In this video, we will talk about diversity, mainly time diversity, and spatial or antenna diversity. So we'll start with something called diversity. Communication over a wireless channel has a poor performance, when compared to wire channel. The reason is mainly because, let's take an example. Let us say we have a transmitter with one antenna, receiver with one antenna. Now because we are considering a wireless channel, we have a channel h in between the transmit antenna and receiver antenna, right? And, there's a high possibility this path, between the transmitter and receiver may be in deep fade, okay. So what exactly we do is, to increase the diversity, what we do is instead of using one antenna in the transmitter and receiver. What we can do is, either we can increase the number of antennas at the transmitter, or at the receiver, or both. Okay, so if you see on the screen, we have two antennas, here on the transmitter and the receiver side. So now what we have, in the single transmit / single receive case, we only have one path. And the possibility or the probability of this single path, going to deep fade is high when compared to the possibility or probability of all the paths going into deep fade, right? This is obvious, therefore, the overall system going to deep fade has lower probability when we are using multiple antennas at the transmitter or receiver or both So diversity can be achieved, either in time, frequency, or space. Like the example that I have shown is, concerning the diversity in space, or we call it antenna diversity. Now we'll start with time diversity. Time diversity is obtained by interleaving and coding over symbols across, coherence time intervals. You see using an example, let us take an example, where we are taking a k bit length input, and we are adding a parity bit of length m bits. Okay, now the total code word length will be what? K plus m. Now, through this example we can see, if K is three let's say 011 is our input bit, we have added m equal to one single-bit parity. So overall the codeword became 0110, so what exactly happened? The parity bit always helps the receiver to detect or correct the bits in error. Okay, so this is a coding part, how the interleaving can be done? We have this coded bit, 0110, which can be interleaved for example in the manner where I am swapping the 1st and 2nd bit, interchanging the 1st and 2nd bit, and keeping the last two bits as it is. This is one kind of interleaved pattern, interleaved pattern can be multiple, infinite number of interleaved patterns. Right? So we have used let's say this one. So, one important thing you should remember is, the receiver always knows the interleaving pattern, Okay? If receiver does not know the interlink pattern then it cannot detect the symbol what has been transmitted after doing some kind of interleaving. It is like we are encrypting, but we are not having a code for to decrypt it, okay? So this is one example of time interleaving. Let us look into a more detail, now I'll consider a code word of length L, let us take an example, where L is equal to four, means the code word length is four. Okay, starting from X one to X L. LX is called to before I said in this example, so we know that because of this channel going to deep fade. If you see, I have taken an example, where the third symbol, right, h_l, We can see on the Y axis and this is the time. So if, we see at this point we are experiencing a deep fade because the channel coefficient is very low at that time, other point it is fine. So if you see, let us take an example where I'm not doing any kind of interleaving, okay? I've just done the coding. In that case, what is happening? We have x_0, x_1, x_2, x_3. X_0, we are transmitting, x_1 we are transmitting, x_2. We are transmitting and x_3 we are transmitting. in this case. What will happen, if we can see from screen h, because of the deep fade, the symbol that has been transmitted in that duration is going bad, right? I cannot detect the symbol because what is happening? this is getting multiplied the symbol, x_2 is getting multiplied with channel, which is very low, now because the channel is very very poor, my symbol is getting corrupted, and I cannot detect this bit without error. Now, let us see if we use interleave whether this will be helpful. This is an example, where what we have done is now, in addition to codeword we are also doing interleaving. So if you see, the symbols this is like some different symbols, some different symbols combination using the interviewing pattern, some kind of pattern. We have obtained this interlocking pattern, okay, which is nothing but a combination of X zero. If you see this is X one, this is X zero, this is X two, this is X three. Similarly, the pattern has repeated. So what will happen, now during the time when the channel is indeed fade, what happened, see only this part will be corrupted right? But if you see, I am able to decode, a few part of X nought, only 1/4 part of X nought, X two, X one, X three is corrupted now, right? If you see this is 1/4 part, 1/4 part, 1/4 part and 1/4 part, only one 1/4 part of X naught, X one, X two, X three is corrupted when compared to the scenario where I was not using inter living the complete X two was getting corrupted. So because the complete x two was getting corrupted there's no option of decoding that without any error, but like, for example in this case only 1/4 part of the bits are corrupted in this case using the bits like parity bits, I can always correct or detect the error at the receiver. So this is the benefit of using interlink, I hope you guys have understood. Now this was about time diversity. Now I'm going into details of spatial diversity, how to obtain special diversity, by placing multiple antennas. Remember the multiple antennas are at either the transmitter, or at the receiver and therefore we call when we are also using multiple transmitter antennas, we call it transmit diversity. When we are using multiple receiver antenna we call it receive diversity. Third scenario can be where we have both, multiple transmit and multiple receive antenna, in this case we'll call it, transmit as well as receive diversity. So let us take an example, this is an example of receive diversity, right, and we will call it a single antenna multiple output system. Okay, this is very well known, whenever we have single antenna at the transmitter will call single input, whenever we have multiple antennas at the receiver multiple output. So overall we call it, single input multiple output; SIMO systems. And like I said, we will achieve a receive diversity. The second case, just the opposite of this where we have multiple transmitter antenna and single receive antenna, obviously we'll call it multiple input, single output, that is MISO. And we are achieving transmit diversity, right? The third case we have both transmitter diversity as well as receive diversity and that's why we are calling it multiple input, multiple output: MIMO. My fourth case is something which is very very important, in the today's 5G, where 5G is using massive MIMO, where instead of only having few antennas at the transmitter is here we are having, a massive number of antennas at the transmitter and the receiver. These numbers can vary from, 32, 64 till 512, also depending on what you can see. We are using either in the sub six regard, or in the millimetre wave. And these things will discuss later in detail. Okay, we have a single antenna transmitter, single antenna receiver, We call it SISO and SIMO. MISO, MIMO and Massive MIMO which has been used in 5G. In this video, we learn about the diversity which helps in reducing the probability of deep fade. We’ve understood two types of diversity here. Time diversity, which is obtained using interleaving and coding. Second, spatial diversity, which is obtained by increasing the number of antennas at the transmitter and receiver. [MUSIC]