Standard Propagation Channel Models for MIMO Communication Systems

Table 7

Open research issues.

S/N.	Open research issues	Brief explanation
1.	Large-scale measurements at mmWave and Terahertz bands	First, conducting large-scale channel measurements is a major concern. It is still very difficult to carry out large-scale channel measurements, due partly to the complex nature of the dynamic environment, especially for outdoor scenarios, and the sophisticated equipment that is required for such measurements in the mmWave [177, 220] and Terahertz (THz) [221] bands. In the case of double-directional channel measurements, switched channel sounders have been used. To the various antenna elements, and in a sequential manner, the sounding signal produced from a single waveform generator is applied. The received signal measurements follow this in a fashion similar to the transmitted signal, and this measurement type has some inherent limitations. First, radio signals transmitted via cables to the antennas from the waveform generator, usually located several miles away, is costly and quite tasking, contributing to large signal attenuation [222]. Fiber optic cables have been deployed for electrooptical conversion and transmission of signals to solve this problem. However, proper calibration and temporal stability are still crucial issues that require further investigation.
2.	Multilink channel measurements	In practice, it can be very tedious to establish critical parameters required for multilink modeling. These parameters include correlation information on the angular spreads and joint cluster identification. Most measurements cannot give the necessary joint double-directional evaluations of multiple transmitter and receiver locations. Consequently, system designs requiring such parameters are seriously hampered. There is a need for a comprehensive model to accommodate these key parameters. Multilink channel models often present some difficulties related to modeling the correlation properties between the different links. This is because the different links appear to identify both joint clusters and disjoint clusters. This scenario can be very worrisome due to the challenging correlated shadowing processes. Future research could explore high-reliability techniques to simplify the modeling of the correlation properties of multilink channels.
3.	High-precision measurements equipment	Mobile devices and applications have witnessed a tremendous surge in usage by millions of subscribers in recent years. This poses the need for large-scale measurements. Larger bandwidth fuels the demand for near precise frequency references, leading to an increase in the amount of data collected, usually in thousands of megabytes. The measurement campaign will most likely continue to be the method of investigating channel models. If reliable information related to the channel is required, the measurement equipment must be well designed with high precision and reliability. Future research could be directed towards more sophisticated, efficient, and high-precision measurement equipment.
4.	Directional characterization	Directional characterization is still a major problem in mmWave channels. Although the IEEE 802.15.3c [223] and IEEE 802.11ad [224] models provide useful information on directional characterization, there is still an urgent need for extensive measurement verification in different environments. Future studies could also be directed towards investigating the impact of moving objects and human beings’ movement obstructing the LoS on the polarization properties. Also, in mmWave and Terahertz channels, the characterization of temporal variations remains an open issue.
5.	Channel model parameterization	Channel model parameterization is still a critical area that needs to be addressed. Most channel models have been parameterized for easy characterization of the stochastic radio channel responses, especially in multilink MIMO systems with spatial correlation properties [92]. For example, it was reported in [171] that the COST 2100 channel model gives practical results, which include several channel properties for proper modeling. However, this model’s extension will require much effort in new field measurements to parameterize adequately and validate the extended model [225]. Future research could also be channeled towards investigating how the various modeling approaches and parameterizations affect the systems’ performance and how they can be designed for the most efficient operation in those channels. A better understanding of the propagation channels will provide a very good platform for efficient system designs, to ease the applicability of channel models, especially in outdoor-to-indoor scenarios.
6.	Channel elevation characteristics	Most findings on elevation characteristics [226, 227] of the channel are based on ray-tracing simulations. There is a need for a model, especially for outdoor to indoor links. This will help the mobile station’s proper positioning since its height is critical to the studies related to wireless channels’ elevation characteristics. Also, there is a need to investigate further how the indoor floor plan and the horizontal positioning of the mobile inside the building impact the channel elevation. A summary of previous research work focusing on 3D modeling, with particular interests in elevation characteristics, is reported [228].
7.	Massive MIMO systems	Conducting large-scale measurements in massive MIMO systems is still a major issue. This is because the temporal variability is relatively small, but many antenna elements require switching. The design of switches for such applications is still a huge issue. It is challenging to design switches for mmWave frequencies [72, 229]. This is why antennas that require mechanical forces to be moved or rotated are in use at such frequencies, and this method is time-consuming and computationally expensive. Future work could explore more efficient and low-cost techniques for conducting large-scale massive MIMO measurements.
8.	Vehicular communications	Several research works related to vehicle-to-vehicle communications have been reported [230–235], especially in the developed cities of the world. Critical parameters, such as path loss and delay spread in different environments, have been reported [236, 237]. The vehicle types in use and the antennas’ positioning call for major concerns [238]. In the first place, there are several hundreds of heavy-duty trucks and buses plowing the highways, but much has not been reported about such vehicles. Besides, there is a need to carefully investigate the impact of such heavy-duty trucks on the transmitters and receivers. We need to know how the shadowing will be impacted if a big truck or a bus is positioned between the transmitter and receiver. Also, we need to investigate how cars’ parking on the sides of the road could impact communication systems. Hopefully, these findings will provide viable insights required for the design of safety-critical applications.
9.	Machine learning-based channel modeling	The design and development of machine learning-based signal prediction algorithms are still a significant issue. This is due partly to the lack of a comprehensive dataset for the development and testing of the models, especially at the Terahertz bands. For example, in [239], the proposed artificial neural network path loss (PL) model’s environmental features could not give the desired descriptions of the actual propagation characteristics due to inadequate information about the height. This ultimately affects the accuracy of the proposed PL prediction models.