5G networks are designed to move data between the base station and user equipment. The same transmitted radio signal also interacts with objects around it. Buildings, vehicles, people and other objects reflect part of the RF energy back towards the receiver. A recent OpenAirInterface implementation shows that these reflections can be processed for radar-like sensing while the 5G base station continues to provide normal communication service to commercial user equipment. So, now let us see if 5G Communication Signals Can Also Sense the Environment along with Accurate LTE RF drive test tools in telecom & Cellular RF drive test equipment and Accurate Wireless Survey Software Tools & Wifi site survey software tools in detail.
This work is a practical example of Integrated Sensing and Communication (ISAC). Instead of deploying a separate radar waveform and separate radio infrastructure, the implementation uses the existing 5G OFDM transmission for both communication and sensing.
Using the Existing 5G OFDM Signal
The interesting part of this implementation is that the 5G waveform itself was not modified for radar operation.
The sensing function was integrated directly into an OpenAirInterface 5G base station. OpenAirInterface already provides a 3GPP-based 5G NR implementation supporting commercial UE connectivity, PDSCH, PDCCH, MIMO, different subcarrier spacing configurations and normal 5G user-plane traffic.
During normal downlink operation, the gNB knows the OFDM symbols that it transmits. The sensing receiver captures the reflected signal and uses knowledge of the transmitted symbols to remove the communication data component. The resulting information can then be processed to determine the characteristics of objects reflecting the RF signal.
The implementation performs this processing in real time rather than collecting IQ data first and analysing it later.
Range and Velocity Detection from 5G
After removing the communication symbols, the system performs range-Doppler processing.
This allows the received reflections to be analysed in two useful dimensions:
Range indicates how far the reflecting object is from the sensing system.
Doppler indicates movement and can be used to estimate the object’s radial velocity.
The reported implementation achieved a nominal 2.57 metre range resolution and 0.28 m/s velocity resolution. It also uses ordered-statistic CFAR detection to identify targets from the processed radar information.
These results are relevant because they were produced using a working 5G base-station implementation rather than a communication waveform generated only for an offline sensing experiment.
Communication Continues While Sensing Runs
One of the main technical questions around ISAC is what happens to normal mobile-network performance when sensing is enabled.
The researchers tested the implementation with a commercial UE connected on the same carrier. Their measurements reported no measurable difference in the estimated downlink throughput when sensing was enabled. The sensing bandwidth also followed the bandwidth allocated by the 5G scheduler.
This is a useful result.
The base station was not switching between a communication mode and a radar mode. Communication and sensing were operating from the same transmitted 5G signal.
From a network implementation point of view, this is much more practical than reserving large amounts of radio resource only for sensing.
Real-Time Processing Is Another Key Requirement
Radar processing can add significant computational load to a base station, particularly when range-Doppler processing is performed continuously.
The implementation therefore measured the processing load of the sensing worker. The reported conservative utilisation bound was 58.4%, with no dropped sensing measurements during the test.
The work also identified a hardware issue related to deterministic transmit-receive phase rotation on the USRP X300 platform. After carrier-frequency alignment was corrected, clutter suppression and coherent processing performance improved considerably.
This part of the result matters because ISAC performance depends not only on algorithms. RF hardware behaviour, phase stability, timing and calibration directly affect sensing accuracy.
Connecting 5G Sensing to O-RAN
The implementation goes beyond target detection at the base station.
A custom E2SM-RADAR service model was developed to send sensing detections and processed information through the O-RAN E2 interface to a near-real-time RAN Intelligent Controller. The researchers demonstrated controller-side functions including object tracking, micro-Doppler processing and classification.
This gives an indication of how sensing could eventually become part of network intelligence rather than remaining a standalone RF experiment.
A future base station could potentially provide communication coverage while also generating information about movement around roads, industrial sites, campuses or other controlled environments.
The main result from this OpenAirInterface work is therefore straightforward: a standard 5G OFDM transmission can carry user traffic and provide radar-like sensing information at the same time, without changing the transmitted waveform and, in the reported test, without a measurable reduction in estimated downlink throughput.
That moves ISAC one step closer from research algorithms towards implementation in a working 5G RAN.
Sources: The technical results are from the August 2026 research paper Real-Time Symbol-Domain OFDM Radar in an OpenAirInterface 5G Base Station With O-RAN Sensing Services, supported with current OpenAirInterface 5G NR implementation documentation.
About RantCell
RantCell is a mobile network testing and monitoring solution designed for telecom operators, enterprises, regulators and engineering teams. It supports RF drive testing, indoor testing, 4G/5G measurements, private network monitoring, automated testing and cloud-based reporting. RantCell helps teams measure network performance from the actual user and device perspective using Android smartphones and supported CPE devices. Also read similar articles from here.