, where R k,av is the average rate that can be supported by user k. Modified-Largest Weighted Delay First (M-LWDF) [Andrews00] i = arg max γ k W k R k, k where W k is the Head-Of-Line (HOL) packet delay for user k, and γ k = C k R k,av. for
Resource in OFDM for Subcarriers User 1 User 2 User 3 Power Available resources: Subcarriers Transmit power Channel is frequency-selective subcarriers not identical.
Algorithms Channel Aware Only (CAO) Scheduling Proportionally Fair (PF) subcarrier allocation [Rhee00] PF subcarrier allocation + power optimization [Shen05] Max utility subcarrier allocation + power optimization [Song05] Channel Aware Queue Aware (CAQA) Scheduling MLWDF for OFDM-TDMA [Andrews00] MLWDF at subcarrier level [Parag05] Our Work Joint Subcarrier and Power (JSPA) approach Optimize power allocation after each subcarrier is allocated for
MLWDF for OFDM-TDMA All subcarriers allocated to a single user in each slot Select user i as: i = arg max γ k W k R k k W k : Head-Of-Line (HOL) packet delay for user k Rk : Rate achievable for user k (water-filling) γk = C k R k,av Ck = log δ k D k to achieve P[delay > D k ] < δ k Throughput optimal single-user scheduling rule Maximum stability: achieves stable queues if any algorithm can achieve it Single-user scheduling in each time slot not optimal for
Subcarrier-wise Approach 1: MLWDF at the subcarrier level [Parag05] For each subcarrier n: i n = arg max γ k W k R k,n k W k : Head-Of-Line (HOL) packet delay for user k R k,n : Rate achievable for user k on subcarrier n Power allocation needed to allocate subcarriers Uniform/fixed power allocation assumption Approach 2: [Song04] Mean packet waiting time instead of Head-Of-Line (HOL) packet delay Other utility functions based on mean packet waiting time for
Joint Subcarrier and Power Start Start for Split power equally amongst subcarriers Allocate all subcarriers to users Optimize power allocation with power Yes Split power equally amongst subcarriers Check if all subcarriers are allocated No Allocate a subcarrier to a user Optimize power allocation with power Update each user s queue Update each user s queue End End
Joint Subcarrier and Power (JSPA) Optimal JSPA too complex Sub-optimal JSPA Power optimization after each subcarrier is allocated leads to better allocation of the remaining suncarriers Power allocation to each user proportional to the number of subcarriers allocated HOL delay is estimated after each subcarrier is allocated Some practical constraints included Discrete-rate constraint: Integer bit M-QAM constellations Extension to band-wise allocation: reduced signaling/feedback for
Simulation Setup for 128 subcarrier OFDM system 12 users, Bernoulli packet arrival, 100 slot buffer 6-tap multipath channel, average channel conditions are different for each user QPSK to 64-QAM
: Throughput vs. Arrival Rate Throughput (Mbps) 7 6 5 4 3 P total = 5 dbw Homogenous rate users CAO+FPA CAO+FPA+PAO CAO+JSPA MLWDF CAQA+FPA CAQA+JSPA for 2 1 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 Arrival rate (Mbps)
: Max. Arrival Rate vs. Transmit Power Arrival rate (Mbps) 7.5 7 6.5 6 5.5 5 4.5 4 3.5 Homogenous rate users CAO+FPA CAO+FPA+PAO CAO+JSPA MLWDF CAQA+FPA CAQA+JSPA for 3 2.5 4 4.5 5 5.5 6 6.5 7 7.5 8 P total in dbw Max. arrival rate for less than 0.5% packets dropped
: Delay Performance P(delay>x) 10 1 10 0 10 1 P total = 8dBW Arrival rate = 3 Mbps Homogenous rate users CAO+FPA CAO+FPA+PAO CAO+JSPA MLWDF CAQA+FPA CAQA+JSPA for 5 10 15 20 25 30 35 40 delay (time slots) Best and worst delay performance among users plotted
: Band-wise Arrival rate (Mbps) 6 5.5 P = 5dBW total 5 4.5 4 L = 1 L = 2 L = 4 L = 8 L = 16 for 3.5 3 1 1.5 2 2.5 3 3.5 Delay spread (rms) µsec L = Number of subcarriers in a sub-band
Resource in CDMA for Spreading codes Power User 1 User 2 Available resources: Spreading codes Transmit power For any given user, all spreading codes are similar (in terms of channel conditions).
Multiuser Scheduling for Most algorithms are single-user scheduling algorithms Proportionally Fair [Tse02] MLWDF [Andrews00] Recent results on multi-user scheduling algorithms Greedy and pairwise greedy allocation [Kumaran05] Gradient-based scheduling [Agrawal04]
: Maximum Supportable Traffic 0.4 MLWDF TWO USER for Fraction of packets dropped 0.3 0.2 0.1 0 0 1 2 3 4 5 6 7 8 9 Average Arrival Rate (Mbps)
: Delay Performance 10 0 MLWDF TWO USER for Prob[delay > d] 10 1 10 2 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 Delay d (in sec)
Dynamic resource allocation is essential to achieve high spectral efficiency Adaptation based on both channel and traffic information Some new results for OFDM and CDMA systems Joint subcarrier and power allocation in OFDM Multiuser scheduling in CDMA Several open problems: Optimality Quantifying the signaling/feedback overhead for
E. F. Chaponniere, P. Black, J. M. Holtzman, and D. Tse, Transmitter directed multiple receiver system using path diversity to equitably maximize throughput, U. S. Patent No. 6449490, September 2002. M. Andrews, K. Kumaran, K. Ramanan, A. Stolyar, R. Vijayakumar, and P. Whiting, CDMA data QoS scheduling on the forward link with variable channel conditions, Bell Labs Technical Memorandum, 2000. W. Rhee and J. M. Cioffi, Increase in capacity of multiuser OFDM system using dynamic subchannel allocation, in Proceedings of the 51st IEEE Vehicular Technology Conference, vol. 2, Spring 2000, pp. 1085 1089. Z. Shen, J. G. Andrews, and B. L. Evans, Adaptive resource allocation in multiuser OFDM systems with proportional rate constraints, IEEE Transactions on Wireless Communications, vol. 4, no. 6, pp. 2726 2737, for
G. Song and Y. Li, Cross-layer optimization for OFDM wireless networks-part II: Algorithm development, IEEE Transactions on Wireless Communications, vol. 4, no. 2, pp. 625 634, March 2005. P. Parag, S. Bhashyam, and R. Aravind, A subcarrier allocation algorithm for OFDMA using buffer and channel state information, in Proceedings of the 62 nd IEEE Vehicular Technology Conference, vol. 1, September 2005, pp. 622 625. G. Song, Y. G. Li, J. L. J. Cimini, and H. Zheng, Joint channel-aware and queue-aware data scheduling in multiple shared wireless channels, in Proceedings of the IEEE Wireless Communications and Networking Conference, vol. 3, March 2004, pp. 1939 1944. for
for R. Agrawal, V. Subramanian, and R. Berry, Joint scheduling and resource allocation in CDMA systems, in 2nd Workshop on Modeling and Optimization in Mobile, Ad Hoc, and Wireless Networks (WiOpt 04), Cambridge, UK, March 2004. K. Kumaran and H. Viswanathan, Joint power and bandwidth allocation in downlink transmission, IEEE Transactions on Wireless Communications, vol. 4, no. 3, pp. 1008 1016, May 2005.