A Multi-Band Full-Duplex Prototype for Integrated Sensing and Communication

A Multi-Band Full-Duplex Prototype for Integrated Sensing and Communication Bixing Yan, Andre Kokkeler, Yang Miao ∗ November 9, 2025 Abstract Integrated sensing and communication (ISAC) has emerged as a key enabler for the next-generation radio network. ISAC systems aim to enable wireless sensing and data transmission functionality in one integrated system. While existing ISAC prototypes in the literature predominantly focus on single-band full-duplex operations with array antennas and spatially separated sensing targets and communication user equipment (UE) to mitigate signal interference, this paper presents a novel multi-band full-duplex ISAC prototype leveraging a software-defined radio (SDR) USRP X440. In this initial implementation, constrained by host PC performance limitations, the proposed system supports a monostatic sensing link at 25.175 GHz with 360 MHz bandwidth for range and velocity estimation alongside a bistatic communication link at 24.25 GHz with 360 MHz bandwidth, each employing distinct orthogonal frequency-division multiplexing (OFDM) waveforms. Experimental results demonstrate successful target detection with a range resolution of 0.4 m and a velocity resolution of 0.2 m/s while maintaining wireless communication performance with minimal signal interference between sensing and communication functionalities. 1 Introduction Integrated sensing and communication (ISAC) systems utilize partially or fully shared radio signals and/or hardware to achieve wireless sensing and wireless data transmission functions. Using orthogonal frequencydivision multiplexing (OFDM) waveforms for both sensing and communication is one solution for tightly integrated ISAC systems [1]. Building ISAC prototypes is the first step in implementing ISAC functionality. Existing ISAC prototypes introduced in [2, 3] predominantly focused on full-duplex sensing and communication within a single frequency band, eliminating signal interference through directional antennas and separately located targets and UEs. These current prototypes also required multiple FPGA modules or software-defined radios (SDR), increasing system complexity. This paper presents an SDR-based multi-band full-duplex ISAC prototype utilizing a single USRP X440, operating at the lower end of 5G NR FR2 or the upper range of mid-band FR3. By allocating separate frequency bands for sensing and communication, the multi-band approach can reduce mutual interference more effectively than single-band systems [1]. Full-duplex operation allows simultaneous transmission and reception, enabling the capture of target echoes in real time without introducing dead zones. As an initial step, the prototype uses two transmit and two receive channels: one transmit–receive pair for monostatic OFDM sensing at 25.175 GHz, and another for bistatic wireless OFDM communication at 24.25 GHz. Due to host PC performance limitations, each channel currently operates with a 360 MHz bandwidth; however, the system is designed to support up to 800 MHz with future upgrades and optimization. Despite the reduced bandwidth, the prototype successfully demonstrates the feasibility of multi-band operation, particularly in mitigating interference between sensing and communication, and validates the performance of both functionalities. ∗University of Twente 1 2 Hardware and software architecture of the proposed prototype The hardware structure of the proposed prototype is shown in Fig. 1. It comprises three main components: the intermediate frequency (IF) band transceiver, implemented with the SDR USRP X440; the local oscillator (LO), realized using a voltage-controlled oscillator (VCO); and the radio frequency (RF) part, where antennas are responsible for transmitting and receiving signals. Figure 1: Hardware architecture of proposed ISAC prototype The USRP X440 features eight TX/RX channels covering 1 MHz to 4 GHz, with bandwidth determined by its FPGA version. For this demonstration, the CG 1600 FPGA configuration enables four active channels across two IF bands: one TX-RX pair at 0.25 GHz and another at 1.175 GHz. A host PC with AMD 7965WX CPU and 100GbE network card controls the USRP X440 via USRP Hardware Driver (UHD) with C++. A VCO generates the 24 GHz LO signal in the LO stage, amplified by an amplifier (P A1in Fig. 1) before a 4-way splitter distributes it to all channels for frequency conversion. I/Q mixers perform up/down conversion in the RF band for 5G NR FR2 compatibility. The system employs two co-located 3 ×12-element array antennas (15 dBi gain each) for transmission (one for sensing with beams facing towards sensing targets, one for communication with beams facing the user equipment (UE)), with a nearby horn antenna (20 dBi gain) receiving monostatic sensing signals. A separate horn antenna connected via a 2 m cable serves as a downlink UE for bistatic communication reception. The prototype