Radar-communication integration, a system that simultaneously enables target detection and information exchange, has become an important research direction in radar and communication technologies in recent years, with vehicle-mounted radar communication systems being one of its key applications. This paper reviews foreign research on radar communication systems, introduces the evaluation indicators, system structure, and waveform design schemes of integrated radar-communication systems, summarizes the simulation performance research results of the system's radar and communication indicators, and analyzes the application prospects and development trends of vehicle-mounted radar communication systems for autonomous driving and intelligent transportation.
The concept of radar communication was proposed around the beginning of the 21st century [1-5], and the concept of radar-communication integration was proposed to adapt to future high-tech warfare. Radar and communication systems, as fundamental components of electronic warfare platforms, play a crucial role in military operations. For a long time, these systems have developed independently, but with technological advancements, the gaps between systems have gradually narrowed. Therefore, the issue of horizontal integration between systems has begun to receive attention, namely, merging existing systems horizontally to make them more versatile and multifunctional. If radar-communication integration can be achieved, it will not only reduce the size and electromagnetic interference of electronic warfare platforms but also improve battlefield command efficiency.
Although radar and communication systems differ significantly in their operating methods, functionalities, and signal characteristics due to their different applications, both systems are fundamentally related to the transmission and reception of electromagnetic waves in space. Structurally, both systems include modules such as antennas, transmitters, receivers, and signal processors. Technologically, the functions of radar, traditionally implemented by hardware devices, are increasingly being replaced by digital signal processing. Simultaneously, the carrier frequencies of communication systems are shifting to the microwave domain, placing them on the same order of magnitude as those used in traditional radar. Therefore, radar and communication systems are converging from their hardware implementation to their software algorithms.
The integration of radar and communication systems is primarily based on sharing the same hardware platform. The simplest approach is time-division sharing, which uses a gating switch to time-multiplex the antenna, transmitter, and receiver hardware platforms of the radar and communication systems. However, in this approach, neither system can continuously occupy resources for extended periods, otherwise it would affect the performance of the other system. Furthermore, the system's performance is limited due to its limited operating time.
Another hardware platform sharing approach is primarily used in phased array radars, where a two-dimensional array is divided into multiple subarrays, each operating independently to perform radar or communication functions. However, due to power limitations of the subarrays, the performance of both the radar and communication systems is affected. Therefore, this integrated technology, which combines hardware sharing with independent radar and communication functions, suffers from resource constraints, impacting both system performance and efficiency. Consequently, recent research on radar-communication integration has focused on signal fusion, i.e., using the same signal on the same hardware platform to achieve both radar and communication functions.
Vehicle-mounted radar communication systems utilize the millimeter-wave radar already installed in vehicles and integrated radar-communication technology. This not only enables vehicle-mounted radar detection and inter-vehicle communication, but also avoids adding extra hardware modules to the vehicle and complicating the vehicle's electromagnetic environment through the introduction of communication functions. This reduces costs and improves spectrum utilization. Therefore, vehicle-mounted radar communication systems represent a significant breakthrough in the transition of integrated radar-communication technology from military to civilian applications.
1. The significance of research on vehicle-mounted radar communication systems
Vehicle-mounted radar communication systems are facing tremendous market opportunities. On the one hand, governments worldwide are placing increasing emphasis on traffic safety, with automotive safety technologies such as automatic emergency braking, forward collision warning, and lane departure warning being increasingly incorporated into relevant laws and regulations. On the other hand, autonomous driving has become a global research hotspot and has even risen to the level of a national strategy within "Made in China 2025."
Currently, advanced driver assistance systems (ADAS) in the industry rely solely on various onboard sensors such as cameras, infrared sensors, and lidar to assist single-vehicle intelligent driving. In contrast, onboard radar communication systems integrate modern communication technology onto the onboard millimeter-wave radar system. While achieving radar detection, they also establish vehicle-to-everything (V2X) communication connections, enabling the vehicle to simultaneously possess functions such as complex environment perception, information sharing, and intelligent decision-making, providing the most effective guarantee for intelligent driving.
