 Let's talk about a quad modulator, the magic box behind every modern radio's data encoding. Hi, I'm Mr. T, the product guy at RF Elements. If you find value in our videos, consider subscribing to our channel, like, or leave a comment. If you're not sure about the basics, check our previous video, where we explained the nuts and bolts of the quad modulation itself. Quad modulator encodes the digital data, meaning 1s and 0s on turn RF wave, that is suitable for wireless transmission. In this process, quad modulator does the bulk of the work. Let's have a look at how it works. QPSK, you can also call 4quam, is a great example to describe quad modulator operation. QPSK can encode 2 information bits per single transmission, or symbol. This gives us 4 possible symbols based on all the combinations of a pair of 1s and 0s. At the input of the modulator, the incoming data stream is divided into 2 branches. Depending on the combination of 1 and 0, the signal determines the amplitude in each of the branches. In other words, the strength of the signal. The carrier signal is then multiplied by this amplitude. This happens in both branches, with one difference. The carrier signals arriving into each branch have the same shape, but are shifted by 90 degrees with respect to each other. We say they are in quadrature, or that they are orthogonal. This is where the names of the branches come from, i for in-phase and q for quadrature. Adding the two carriers together gives us the resulting modulated signal ready for transmission. Using a few mathematical equations, we can look at the modulated signal through an IQ diagram, where the x-axis corresponds to the in-phase part and the y-axis corresponds to the quadrature branch. With 4quam, the diagram has 4 symbols mapped into it. Each of the symbols corresponds to one of the possible combinations of the two data bits. Connecting the center of the graph with the symbol point tells us what is the amplitude of the resulting signal, while the counterclockwise angle from the x-axis corresponds to the phase of each symbol. The resulting 4quam signal has one possible amplitude and four possible phase values uniquely identifying each symbol. For higher order of modulation depth, such as 16quam, the input signal data is chopped into 4-bit symbols. Two bits for the i-branch and two for the q-branch, producing four possible amplitudes in both branches, giving us 16 possible symbols or the combinations of amplitude and phase of the resulting signal when looking at the constellation diagram of 16quam. With growing modulation depth, the principle doesn't change. With 256quam, the symbol is 8-bits long, giving 4-bits for i and 4-bits for q-branch, each with 16 different amplitudes, resulting into 256 combinations of phase and amplitude of the resulting signal. With 1024quam, and I'm sure you can guess it by now, the symbol is 10-bits long, with 5-bits or 32 possible amplitudes in each branch, so all the possible phase and amplitude combinations give us 1024 uniquely identifiable signals. The rest of the modulator is pretty much the same. The carriers and quadrature are multiplied by the amplitudes and added at the output. Spectral efficiency is a measure of how much data we can transfer using the same bandwidth but different modulations. With quam modulation, the growing number of bits per symbol indicates growing spectral efficiency, since we pack ever increasing amounts of information bits into the same space. The trade-off here is that the growing modulation depth requires increasingly higher SNR levels. This confuses a lot of wisps when buying the newest radios that are for example capable of working at 1024quam or the MCS-10 and 11. The maximum achievable data throughput is higher at MCS-10, but the maximum distance at which this is possible is very small. This is because the high quam depth requires very high SNR levels, so the output amplifier of the radio needs to work in a linear regime. Practically, it means that the output power of a radio working at 1024quam has to be close to the lower end of the output power range, typically around 20dbm. The rule of thumb is that the bigger the MCS-10 rate, the smaller the distance at which you can achieve it. If you want to know more about this principle, check our previous video. Additionally, in unlicensed-with-networks, the biggest issue of RF noise complicates this even more. Since SNR is the signal strength minus the noise floor, it is decreased by the amount of the interference an antenna collects. Therefore, to leverage the highest MCS levels, you need to avoid collecting the ambient noise as much as possible. For example, by using our horn sector antennas that do not have any side lobes, maximizing the SNR the radio is working with. Did you find this video useful? If so, consider subscribing to our channel, like, leave a comment or check out some of our older episodes.