QoS Support in 3G/LTE Femtocells
Project Funded By Argela
In the near future, femtocells are likely to enter homes to increase coverage, spectrum efficiency, as well as data rates.
A typical femtocell configuration is a simplified node-B (3G/LTE base station). A femtocell is likely to serve multiple clients and/or services for one client. A scheduling module in the MAC-hs unit of a femtocell is very likely to increase performance dramatically since each service served by the femtocell is likely to have very different QoS/QoE requirements. Traditionally a 3G/LTE scheduler uses the channel quality information to schedule users in time. While users with better channel conditions are favored, a compromise is usually made by the scheduler to provide some notion of fairness. Ultimately, the primary goal is spectral efficiency – QoS plays a distant second role, if any (and usually none). However, the femtocell setup is dramatically different. In the scheduler design, channel quality is no longer the primary parameter here. Femtocell schedulers need to be QoS/QoE based without entirely compromising the spectrum efficiency. This project is concerned with the design of schedulers that enable QoS differentiation in femtocells.
Wireless Broadcasting in the Broadband Regime
The work we have presented in [2] has focused on the design of a wireless air interface where the bandwidth is limited to current 2G and 3G wireless system specifications of 1.25 MHz. In this light, we investigated whether scalable video coding (SVC) algorithms are useful in broadcasting. When the bandwidth is increased, as will be the case in 4G networks, the air interface we have designed will need a complete re-design. The additional bandwidth dimension in the resource allocation problem will require the design of two-dimensional schedulers that will allocate system resources to various video quality layers in the code space as well as bandwidth. We propose to investigate OFDM-like air interface designs for efficient video broadcasting in this scenario.
Wireless Broadcasting over an Ad-Hoc Network
We consider the transmission of a signal from a source peer to every other peer in an ad hoc wireless network. The common practice in this broadcasting problem is to send the transmission to all nodes in a single hop and to fix a data rate such that every user can receive the data correctly. In order to increase the transmission data rate, which is a nonlinear function of the transmission power, distance and the interference observed, multiple hops can be introduced at the expense of an increase in the maximum observed delay in the system. In this scheme, peers who are capable of decoding the transmitted signal may re-broadcast it. However, re-broadcasting peers cause interference to one another while trying the capture the source’s transmission which leads to a reduction in the data rate. Then it becomes important to determine which peers should re-broadcast. This leads to the problem of finding the subset of cooperating peers and the number of hops to maximize the data rate and to minimize the maximum observed delay in the system.
