During link budget design, we maintain a link margin (that covers fade margin, and any other losses). Does link/fade margin depend on the technology being used. In fact I am working on Rotating Polarization Wave (RPW) based LPWAN [1]. The error performance of RPW is inferior to LoRa in AWGN, but superior to LoRa under Rayleigh fading environment. I have compared the two LPWANs for a BER of 10^-5. A packet error rate of 1% is achieved with RPW at a received power level of -120 dBm compared to -109 dBm for LoRa.
[1] Article Quaternion Model of Higher Order Rotating Polarization Wave ...
I think you will find that RP-MPSK uses polarization diversity to increase the diversity gain.
Polarization is a hardware thing, it is not just modulation. To use polarization diversity the extra channel (polarization) has to already be there in the hardware, and could be used for various modulation schemes, for example to double the data-rate, with each channel on one of the polarizations and so not using polarization diversity on each channel, but with the two channels having different fading characteristics.
Polarization is one of the things that affects fading, as also are frequency and antenna position. Adding diversity in any of these multiplies the number of channels available, each with (hopefully) independent fading characteristics. Different modulation schemes can use different combinations of these channels to try to get the optimum data rate or fade margin in the particular environment.
Dear Malcolm White thanks for the explanation. So in short, it is correct to say that the fading loss of LoRa and RPW is different, and therefore, the additional fade margin achieved using RPW can be utilized for other benefits for example increase the range of communication, provided transmit power is the same for both LoRa and RPW.
Dear Zaid,
The modulation method determines the signal power distribution and its dynamic range. Received power is an average measure that may not highlight the effect of such dynamic power distribution which impacts the BER. This needs to be factored along with the fading model for both scenarios for a fair comparison.
Rgds
Dear Shiv Prasad Tripathy ,
So I should consider both BER vs SNR as well as Packet Error Rate (PER) vs Received Power (Pr) for a fair comparison?
The difficulty I have in comparing the two is the definition of sensitivity which is, to the best of my understanding, defined in a noise environment only. The link budget is entirely based on the receiver sensitivity, and hence the allowable margin. While in the case of RPW, the improvement is in performance over fast/Rayleigh fading.
Now, I have a few questions to ask:
1. The receiver sensitivity is estimated from the following equation:
sensitivity dBm = -174 dBm + SNR_min + 10*log(bandwidth) + NF
Would it be meaningful and theoretically acceptable that I define the sensitivity in terms of minimum SNR for correct signal detection over fading channel instead of noise?
2. The link budget equation is:
Pr dBm = Pt + Gains - Path Loss (PL) - Other losses (fading loss, implementation loss, mismatch loss etc)
Now if I use free space path loss (FSPL), I can consider all other losses such as shadowing and fading separately. But on the other hand, If I consider Okumura Hata (LoS PL model), do I still need to consider fading loss? I believe the other type of loss associated with fading i.e. shadowing is already considered in this model since it covers a wide variety of environments (urban, rural, etc.). If I can consider fading loss separately, then I can compare the two modulations based on the link budget equation. And hence I can say that the coverage range of RPW is better than LoRa over fading in (for example ) an obstructive environment, or, it offers a better reliability than LoRa over the same range.
I have gone through some books to understand this, but I could not conclude anything.
Can you through some light on this or mention a specific reference that can answer my questions?
I think there are two parts to the problem - one relates to modulation, and the other to link budget - and both need to be considered together for an end-to-end performance.
Point 1: To do a comparison on digital streams, Eb/No for the different modulation schemes may be used based on desirable BER. I think tests should be on raw PRBS data without forward error correction/ retransmission, for a ‘rf link/path level’ comparison.
Point 2: Considering fading loss independently does allow comparison of modulation schemes. But one must define the yardstick for comparison, like identical bit rates or identical bandwidth, same BER, and so on.
Just my thoughts.
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Waves