Vehicle-mounted radar communication systems not only possess the detection advantages of millimeter-wave radar—namely, all-weather, all-time, high-precision, and high-resolution capabilities in vehicle environmental perception—but also enable the transmission and sharing of information about the vehicle itself and its surroundings through a vehicle-to-everything (V2X) network established by the system. This allows vehicles to gain environmental awareness beyond visual range, thus endowing them with "visual + auditory" capabilities. Therefore, vehicle-mounted radar communication systems are the most fundamental and effective means of supporting intelligent driving and smart transportation.
Furthermore, with the advent of the 5G era, communication frequency bands are no longer limited to below 6GHz, but have expanded to the microwave bands ranging from tens of gigahertz to hundreds of gigahertz. Vehicle-mounted radar communication systems can use globally unified spectrum, namely the 24GHz, 77GHz, and 79GHz bands. These bands are close to the high-frequency communication bands of 5G and microwave communication. Therefore, based on the common origins of radar and communication technologies, researching vehicle-mounted radar communication systems not only makes it possible for vehicles to establish vehicle-to-vehicle communication networks via vehicle-mounted radar, but also provides technical accumulation for the research of 5G high-frequency communication technology.
2. Current Research Status of Vehicle-Mounted Radar Communication Systems
2.1 Evaluation Indicators for Radar-Communication Integration
Evaluation metrics for integrated radar and communication systems include radar metrics and communication metrics. Radar metrics typically relate to the radar's measurement requirements for various aspects of a target, such as distance, velocity, and angle, and mainly include measurement range, measurement accuracy, and resolution. Communication metrics primarily include signal-to-noise ratio and data rate, and the introduction of communication functionality should not reduce the radar's detection performance.
The following mainly introduces radar specifications.
2.1.1 Distance
(1) Radar range
The effective range of a radar can be derived from the radar equations. These equations relate the radar's effective range to various factors such as radar transmission, reception, antenna, and environment, reflecting the degree to which each radar parameter affects the radar's effective range. The basic radar equations are:
Where R_max is the maximum range of the radar, P_t is the radar transmit power, G_t and G_r are the gains of the transmit and receive antennas, respectively, σ is the radar cross-section, and S_sim is the minimum detectable signal power of the radar receiver.
(2) Distance measurement range
The ranging range includes the minimum measurable distance and the maximum single-value ranging range.
The minimum measurable range refers to the closest target distance that a radar can measure. For pulse radar, the transmitting and receiving antennas are shared. During the duration of the transmitted pulse width τ, the receiver cannot receive the target echo. After the transmitted pulse ends, switching the antenna from transmit to receive mode also requires a certain time t_0, during which the receiver cannot receive the target echo. Therefore, the minimum measurable range of the radar is:
The maximum single-value ranging range of a radar is determined by the pulse repetition period T_m. To ensure single-value ranging, T_m should typically be selected as ≥ 2R_max/c, where R_max is the maximum effective range of the target. When the radar repetition frequency cannot meet the requirements for single-value ranging, range ambiguity will occur.
(3) Distance resolution
Range resolution typically refers to the minimum distinguishable distance between two point targets of the same size in the same direction. For simple pulse radar signals, the narrower the pulse, the better the range resolution. For complex pulse compression signals, the effective bandwidth B of the radar signal determines the range resolution; the wider the effective bandwidth, the better the range resolution. Range resolution can be expressed as:
(4) Distance measurement accuracy
Ranging accuracy refers to the accuracy with which radar measures the distance to a target, and is generally expressed as root mean square error (RMSE). Theoretically, the minimum RMS error of the distance to a single strong scattering point can be expressed as:
Where E/N_0 is the signal-to-noise ratio. It can be seen that the ranging accuracy of a radar is inversely proportional to both the signal bandwidth and the signal-to-noise ratio.
