I provide a simple analogy. Suppose two persons P1 and P2 own a room reserved for conservation and no others are allowed inside. These two persons are part of primary system. Now two other persons S1 and S2 who do not have their room enters the primary room. These two persons are part of secondary system. Noted that P's do not want to talk with S's and vice versa. Now what S1 and S2 requires to start conversation. Since the room is allocated to the P's they will not talk untill both P1 and P2 stops talking or talk in such a small voice so that P's do not get disturbed.

Now in mathematical form. If only primary are active, the received signal at primary receiver is Ypr = Hp Xp+Wp where Hp is direct link between primary Tx and primary Rx, and Xp is symbol transmitted from primary Tx. Do you know why Xp is received at primary receiver? It is because the receiver is tuned to the frequency of transmitted symbol. Now a secondary Tx wants its symbol Xs to be transmitted to its secondary Rx. But at what frequency it will transmit Xs? Secondary do not have their own bandwidth. Suppose it uses the same frequency of primary system. In this case, the primary will be Ypr = Hp Xp+HpsXs+Wp adding the interference HpsXs which will affect the primary data rate. Similarly Ysr = HsXs+HspXp+Ws will be received signal at secondary adding the interference HspXp from primary. So simultaneous use of same bandwidth will add the interference. So one way is to detect whether primary is transmitting or not. Once this is detected, secondary starts and receives Ysr =HsXs+Ws and stops as soon as primary become active

From these two examples, you observe that, primary and secondary systems are unique and have their own system parameters. The whole concept can be defined as the coexistence of two systems by utilizing the same BW and nothing else.

Another example: You know FM radio is allocated in the region of 88-108 MHz and all bandwidth is fully occupied from different FM broadcaster.

Now one FM broadcaster uses frequency 93.5 MHz Red FM. You tune your receiver at 93.5 MHz and listen music etc. Suppose you want your own FM broadcasting system but no BW is available from 88-108 MHz. If you use the same freq 93.5 MHz, FM receiver tuned at 93.5 MHz will play two sounds one from Red FM and other your FM and will cause interference. So what you will do, you will keep tracking when Red FM stops broadcasting and you will start your FM station to broadcast and users tuned at 93.5 MHz will listen the audio played by you.

Notice: The examples given above are just to explain you the concept of cognitive system through analogy.

In a communication system perspective, think the network provider having a BS serving a number of users with a allocated BW. In the cognitive system, we denote this system as primary. Now another network provider (secondary system having a BS and number of users) wants to serve another set of users but does not have BW usually bought in an auction. The primary receivers receives only those signal which are tuned to frequency of their allocated BW. The secondary transmitter needs a BW for its transmission. If the secondary system uses the same BW, the primary receivers will receive the signal from secondary transmitter acting as an interference. So the secondary system will wait until the primary system stops transmitting (spectrum detection ) and will serve its user using the BW of primary system. Or, secondary system will transmit simultaneously along with primary transmitter without increasing the interference at the primary receiver above a predefined threshold.

So the secondary BS needs a dedicated connection to primary BS in order to get the call done.

Or the secondary works irrespective the primary?

I provide a simple analogy. Suppose two persons P1 and P2 own a room reserved for conservation and no others are allowed inside. These two persons are part of primary system. Now two other persons S1 and S2 who do not have their room enters the primary room. These two persons are part of secondary system. Noted that P's do not want to talk with S's and vice versa. Now what S1 and S2 requires to start conversation. Since the room is allocated to the P's they will not talk untill both P1 and P2 stops talking or talk in such a small voice so that P's do not get disturbed.

Now in mathematical form. If only primary are active, the received signal at primary receiver is Ypr = Hp Xp+Wp where Hp is direct link between primary Tx and primary Rx, and Xp is symbol transmitted from primary Tx. Do you know why Xp is received at primary receiver? It is because the receiver is tuned to the frequency of transmitted symbol. Now a secondary Tx wants its symbol Xs to be transmitted to its secondary Rx. But at what frequency it will transmit Xs? Secondary do not have their own bandwidth. Suppose it uses the same frequency of primary system. In this case, the primary will be Ypr = Hp Xp+HpsXs+Wp adding the interference HpsXs which will affect the primary data rate. Similarly Ysr = HsXs+HspXp+Ws will be received signal at secondary adding the interference HspXp from primary. So simultaneous use of same bandwidth will add the interference. So one way is to detect whether primary is transmitting or not. Once this is detected, secondary starts and receives Ysr =HsXs+Ws and stops as soon as primary become active

From these two examples, you observe that, primary and secondary systems are unique and have their own system parameters. The whole concept can be defined as the coexistence of two systems by utilizing the same BW and nothing else.

Another example: You know FM radio is allocated in the region of 88-108 MHz and all bandwidth is fully occupied from different FM broadcaster.

Now one FM broadcaster uses frequency 93.5 MHz Red FM. You tune your receiver at 93.5 MHz and listen music etc. Suppose you want your own FM broadcasting system but no BW is available from 88-108 MHz. If you use the same freq 93.5 MHz, FM receiver tuned at 93.5 MHz will play two sounds one from Red FM and other your FM and will cause interference. So what you will do, you will keep tracking when Red FM stops broadcasting and you will start your FM station to broadcast and users tuned at 93.5 MHz will listen the audio played by you.

Notice: The examples given above are just to explain you the concept of cognitive system through analogy.

@Sm Zafaruddin - What if primary user again starts communication (broadcasting), primary user will not wait or should not experience interference because of secondary users. And how much time will secondary user will take to vaccant/empty the band they occupied

Initially, secondary transceiver only has a single task: keep tracking the activity of primary system. However, after detecting the vacant band, secondary system has two tasks: 1. Keep tracking the activity of primary system. 2.Using the vacant bandwidth for its transmissions . Once secondary detects that primary starts again, secondary transmissions should stop immediately to avoid interference on primary. Please go through the first analogy again. What happens if P's start talking again? Secondary S's should stop immediately.

There are no services provided by network providers in the unlicensed free bands/channels. The white holes can only be used by opportunistic users to transmit/receive their data subject to interference and jam by other competitors trying to get access to same bands/channels. Unless you jump into licensed bands/channels then the cognitive radio has to keep sweeping the entire operational spectrum to release the occupied bands/channels upon PU detection. In either case, no services entertainment for SU expected unless centralized by spectrum broker for spectrum management, supervision and control. Hence SU to SU communication can be expected to have some high level of services and online applications upgrade.

Can you further elaborate the difference between no services, data transmissions, and high level services? I think no excludes everything neither high level services nor data transmissions.

I have explained the cognitive concept purely through signal processing in the physical layer. In the absence of PU transmissions, secondary users indeed transmit Xs which will always correspond to a service provided by a secondary system to its own users. Type of services can be any depending on the configuration of the primary network. I can not think of a user without a network provider/system or I have misunderstood the term ''user'' in your context. Thank you.

There are three different ways secondary user can access the network resources.

they are

Overlay techniques

Underlay techniques

interweave technques

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.94.3568&rep=rep1&type=pdf

The above link gives you models considered for all three cases

@ Sm Zafaruddin, If the primary users are not communicating in the allocated band. Then Secondary user identify that particular band by means of spectrum sensing and starts the communication.

But how the secondary user transmitter conveys that band information (parameters like Carrier freq, bandwidth etc etc) to its own (secondary user) receiver.?