This paper introduces a new coded modulation framework for massive MIMO systems that leverages low-resolution analog-to-digital converters (ADCs) alongside superposition-based 64-ary QAM modulation to increase spectral utilization efficiency. To address the complexity of signal detection, the proposed approach reformulates the original 64-ary QAM channel into an equivalent BPSK model. This reformulation supports a two-layer Tanner graph structure for the combined tasks of signal detection and channel decoding, with computational complexity that grows quadratically with the number of transmit antennas. Protograph LDPC (PLDPC) codes are adopted in the system due to their excellent error correction performance, and integrates an “equal-weight” superposition constellation, which delivers a 0.7–1 dB gain in performance compared to the traditional “equal-distance” design under coarse quantization. Furthermore, the paper introduces a modified PEXIT analysis method that holistically accounts for the combined effects of MIMO fading, quantization noise from ADCs, and LDPC decoding characteristics. Simulation results confirm the reliability of the analytical approach and highlight the critical role of optimized power allocation in maintaining high energy and spectral efficiency in emerging 5G and 6G networks utilizing lower solution receivers.