operates in two RF bands: 24.25 GHz for wireless communication and 25.175 GHz for monostatic sensing. The software architecture of the proposed prototype comprises three main components: signal generation for sensing and/or communication, sensing algorithms, and communication receiver algorithms. Signal generation is handled using MATLAB, allowing customization of distinct waveforms for the communication and sensing channels. The sensing algorithms, implemented in Python as illustrated in Fig. 2, are used for post-processing to reduce CPU load during continuous streaming with the USRP. These algorithms focus on estimating target range and velocity using a standard FFT-based approach with synchronized transmit and receive signals. The communication receiver is also implemented in Python for offline processing and includes basic OFDM demodulation algorithms. 2 Figure 2: ISAC prototype’s signal processing architecture 3 Proof-of-Concept Trial and Example Results Indoor measurements evaluated the prototype’s performance using a 30cm ×30cm metal plate as a stable sensing target positioned at 1.3 m or 1.9 m from the sensing transceiver. The communication receiver was 2 m away from the communication transmitter. The scenario is shown in Fig. 3. Two distinct OFDM signals were employed for sensing and communication, respectively. Both signals utilized 4096 subcarriers over a 360 MHz bandwidth, enabling a range resolution of 0.4 m. For velocity estimation, 2000 OFDM symbols were used, achieving a velocity resolution of 0.2 m/s. Figure 3: Indoor measurement scenario of the ISAC prototype Communication performance was evaluated by varying the sensing frequency band and assessing its interference with the communication channel through the bit error rate (BER). Three scenarios were tested: sensing and communication sharing the same frequency band, sensing and communication operating on different frequency bands, and communication without sensing signals. In all cases, the sensing target was positioned 1.3 m from the sensing transceiver. A total of 3.8 million bits were transmitted, and the results are presented in Fig. 4. The results indicate that assigning separate frequency bands for sensing and communication effectively minimizes interference, as the BER matches that of the communication-only scenario. Conversely, using the same frequency band for both functionalities results in severe interference from the sensing channel, rendering communication unreliable. 3 Figure 4: BER performance comparison for multi-band ISAC system showing (a) bit error rates and (b) absolute error counts across three scenarios: same frequency operation, different frequency operation (925 MHz separation), and communication-only baseline. Sensing performance for range and velocity estimation for the target at two different locations is shown in Fig. 5. From the figure, it is clear that the target was detected successfully with correct range and velocity estimation results. Also, the communication receiver antenna can be seen as a second target in Fig. 5a in the bin around 2 m and 0 m/s. In Fig. 5b, it does not appear as a distinct second target. This is because the distance between the antenna and the main target is smaller than the system’s range resolution, making them indistinguishable in the range domain. (a) Target at 1.3 m range and 0 m/s velocity (b) Target at 1.9 m range and 0 m/s velocity Figure 5: Range-Doppler maps showing target detection at different locations. 4 Conclusion This paper presents a full-duplex integrated sensing and communication (ISAC) prototype developed using the USRP X440 platform. The system currently operates in two frequency bands (24.25 GHz and 25.175 GHz) and supports two transmit and two receive channels. Each channel employs a bandwidth of 360 MHz, enabling a range resolution of 0.4 meters and a velocity resolution of 0.2 m/s for monostatic OFDM-based sensing. Simultaneously, it supports bistatic wireless communication with minimal interference from the sensing channels. Future work will enhance the host PC’s performance to increase bandwidth (up to 800 MHz 4 per channel) and explore advanced signal synchronization and angle estimation algorithms. Additionally, the system will be evaluated in more complex scenarios involving multiple and dynamic targets. References [1] C. Sturm et al., ”An OFDM System Concept for Joint Radar and Communications Operations,” VTC Spring 2009 - IEEE 69th Vehicular Technology Conference, Barcelona, Spain, 2009, pp. 1-5. [2] Q. Zhang et al., ”Sensing and Communication Integrated System for Autonomous Driving Vehicles,” IEEE INFOCOM 2020 - IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS), Toronto, ON, Canada, 2020, pp. 1278-12793. [3] C. D. Ozkaptan et al., ”Software-Defined MIMO OFDM Joint Radar-Communication Platform with Fully Digital mmWave Architecture,” 2023 IEEE 3rd International Symposium on Joint Communications & Sensing (JC&S), Seefeld, Austria, 2023, pp. 1-6. 5