What the paper shows: The authors study integrated sensing and communication (ISAC) systems where both the communication receiver and the sensing receiver use 1-bit analog-to-digital conversion. Working in a simple but common wireless model, they derive information-theory limits and find that, under one realistic channel-knowledge assumption, there is no fundamental trade-off between sensing and communication. Both tasks can reach their respective information limits simultaneously with a particular low-complexity signaling rule.

What the system looks like: The transmitter sends a complex-valued signal subject to an average power constraint. The same transmitted signal is used for two separate links: a communication link to a receiver that decodes a message, and a monostatic sensing link where the reflected signal is used to estimate a parameter of the environment. Both channels are modeled as complex Gaussian fading with additive Gaussian noise. Each receiver applies a 1-bit I/Q quantizer, meaning the receiver records only the sign of the real and imaginary parts of the received signal, so each sample takes one of four possible values. The authors measure communication performance with communication mutual information (CMI) and sensing performance with sensing mutual information (SMI), two standard information-theory metrics that count how many bits of information the quantized outputs convey.

Main result when the receiver knows the channel: When the communication channel state is known at the communication receiver (but not at the transmitter), the paper characterizes the joint capacity region for communication and sensing. The surprising conclusion is that there is no inherent trade-off: one can achieve the maximum communication rate and the maximum sensing rate at the same time. The input distribution that achieves both limits has constant amplitude and a specific rotational symmetry: intuitively, the transmitter uses symbols that lie on a circle and are rotated by 90 degrees (π/2) steps. This structure matches the coarse 1-bit I/Q quantization and makes efficient use of the limited receiver resolution.