This paper introduces the performance characteristics of the currently popular single-chip RF transceiver nRF24E1 and the special chip TMC2023 with related computational functions; it elaborates on the relevant computational theories in signal processing, and combines these theories with a circuit based on the two chips mentioned above, applying it to an automotive collision avoidance system to enhance its collision avoidance capability. With the development of the times and the progress of society, more and more cars have entered ordinary households. Although road conditions are constantly improving, the current situation of traffic congestion on highways remains unavoidable. Coupled with the gradual increase in vehicle speed, serious traffic accidents occur constantly, causing huge losses of life and property to people and society. An automotive collision avoidance system is a detection device that can issue visual and auditory warning signals to the driver in advance. It is usually installed on a car and can detect pedestrians, vehicles, or surrounding obstacles attempting to approach the vehicle; it can issue early warnings to the driver and passengers of an impending collision, prompting the driver, or even bypassing the vehicle, to take emergency measures to handle special dangerous situations and avoid losses. Currently, although countries are researching collision avoidance systems (internationally known as active safety systems), how to better solve the problem of false alarms remains a challenge for researchers. International researchers have reached a consensus through extensive experimental studies that, in order to effectively solve the above problems, the collision avoidance system must have the following functions: (1) It must have angle measurement capability, and the azimuth information of the target is essential for removing false alarms; (2) It should be easy to generate complex transmission signals with strong anti-interference performance, and cooperate with real-time and efficient signal processing and target detection algorithms to remove false alarms. Only when the above two points are closely combined can the reliability of the car collision avoidance system be guaranteed. 1 Introduction to the features of TMC2023 chip and nRF24E1 chip TMC2032 is a new type of all-digital correlator circuit, whose correlation word length and correlation threshold are programmable. This chip is a single-chip 64-bit CMOS all-digital correlator large-scale integrated circuit launched by TRW in recent years. It has three independent clock 8-bit shift registers (random data register A, local code register B and mask code register M); and a 7-bit register to load preset threshold values. Random data of any length between 0 and 64 is correlated with the local code, output as a 7-bit BCD code with a tri-state buffer, and compared with a preset threshold value in a comparator. If the correlation value is greater than or equal to the threshold value, the flag bit changes from low to high. Due to the use of advanced high-speed CMOS manufacturing technology, the parallel correlation rate reaches over 30MHz. It can be widely used in synchronization, matched filtering, error detection, recording, and barcode recognition, and is especially suitable for radar signal recognition. The nRF24E1 is a wireless RF transceiver chip with an operating frequency of up to 2.4GHz, a channel operation time of less than 200μs, a data rate of 1Mbps, and does not require an external SAW filter. It is the world's first globally universal low-cost RF system-on-a-chip. It internally incorporates an 8051-compatible microprocessor and a 10-bit 9-input A/D converter, and can operate stably between 1.9V and 3.6V; it also includes an internal voltage regulator and VDD voltage monitor. The wireless transceiver section has the same functions as the nRF2401. This function is initiated by the internal parallel port and internal SPI. Each signal to be transmitted can be programmed as an interrupt for the processor or transmitted to the microprocessor via the GPIO port. The nRF24E1 chip can achieve wireless communication within the globally common frequency band of 2.4–2.5 GHz. The transceiver section includes a frequency divider, amplifier, regulator, and two transceiver units. Output power, frequency band, and other RF parameters can be easily adjusted through RF registers. In transmit mode, the current consumption is only 10.5 mA; in receive mode, the current consumption is only 18 mA, resulting in very low power consumption. 2 System Structure The entire information acquisition system consists of five sets of RF transmitting and receiving devices. The basic circuitry of each transmitting and receiving unit is the same. These five transceiver devices are connected to a DSP central processing unit, which is responsible for calculating the data transmitted and making decisions based on the actual situation. The structure of each transmitting and receiving device is shown in Figure 1. First, a high-frequency electromagnetic wave is generated by an RF transmitting circuit based on the nRF24E1 core. Then, a modulation signal from the related computing chip TMC2032 modulates the wave, producing an RF signal different from other RF transceiver units, thus preparing for reception. To ensure the electromagnetic wave signal can travel a sufficiently long distance, the modulated signal needs to be amplified; this is accomplished by a power amplifier circuit. Finally, this signal is transmitted into the air. [align=center] Figure 1: Core Circuit of RF Transceiver[/align] When the transmitted electromagnetic wave encounters an obstacle and returns, it is first identified by the related computing chip TMC2032. If it was transmitted from the same group of transmitters, it is received and further transmitted to the RF receiving section; otherwise, it is rejected. Then, the receiving section calculates the propagation time based on the phase shift generated by the electromagnetic wave propagating in the air, and then calculates the distance between the obstacle and the group of transceivers. Finally, this distance information is sent to the central processing unit. The central processing microcontroller simultaneously calculates the distance information from the five groups of RF transceiver units to determine the spatial orientation of the obstacle relative to the vehicle. At this point, the obstacle information collection is basically complete. The remaining task is to transmit this comprehensive information to a higher-level central processing unit for final decision-making. 