2.1.2 Speed
Based on the Doppler frequency, where v_r is the radial velocity, the velocity measurement accuracy can be expressed as:
The speed resolution is:
Where τ is the signal duration, which is proportional to the signal width. It can be seen that both speed measurement accuracy and speed resolution are inversely proportional to the signal width, and the shorter the signal wavelength, the higher the speed measurement accuracy and speed resolution.
2.1.3 Angle
The angle measurement is related to the antenna aperture. If the antenna's half-power beamwidth is:
The measurement accuracy of azimuth or elevation angle can then be expressed as:
2.2 Waveform Design of Radar-Communication Integrated System
The biggest challenge in integrated radar and communication systems is finding a suitable signal that can simultaneously perform information transmission and radar detection functions. Both radar and communication parameters are related to channel characteristics, the most important being Doppler frequency and maximum multipath delay. Since the echo travels twice the propagation path, these characteristics have a greater impact on radar. Besides the physical characteristics of the channel, there are also some limitations specific to radar performance, mainly related to the radar's ambiguity function.
The goal of traditional radar waveform design is to obtain a waveform with optimal autocorrelation characteristics to ensure radar detection performance. The selection of radar waveforms must consider three performance factors: target range, Doppler effect, and azimuth angle. For vehicle-mounted radar, in densely trafficked areas, the waveform should effectively combat interference and noise. Key performance indicators for communication include coverage, latency, data rate, and system capacity. The selection of communication waveforms must ensure resistance to various channel fading and multi-user interference to correctly demodulate and decode communication information.
Considering the existing radar implementation technology and existing communication technology, the main research directions of vehicle-mounted radar communication integration signals are: radar communication based on linear frequency modulation [4], radar communication based on spread spectrum [6-9], and radar communication based on OFDM [10-11]. Of course, these technologies can also be further combined with multi-antenna [7], beamforming and other technologies.
2.3 Radar Communication Based on Linear Frequency Modulation
Radar communication based on linear frequency modulation is mainly divided into two categories: schemes based on quasi-orthogonal waveform superposition [4-5] and schemes based on a single waveform [11-13]. The single waveform scheme can be further divided into two categories: schemes based on waveform separation [9, 14] and schemes based on fractional Fourier transform [6].
2.3.1 Scheme based on quasi-orthogonal waveform superposition
In the quasi-orthogonal waveform superposition scheme [4-5], radar and communication signals use mutually "orthogonal" waveforms. For example, radar detection can use a down-chirp signal (frequency decreases linearly with time), and communication data can use an up-chirp signal (frequency increases linearly with time). Two "orthogonal" matched filters are used to extract the desired signals respectively. Data between users can be distinguished by different frequency modulation slopes, different transmission times, and different starting frequencies. In this scheme, the radar signal is:
The communication signal is:
Radar signals and communication signals are essentially orthogonal within a single radar pulse.
Figure 1. Schematic diagram of the scheme based on quasi-orthogonal waveform superposition.
(DQPSK: Quad-phase relative phase shift keying)
Figure 1 shows a schematic diagram of the scheme based on quasi-orthogonal waveform superposition.
2.3.2 Waveform Separation Scheme
Figure 2. Schematic diagram of the waveform separation-based scheme
Figure 2 shows a schematic diagram of the waveform separation scheme [13]. At the transmitting end, the encoded communication information is modulated onto the radar waveform and transmitted. At the receiving end, the radar signal and the communication signal are separated by a separator and then processed separately. Separation methods include homomorphic filtering, whitening, etc.
2.3.3 Scheme based on fractional Fourier transform
Figure 3. Schematic diagram of the scheme based on fractional Fourier transform.
(FRFT: Fractional Fourier Transform)
Figure 3 shows a schematic diagram of the fractional Fourier transform (FRFT) scheme [11]. The radar signal and the communication signal are the same. The communication data is modulated on the Chirp signal with different initial frequencies. The receiver uses the fractional Fourier transform to extract the communication data and the radar signal respectively.
