Fuchao He, Zheng Shi, Guanghua Yang, Xiaofan Li, Xinrong Ye, Shaodan Ma
{"title":"HARQ-IR 辅助短分组通信:BLER 分析和吞吐量最大化","authors":"Fuchao He, Zheng Shi, Guanghua Yang, Xiaofan Li, Xinrong Ye, Shaodan Ma","doi":"arxiv-2312.04377","DOIUrl":null,"url":null,"abstract":"This paper introduces hybrid automatic repeat request with incremental\nredundancy (HARQ-IR) to boost the reliability of short packet communications.\nThe finite blocklength information theory and correlated decoding events\ntremendously preclude the analysis of average block error rate (BLER).\nFortunately, the recursive form of average BLER motivates us to calculate its\nvalue through the trapezoidal approximation and Gauss-Laguerre quadrature.\nMoreover, the asymptotic analysis is performed to derive a simple expression\nfor the average BLER at high signal-to-noise ratio (SNR). Then, we study the\nmaximization of long term average throughput (LTAT) via power allocation\nmeanwhile ensuring the power and the BLER constraints. For tractability, the\nasymptotic BLER is employed to solve the problem through geometric programming\n(GP). However, the GP-based solution underestimates the LTAT at low SNR due to\na large approximation error in this case. Alternatively, we also develop a deep\nreinforcement learning (DRL)-based framework to learn power allocation policy.\nIn particular, the optimization problem is transformed into a constrained\nMarkov decision process, which is solved by integrating deep deterministic\npolicy gradient (DDPG) with subgradient method. The numerical results finally\ndemonstrate that the DRL-based method outperforms the GP-based one at low SNR,\nalbeit at the cost of increasing computational burden.","PeriodicalId":501433,"journal":{"name":"arXiv - CS - Information Theory","volume":"26 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2023-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"HARQ-IR Aided Short Packet Communications: BLER Analysis and Throughput Maximization\",\"authors\":\"Fuchao He, Zheng Shi, Guanghua Yang, Xiaofan Li, Xinrong Ye, Shaodan Ma\",\"doi\":\"arxiv-2312.04377\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"This paper introduces hybrid automatic repeat request with incremental\\nredundancy (HARQ-IR) to boost the reliability of short packet communications.\\nThe finite blocklength information theory and correlated decoding events\\ntremendously preclude the analysis of average block error rate (BLER).\\nFortunately, the recursive form of average BLER motivates us to calculate its\\nvalue through the trapezoidal approximation and Gauss-Laguerre quadrature.\\nMoreover, the asymptotic analysis is performed to derive a simple expression\\nfor the average BLER at high signal-to-noise ratio (SNR). Then, we study the\\nmaximization of long term average throughput (LTAT) via power allocation\\nmeanwhile ensuring the power and the BLER constraints. For tractability, the\\nasymptotic BLER is employed to solve the problem through geometric programming\\n(GP). However, the GP-based solution underestimates the LTAT at low SNR due to\\na large approximation error in this case. Alternatively, we also develop a deep\\nreinforcement learning (DRL)-based framework to learn power allocation policy.\\nIn particular, the optimization problem is transformed into a constrained\\nMarkov decision process, which is solved by integrating deep deterministic\\npolicy gradient (DDPG) with subgradient method. The numerical results finally\\ndemonstrate that the DRL-based method outperforms the GP-based one at low SNR,\\nalbeit at the cost of increasing computational burden.\",\"PeriodicalId\":501433,\"journal\":{\"name\":\"arXiv - CS - Information Theory\",\"volume\":\"26 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2023-12-07\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - CS - Information Theory\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2312.04377\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - CS - Information Theory","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2312.04377","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
摘要
有限块长信息论和相关解码事件极大地阻碍了对平均块误码率(BLER)的分析。幸运的是,平均误码率的递归形式促使我们通过梯形近似和高斯-拉盖尔正交来计算其值。此外,我们还进行了渐近分析,得出了高信噪比(SNR)下平均误码率的简单表达式。然后,我们研究了通过功率分配最大化长期平均吞吐量(LTAT),同时确保功率和 BLER 约束。为了提高可操作性,我们采用了渐近 BLER,通过几何编程(GP)来解决这个问题。然而,由于这种情况下的近似误差较大,基于 GP 的解决方案在低信噪比时会低估 LTAT。另外,我们还开发了一种基于深度强化学习(DRL)的框架来学习功率分配策略,特别是将优化问题转化为受约束马尔可夫决策过程,并通过将深度确定性策略梯度(DDPG)与子梯度法相结合来解决该问题。数值结果最终证明,在低信噪比条件下,基于 DRL 的方法优于基于 GP 的方法,尽管代价是增加了计算负担。
HARQ-IR Aided Short Packet Communications: BLER Analysis and Throughput Maximization
This paper introduces hybrid automatic repeat request with incremental
redundancy (HARQ-IR) to boost the reliability of short packet communications.
The finite blocklength information theory and correlated decoding events
tremendously preclude the analysis of average block error rate (BLER).
Fortunately, the recursive form of average BLER motivates us to calculate its
value through the trapezoidal approximation and Gauss-Laguerre quadrature.
Moreover, the asymptotic analysis is performed to derive a simple expression
for the average BLER at high signal-to-noise ratio (SNR). Then, we study the
maximization of long term average throughput (LTAT) via power allocation
meanwhile ensuring the power and the BLER constraints. For tractability, the
asymptotic BLER is employed to solve the problem through geometric programming
(GP). However, the GP-based solution underestimates the LTAT at low SNR due to
a large approximation error in this case. Alternatively, we also develop a deep
reinforcement learning (DRL)-based framework to learn power allocation policy.
In particular, the optimization problem is transformed into a constrained
Markov decision process, which is solved by integrating deep deterministic
policy gradient (DDPG) with subgradient method. The numerical results finally
demonstrate that the DRL-based method outperforms the GP-based one at low SNR,
albeit at the cost of increasing computational burden.