Problem with getting correct SER (path loss in A&F)

I am generated three Rayleigh channel gains including path loss assuming all fixed links
PL1=108; % Path loss for BS-RS
PL2=102;% Path loss for RS-MS
PL3=118;% Path loss for BS-MS
%% Channel gains
H_BS1_RS1=(1/sqrt(2)).*[randn(1,N)+j*randn(1,N)];
channel_BS1_RS1=H_BS1_RS1.*(db2mag(-PL1));
H_RS1_MS1=(1/sqrt(2)).*[randn(1,N)+j*randn(1,N)];
channel_RS1_MS1=H_RS1_MS1.*(db2mag(-PL2));
H_BS1_MS1=(1/sqrt(2))*[randn(1,N)+j*randn(1,N)];
channel_BS1_MS1=H_BS1_MS1.*(db2mag(-PL3));
nbr=(1/sqrt(2)).*(randn(1,N)+j*randn(1,N));%% awgn noise for BS-RS link

data_BS1_RS1=sqrt(Pb/2).*data_BS1.*channel_BS1_RS1+(db2mag(-PL1)).*10^(-snr_db(i)/20)*nbr; and so on

But i am not able to get correct SER. In single link path loss has no effect on SER but in A&F path loss produce worst result: Without Path loss code is working fine.
Can any body have some idea: