Passive Internet-of-Things (IoT), a new paradigm based on battery-free devices, is a promising technology to enable several use cases that require connectivity with stringent requirements in terms of cost, complexity, and energy efficiency. These use cases span critical sectors, such as healthcare, transportation, and agriculture. Passive IoT relies on the development of technologies such as radio frequency (RF) energy harvesting, low-power computing, and backscatter communication. Particularly, backscatter communication allows devices to modulate its information on external RF signals that are backscattered to the receiver or reader.

BC considers the following elements: a carrier emitter (CE), a reader, and a backscatter device (BD). The main BC configurations are monostatic BC (MoBC), ambient BC (AmBC) and bistatic BC (BiBC). In a MoBC setup, the CE and reader are co-located and share parts of the same infrastructure. A monostatic system suffers from round-trip path loss, and requires full-duplex technology if the same antennas are simultaneously used for transmission and reception. In an AmBC setup, CE and reader are in different locations, while the CE is not considered dedicated. AmBC uses ambient sources to transmit information, such as Wi-Fi, Bluetooth, and TV signals. In BiBC, the CE and reader are also spatially separated from each other, but there is a dedicated CE. In addition, BiBC can operate in half duplex mode, thus avoiding the complexity associated to the full-duplex operation.

Due to the double path-loss effect on the two-way backscatter link, the received backscattered signal is typically weak compared to the direct link interference (DLI) from a CE. This requires a high dynamic range of the circuitry in the reader. As a result, a high-resolution analog-to-digital converter (ADC) is required to detect the weak backscattered signal under heavy DLI; this represents a great limitation because ADCs are major power consumers. Nonetheless, the benefits provided by multiple-antenna and distributed multiple-input multiple-output (MIMO) technologies can be explored to circumvent the limitations of BiBC, which is the main focus of this thesis.

In this context, the contributions of this thesis are two-fold. First, we propose a novel transmission scheme that includes a protocol for channel estimation at the multiple-antenna CE as well as a transmit beamformer design to suppress the DLI between the two ends of a bistatic link (namely CE and reader) and increase the detection performance of the BD symbol. Further, we derive a generalized log-likelihood ratio test (GLRT) to detect the symbol/presence of the BD and provide an iterative algorithm to estimate the unknown parameters in the GLRT. Simulation results show that the required dynamic range of the system is significantly decreased while the detection performance of the BD symbol is increased, by the proposed algorithm compared to a system not using beamforming at the CE.

For the second part, we consider BiBC in the context of cell-free MIMO networks by exploring the optimal selection of CE and reader among multiple access points, leveraging prior knowledge about the area where the BD is located. First, a maximum a posteriori probability (MAP) detector to decode the BD information bits is derived. Then, the exact probability of error for this detector is obtained. In addition, an algorithm to select the best CE-reader pair for serving the specified area is proposed. Finally, simulation results show that the error performance of the backscatter communication (BC) is improved by the proposed algorithm compared to the benchmark scenario.