3. Transceiver Unit Layout and Calculation Principle During vehicle operation, the system needs to determine the spatial orientation of obstacles in front, behind, to the left, right, and above/below to avoid them. For obstacles behind, only their distances from the vehicle in front, behind, to the left, and to the right need to be determined. Therefore, three radio frequency transceiver systems are installed at the front of the vehicle, arranged in a triangular configuration perpendicular to the horizontal plane. Two radio frequency transceiver systems are installed at the rear of the vehicle, arranged horizontally. The entire transceiver system installation is shown in Figure 2. The following is a simplified description of the process for calculating obstacle distances using the radio frequency transceiver system. [align=center]Figure 2 Schematic diagram of sensor distribution in a car collision avoidance system Figure 3 Schematic diagram of obstacle distance calculation[/align] The obstacle distance calculation schematic diagram is shown in Figure 3. Points A, B, and C represent three ultrasonic sensors installed at the front of the vehicle; point E represents the obstacle; EF represents the distance from point E to the horizontal plane, FG represents the distance from the obstacle to the front plane of the vehicle, and AG represents the distance from the obstacle to the side of the vehicle. The requirement is to find the three straight line segments EF, FG, and AG that represent the spatial relative position of the obstacle and the vehicle. The solution is as follows: In ΔABC, construct BD and AC, and connect ED and FD. The area SΔABC can then be calculated as: SΔABC = where S = 1/2(T1 + T2 + T3), and T1, T2, and T3 represent the lengths of line segments AB, BC, and AC, respectively. Therefore, BD = 2SΔABC/AC. In ΔADE, replace AE, BE, CE, AB, BC, and AC with the distances S1, S2, and S3 that the RF transceiver system can detect, and the known distances T1, T2, and T3, respectively. Then, the values ​​of the three distances EF, FG, and AG can be obtained. 4. Related Algorithms Used With the widespread application of radio frequency in daily life, people have gradually discovered certain defects in RF ranging: ① The effective operating distance is relatively short, and increasing the transmission power alone to increase the measurement distance is very limited; ② The ranging accuracy mainly depends on the signal-to-noise ratio of the echo signal. Under a certain signal-to-noise ratio, increasing the gain of the pre-amplifier circuit alone to improve the measurement accuracy is also very limited. To address the aforementioned issues, a pseudocode-based radio frequency transmission and reception system was envisioned for automotive collision avoidance systems. White noise instantaneously follows a Gaussian distribution (normal distribution). Its power spectral density is uniform across a wide frequency band, and its autocorrelation function has a delta function shape. Although pseudo-random codes have only two levels, they exhibit correlation characteristics similar to white noise, except that their amplitude probability distribution no longer follows a Gaussian distribution. Therefore, white noise can be described using the balance, run, and correlation characteristics of pseudo-random sequences. Pseudo-random coding is implemented using logical operations, and the signal's autocorrelation function satisfies: It can be seen that when P is sufficiently large, the autocorrelation coefficient exhibits a sharp two-level characteristic, approaching a delta function. In ultrasonic ranging based on pseudo-random codes, the sharp characteristic of the pseudocode autocorrelation function is utilized to measure the delay between the transmitted and received codes, thereby improving measurement accuracy. m-sequence pseudo-random codes are the longest-period sequences generated by linear shift registers. Due to their excellent correlation characteristics and ease of generation, they have been widely used. According to the definition of correlation function, let two time functions be x1(t) and x2(t). Then, x1(t) is called the autocorrelation function of x1(t), and x2(t) is called the cross-correlation function of x1(t) and x2(t). In signal detection theory, there are two types of problems: one is signal detection, which involves determining the presence or absence of a signal based on a received mixed signal (signal plus noise or pure noise); the other is parameter estimation, which involves estimating certain parameters (e.g., amplitude, phase, frequency, pulse amplitude, etc.) or waveform of the signal based on the detection of its presence or absence. To improve anti-interference capability, it is necessary to find the optimal signal reception method under interference conditions. The correlation function of a periodic signal is still a periodic function, while the correlation function of interference noise is a delta function. Based on these differences, a correlator can be used to detect periodic signals mixed with noise interference. This method of detecting signals using differences in time-domain characteristics is called correlation reception. Depending on the reference signal, correlation reception is further divided into autocorrelation reception and cross-correlation reception. Autocorrelation reception, when the input waveform (or data) is unknown, uses an autocorrelation unit to calculate the autocorrelation function. Cross-correlation reception, when the reference signal is known, uses a correlator to calculate the cross-correlation function between the input waveform (or data) and the local signal. In this design, the reference signal is the local code, so cross-correlation reception is used. In radio frequency ranging systems, it is necessary not only to detect the presence or absence of echo signals but also to accurately measure the delay between the echo and transmitted signals. This is to accurately determine the time taken for radio wave propagation and thus calculate the distance between the obstacle and the vehicle. This paper utilizes advanced radio frequency technology and a stable and reliable correlation algorithm to give the car strong collision avoidance capabilities. The above overall design scheme has been implemented on an experimental vehicle and, to a certain extent, solves the two key problems that have always plagued researchers: it possesses relatively sensitive angle measurement capabilities and strong anti-interference capabilities, thus enabling the car to have strong collision avoidance capabilities. Nevertheless, because human life is at stake during vehicle operation, its performance still needs further improvement to enhance the absolute safety of the entire system and promote its widespread adoption. References 1 Hu Shuhao. Practical Radio Frequency Technology. Beijing: Electronic Industry Press, 2004 2 Ren Guanzhong, Ning Yonglan. Phase Measurement Technology. Electrical Measurement & Instrumentation, 2002(9): 41-60 3 Zhang Ming, Liu Yinfeng, Huang He. Design of PC Wireless Transceiver Module Based on nRF401. Electronic Technology Application, 2002;28(4):76-79 4 TRW Product Summary. TRW LS1 Products Inc, 2000