The answer is pretty straight forward. Due to the material of radome in the far field, there is a material loss associated with it due to which its insertion loss increases (or in other words, its gain is reduced). However, return loss of the antenna has little to do with radome but still antenna can translated to a lower frequency in terms of return loss bandwidth. Attached application note might help more.
In simple terms I think you can say that Insertion loss is the loss between two ports (How much power is lost between two ports while transmission of signal from one port to another).
By applying same analogy I am not able to relate the two as if how this radome is causing insertion loss.
True that insertion loss is the loss of power between two ports. In case of transmit and receive antennas, the insertion loss comes from the wireless channel between the two antennas. So when you introduce a radome in the channel. The insertion loss of the channel will be increased due to material properties of the radome; just like any other material if you place between the TX and RX antennas.
However, if this is not the scenerio in your case then I cannot say how radome would affect the insertion loss?
A radome will reflect some power and absorb some power. The reflection and absorption result in transmission loss, or insertion loss, or loss of gain, which is part of S21. Depending on the shape of the radome, the reflection may or may not be reflected back to the antenna. If any is, the reflection from the radome will be received by the antenna and so affect S11, the return loss.
In a measurement of a radome, or a simulation, S21 refers to the loss between a plane wave excited one side of the radome and a plane wave received on the other side.
The reflection from a radome is not necessarily the same as S11 for the radome illuminated by its antenna. For example, a slanted radome can be designed so that its reflection goes into an absorbing load so that it does not reach the antenna. The radome still has a reflection, but it does not contribute to the S11 of the antenna.
Can you please tell me about what the physical quantity / significance of measuring isolation in base station antenna.How does it actually measures isolation ?
For example -
We measure isolation (S21) between two ports of one column which is defined by M45 (Port 1 )and P45 (Port 2).
According to what I think , it is the measurement of how the two X pol are isolated .We need to get better isolation in order to reduce crosstalk etc.
We can achieve the desired isolation by putting isolation chokes.So how does the chokes help in getting better isolation.
Also if we put some horizontal chokes the gain of the antenna reduces . What is the reason behind this?
This is partly guesswork, but in a base-station some antennas will be transmitting and some receiving. You do not want to receive your own transmissions, so isolation between tx and rx is important. One way that signals get from tx to rx is through currents in the common ground structure. Chokes can be used to reduce the ground currents, however, ground currents are important in many antennas and so stopping ground currents may reduce the gain.
Isolation between two ports in one column may refer to two antennas in a vertical array. If these have cross-talk when they are both fed at once, it looks like a bad match, so isolation is important here, too, unless the antennas are tuned to take account of cross-talk. However, the level of cross-talk that is acceptable between antennas within an array is much higher than would be acceptable between tx and rx.