wasm
Moderator
Italy
8 Posts |
Posted - 24/11/2004 : 14:19:07
|
UWB transmissions can be put into practice with different kinds of spreading approaches, based on time repetition of single pulses.
In classical DS approach, a narrowband signal is generated by a square wave rb(t) equal to the length of the data bit Tb. A second square wave rc(t) is then generated based on a random code (whose time extension is Tb) and it is multiplied with the single square.
The processing gain of this system can be defined as PG = Tb/Tc with Tc = Tb/Nc (Nc the number of chips of the random code), while the bandwidth of the two process can be approximated as Bb=1/Tb and Bc=1/Tc according to the Shannon bandwidth that practically is the bandwidth between first lobe of the sinc (Fourier(rect) = sinc). When raised cosine waveform are used instead of a rectangle this assumption becomes still more valid.
With these assumption the effective signal-to-noise ratio can be written
SNR = Ps/(N0 Bc) = Eb/(N0 Bc Tb) --> Eb/(N0 PG)
In common wideband systems the time ratio defined as the processing gain and the bandwidth ratio Bb/Bc are inverses of each other, in UWB systems they are independent properties. This means that the formula before the arrow for calculating the SNR is still valid also for UWB systems, while the formula after the arrow cannot be used anymore.
In DSSS spreading technique, chips are generated by digital circuits, I mean that single chip it is generated by the state or state-transition of a digital circuit, so to obtain a narrower “pulse” you have to increase the speed of the digital section… Once you have a so fast digital processor, to gain all the advantages of the DSSS you must use continuous PN sequence, in which there is no pause between two adjacent chips.
In DS-UWB the chip shape (and then the frequency spreading) is produced by an analog gated circuit. To the digital section it is devolved only the function to drive this “analog” pulse generator, so you can choose a digital processor that satisfy the modulation bandwidth requirements but not the spreading rate as required in the DSSS.
Slower digital section it is (can be) required in UWB systems
Obviously when no continuous pulse spreading code it is chosen, also other spreading techniques are available as for instance TH. Successively, when it is preferable to use one techniques instead of another one will be explained.
Now we known that the advantages of the UWB systems resides in the UWB pulse, the next step is to discover the UWB pulse property and the different spreading approach properties… |
Edited by - wasm on 24/11/2004 14:26:00 |
|