During the iteration, cvx can be solved normally in the first few times, but it suddenly becomes "failed" or "infeasible" and appears “{invalid} .* {convex}"

[lee_chloe]

Disciplined convex programming error:
Cannot perform the operation: {invalid} .* {convex}

• z = feval( oper, x, y );

main
SCB(k) <= f_k(k)/phi(k) + BT/log(2)/EPS(log(q_kj(k,1))+1 -q_kj(k,1)e_kj(k,1))+BT/log(2)/EPS*(log(q_kj(k,2))+1 -q_kj(k,2)*e_kj(k,2));

[Mark_L_Stone]

What's wrong with cvx?

Elaborating on @jackfsuia 's answer, if one iteration of CVX fails due to being infeasible or solver numerical difficulty, the variable values are populated with NaN, which are then fed in as input data to the next iteration. CVX considers NaN input data to be invalid, hence the error message. You should ask not what is wrong with CVX, but what is wrong with crude, unsafeguarded SCA.