A Least Bit Error Adaptive Array for Multi-Level Modulations

Co-channel interference (CCI) is one of the most horrible obstacles for radio communications systems to increase system throughput. Adaptive arrays are well known to mitigate the performance degradation caused by CCI [1]. Especially, adaptive arrays based on minimum mean square (MMSE) criteria have been investigated intentionally because they can suppress the CCI sufficiently even if direction of arrival (DOA) of CCI varies dynamically. Actually, when the number of arrays is NR, NR−1 signals of CCI can be only nulled by the MMSE adaptive arrays. Hence, if more than NR signals of CCI are received, the transmission performance of the MMSE adaptive arrays degrades seriously. On the other hand, adaptive arrays based on minimum bit error rate (BER) criteria are known to be able to mitigate more than NR interference signals [2, 3, 4, 5]. However, these adaptive arrays have been applied for only the binary phase shift keying (BPSK) modulation, even though multi-level modulations are indispensable for increasing the system throughput, nowadays. This paper proposes a novel algorithm for multilevel modulation, especially, amplitude phase shift keying (APSK) based on the minimum BER criteria. The proposed algorithm is defined to minimize the BER function of the APSK. Therefore, the proposed algorithm is named as the least bit error rate (LBER) algorithm for multi-level modulation. Moreover, we derive an equation that is satisfied by the weight vector to which the proposed algorithm converges. Furthermore, the performance of the adaptive array based on the proposed algorithm is evaluated by computer simulation that the proposed adaptive arrays achieves superior performance even when the number of interference signals is equal to or more than the number of the arrays.