2.3.4 Spread Spectrum-Based Radar Communication
To achieve better communication performance, a spread spectrum signal with good autocorrelation characteristics can be used as the integrated radar and communication signal. The system transmits only one spread spectrum signal. On the one hand, the system uses the echo of its own transmitted signal for target detection, realizing radar functions; on the other hand, the system uses this transmitted signal to transmit communication data to other systems. Data between users is distinguished by different spreading codes.
The signal received by a radar [7] is:
The result is obtained by correlating the local code with the received radar signal:
After simplifying the formula, the relevant peak value can be obtained when τ=2R/c_0-(ki)T. At this time, k=i+2R/(c_0*T), and the distance of the target and the spreading code used are known. The data can be demodulated and the target speed can be obtained (derived from the code phase).
2.3.5 OFDM-based radar communication
OFDM signal[10] is also one of the research contents of waveform design of radar communication integrated system. As a radar signal, OFDM signal has a thumbtack-shaped ambiguity function, and has high resolution of range and Doppler, but no range-Doppler coupling problem. It can independently process range and Doppler information.
However, OFDM signals are more sensitive to Doppler frequency shift, which can disrupt the orthogonality between echo subcarriers, thus requiring frequency offset estimation and compensation. In addition, OFDM signals have a high peak-to-average power ratio (PAPR). To obtain higher transmit power, the PAPR of the signal should be reduced as much as possible, and a linear amplifier with a large dynamic range should be used.
Figure 4. Schematic diagram of OFDM-based scheme
(FFT: Fast Discrete Fourier Transform; IFFT: Inverse Discrete Fourier Transform)
The radar communication based on OFDM signals is illustrated in Figure 4 [10]. The radar signal and the communication signal are the same, and the echo received by a radar is [7]:
Divide it by the locally transmitted signal, and then perform a series of Inverse Discrete Fourier Transform (IDFT) / Discrete Fourier Transform (DFT) operations to obtain the distance (R) and velocity (f_D).
After receiving the above signals, other radars can obtain the communication data by performing Fast Fourier Transform (FFT) operations, demodulation, and decoding.
2.4 Simulation/Test Results
Currently, research on radar-communication integrated signals mainly focuses on three types: frequency-modulated continuous wave, spread spectrum signal, and orthogonal frequency division multiplexing (OFDM) signal. Related research and simulations are also mostly based on these three types.
2.4.1 Radar Performance
2.4.1.1 Simulation/Test Results of the Linear Frequency Modulation Scheme
In reference [4], the authors used Up-Chirp signals (radar) and Down-Chirp signals (communication) (which are basically orthogonal). The chirp signal had a modulation frequency of 40 MHz/μs, a system bandwidth of 80 MHz, a pulse time of 2 μs (the time required for the frequency to go from the lowest to the highest), a processing gain (time-width-bandwidth product) of 22 dB, and a data modulation method of π/4-DQPSK. The radio frequency was 10 GHz.
The simulation results in reference [4] show that when the signal-to-interference-plus-noise ratio (SINR) exceeds 15dB, the detection probability can reach 85% (or higher), thus most targets can be detected.
In reference [5], the authors used Up-Chirp signals (radar) and Down-Chirp signals (communication) (which are basically orthogonal). The chirp signal had a modulation frequency of 1 GHz/μs, a carrier frequency of 750 MHz, a system bandwidth of 500 MHz, a pulse time of 0.5 μs (the time required for the frequency to go from the lowest to the highest), and a processing gain (time-width-bandwidth product) of 24 dB. The data modulation method was binary phase shift keying (BPSK), the radio frequency was 75 MHz, and the transmit power was 27 dBm.
The test results from reference [5] show that its radar communication system can reliably distinguish two targets 63cm apart at a distance of 10m. In addition, reference [5] also mentions that its target detection probability is 99%.
