The fifth-generation (5G) cellular standard takes the LTE foundation and pushes it into territory that earlier generations could not reach: peak downlink data rates of 20 Gbps under the 3GPP’s enhanced Mobile Broad Band (eMBB) service class, end-to-end latencies below one millisecond under Ultra-Reliable Low-Latency Communications, and Block Error Rates that have to settle at 10⁻⁴–10⁻⁶ once HARQ retransmissions converge. Within those targets, two metrics — throughput and Bit Error Rate — act as the make-or-break indicators of physical-layer quality. This paper presents a review of the techniques that have been reported in the literature to raise throughput, suppress BER, or find workable trade-offs between the two. The coverage spans modulation design, massive Multiple-Input Multiple-Output and adaptive beamforming, Orthogonal Frequency Division Multiplexing with Peak-to-Average Power Ratio mitigation, MIMO-OFDM with Index Modulation, channel estimation and hybrid detection, Non-Orthogonal Multiple Access, Low-Density Parity-Check coding, and the newer class of machine-learning-driven slicing and resource-allocation algorithms. Verified quantitative findings drawn from the surveyed work include a 60% throughput increase when Weighted Overlap-and-Add filtering is integrated into Maximum Ratio Transmission at the transmitter, a 30% BER reduction accompanied by a 25% spectral-efficiency gain when Fractional Least Mean Square beamforming is used in place of conventional LMS in a 4×4 MIMO setup, and an up-to-20% throughput gain from Block Markov Superposition Transmission-enhanced 5G LDPC coding at a constant BLER target. The paper concludes by highlighting open problems around energy efficiency, high-mobility CSI robustness, and ITU-T-aligned slice enforcement that will determine the trajectory of 5G and early 6G research.
 

5G,throughput, bit error rate,massive MIMO, beamforming,OFDM, PAPR, NOMA, LDPC, network slicing, dynamic spectrum sensing.