In odd slot, source S transmits its symbol to D and to both relays A and B. The received signal at D is given by

Where n is the time slot, and are the data symbols of PU and SU. the amplification factor satisfying the power constraint expressed as below

The received signal at relays A and B during nth slot is

Similarly in even time slot, the received signal at D is

The received signal at relays A and B during (n-1) th slot is

IRI mitigation – ZF SIC

In this section, we propose ZF-SIC for interference cancellation at D. The extraction of PU symbols considering the received signal at D in time slot t1 and t2 is given by

In t1,

In t2,

Extracting one of the symbols with better channel gain from received signal , we obtain the other symbol S2 by iteratively subtracting with . So, we have

Finally, S2 is obtained after equalization by zero forcing method as given below

where the equalized symbol

Optimum power allocation

In this section, power is allocated to secondary relays appropriately to satisfy the interference constraint on PU destination and also for guaranteed QOS for PU.

BER Minimization

For optimum power allocation in t2 and t3, the sum of individual powers of PU (Ps) and SU (PA and PB) should be equal to the total power constraint Ptotal. In t2 and t3, optimal power constraint is as follows

The individual powers of the transmitters in the proposed model is found as below

+

+

Then, the individual powers of , and is obtained as PS=0.657W, PA=0.342W for t2 and PS=0.686W, PB =0.314W for t3.

Rate Maximization

The maximum achievable rate in bits/sec/hertz based on the proposed scheme SENT after MRC is given by

where is rate at D through direct link communication given by

is the SNR of direct link. and is the rate achieved through relay A and B at D and ,, are the MRC constants.

where is the SNR between source S to relay A and

is the SNR between A and destination D.

Similarly for SU, the rate is expressed as

where is the SNR between A and B.

Relay based routing performance

Fig.6 Number of SU nodes Vs PDR

Computed rates are used for SU data transmission and it is inferred from Fig.6 that as number of relay nodes increases, PDR performance is improved by 24% when compared to CAODV.