Galati Giordano Lorenzo
Present position: Temporary Researcher at DEI - Politecnico di Milano
|Thesis title:||New Advanced Techniques for Radio Resource Management|
In the next future, wireless communication systems will move towards the achievement of high transmission rates for both fixed and mobile users. The main topics of this thesis are the investigation of adaptive radio resource allocation strategies in the presence of a finite number of modulation and coding profiles and the evaluation of their impact on the interference distribution in a multi-user/multi-cellular wireless network based on Orthogonal Frequency Division Multiplexing Access (OFDMA). These kinds of allocation procedures are based on the assumption of having a perfect knowledge of the Channel State Indicator (CSI) and they involve many degrees of freedom (i.e. frequency, time, space, modulation, coding and power). Each allocation algorithm could be built following many different strategies and according to the required Quality of Service (QoS), fairness and complexity criteria.
The content of the thesis mainly covers the following points:
1. Investigation of the resource allocation approaches operating in the frequency domain, namely suited to OFDMA systems. Proposal of novel algorithms, comparison with existing methods and impact on the system performance (coverage and throughput) in a multi-cellular scenario. Moreover, advantages and drawbacks of the strategies based on the maximization of the transmission rate or on the minimization of the transmission power.
2. Resource allocation approaches operating in two dimensions, frequency (i.e. sub-channels) and time slots. Different classes of algorithms are implemented: they are based on heuristic approaches and they extend some of the OFDMA sub-optimal solutions to the two dimensional context. This part was validated by means of a dynamic system simulator for IEEE 802.16 in which traffic model, QoS and scheduling are properly implemented.
3. Investigation on the relation between resource allocation (in terms of frequency, power and fairness criteria) and generated interference inside a wireless multi-cellular network. This relation is not trivial since, when allocation algorithms allow a reduction of the average transmitted power in order to adapt it to the nearest modulation and coding scheme, the overall network interference tends to decrease, modifying the SINRs distribution and causing a new update in the assignment of the available resources. The system acts as a loop that converges to a stable solution after a certain number of iterations.