Dynamic range of Power Amplifier.

Is there anyone here has detailed information about dynamic range in PA?I know that it is the ratio of Maximum Input Power of Amplifier over Minimum Input Power of Amplifier.If the output of my OFDM system connects to an Power Amplifier (regardless of D/A or other modules), how should I measure the dynamic range? Thanks.

The large dynamic range of OFDM systems presents a challenge for PA design.
Power Amplifiers have both linear and non-linear regions where the non-linear regions occur for large output powers (near saturation). To minimize the amount of distortion and to reduce the amount of out-of-band energy generated by the OFDM PA, need to operate as much as possible in the linear region. With this large dynamic range, means that OFDM must keep its average power well below the non-linear region in order to accommodate the signal power peaks.
Theoretically OFDM can get a peak-to-average of ~17dB, in case that all the subcarriers are ON at the same time (an extremely rare event). Must have in mind this exception analyzing the dynamic range.

Hi Vfone,
I have done some modifications on the OFDM system, and I able to reduce 3dB dynamic range. But, the PAPR of the system is remain the same due to some reasons. Is this improvement bring any benefits towards power amplifier? I know that PAPR is the ratio between Max. Power and Average Power, while dynamic range is the ratio between Max. Power and Min. Power. Which parameters is much more concerns during the design? This is because most of the authors concern about PAPR(or crest factor) while nearly non of them discuss about dynamic range. Because from my understanding, low PAPR doesn't means to have low dynamic range. Am i right? thx.

I know that in OFDM using special block codes as the subcarrier modulation rather than allowing the data to modulate the subcarriers directly can significantly reduce the peak-to-average ratio.
About the specification of the dynamic range for OFDM PA probably this is not written somewhere, or at least not for WLAN applications. Is not like CDMA where the transmitter dynamic range is specified very clear (74dB), because CDMA systems are based on changing quickly the output power levels.
You can calculate the PA dynamic range, when you know the Output Power range of the system (for WLAN aprox. +10dBm to +22dBm) and PAR (peak-to-average ratio). For example for proper operation in 802.11g mode, an amplifier requires a setback of –6dB to -8dB down from the P1dB compression point of the amplifier. This is required to keep the amplifier operating in its linear range.

Hi Vfon,
Thx. Actually, I am designing the OFDM system by using computer software and running simulation on it to check the PAPR or dynamic range for my system before and after modifications. I realised that from my simulation (I ploted the output of my OFDM system as 'power vs time'), the PAPR doesn't change due to some reasons. However, I reduced the maximum power which is about half of it. I am so curious to know whether it is a improvement, or just meaningless. If I said, having low PAPR doesn't mean that having low dynamic range, do you agree with me? Also, if I reduced my average power (at the output of OFDM) is there any disadvantages towards power amplifier(regardless of BER)? Once again thank you.

Probably in your simulation you didn’t use the right algorithm for reducing PAR. There are more options in reducing PAR in OFDM: controlling the amplitude and phase of subcariers, clipping the signal, or replacing peaks with smooth but lower amplitude pulse. In all cases should take care about extra distortion that is introduced.
The main disadvantage when baking off the average power is that you will get lower Power Added Efficiency.

What you mean is, the system after modifications is not really in the right way. And having low PAPR doesn't mean that having low dynamic range is not right as well??

I was thinking to this because you said that couldn’t minimize PAR of your signal in your simulation. Definitely should work somehow.
The PA dynamic range has two sides: upper side and lower side.
For upper side the average power theoretically shall be backed-off with maximum PAR (17dB that is worst case in OFDM). But as I said, after practical measurements a 6 to 8dB back-off power seems to be ok to meet the spec requirements.
For example if the P1dB of your PA is 27dBm, the upper side of the average power will be 20dBm.
The lower side is dictated by the lowest average power of the system spec. Usually in lower side the Power Amplifiers have more margin and is not a problem to meet the requirements.