The evolution of mobile networks from 4G LTE to 5G NR is driven by an exponential growth in data traffic, propelled by the ubiquity of high-definition video streaming, the proliferation of Internet of Things (IoT) devices, and the emergence of ultra-low latency applications [1] [2]. To overcome the spectral efficiency constraints inherent to traditional multiplexing schemes (FDM, TDM, CDMA), Orthogonal Frequency Division Multiplexing (OFDM) has established itself as the core physical layer standard [3]-[5]. Utilizing subcarriers that overlap at right angles, OFDM enhances spectral efficiency and offers strong protection against multipath fading that varies with frequency.

Nonetheless, moving to 5G NR poses considerable operational and financial difficulties, especially in Sub-Saharan Africa, where mobile networks are mainly designed for 4G. For a company such as Telecel Faso, which is the third-largest provider in Burkina Faso with close to 3 million subscribers, transitioning to 5G NR involves a practical approach focused on optimizing capital expenditure and ensuring compatibility with existing systems. The deployment of late-generation Huawei radio equipment (such as the BBU BTS5900 and advanced RRUs) enables a Non-Standalone (NSA) migration path via a straightforward software upgrade [6] [7]. Consequently, the core research challenge shifts from determining whether 5G NR is achievable to precisely characterizing the physical-layer requirements and economic viability of its integration.

While the scientific literature extensively addresses physical-layer bottlenecks in OFDM systems—such as Peak-to-Average Power Ratio (PAPR) reduction through clipping, SLM, or PTS techniques [8] [9], Doppler shift mitigation using flexible 5G NR numerologies, and spatial diversity gains via MIMO technologies [3] [5]—few studies holistically combine these physical parameters within a simulation framework tailored to the operational realities of a West African mobile operator.

This paper presents a techno-economic feasibility study for the deployment of 5G NR at Telecel Faso, supported by a MATLAB-based simulation of an end-to-end OFDM transmission chain. The study sequentially quantifies the energy efficiency gains achieved through PAPR reduction, link robustness under high highway mobility scenarios, and diversity gains in dense urban channels. These physical performance metrics are then translated into a practical deployment roadmap.

The remainder of this paper is organized as follows: Section II outlines the system model and MATLAB simulation methodology; Sections III through V we adress analyze the empirical results regarding PAPR, Doppler tolerance, and MIMO diversity; In Section VI, we present the techno-economic feasibility assessment for Telecel Faso, followed by concluding remarks in Section VII.

The advent of fifth-generation mobile networks (5G NR) marks a major technological breakthrough, promising ultra-high data rates, reduced latency, and unprecedented spectral flexibility. However, in sub-Saharan Africa, simply copying and pasting Western or Asian models cannot achieve the deployment of these technologies. For service providers such as Telecel Faso in Burkina Faso (with around 3 million customers), the transition to 5G NR must reconcile radio performance requirements with the economic constraints of existing infrastructure (Huawei BBU BTS5900 and RRU equipment). In practice, this transition faces three major technological hurdles:

Energy inefficiency (high PAPR).Vulnerability to mobility (Doppler effect).Urban error floor (multiple paths).

To address these shortcomings, this article uses Modeling and MATLAB simulation to evaluate an OFDM transmission chain adapted to the context of Telecel Faso.

Characterizing the mobile radio channel represents an indispensable prerequisite for cellular network dimensioning. Unlike stable, guided wired links [8], the wireless environment imposes severe propagation constraints: path loss degrades the signal-to-noise ratio (SNR) as a function of distance; multipath propagation in urban environments causes frequency-selective fading and inter-symbol interference (ISI), while user mobility induces Doppler spread that disrupts frequency synchronization. To address this threefold physical challenge, Cyclic Prefix Orthogonal Frequency Division Multiplexing (CP-OFDM) was standardized by 3GPP as the baseline waveform for 4G LTE and 5G NR down link transmissions [2] [10].

Conceptually, an OFDM system divides the channel bandwidth into a large number of orthogonal subcarriers, efficiently synthesized via Inverse Fast Fourier Transform (IFFT) at the transmitter and extracted via FFT at the receiver [4]. Strict orthogonality between two subcarriers with frequencies f k and f l is guaranteed over a symbol duration T S provided their subcarrier spacing Δ f satisfies Δ f = 1 / T S . This managed overlap removes the passive guard bands needed by traditional FDM, thus enhancing spectral efficiency.

To reduce the delay spread of the channel, a cyclic prefix (CP) is added to the beginning of every symbol. By absorbing inter-symbol interference, the CP converts the linear convolution of the channel into a circular convolution, reducing receiver equalization to a straightforward complex division per subcarrier.

Table 1 below summarizes the main differences between 4G LTE and 5G NR.

Table 1.Summary Comparison of 4G LTE/5G NR Parameters.

Furthermore, OFDM inherently aligns with multi-antenna (MIMO) architectures. Due to the fact that every subcarrier undergoes uniform fading, the spatial processing needed to merge the streams from multiple antennas is greatly simplified, making it possible to implement large-scale MIMO systems that are vital for increasing capacity in crowded urban environments.

However, there are two significant physical constraints that practical OFDM setups encounter. The first issue is that the coherent addition of subcarriers results in a high Peak-to-Average Power Ratio (PAPR), pushing power amplifiers into nonlinear areas of operation, which presents a major financial and energy efficiency challenge for base station implementations. The second issue is that high sensitivity to frequency shifts caused by Doppler effects disrupts the orthogonality between subcarriers, leading to inter-carrier interference (ICI) in situations with high mobility.