2.4.1.2 Simulation/Test Results of Direct Sequence Spread Spectrum Scheme
In reference [11], the simulation settings used by the authors were: m-sequences for spread spectrum (SF=15, 31, 63, 127, 255); chip rate of 48 MCps; signal bandwidth of 96 MHz; data length of 256 symbols; and data modulation method of BPSK. From the simulation results in reference [11], it can be seen that when SINR exceeds 0 dB, the peak sidelobe (PSL) (SF=255) reaches 40 dB, thus effectively distinguishing two different targets.
2.4.1.3 Simulation/Test Results of OFDM Scheme
In reference [7], the simulation settings used by the authors were: carrier frequency of 5.9 GHz, number of full-phase OFDM subcarriers of 512, CP length of 1.4 μs, full-phase OFDM symbol length after adding CP of 23.8 μs, system bandwidth of 91.5 MHz, number of full-phase OFDM symbols in one frame of 177, time length of one frame of 4.25 ms, and subcarrier spacing of 180 kHz. From the simulation results of reference [7], it can be seen that when SINR exceeds 0 dB, the mean square error (MSE) of distance is almost close to 0, so that two different targets can be effectively distinguished; when SINR exceeds 0 dB, the MSE of Doppler frequency shift is about 100 Hz (equivalent to 5 m/s, 18 km/h), so that two different moving speeds can be effectively distinguished.
The simulation results above show that the three radar communication schemes can effectively detect targets.
2.4.2 Communication performance
2.4.2.1 Simulation/Test Results of the Linear Frequency Modulation Scheme
In reference [4], the authors used Up-Chirp signals (radar) and Down-Chirp signals (communication) (which are basically orthogonal). The chirp signal had a modulation frequency of 40 MHz/μs, a system bandwidth of 80 MHz, a pulse time of 2 μs (the time required for the frequency to go from the lowest to the highest), a processing gain (time-width-bandwidth product) of 22 dB, and a data modulation method of π/4-DQPSK. The radio frequency was 10 GHz.
The simulation results in reference [4] show that when the SINR exceeds 11dB, the bit error rate (BER) is less than 0.1%, which can meet the general communication performance requirements.
2.4.2.2 Simulation/Test Results of Direct Sequence Spread Spectrum Scheme
In reference [7], the simulation settings used by the authors are as follows: the carrier frequency is 2MHz, the sampling frequency is 20MHz, the m-sequence is used for spreading, the spreading factor is 15 or 31, the chip width is 1μs, the data modulation method is differential coherent binary phase shift keying (DBPSK), and the data length is 2000 symbols.
The simulation results in reference [7] show that when SINR exceeds 3dB, BER (SF=15) is less than 0.1%, which can meet the general communication performance requirements.
2.4.2.3 Simulation/Test Results of OFDM Scheme
In reference [10], the simulation settings used by the authors are as follows: the carrier frequency is 5.9 GHz, the number of full-phase OFDM subcarriers is 512, the CP length is 1.4 μs, the full-phase OFDM symbol length after adding CP is 23.8 μs, the system bandwidth is 91.5 MHz, the number of full-phase OFDM symbols in one frame is 177, the time length of one frame is 4.25 ms, and the subcarrier spacing is 180 kHz.
The simulation results in reference [10] show that when SINR exceeds 8.2dB, BER is less than 0.1%, which can meet the general communication performance requirements.
The simulation results above show that the three radar communication schemes can transmit data well under relatively low SINR.
2.5 Test/Experiment System
2.5.1 Radar Communication Experimental System Based on Linear Frequency Modulation
Figure 5. Radar communication experimental system based on linear frequency modulation
(LHCP: Left-handed circular polarization; PRBS: Pseudo-random binary sequence; RHCP: Right-handed circular polarization)
As shown in Figure 5 [5]: The operating frequency of the system is 750MHz, the bandwidth is 500MHz, the range resolution is 63cm, the radar detection probability is 99%, and the false alarm rate is 7%. The BER at a rate of 1Mbit/s is 0.002 (at this time, the radar pulse repetition frequency is 150kHz and the radar pulse time width is 1.5ns).
