无源感知作为物联网的连接引擎，因其能量自供给、免维护、兼具感知计算通信等优点受到广泛研究。在感知数据不断增大、传输需求动态变化的场景中，无源感知系统由于采用单一通信方式，其环境适应能力受限，存在吞吐量受限、能量利用率低下、频谱利用率不高、数据传输实时性较差等问题，无法满足在有限带宽内进行数据快速大量传输的需求。本项目拟以多调制适配作为切入点提高无源系统的能量利用率、频谱利用率以及吞吐量，基于神经网络方法对信道进行预测，针对不同信道状况自适应选择调制方式，解决“抗噪性能较差”的问题；基于能量和相位角度特征设计合理的聚类算法最大程度的提高收敛速度、聚类效率，解决“解码效率较低”的问题；通过建立链路能量预算得出射频前端的最优阻抗配置，解决“能量效率不高”问题。在此基础上，本项目将设计并实现无源感知多调制适配系统原型，充分验证方案的有效性，为无源系统技术的成熟应用提供有力支撑。

As the connection engine of the Internet of things, passive sensing system has attracted more attention and research due to its advantages of self-supply of energy, maintenance-free and perceptual computing communication. However, in the scenario where the perception data keeps increasing and the transmission demand changes dynamically, the environment adaptability of the passive perception system is limited due to the single modulation mode. There are problems such as limited throughput, low energy utilization rate, low frequency spectrum utilization rate and poor real-time performance of data transmission, which cannot meet the demand for fast and large data transmission within limited bandwidth. Such problems make it impossible to meet the demand for rapid mass transfer of data within a limited bandwidth. This project intends to adopt multi-modulation adaptation as the breakthrough point to improve the energy efficiency, spectrum efficiency and throughput of passive systems. The channel is predicted based on neural network, and the modulation mode is adaptively selected for different channel conditions to solve the problem of “poor anti-noise performance”. Based on the energy and phase angle characteristics, a reasonable clustering algorithm is designed to maximize the convergence speed and clustering efficiency, so that the problem of “low decoding efficiency” be solved. The optimal impedance configuration of the RF front-end is obtained by establishing the link energy budget to solve the problem of “energy efficiency is not high”. On this basis, the project will design and implement a prototype of passive sensing multi-modulation adaptation system, fully verify the effectiveness of the scheme, and provide strong support for the maturity of passive system technology.