Cooperation Versus Multiplexing: Multicast
Scheduling Algorithms for OFDMA Relay Networks
Abstract
With the next-generation cellular networks making atransition toward smaller cells, two-hop orthogonal frequency-divisionmultiple access (OFDMA) relay networks have become adominant, mandatory component in the 4G standards (WiMAX802.16j, 3GPP LTE-Adv). While unicast flows have received reasonableattention in two-hop OFDMA relay networks, not muchlight has been shed on the design of efficient scheduling algorithmsfor multicast flows. Given the growing importance of multimediabroadcast and multicast services (MBMS) in 4G networks, thelatter forms the focus of this paper. We show that while relaycooperation is critical for improving multicast performance, itmust be carefully balanced with the ability to multiplex multicastsessions and hence maximize aggregate multicast flow. To this end,we highlight strategies that carefully group relays for cooperationto achieve this balance. We then solve the multicast schedulingproblem under two OFDMA sub channelization models. We establishthe NP-hardness of the scheduling problem even for thesimpler model and provide efficient algorithms with approximationguarantees under both models. Evaluation of the proposedsolutions reveals the efficiency of the scheduling algorithms as wellas the significant benefits obtained from the multicasting strategy.
Architecture
Algorithm
Efficient Scheduling Algorithm:
The designof efficient scheduling algorithms for multicast traffic forms avital component of MBMS and in turn forms the focus of thispaper. Multicasting in two-hop relay networks is significantly different from the conventional cellular multicast: The broadcastadvantage of multicast data is significantly diminishedon the access (second) hop [Fig. 1(a)], where they becomeequivalent to multiple uni-cast transmissions from different RSto mobile stations (MS), thereby requiring more transmissionresources. Relay cooperation mechanisms allow multiple RSto simultaneously transmit the multicast data on the sametransmission resource. This helps retain the broadcast natureof the traffic on the access hop, making cooperation a criticalcomponent in improving multicast performance.
Existing System
WITH the next-generation wireless networks movingtoward smaller (micro, pico) cells for providing higherdata rates, there is a revived interest in multihop wirelessnetworks from the perspective of integrating them with cellularnetworks. With a decrease in cell size, relay stations (RS) arenow needed to provide extended coverage. In this context,two-hop relay-enabled wireless networks havebecome a dominant, mandatory component in the 4G standards (WiMAX 802.16m, 3GPP LTE-Adv) due to the plethoraof envisioned applications (hotspots, office buildings, undergroundtunnel access, etc.) they support.
Disadvantages:
1. Only return a single channel quality value.
2. No multicast scheduling.
Proposed System
Evaluation of the proposedsolutions reveals the efficiency of the scheduling algorithms as well
as the significant benefits obtained from the multicasting strategy. We evaluate the proposed solutions in an event-driven simulator that incorporates realistic physical-layer effects.
Advantages:
- multicast scheduling,
- Multiuser diversity.
Modules:
1. Two-Hop relay networks
2. Relay Cooperation
3. Multicasting
4. OFDMA
1. Two-Hop relay networks:
Multicasting in two-hop relay networks is significantly different from the conventional cellular multicast: The broadcastadvantage of multicast data is significantly diminishedon the access where they becomeequivalent to multiple unicast transmissions from different RSto mobile stations (MS), thereby requiring more transmissionresources. Relay cooperation mechanisms allow multiple RSto simultaneously transmit the multicast data on the sametransmission resource. This helps retain the broadcast natureof the traffic on the access hop, making cooperation a criticalcomponent in improving multicast performance.
2. Relay Cooperation:
We show that there existsa subtle tradeoff between cooperation gains and the ability tomultiplex multicast sessions effectively, both of which are essentialfor maximizing the aggregate multicast system performance.We highlight how strategies that carefully group relaysfor cooperation are needed to address this tradeoff effectively.We then solve the core multicast scheduling problem, which requiresdetermining the allocation of subchannels to multicastsessions on both the relay and access hops such that both cooperationand multiplexing gains are leveraged to maximize themulticast system performance. In the process, motivated by recentrelay standards [1], [2], [9], we consider two models forhow subcarriers are grouped to form a subchannel in OFDMA:distributed (DP) and contiguous (CP) permutations.
3.Multicasting:
Unlike unicast works, the OFDMA schedulingworks on multicast data have largely been restricted to one-hop cellular networks. These solutions cannot be directlycarried over to relay networks, where the nature of multicasttraffic and its broadcast advantage is significantly altered onthe access hop. Multicasting with relays has received increasedattention recently. Information-theoretic works havelooked at capacity bounds for a multicast system with relays.
4. OFDMA:
Orthogonal frequency-division multiple access (OFDMA)has become the popular choice for air interface technologyin 4G networks. The entire spectrum is divided into multiple carriers (subchannels), allowing for multiple users to operatein tandem. This leads to several physical-layer and schedulingbenefits. The two-hop network model coupled withOFDMA provides several diversity (multiuser, channel, andcooperative) gains that can be leveraged through intelligentscheduling.
HARDWARE & SOFTWARE REQUIREMENTS:
HARDWARE REQUIREMENTS:
System: Pentium IV 2.4 GHz.
Hard Disk: 40 GB.
Floppy Drive: 1.44 Mb.
Monitor: 15 VGA Color.
Mouse: Logitech.
Ram: 512 MB.
SOFTWARE REQUIREMENTS:
Operating system : Windows XP Professional.
Coding Language: C#.NET