Aiming at the problems in deep space communications such as time extension, high bit error rate, channel asymmetry, high redundancy of short code length LT codes, and low decoding success rate, a deep space data transmission based on optimized LT (OLT) codes is proposed. Strategy. Firstly, an OLT code is proposed by adjusting the degree distribution function, adopting a novel packet selection strategy and a joint decoding algorithm. Then, a deep space data transmission strategy based on OLT codes is proposed; the sending end encodes the data files with OLT codes and sends them; at the receiving end, a joint decoding algorithm is used to decode the received encoded packets to obtain the original data. The simulation results show that compared with LT codes, OLT codes can improve the decoding success rate and reduce redundancy; compared with CFDP, this strategy can effectively reduce delays and improve the effectiveness and reliability of data transmission; in high loss scenarios In the case of packet rate, the advantages of this strategy are more obvious.
Deep space communication is the foundation of deep space exploration. Future deep space exploration missions require advanced data transmission strategies to ensure the reliable transmission of large amounts of data. Due to the long transmission distance of deep space communication and the low signal-to-noise ratio, the link environment is harsh[“, resulting in long deep space data transmission time, high bit error rate, channel asymmetry and other shortcomings. Therefore, traditional data transmission strategies are no longer applicable in deep space. [2-4]. The deep space data transmission strategy based on the improved transmission control protocol [5-8] adopts a new initial state algorithm, an improved congestion control mechanism and a packet loss processing strategy. Although it can effectively improve the throughput and link utilization, but cannot cope with delays and high bit error rates. The file transfer protocol (CCSDS FileDelivery Protocol, CFDP) proposed by the Consultative Committee for Space Data Systems (CCSDS) can be well adapted to deep Empty environment [°, reducing the interaction process; but it does not process multiple data packets, and still requires feedback and retransmission, resulting in too large a delay.
In recent years, fountain code, as a forward error correction grouping technology that does not require feedback links and can approach the Shannon limit with any probability, has been applied to deep space data transmission strategies.
(10-13] to reduce the number of retransmissions and delays. LT code is a fountain coding scheme with practical performance1, but it has large redundancy in short code lengths, high coding and decoding complexity, and The commonly used Belief Propagation (BP) algorithm is not efficient, and the Gaussian Elimination (GE) algorithm is too complex. Literature [15] uses the correlated column compensation coding algorithm and the incremental Gaussian elimination algorithm to solve short-term problems. The improvement of the code length LT code has improved the decoding success rate of the LT code to a certain extent, but the coding and decoding complexity is high, and no specific data transmission strategy is given.
In response to the above problems, this article first analyzes the encoding and decoding algorithm of LT code, and discusses the application performance of LT code in deep space communication. Then the short code length LT code is optimized to obtain the optimized LT (Optimized LT, OLT) code: on the premise of maintaining the code-rate characteristics of the LT code, a new packet selection strategy is proposed, and the degree distribution function is adjusted. , fully ensuring full coverage of the original data; then analyzing the shortcomings of the existing decoding algorithm, using the BP algorithm and the GE algorithm for joint decoding to obtain higher decoding performance. Finally, a deep space data transmission strategy based on OLT codes is proposed and simulated and analyzed.
1LT code and its deep space application performance analysis 1.1 LT code encoding and decoding algorithm
The encoding method of LT code is: 1) Assume that the source file size is k*L, divide it into h data packets, and the size of each data packet is L; 2) Select a degree value from the given degree distribution function d. Then randomly select l data packets from k data packets for XOR to obtain a coded packet; 3) Repeat the previous step to obtain a certain number of coded packets.
The BP algorithm commonly used in LT codes is an iterative algorithm. As shown in Figure 1, at each step of decoding, the decoder looks for encoding packets with degree 1 in the encoding packet set. The set of encoding packets with degree 1 is called the decoding fluctuation set. Obviously, decoding fluctuations Data packets connected by centralized encoding packets can be directly decoded. At each step of the iteration, the decoder XORs a decoded data packet with all the encoding packets connected to it, and the calculation result replaces the value of the corresponding encoding packet. After completion, the corresponding connection relationship is deleted. If a coding packet with degree 1 is generated, add the decoding fluctuation set. Repeat the above process until there is no encoding packet with degree 1. If all data packets are recovered, the decoding is successful, otherwise the decoding fails.