Motivation

SDR presents a hard real-time computing challenge. The first generation of SDR mobile terminals (SDR-MTs) will be very limited in computing resources (processing powers, interprocessor bandwidths, memory, power, etc.) and will, most probably, not be capable of supporting more than one RAT implementation at a time. The flexibility of these terminals is then a function of the capability of their reconfiguration managers, which need to track the states of the computing resources. On the contrary, the computing resources of SDR base stations (SDR-BSs) are less limited. For example, there are no space and energy limitations, so that as many processing devices as necessary can be employed. Nevertheless, an optimization of computing resources would be highly desirable to reduce the operational cost of SDR-BSs. The potentially large number of users and the platforms’ high degrees of flexibility, modularity, and reconfigurability make the computing resource management at SDR-BSs equally important though more complex than that at SDR-MTs. In general, the higher the flexibility, the more complex becomes the (computing resource) management.

An SDR processing chain, SDR application or waveform, is the part of an SDR transceiver that is implemented in software. It can be understood as a set of concurrent processes that continuously process and propagate real-time data. Such a processing chain is not specifically tailored but rather executable on any general-purpose platform with sufficient computing capacity. We therefore consider general-purpose computing methods practical for SDR systems because of the similarities between future’s SDR applications and platforms and today’s general-purpose computing applications and platforms. We particularly believe that the introduction of appropriate mapping and scheduling techniques, which are indispensable for the dynamic switch between RATs, will leverage the design of SDR platforms and applications. Mapping describes the process of assigning software modules to hardware resources, whereas scheduling determines their execution times. We consider them to be two complementary computing resource management methods. Wireless or SDR systems, however, reveal specific aspects, essentially regarding flexibility and efficiency, that have not been jointly considered so far in heterogeneous computing:

1. Time slot based division of the transmission medium (radio time slot);
2. Continuous data transmission and reception;
3. RAT-specific quality of service (QoS) targets;
4. Real-time computing requirements and limited computing resources;
5. Different constraints and computing loads for different RATs and radio conditions;
6. Dynamic reconfiguration of the protocol stack, either partial or total;
7. Heterogeneous multiprocessor execution platforms.

Our computing resoure management proposal consist of several modules. These are illustrates in the figure and are described in the corresponding links. The basis for the resource management is the time management?. The SDR computing system modeling track the SDR platforms' available computing resource and the SDR applications' computing requirements (computing resource models?). These models are used by a general-purpose algorithm and a specific cost function, which manages (keeps track of and updates) the available computing resources and requirements (computing resource management?).

titulo 2

hola

hola adios

In this page you will find information related with the computing resources management framework incorporated to PHAL-OE

kkljklk

EstoEsUnLink?