Abstract: Abstract: Multi-machine harvesting has been the popular form to improve the operation efficiency of the combine harvester, particularly with agricultural intensification, scale, and industrialization. The demand can also be required for the Internet of Things (IoT) and wireless communication of the combine harvester group. However, the current communication technologies of the IoT cannot fully meet the unstructured and large-scale agricultural scenarios, due mainly to the industrial design with the relatively short communication distance, low reliability, and high cost. Fortunately, the LoRa technology has unique advantages in the transmission distance, transmission rate, node capacity, and deployment mode. Moreover, the lower cost is suitable for the large-scale deployment of agricultural vehicles. At the same time, the parameters of LoRa technology can fully meet the needs of wireless communication in the operation scenario of combine harvester group. In addition, the 2.4 GHz band can further improve the performance of LoRa technology and regional universality. Therefore, wireless communication can be expected to apply to the scenario of combine harvester group. In this study, a communication network structure was proposed for the working mode of combine harvester group using the 2.4 GHz LoRa technology. Firstly, the LoRa network structure was constructed for the joint harvester cluster in the form of host and slave. The working mode of each node was then set using the cooperative operation scenario of the harvesters group. A radio frequency chip model (SX1280) was selected to simulate the transceiver mode, according to the application scenario. The optimal path attenuation model was chosen for the maximum communication distance between the master and slave. Secondly, a simulation model was established to optimize the application parameters of LoRa using NS-3 network platform, including the bandwidth, coding rate, spreading factor, transmission power, and frequency. The static and moving state of network nodes were used for the simulation scenarios of combine harvester communication network. Once the network nodes were fixed, a circular area with a radius of 3 km was selected as the deployment range of all network nodes. All network nodes were deployed in the circular range, according to the law of uniform distribution. The allocation scheme of spread spectrum factor was selected to accommodate the number of nodes and the minimum transmission interval in the simulated communication scenario of combine harvester cluster. Finally, the hardware and software were designed for the hardware terminal in the communication protocol of the combine harvester group. Both simulation and field experiments were conducted to comprehensively evaluate the effectiveness of the new LoRa for the combine harvesters group. The simulation results show that the spread spectrum factor with the specific proportion improved the network communication performance in the application scenario. When more than 90% of the reception success rate was guaranteed, the 1 625 kHz bandwidth supported by the 25 nodes to communicate at a 1 s transmission interval in a static scenario. In the dynamic scenario, the communication delay was less than 10 ms under 1 625 kHz bandwidth. The field test results show that the overall success rates of communication were 99.3%, and 92.5% under the "Report" and "Request" modes, respectively. The average response time was 123.07 ms. Consequently, the LoRa technology was applied for wireless communication between the combine harvesters, which can fully meet the requirements of joint harvester cluster communication for the number of accommodated devices and transmission interval. The number of nodes in the cluster can also be adjusted appropriately, according to the requirements for the communication interval. The Report mode supported the higher communication frequencies suitable for the reporting data from the combined harvesters, while the Request mode was suitable for the lower frequency, and responsive communications. The finding can provide a feasible solution to the mutual communication of the combine harvester clusters during cooperative operation.