As a result, moving from a fixed 4G LTE framework to a more adaptable 5G NR structure necessitates addressing these physical obstacles while maintaining compatibility with existing systems. Optimizing these physical layer parameters in Table 1 forms the core of the techno-economic feasibility study conducted on Telecel Faso’s network.

To translate the physical principles of CP-OFDM discussed in Section II into quantifiable performance indicators, an end-to-end transmission chain was implemented within the MATLAB environment [11]. The decision to build a custom algorithmic framework—bypassing proprietary, commercial 5G toolboxes—serves a dual scientific and operational objective: it ensures granular control over every processing block (QAM mapping, IFFT synthesis, CP insertion, channel modeling, FFT, equalization, and demodulation) while faithfully replicating the core processing sequences executed by the Digital Signal Processors (DSP) within the Huawei BBU BTS5900 baseband units [12] currently deployed across Telecel Faso’s network.

To balance computational efficiency with physical accuracy, the baseline physical model employs a 64-point FFT core configured with 52 active subcarriers (a scaled, representative model of the full-scale 2048-point LTE standard), appended with a 16-sample cyclic prefix ( C P ≈ 25 % ). The physical layer chain supports QPSK, 16-QAM, and 64-QAM modulation schemes.

Performance evaluation is conducted across two complementary propagation environments:

An Additive White Gaussian Noise (AWGN) channel: serving as the theoretical baseline for PAPR characterization and constellation decision threshold validation.A 3-tap frequency-selective Rayleigh fading channel: defined by relative delay profiles of [ 0 , 1 , 3 ] samples and power gain profiles of [ 1 , 1.5 , 0.25 ] , accurately capturing the multipath reflection profile characteristic of Bobo-Dioulasso’s dense urban topology.

To evaluate Doppler shift resilience and lay the groundwork for 5G NR migration dimensioning, the four 3GPP-standardized numerologies μ ∈ { 0 , 1 , 2 , 3 } , corresponding to subcarrier spacings Δ f from 15 to 120 kHz, were integrated into the simulator. Each operational scenario undergoes Monte Carlo evaluation over 1,000 consecutive OFDM symbols per Signal-to-Noise Ratio (SNR) step, swept across a range of 0 to 30 dB. Table 2 summarizes the complete parameter set governing the simulation platform. They are aligned with the reference 5G NR numerology ( μ = 0.15 kHz subcarrier spacing, 20 MHz channel bandwidth).

Table 2. Main parameters of simulated OFDM chain.

Figure 1 shows the general architecture of simulated OFDM chain.

The sampling rate of 30.72 MHz corresponds to a 20 MHz bandwidth, that is, a ratio fs/B of 1.536. The 3.5 GHz carrier lies in band n78, which constitutes one of the principal axes of worldwide 5G NR deployment and is envisaged for the deployment of Telecel Faso. The Doppler model retained is Jakes’ model, conventionally employed for mobile channels. At 500 km/h and 3.5 GHz, the maximum Doppler frequency reaches approximately 1.62 kHz, that is, nearly 11% of the subcarrier spacing for µ = 0. This ratio regulates the interference between carriers and the resilience of the system at elevated speeds. The power-delay profile is adjusted so that the total of the powers equals one, ensuring a consistent reference SNR no matter the setup.

In exploring each physical limitation discussed in the following sections (PAPR, Doppler shift, and MIMO application), the approach follows a thorough four-step process: first, empirical observation of the physical effect through simulation; second, application of a specific mitigation strategy; third, quantitative assessment of the resulting compromises; and lastly, development of practical engineering suggestions. This organized method provides a consistent foundation for evaluating physical limitations before they are integrated into the techno-economic viability analysis in Section VII.

The preceding results allow for the definition of an adaptive zoning model, summarized in Table 4, which pairs each geographic zone type with a specific numerology, antenna configuration, and 5G NR service mix aligned with the measured empirical constraints.

This paper presented an in-depth simulation-based characterization of an OFDM transmission chain operating under LTE/5G NR requirements to resolve three fundamental physical layer bottlenecks: Peak-to-Average Power Ratio (average PAPR of 6.7 dB, reduced by 2.5 dB via an optimal 8 dB clipping threshold), Doppler tolerance (link integrity guaranteed up to 350 km/h using numerology μ = 2), and spatial diversity exploitation (15 to 18 dB gain delivered by the 2 × 2 Alamouti MIMO scheme over a Rayleigh fading channel). The underlying techno-economic evaluation demonstrates that a Phase-1 5G\NR deployment is immediately actionable for Telecel Faso via a software-only upgrade of existing radio infrastructure, requiring an initial investment estimated at 10 million XOF.

Beyond its direct operational impact on the Burkinabè market, this research establishes that a contextualized, adaptive deployment model—in contrast to uniform rollouts modeled after mature markets—represents the most viable trajectory for mobile operators in developing nations. Future research avenues emerging from this study, including artificial intelligence-driven channel estimation [16], extension into millimeter-wave spectrum bands [17], and network slicing modeling for URLLC applications in telemedicine and connected agriculture, set the groundwork for next phases of this infrastructure evolution.

The authors express their gratitude to the Technical Directorate of Telecel Faso in Bobo-Dioulasso for providing access to radio site infrastructure and operational network data, which proved essential for contextualizing and validating this techno-economic study.

Conceptualization, methodology, and formal analysis: Wendpuiré Ousmane Compaoré. Research and data management: Izak Kévin Compaoré and Maxime Kafando. Analytical review: Izak Kévin Compaoré and Maxime Kafando. Supervision: Zacharie Koalaga. Writing, revision, and editing: Wendpuiré Ousmane Compaoré, Izak Kévin Compaoré, and Zacharie Koalaga. All authors have read and approved the final version of manuscript.