The spectrum fragmentation below 6 GHz makes current wireless communication systems not suitable to support user data rates of Gb/s and above in a commercially viable manner. Such bandwidth shortage has motivated the exploration of vacant, unlicensed spectrum at higher frequencies, and in particular in the millimeter-wave (mmWave) frequency bands which the future fifth-generation (5G) wireless standard is expected to exploit. Due to the higher propagation loss and unfavorable atmospheric absorption, data transmission over relatively long distances represents a serious challenge at mmWaves. The rather short wavelength, however, allows more antenna elements to be integrated into devices and base stations operating in this band, thus enabling the implementation of large-scale multiple-input multiple-output (MIMO) and adaptive antenna arrays to overcome range limitations. Nevertheless, traditional MIMO processing performed in sub-6 GHz systems is, at present, impractical at mmWaves because the high cost, power consumption, and complexity of mixed-signal hardware at such frequencies hinder the use of a dedicated RF chain for each antenna element. For this reason, most of the literature proposes analog/RF beamforming architectures which rely on networks of RF phase shifters to control the phase of the signal at each antenna element. Compared to fully-digital beamforming architectures which enable to precisely control both phase and amplitude of the transmitted/received signals, analog beamforming is sub-optimal because of the constant amplitude and the low phase resolution constraints affecting the RF phase shifters currently available at mmWaves.
Hybrid analog-digital architectures aim at approaching the performance of fully-digital beamforming by dividing the precoding/combining operations between the analog and digital domains and adopting fewer RF chains compared to pure digital architectures. The availability of multiple RF chains allows hybrid analog-digital architectures to shape beam patterns very close to those attained by a fully-digital architecture, i.e., with limited overlap between adjacent beams and excellent flatness over the covered sectors. Despite the substantial research interest recently gained by hybrid beamforming for mmWave systems, most of the literature has until now proposed high complexity solutions relying on unrealistic hardware, e.g., assuming the availability of a considerable number of RF chains and RF phase shifters with large (or even infinite) number of quantization bits. Our research work in this field aims at addressing relevant aspects of the initial access procedure at mmWave frequencies when hybrid analog-digital architectures are exploited at both the base station (BS) and mobile station (MS). As discussed in our recent publications , , the constant amplitude and low phase resolution constraints of RF phase shifters make analog beamforming architectures unable to (1) shape flat-top beams with limited overlap as required for a precise angular domain partitioning, (2) scan multiple spatial directions simultaneously, (3) achieve multiplexing gain by transmitting over multiple parallel data streams. On the contrary, splitting the beamforming operations between the analog and the digital domains and exploiting more than one RF chain, hybrid analog-digital architectures allow to overcome such limitations.
In , , we formulate an optimization problem to approximate fully-digital, multi-beam, multi-beamwidth beam patterns by means of a hybrid analog-digital architecture requiring a number of RF chains much lower than the number of antenna elements and only 2-bit RF phase shifters. We propose and implement two algorithms for the synthesis of hybrid codebooks: (i) a variant of the classical OMP algorithm which exploits a dynamic dictionary learning (DDL) mechanism  and (ii) a greedy geometric approach based on a low-complexity, dictionary-free algorithm . Both solutions allow to synthesize beam patterns with reduced sidelobe level, excellent flatness over the covered sector, and limited overlap with adjacent beams. In , we leverage our codebooks in the framework of a sequential adaptive beam training protocol to estimate the most promising angle-of-departure and angle-of-arrival of the mm-wave channel. In , instead, we propose a beam search strategy which effectively accelerates the link establishment by exploiting the ability of mm-wave devices to receive from multiple directions simultaneously. Compared to the literature, which proposes complex algorithms based on the use of RF phase shifters with very high (or even infinite) resolution, our approaches provide significantly better performance with lower complexity hardware.
- D. De Donno, J. Palacios, D. Giustiniano, and J. Widmer (May 2016)
Hybrid Analog-Digital Beam Training for mmWave Systems with Low-Resolution RF Phase Shifters (Paper) [PDF ]
In: International Workshop on 5G RAN Design in co-location with IEEE ICC 2016, 23-27 May 2016 , Kuala Lumpur, Malaysia.
- J. Palacios, D. De Donno, D. Giustiniano, and J. Widmer (September 2016)
Speeding Up mmWave Beam Training through Low-Complexity Hybrid Transceivers (Paper) [PDF ]
In: The 27th Annual IEEE International Symposium on Personal Indoor and Mobile Radio Communications (IEEE PIMRC 2016), 4-7 September 2016, Valencia, Spain.