Adaptive Quantization Transmission Mechanism to Optimize SE in ISAC Systems

The sixth-generation (6 G) integrated sensing and communication (ISAC) paradigm is aimed at simultaneously delivering high-capacity communication and high-accuracy sensing for emerging intelligent network services. A major challenge arises from the separation of sensing measurements and processing nodes, as performing quantization during data transmission fundamentally limits the accuracy of sensing; however, this topic has not been adequately modeled in prior studies. To address this issue, we derive analytical approximations of the quantization noise introduced under uniform and non-uniform quantization schemes. Afterward, we establish a closed-form expression for the target detection probability that explicitly incorporates both the sensing signal-to-noise ratio (SSNR) and signal-to-quantization noise ratio. On the basis of these results, we propose an adaptive quantization transmission mechanism (AQTM) that dynamically selects quantization strategies and bit resolutions to balance sensing performance with spectrum efficiency (SE). The simulation results match the quantizer approximation results when the number of uniform and non-uniform quantization bits is greater than 3 and 5, respectively. The target detection probability achieved in the simulation is consistent with the corresponding closed-form expression. Compared with a 16-bit baseline, the introduction of the AQTM in the prototype reduces sensing data transmission resources by 62.54% at an uplink communication signal-to-interference-plus-noise ratio (SINR) of 15 dB and an SSNR of 6.8 dB and by 68.7% at an uplink communication SINR of 5 dB and an SSNR of 7.5 dB. These results validate our theoretical framework and show the practical value of the AQTM in significantly increasing spectral efficiency.