An improved radius adaptive K-Best algorithm for MIMO system

The K-Best algorithm can achieve high BER performance with a large K value, but the complexity is very high. The RAKB algorithm reduces the complexity and gets similar BER performance compares to the K-Best algorithm. However, with the antennas increase, the BER performance improvement of the RAKB algorithm is not obvious. This paper presents an improved algorithm which is based on the RAKB algorithm. Unlike the invariant K value used in the K-Best algorithm and the RAKB algorithm, we set a variable K value and bring in the SQRD method in the proposed algorithm. Simulation results show that with the number of antennas increase, the proposed algorithm has a noticeable BER performance improvement compares to the RAKB algorithm. Meanwhile, it may also reduce the complexity in terms of the visited nodes.