Nowadays, the Internet‐of‐Things (IoT)‐related applications require the design of low‐cost and low‐power wireless sensor nodes. Although backscatter radio communication is a mature technology used in radio frequency identification (RFID) applications, ambient backscattering is a novel approach taking advantage of ambient signals to simplify wireless system topologies to just a sensor node and a receiver circuit eliminating the need for a dedicated carrier source. This work introduces two novel wireless tags that utilize broadcast frequency modulated (FM) signals for backscatter communication. The first tag uses binary modulation and the second, a high‐order modulation technique for the communication. Each proof‐of‐concept tag consists of an ultra‐low power microcontroller (MCU) and a radio frequency (RF) front‐end for wireless communication. The MCU can accumulate data from sensors through an analog‐to‐digital converter (ADC), while it transmits the information back to the receiver through the front‐end. In the first tag, the front‐end uses On‐Off keying (OOK) binary modulation and FM0 encoding on ambient FM signals. The second design utilizes the four‐pulse amplitude modulation (4‐PAM) and uses a digital‐to‐analog converter (DAC) for the RF front‐end control. The receiver consists of a commercial low‐cost software defined radio (SDR) that downconverts the received signal to baseband and decodes it using a suitable signal processing algorithm. Both prototype tags were demonstrated indoors using an FM station at 34.5 km away. The first tag with binary modulation was tested for a 5 m tag‐reader distance, and it was demonstrated by backscattering information at 2.5 kbps featuring an energy per packet of 36.9 μJ. Using the 4‐PAM, the second prototype was demonstrated over a 2 m tag‐reader distance for a low bit rate of 328 bps with power consumption only 27 μW