Spatial Channel Model (SCM)

The SCM operates on a drop-based simulation methodology where the channel is considered static for a defined period before being randomly re-initialized.

• A drop represents a segment of time where large-scale parameters (path loss, shadow fading, angular spread) remain constant.
• Within a drop, small-scale fading is generated by summing contributions from multiple scatterer clusters.
• This approach decouples system-level mobility from physical layer fading, making it computationally efficient for testing link adaptation and scheduling algorithms.
• The model defines specific user environments: Urban Macro, Urban Micro, and Suburban Macro, each with distinct path loss models and angular spread distributions.

The SCM synthesizes the MIMO channel matrix by modeling the physical propagation as a superposition of N = 6 discrete clusters, each representing a group of unresolvable multipath components.

• Each cluster is defined by a unique Angle of Departure (AoD) and Angle of Arrival (AoA) , generated from wrapped Gaussian or Laplacian distributions.
• A single cluster is further decomposed into 20 sub-paths to simulate intra-cluster angular spread.
• The phase of each sub-path is randomized uniformly, producing Rayleigh or Rician fading statistics at the receiver.
• This geometry-based approach accurately captures the spatial correlation between antenna elements, which is critical for evaluating beamforming and spatial multiplexing gains.

The primary output of the SCM is a complex-valued channel coefficient matrix H for each user, capturing the spatial signature between transmit and receive antenna arrays.

• The model generates an n × m matrix where n is the number of receive antennas and m is the number of transmit antennas.
• Coefficients are derived using the steering vector of the antenna array geometry, incorporating the phase shifts induced by the angle of arrival.
• The SCM explicitly models cross-polarization discrimination (XPD) , accounting for power leakage between vertical and horizontal polarizations.
• This correlated output enables realistic testing of Precoding Matrix Indicator (PMI) selection and rank adaptation algorithms in closed-loop MIMO systems.

The SCM integrates empirical large-scale propagation models to accurately set the macroscopic power level before applying small-scale fading.

• Path loss is calculated using the COST 231 Hata model for Urban Macro cells and the COST 231 Walfisch-Ikegami model for Urban Micro cells, with distance-dependent log-distance formulas.
• Shadow fading is modeled as a log-normal random process with a standard deviation of 8 dB for macro cells and 10 dB for micro cells.
• Crucially, the SCM specifies a site-to-site correlation of 1.0 for shadow fading, meaning all users connected to the same base station experience the same shadow fade offset.
• This realistic power scaling ensures that system-level simulations accurately reflect cell-edge user performance and inter-cell interference dynamics.

The SCM defines detailed antenna characteristics to accurately simulate the spatial filtering effects of practical base station and user equipment arrays.

• The base station antenna pattern is modeled using a sectorized gain pattern with a defined 3 dB beamwidth (typically 70 degrees) and a maximum attenuation of 20 dB at the sector boundaries.
• The model supports dual-polarized antennas with configurable slant angles (+45°/-45°), generating four distinct polarization sub-channels per link.
• The user equipment antenna is modeled as an omnidirectional pattern with a gain of 0 dBi, reflecting typical mobile handset characteristics.
• These detailed antenna models are essential for evaluating hybrid beamforming architectures and the performance of Type-II codebook feedback schemes.

The SCM was developed by the 3GPP-3GPP2 Spatial Channel Model Ad Hoc Group and published in 2003 as a joint standard for MIMO system evaluation.

• The original SCM was designed for a maximum bandwidth of 5 MHz and a carrier frequency of 2 GHz, reflecting early 3G deployment parameters.
• The SCM Extension (SCME) was later developed to support higher bandwidths up to 100 MHz by introducing intra-cluster delay spread and time-varying Doppler spectra.
• SCME enables the modeling of frequency selectivity within a cluster, which is critical for evaluating Orthogonal Frequency Division Multiplexing (OFDM) systems.
• The methodology directly influenced the WINNER II and QuaDRiGa channel models, which extend the geometry-based stochastic approach to millimeter-wave and massive MIMO scenarios.