Hybrid Beamspace DOA Estimation for RF-Chain-Limited mmWave Passive Arrays

Hybrid convolutional beamspace (CBS) is proposed for direction-of-arrival (DOA) estimation based on millimeter-wave (mmWave) passive arrays. In hybrid processing, an analog combiner is used to reduce the number of radio-frequency (RF) chains thereby lowering hardware cost. Then, a digital combiner and digital processor are used to estimate DOAs. In hybrid CBS, the analog combiner is equivalent to a filter followed by a decimator (downsampler). This decimation reduces dimensionality and complexity. Thus, it is especially attractive for large arrays, often used for mmWaves. It is shown that a CBS filter with arbitrary coefficients is implementable despite the unit-modulus constraints imposed by analog combiners using phase shifters. A new scheme of CBS is proposed based on nonuniform decimation and difference co-array method. The retained samples correspond to the sensor locations of a virtual sparse array, dilated by an integer factor, which results in larger co-array aperture and thus better estimation performance. Co-array method also allows us to identify more DOAs than RF chains. Besides, with the use of random or deterministic filter delays that vary with snapshots, a new method is proposed to decorrelate sources for the co-array method to work. The Cramer-Rao bound for DOA estimates based on any beamspace output, including CBS, is also derived. The design stage of Capon-CBS, a variant of CBS based on Capon beamforming, is also realized in a hybrid analog and digital architecture. By simulations, hybrid CBS is shown to yield smaller estimation errors, higher DOA resolution, and lower hardware complexity than previous methods.