2.5.2 Radar Communication Test System Based on Direct Sequence Spread Spectrum
Figure 6. Radar communication test system (signal processing board) based on direct sequence spread spectrum.
Figure 6 shows the radar communication test system (signal processing board) based on direct sequence spread spectrum (Nanjing University of Science and Technology), which consists of field programmable gate array (FPGA), analog-to-digital converter (A/D), digital-to-analog converter (D/A) and the like [8]. The system uses 30MHz intermediate frequency, 31-bit m-sequence spread spectrum, communication rate of 129kbit/s, and PSL of 13dB.
2.5.3 OFDM-based radar communication experimental system
Figure 7 OFDM Ultra-Wideband Synthetic Aperture Radar Test System
The University of Miami in the United States developed an ultra-wideband synthetic aperture radar and integrated it into a communication and radar system. Figure 7 shows the OFDM ultra-wideband synthetic aperture radar experimental system they developed in the laboratory [15-16].
2.6 Summary
The simulations and experimental systems above demonstrate that vehicle-mounted radar communication systems can utilize various signals. The simplest approach is to use the most commonly used radar signal—Frequency Modulated Continuous Wave (FMCW)—with communication information directly modulated onto this signal. Alternatively, existing communication signals such as spread spectrum and OFDM signals can be used. Simulation verification shows that…
The radar operating range of the 24GHz vehicle-mounted radar communication system can reach 100m, the communication range is over 500m, and the data transmission rate can reach up to 20Mbit/s (using OFDM signal).
Using a 77GHz vehicle-mounted radar communication system, the radar's ranging range and effective communication distance are basically equivalent, reaching 250m, with a peak data rate of 20Mbit/s (using OFDM signals), while the resolution and accuracy of distance and speed are far higher than those of the 24GHz system.
Among them, the distance resolution can be less than 1m, and the speed measurement range can reach ±200km/h. As for the latency index, there are no other network latencies except for propagation latency and system processing latency, which can meet the latency requirements for automotive safety [17-19].
Therefore, based on the simulation results, the vehicle-mounted radar communication system can achieve vehicle-to-everything (V2X) communication without sacrificing radar performance. It can not only provide various driving assistance functions for vehicles, but also obtain road information from a longer perspective, thus meeting the needs of intelligent driving for the fusion of sensor perception information and network information.
3. Conclusion
Driven by technological innovation, the integration of communications, the Internet and various industries is developing rapidly, and the era of the Internet of Everything has begun. This includes not only connections between people and between people and things, but also connections between things, with the Internet of Vehicles being an important component.
For the automotive industry, with the increasing demands of users for driving comfort and safety, autonomous driving has become one of the most sought-after goals. Currently, the overall level of the autonomous driving industry is at Level 1/Level 2 (according to the US NTHSA or SAE standards) or driver assistance (according to the Chinese SAE standards). In China, the penetration rate of most major functions of advanced driver assistance systems in new cars is less than 10%, except for electronic stability control systems.
According to information released by the China Society of Automotive Engineers, by 2020, China will promote the application of partially automated driving, which is mainly based on autonomous environmental perception and supplemented by connected information services. By 2025, the focus will be on forming connected environmental perception capabilities and achieving highly automated driving under complex working conditions. By 2030, fully automated driving will be achieved through V2X collaborative control.
Therefore, a key issue is how to rapidly and effectively advance from the current low-penetration stage of driver assistance to the stage of autonomous driving with autonomous environmental perception and connectivity. Through onboard radar communication systems, vehicles can not only perceive their surroundings through their own radar detection capabilities, but also establish communication networks with other vehicles, obtaining broader area information through collaborative communication. The fusion of near and far information not only provides strong protection for the vehicle's own safe driving, but also enables intelligent driving across all roads and improves overall traffic efficiency.
Therefore, vehicle-mounted radar communication systems will become one of the most critical sensors in the autonomous driving industry, accelerating the industrialization of intelligent connected vehicles and thereby improving the overall level of the autonomous driving industry.
